OS/2 Professional

home *** CD-ROM | disk | FTP | other *** search

/ OS/2 Professional / OS2PRO194.ISO / os2 / info / prgramer / gnuemac / elisp.inf (.txt) next >

Wrap

OS/2 Help File | 1994-01-21 | 581.6 KB | 22,635 lines

ΓòÉΓòÉΓòÉ 1. Title page ΓòÉΓòÉΓòÉ GNU Emacs Lisp Reference Manual GNU Emacs Version 18 for Unix Users Edition 1.03, December 1990 by Bil Lewis, Dan LaLiberte, and the GNU Manual Group Copyright (C) 1990 Free Software Foundation, Inc. This is edition 1.03 of the GNU Emacs Lisp Reference Manual, for Emacs Version 18, December 1990. ISBN 1-882114-10-8. Published by the Free Software Foundation, 675 Massachusetts Avenue, Cambridge, MA 02139 USA Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies. Permission is granted to copy and distribute modified versions of this manual under the conditions for verbatim copying, provided that the entire resulting derived work is distributed under the terms of a permission notice identical to this one. Permission is granted to copy and distribute translations of this manual into another language, under the above conditions for modified versions, except that this permission notice may be stated in a translation approved by the Foundation. Cover art by Etienne Suvasa. ΓòÉΓòÉΓòÉ 2. Top ΓòÉΓòÉΓòÉ This Info file contains edition 1.03 of the GNU Emacs Lisp Reference Manual. ΓòÉΓòÉΓòÉ 3. GNU Emacs General Public License ΓòÉΓòÉΓòÉ (Clarified 11 Feb 1988) The license agreements of most software companies keep you at the mercy of those companies. By contrast, our general public license is intended to give everyone the right to share GNU Emacs. To make sure that you get the rights we want you to have, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. Hence this license agreement. Specifically, we want to make sure that you have the right to give away copies of Emacs, that you receive source code or else can get it if you want it, that you can change Emacs or use pieces of it in new free programs, and that you know you can do these things. To make sure that everyone has such rights, we have to forbid you to deprive anyone else of these rights. For example, if you distribute copies of Emacs, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must tell them their rights. Also, for our own protection, we must make certain that everyone finds out that there is no warranty for GNU Emacs. If Emacs is modified by someone else and passed on, we want its recipients to know that what they have is not what we distributed, so that any problems introduced by others will not reflect on our reputation. Therefore we (Richard Stallman and the Free Software Foundation, Inc.) make the following terms which say what you must do to be allowed to distribute or change GNU Emacs. ΓòÉΓòÉΓòÉ 3.1. Copying Policies ΓòÉΓòÉΓòÉ 1. You may copy and distribute verbatim copies of GNU Emacs source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each file a valid copyright notice ``Copyright (C) 1988 Free Software Foundation, Inc.'' (or with whatever year is appropriate); keep intact the notices on all files that refer to this License Agreement and to the absence of any warranty; and give any other recipients of the GNU Emacs program a copy of this License Agreement along with the program. You may charge a distribution fee for the physical act of transferring a copy. 2. You may modify your copy or copies of GNU Emacs source code or any portion of it, and copy and distribute such modifications under the terms of Paragraph 1 above, provided that you also do the following: o cause the modified files to carry prominent notices stating who last changed such files and the date of any change; and o cause the whole of any work that you distribute or publish, that in whole or in part contains or is a derivative of GNU Emacs or any part thereof, to be licensed at no charge to all third parties on terms identical to those contained in this License Agreement (except that you may choose to grant more extensive warranty protection to some or all third parties, at your option). o if the modified program serves as a text editor, cause it, when started running in the simplest and usual way, to print an announcement including a valid copyright notice ``Copyright (C) 1988 Free Software Foundation, Inc.'' (or with the year that is appropriate), saying that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License Agreement. o You may charge a distribution fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. Mere aggregation of another unrelated program with this program (or its derivative) on a volume of a storage or distribution medium does not bring the other program under the scope of these terms. 3. You may copy and distribute GNU Emacs (or a portion or derivative of it, under Paragraph 2) in object code or executable form under the terms of Paragraphs 1 and 2 above provided that you also do one of the following: o accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Paragraphs 1 and 2 above; or, o accompany it with a written offer, valid for at least three years, to give any third party free (except for a nominal shipping charge) a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Paragraphs 1 and 2 above; or, o accompany it with the information you received as to where the corresponding source code may be obtained. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form alone.) For an executable file, complete source code means all the source code for all modules it contains; but, as a special exception, it need not include source code for modules which are standard libraries that accompany the operating system on which the executable file runs. 4. You may not copy, sublicense, distribute or transfer GNU Emacs except as expressly provided under this License Agreement. Any attempt otherwise to copy, sublicense, distribute or transfer GNU Emacs is void and your rights to use GNU Emacs under this License agreement shall be automatically terminated. However, parties who have received computer software programs from you with this License Agreement will not have their licenses terminated so long as such parties remain in full compliance. 5. If you wish to incorporate parts of GNU Emacs into other free programs whose distribution conditions are different, write to the Free Software Foundation. We have not yet worked out a simple rule that can be stated here, but we will often permit this. We will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software. Your comments and suggestions about our licensing policies and our software are welcome! Please contact the Free Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139. ΓòÉΓòÉΓòÉ 3.2. NO WARRANTY ΓòÉΓòÉΓòÉ BECAUSE GNU EMACS IS LICENSED FREE OF CHARGE, WE PROVIDE ABSOLUTELY NO WARRANTY, TO THE EXTENT PERMITTED BY APPLICABLE STATE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING, FREE SOFTWARE FOUNDATION, INC, RICHARD M. STALLMAN AND/OR OTHER PARTIES PROVIDE GNU EMACS ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE GNU EMACS PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW WILL FREE SOFTWARE FOUNDATION, INC., RICHARD M. STALLMAN, AND/OR ANY OTHER PARTY WHO MAY MODIFY AND REDISTRIBUTE GNU EMACS AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY LOST PROFITS, LOST MONIES, OR OTHER SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH PROGRAMS NOT DISTRIBUTED BY FREE SOFTWARE FOUNDATION, INC.) THE PROGRAM, EVEN IF YOU HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES, OR FOR ANY CLAIM BY ANY OTHER PARTY. ΓòÉΓòÉΓòÉ 4. Introduction ΓòÉΓòÉΓòÉ Most of the GNU Emacs text editor is written in the programming language called Emacs Lisp. You can write new code in Emacs Lisp and install it as an extension to the editor. However, Emacs Lisp is more than a mere ``extension language''; it is a full computer programming language in its own right. You can use it as you would any other programming language. Because Emacs Lisp is designed for use in an editor, it has special features for scanning and parsing text as well as features for handling files, buffers, displays, subprocesses, and so on. Emacs Lisp is closely integrated with the editing facilities; thus, editing commands are functions that can also conveniently be called from Lisp programs, and parameters for customization are ordinary Lisp variables. This manual describes Emacs Lisp, presuming considerable familiarity with the use of Emacs for editing. (See The GNU Emacs Manual, for this basic information.) Generally speaking, the earlier chapters describe features of Emacs Lisp that have counterparts in many programming languages, and later chapters describe features that are peculiar to Emacs Lisp or relate specifically to editing. This is edition 1.02. ΓòÉΓòÉΓòÉ 4.1. Caveats ΓòÉΓòÉΓòÉ This manual has gone through numerous drafts. It is nearly complete but not flawless. There are a few sections which are not included, either because we consider them secondary (such as most of the individual modes) or because they are yet to be written. Because we are not able to deal with them completely, we have left out several parts intentionally. This includes most references to VMS and all information relating to X Windows and Sunview. The functions for using X Windows will in any case be entirely different in version 19. (The Free Software Foundation expends no effort on support for Sunview, since we believe users should use free X Windows rather than proprietary window systems.) The manual should be fully correct in what it does cover, and it is therefore open to criticism on anything it says---from specific examples and descriptive text, to the ordering of chapters and sections. If something is confusing, or you find that you have to look at the sources or experiment to learn something not covered in the manual, then perhaps the manual should be fixed. Please let us know. As you use this manual, we ask that you send corrections as soon as you find them. If you think of a simple, real life example for a function or group of functions, please make an effort to write it up and send it in. Please reference any comments to the node name and function or variable name, as appropriate. Please mail comments and corrections to lisp-manual-bugs@prep.ai.mit.edu. --Bil Lewis, Dan LaLiberte, Richard Stallman ΓòÉΓòÉΓòÉ 4.2. Lisp History ΓòÉΓòÉΓòÉ Lisp (LISt Processing language) was first developed in the late 1950s at the Massachusetts Institute of Technology for research in artificial intelligence. The great power of the Lisp language makes it superior for other purposes as well, such as writing editing commands. Dozens of Lisp implementations have been built over the years, each with its own idiosyncrasies. Many of them were inspired by Maclisp, which was written in the 1960's at MIT's Project MAC. Eventually the implementors of the descendents of Maclisp came together and developed a standard for Lisp systems, called Common Lisp. GNU Emacs Lisp is largely inspired by Maclisp, and a little by Common Lisp. If you know Common Lisp, you will notice many similarities. However, many of the features of Common Lisp have been omitted or simplified in order to reduce the memory requirements of GNU Emacs. Sometimes the simplifications are so drastic that a Common Lisp user might be very confused. We will occasionally point out how GNU Emacs Lisp differs from Common Lisp. If you don't know Common Lisp, don't worry about it; this manual is self-contained. ΓòÉΓòÉΓòÉ 4.3. Conventions ΓòÉΓòÉΓòÉ This section explains the notational conventions that are used in this manual. You may want to skip this section and refer back to it later. ΓòÉΓòÉΓòÉ 4.3.1. Some Terms ΓòÉΓòÉΓòÉ Throughout this manual, the phrases ``the Lisp reader'' and ``the Lisp printer'' are used to refer to those routines in Lisp that convert textual representations of Lisp objects into actual objects, and vice versa. See Printed Representation, for more details. You, the person reading this manual, are assumed to be ``the programmer'' and are addressed as ``you''. ``The user'' is the person who uses the code that you write. Examples of Lisp code appear in this font or form: (list 1 2 3). Names that represent arguments or metasyntactic variables appear in this font or form: first-number. ΓòÉΓòÉΓòÉ 4.3.2. nil and t ΓòÉΓòÉΓòÉ In Lisp, the symbol nil is overloaded with three meanings: it is a symbol with the name `nil'; it is the logical truth value false; and it is the empty list---the list of zero elements. When used as a variable, nil always has the value nil. As far as the Lisp reader is concerned, `()' and `nil' are identical: they stand for the same object, the symbol nil. The different ways of writing the symbol are intended entirely for human readers. After the Lisp reader has read either `()' or `nil', there is no way to determine which representation was actually written by the programmer. In this manual, we use () when we wish to emphasize that it means the empty list, and we use nil when we wish to emphasize that it means the truth value false. That is a good convention to use in Lisp programs also. (cons 'foo ()) ; Emphasize the empty list (not nil) ; Emphasize the truth value false In context where a truth value is expected, any non-nil value is considered to be true. However, t is the preferred way to represent the truth value true. When you need to choose a value which represents true, and there is no other basis for choosing, use t. The symbol t always has value t. In Emacs Lisp, nil and t are special symbols that always evaluate to themselves. This is so that you do not need to quote them to use them as constants in a program. An attempt to change their values results in a setting-constant error. See Accessing Variables. ΓòÉΓòÉΓòÉ 4.3.3. Evaluation Notation ΓòÉΓòÉΓòÉ When you evaluate a piece of Lisp code, it produces a result. In the examples in this manual, this is indicated with `=>': (car '(1 2)) => 1 You can read this as ``(car '(1 2)) evaluates to 1''. When a form is a macro call, it expands into a new form for Lisp to evaluate. We show the result of the expansion with `->'. We may or may not show the actual result of the evaluation of the expanded form. (third '(a b c)) -> (car (cdr (cdr '(a b c)))) => c Sometimes to help describe one form we show another form which produces identical results. The exact equivalence of two forms is indicated with `=='. (make-sparse-keymap) == (list 'keymap) ΓòÉΓòÉΓòÉ 4.3.4. Printing Notation ΓòÉΓòÉΓòÉ Many of the examples in this manual print text when they are evaluated. If you execute the code from an example in a Lisp Interaction buffer (such as the buffer `*scratch*'), the printed text is inserted into the buffer. If the example is executed by other means (such as by evaluating the function eval-region), the text printed is usually displayed in the echo area. You should be aware that text displayed in the echo area is truncated to a single line. In examples that print text, the printed text is indicated with `-|', irrespective of how the form is executed. The value returned by evaluating the form (here bar) follows on a separate line. (progn (print 'foo) (print 'bar)) -| foo -| bar => bar ΓòÉΓòÉΓòÉ 4.3.5. Error Messages ΓòÉΓòÉΓòÉ Some examples cause errors to be signaled. In them, the error message (which always appears in the echo area) is shown on a line starting with `error-->'. Note that `error-->' itself does not appear in the echo area. (+ 23 'x) error--> Wrong type argument: integer-or-marker-p, x ΓòÉΓòÉΓòÉ 4.3.6. Buffer Text Notation ΓòÉΓòÉΓòÉ Some examples show modifications to text in a buffer, with ``before'' and ``after'' versions of the text. In such cases, the entire contents of the buffer in question are included between two lines of dashes containing the buffer name. In addition, the location of point is shown as `-!-'. (The symbol for point, of course, is not part of the text in the buffer; it indicates the place between two characters where point is located.) ---------- Buffer: foo ---------- This is the -!-contents of foo. ---------- Buffer: foo ---------- (insert "changed ") => nil ---------- Buffer: foo ---------- This is the changed -!-contents of foo. ---------- Buffer: foo ---------- ΓòÉΓòÉΓòÉ 4.3.7. Format of Descriptions ΓòÉΓòÉΓòÉ Functions, variables, macros, commands, user options, and special forms are described in this manual in a uniform format. The first line of a description contains the name of the item followed by its arguments, if any. The category---function, variable, or whatever---appears at the beginning of the line. The description follows on succeeding lines, sometimes with examples. ΓòÉΓòÉΓòÉ 4.3.7.1. A Sample Function Description ΓòÉΓòÉΓòÉ In a function description, the name of the function being described appears first. It is followed on the same line by a list of parameters. The names used for the parameters are also used in the body of the description. The appearance of the keyword &optional in the parameter list indicates that the arguments for subsequent parameters may be omitted (omitted parameters default to nil). Do not write &optional when you call the function. The keyword &rest (which will always be followed by a single parameter) indicates that any number of arguments can follow. The value of the single following parameter will be a list of all these arguments. Do not write &rest when you call the function. Here is a description of an imaginary function foo: -- Function: foo integer1 &optional integer2 &rest integers The function foo subtracts integer1 from integer2, then adds all the rest of the arguments to the result. If integer2 is not supplied, then the number 19 is used by default. (foo 1 5 3 9) => 16 (foo 5) => 14 More generally, (foo w x y...) == (+ (- x w) y...) Any parameter whose name contains the name of a type (e.g., integer, integer1 or buffer) is expected to be of that type. A plural of a type (such as buffers) often means a list of objects of that type. Parameters named object may be of any type. (See Types of Lisp Object, for a list of Emacs object types.) Parameters with other sorts of names (e.g., new-file) are discussed specifically in the description of the function. In some sections, features common to parameters of several functions are described at the beginning. See Lambda Expressions, for a more complete description of optional and rest arguments. Command, macro, and special form descriptions have the same format, but the word `Function' is replaced by `Command', `Macro', or `Special Form', respectively. Commands are simply functions that may be called interactively; macros process their arguments differently from functions (the arguments are not evaluated), but are presented the same way. Special form descriptions use a more complex notation to specify optional and repeated parameters because they can break the argument list down into separate arguments in more complicated ways. `[optional-arg]' means that optional-arg is optional and `repeated-args...' stands for zero or more arguments. Parentheses are used when several arguments are grouped into additional levels of list structure. Here is an example: -- Special Form: count-loop (var [from to [inc]]) body... This imaginary special form implements a loop that executes the body forms and then increments the variable var on each iteration. On the first iteration, the variable has the value from; on subsequent iterations, it is incremented by 1 (or by inc if that is given). The loop exits before executing body if var equals to. Here is an example: (count-loop (i 0 10) (prin1 i) (princ " ") (prin1 (aref vector i)) (terpri)) If from and to are omitted, then var is bound to nil before the loop begins, and the loop exits if var is non-nil at the beginning of an iteration. Here is an example: (count-loop (done) (if (pending) (fixit) (setq done t))) In this special form, the arguments from and to are optional, but must both be present or both absent. If they are present, inc may optionally be specified as well. These arguments are grouped with the argument var into a list, to distinguish them from body, which includes all remaining elements of the form. ΓòÉΓòÉΓòÉ 4.3.7.2. A Sample Variable Description ΓòÉΓòÉΓòÉ A variable is a name that can hold a value. Although any variable can be set by the user, certain variables that exist specifically so that users can change them are called user options. Ordinary variables and user options are described using a format like that for functions except that there are no arguments. Here is a description of the imaginary electric-future-map variable. -- Variable: electric-future-map The value of this variable is a full keymap used by electric command future mode. The functions in this map will allow you to edit commands you have not yet thought about executing. User option descriptions have the same format, but `Variable' is replaced by `User Option'. ΓòÉΓòÉΓòÉ 4.4. Acknowledgements ΓòÉΓòÉΓòÉ This manual was written by Robert Krawitz, Bil Lewis, Dan LaLiberte, Richard M. Stallman and Chris Welty, the volunteers of the GNU manual group, in an effort extending over several years. Robert J. Chassell helped to review and edit the manual, with the support of the Defense Advanced Research Projects Agency, ARPA Order 6082, arranged by Warren A. Hunt, Jr. of Computational Logic, Inc. Corrections were supplied by Karl Berry, Bard Bloom, David Boyes, Alan Carroll, David A. Duff, Beverly Erlebacher, David Eckelkamp, Eirik Fuller, Eric Hanchrow, George Hartzell, Nathan Hess, Dan Jacobson, Jak Kirman, Bob Knighten, Frederick M. Korz, Joe Lammens, K. Richard Magill, Brian Marick, Roland McGrath, Skip Montanaro, John Gardiner Myers, Arnold D. Robbins, Raul Rockwell, Shinichirou Sugou, Kimmo Suominen, Edward Tharp, Bill Trost, Jean White, Matthew Wilding, Carl Witty, Dale Worley, Rusty Wright, and David D. Zuhn. ΓòÉΓòÉΓòÉ 5. Lisp Data Types ΓòÉΓòÉΓòÉ A Lisp object is a piece of data used and manipulated by Lisp programs. For our purposes, a type or data type is a set of possible objects. Every object belongs to at least one type. Objects of the same type have similar structures and may usually be used in the same contexts. Types can overlap, and objects can belong to two or more types. Consequently, we can ask whether an object belongs to a particular type, but not for ``the'' type of an object. A few fundamental object types are built into Emacs. These, from which all other types are constructed, are called primitive types. They include integer, cons, symbol, string, vector, subr and several special types, such as buffer, that are related to editing. (See Editing Types.) While an object may be a member of more than one type, every object is a member of exactly one primitive type. The primitive type of an object is stored with the object's data. A Lisp function is provided for each primitive type to check whether an object is a member of that type. Note that Lisp is unlike many other languages in that Lisp objects are ``self-typing'': the primitive type of the object is implicit in the object itself. For example, if an object is a vector, it cannot be treated as a number because Lisp knows it is a vector, not a number. In most languages, the programmer must declare the data type of each variable, and the type is known by the compiler but not represented in the data. Such type declarations do not exist in Emacs Lisp. This chapter describes the purpose, printed representation, and read syntax of each of the standard types in GNU Emacs Lisp. Details on how to use these types can be found in later chapters. ΓòÉΓòÉΓòÉ 5.1. Printed Representation and Read Syntax ΓòÉΓòÉΓòÉ The printed representation of an object is the format of the output generated by the Lisp printer (the function print) for that object. The read syntax of an object is the format of the input accepted by the Lisp reader (the function read) for that object. Most objects have more than one possible read syntax. Some types of object have no read syntax; except in these cases, the printed representation of an object is also a read syntax for it. In other languages, an expression is text; it has no other form. In Lisp, an expression is primarily a Lisp object and only secondarily the text that is the object's read syntax. Often there is no need to emphasize this distinction, but you must keep it in the back of your mind, or you will occasionally be very confused. Every type has a printed representation. Some types have no read syntax, since it may not make sense to enter objects of these types directly in a Lisp program. For example, the buffer type does not have a read syntax. Objects of these types are printed in hash notation: the characters `#<' followed by a descriptive string (typically the type name followed by the name of the object), and closed with a matching `>'. Hash notation cannot be read at all, so the Lisp reader signals the error invalid-read-syntax whenever a `#' is encountered. (current-buffer) => #<buffer objects.texi> When you evaluate an expression interactively, the Lisp interpreter first reads the textual representation of it, producing a Lisp object, and then evaluates that object (see Evaluation). However, evaluation and reading are separate activities. Reading returns the Lisp object represented by the text that is read; the object may or may not be evaluated later. See Input Functions, for a description of read, the basic function for reading objects. ΓòÉΓòÉΓòÉ 5.2. Comments ΓòÉΓòÉΓòÉ A comment is text that is written in a program only for the sake of humans that read the program, and that has no effect on the meaning of the program. In Lisp, a comment starts with a semicolon (`;') and continues to the end of line. Comments are discarded by the Lisp reader, and do not become part of the Lisp objects which represent the program within the Lisp system. We recommend these conventions for where to put comments and how to indent them: `;' Comments that start with a single semicolon, `;', should all be aligned to the same column on the right of the source code. Such comments usually explain how the code on the same line does its job. In Lisp mode and related modes, the M-; (indent-for-comment) command automatically inserts such a `;' in the right place, or aligns such a comment if it is already inserted. (The following examples are taken from the Emacs sources.) (setq base-version-list ; there was a base (assoc (substring fn 0 start-vn) ; version to which file-version-assoc-list)) ; this looks like ; a subversion `;;' Comments that start with two semicolons, `;;', should be aligned to the same level of indentation as the code. Such comments are used to describe the purpose of the following lines or the state of the program at that point. For example: (prog1 (setq auto-fill-hook ... ... ;; update mode-line (set-buffer-modified-p (buffer-modified-p)))) These comments are also written before a function definition to explain what the function does and how to call it properly. `;;;' Comments that start with three semicolons, `;;;', should start at the left margin. Such comments are not used within function definitions, but are used to make more general comments. For example: ;;; This Lisp code is run in Emacs when it is to operate as ;;; a server for other processes. `;;;;' Comments that start with four semicolons, `;;;;', should be aligned to the left margin and are used for headings of major sections of a program. For example: ;;;; The kill ring The indentation commands of the Lisp modes in Emacs, such as M-; (indent-for-comment) and TAB (lisp-indent-line) automatically indent comments according to these conventions, depending on the the number of semicolons. See Manipulating Comments. Any character may be included in a comment, but it is advisable to precede a character with syntactic significance in Lisp (such as `\' or unpaired `(' or `)') with a `\', to prevent it from confusing the Emacs commands for editing Lisp. ΓòÉΓòÉΓòÉ 5.3. Programming Types ΓòÉΓòÉΓòÉ There are two general categories of types in Emacs Lisp: those having to do with Lisp programming, and those having to do with editing. The former are provided in many Lisp implementations, in one form or another. The latter are unique to Emacs Lisp. ΓòÉΓòÉΓòÉ 5.3.1. Number Type ΓòÉΓòÉΓòÉ Integers are the only kind of number in GNU Emacs Lisp, version 18. The range of values for integers is -8388608 to 8388607 (24 bits; i.e., -2**23 to 2**23 - 1) on most machines, but is 25 or 26 bits on some systems. It is important to note that the Emacs Lisp arithmetic functions do not check for overflow. Thus (1+ 8388607) is -8388608 on 24-bit implementations. Version 19 supports floating point numbers. The read syntax for numbers is a sequence of (base ten) digits with an optional sign. The printed representation produced by the Lisp interpreter never has a leading `+'. -1 ; The integer -1. 1 ; The integer 1. +1 ; Also the integer 1. 16777217 ; Also the integer 1! (on a 24-bit or 25-bit implementation) See Numbers, for more information. ΓòÉΓòÉΓòÉ 5.3.2. Character Type ΓòÉΓòÉΓòÉ A character in Emacs Lisp is nothing more than an integer. In other words, characters are represented by their eight-bit ASCII values. For example, the character A is represented as the integer 65. If an arbitrary integer is used as a character, only the lower eight bits are significant. Individual characters are not often used in programs. It is far more common to work with strings, which are sequences composed of characters. See String Type. Since characters are really integers, the printed representation of a character is a decimal number. This is also a possible read syntax for a character, but it is not convenient for that purpose. Therefore, Emacs Lisp provides a variety of read syntax formats, such as `?A' and `?\101', for entering characters. They all start with a question mark. The usual read syntax for alphanumeric characters is a question mark followed by the character; thus, `?A' for the character A, `?B' for the character B, and `?a' for the character a. For example: ?Q => 81 ?q => 113 The characters backspace, tab, newline, vertical tab, formfeed, return, and escape may be represented as `?\b', `?\t', `?\n', `?\v', `?\f', `?\r', `?\e', respectively. Those values are 8, 9, 10, 11, 12, 13, and 27 in decimal. Thus, ?\b => 8 ; backspace, BS, C-h ?\t => 9 ; tab, TAB, C-i ?\n => 10 ; newline, LFD, C-j ?\v => 11 ; vertical tab, C-k ?\f => 12 ; formfeed character, C-l ?\r => 13 ; carriage return, RET, C-m ?\e => 27 ; escape character, ESC, C-[ ?\\ => 92 ; backslash character, \ Control characters may be represented using yet another read syntax. This consists of a question mark followed by a backslash, caret, and the corresponding non-control character, in either upper or lower case. For example, either `?\^I' or `?\^i' may be used as the read syntax for the character C-i, the character whose value is 9. Instead of the `^', you can use `C-'; thus, `?\C-i' is equivalent to `?\^I', and `?\C-i' is equivalent to `?\^i'. For example: ?\^I => 9 ?\C-I => 9 The DEL key can be considered and written as Control-?: ?\^? => 127 ?\C-? => 127 When you represent control characters to be found in files or strings, we recommend the `^' syntax; but when you refer to keyboard input, we prefer the `C-' syntax. This does not affect the meaning of the program, but may guide the understanding of people who read it. A meta character is a character that has its eighth bit set. The read syntax for these characters is question mark, backslash, `M-', and the corresponding seven-bit character. For example, `?\M-A' stands for M-A, the character with code 193 (301 octal). The seven-bit character can be written directly or specified by any of the `\' escapes described above or below. Thus, M-A can be written as `?\M-A', or as `?\M-\101', or as `?\301'. Likewise, the character whose decimal printed representation is 130 (202 octal) can use any one of the following for its read syntax: `?\202', `?\M-\C-b', `?\C-\M-b', or `?\M-\002'. For example: ?\C-\M-b => 130 ?\M-\C-b => 130 ?\^\M-b => 130 ?\M-\002 => 130 Finally, the most general read syntax consists of a question mark followed by a backslash and the ASCII code for the character in octal (up to three octal digits); thus, `?\101' for the character A, `?\001' for the character C-a, and ?\002 for the character C-b. Although this syntax can represent any character, it is preferred only when the precise octal value is more important than the ASCII representation. (These sequences which start with backslash are also known as escape sequences, because backslash plays the role of an escape character, but they have nothing to do with the character ESC.) ?\012 => 10 ?\n => 10 ?\C-j => 10 ?\101 => 65 ?A => 65 ?\301 => 193 ?\M-A => 193 ?\M-\101 => 193 A backslash is allowed, and harmless, preceding any character without a special escape meaning; thus, `?\A' is equivalent to `?A'. There is no reason to use a backslash before most such characters. However, any of the characters `()\|;'`"#.,' should be preceded by a backslash to avoid confusing the Emacs commands for editing Lisp code. Whitespace characters such as space, tab, newline and formfeed should also be preceded by a backslash. However, it is cleaner to use one of the easily readable escape sequences, such as `\t', instead of an actual control character such as a tab. ΓòÉΓòÉΓòÉ 5.3.3. Sequence Types ΓòÉΓòÉΓòÉ A sequence is a Lisp object that represents an ordered set of elements. There are two kinds of sequence in Emacs Lisp, lists and arrays. Thus, an object of type list or of type array is also considered a sequence. Arrays are further subdivided into strings and vectors. Vectors can hold elements of any type, but string elements must be characters---integers from 0 to 255. Lists, strings and vectors are different, but they have important similarities. For example, all have a length l, and all have elements which can be indexed from zero to l minus one. Also, several functions, called sequence functions, accept any kind of sequence. For example, the function elt can be used to extract an element of a sequence, given its index. See Sequences Arrays Vectors. It is impossible to read the same sequence twice, in the sense of eq (see Equality Predicates), since sequences are always created anew upon reading. There is one exception: the empty list () is always read as the same object, nil. ΓòÉΓòÉΓòÉ 5.3.4. List Type ΓòÉΓòÉΓòÉ A list is a series of cons cells, linked together. A cons cell is an object comprising two pointers named the car and the cdr. Each of them can point to any Lisp object, but when the cons cell is part of a list, the cdr points either to another cons cell or to the empty list. See Lists, for functions that work on lists. The names car and cdr have only historical meaning now. The original Lisp implementation ran on an IBM 704 computer which divided words into two parts, called the ``address'' part and the ``decrement''; car was an instruction to extract the contents of the address part of a register, and cdr an instruction to extract the contents of the decrement. By contrast, ``cons cells'' are named for the function cons that creates them, which in turn is named for its purpose, the construction of cells. Because cons cells are so central to Lisp, we also have a word for ``an object which is not a cons cell''. These objects are called atoms. The read syntax and printed representation for lists are identical, and consist of a left parenthesis, an arbitrary number of elements, and a right parenthesis. Upon reading, any object inside the parentheses is made into an element of the list. That is, a cons cell is made for each element. The car of the cons cell points to the element, and its cdr points to the next cons cell which holds the next element in the list. The cdr of the last cons cell is set to point to nil. A list can be illustrated by a diagram in which the cons cells are shown as pairs of boxes. (The Lisp reader cannot read such an illustration; unlike the textual notation, which can be understood both humans and computers, the box illustrations can only be understood by humans.) The following represents the three-element list (rose violet buttercup): ___ ___ ___ ___ ___ ___ |___|___|--> |___|___|--> |___|___|--> nil | | | | | | --> rose --> violet --> buttercup In the diagram, each box represents a slot that can refer to any Lisp object. Each pair of boxes represents a cons cell. Each arrow is a reference to a Lisp object, either an atom or another cons cell. In this example, the first box, the car of the first cons cell, refers to or ``contains'' rose (a symbol). The second box, the cdr of the first cons cell, refers to the next pair of boxes, the second cons cell. The car of the second cons cell refers to violet and the cdr refers to the third cons cell. The cdr of the third (and last) cons cell refers to nil. Here is another diagram of the same list, (rose violet buttercup), sketched in a different manner: --------------- ---------------- ------------------- |car |cdr | |car |cdr | |car |cdr | | rose | o---------->| violet | o---------->| buttercup | nil | | | | | | | | | | --------------- ---------------- ------------------- A list with no elements in it is the empty list; it is identical to the symbol nil. In other words, nil is both a symbol and a list. Here are examples of lists written in Lisp syntax: (A 2 "A") ; A list of three elements. () ; A list of no elements (the empty list). nil ; A list of no elements (the empty list). ("A ()") ; A list of one element: the string "A ()". (A ()) ; A list of two elements: A and the empty list. (A nil) ; Equivalent to the previous. ((A B C)) ; A list of one element ; (which is a list of three elements). Here is the list (A ()), or equivalently (A nil), depicted with boxes and arrows: ___ ___ ___ ___ |___|___|--> |___|___|--> nil | | | | --> A --> nil ΓòÉΓòÉΓòÉ 5.3.4.1. Dotted Pair Notation ΓòÉΓòÉΓòÉ Dotted pair notation is an alternative syntax for cons cells that represents the car and cdr explicitly. In this syntax, (a . b) stands for a cons cell whose car is the object a, and whose cdr is the object b. Dotted pair notation is therefore more general than list syntax. In the dotted pair notation, the list `(1 2 3)' is written as `(1 . (2 . (3 . nil)))'. For nil-terminated lists, the two notations produce the same result, but list notation is usually clearer and more convenient when it is applicable. When printing a list, the dotted pair notation is only used if the cdr of a cell is not a list. Box notation can also be used to illustrate what dotted pairs look like. For example, (rose . violet) is diagrammed as follows: ___ ___ |___|___|--> violet | | --> rose Dotted pair notation can be combined with list notation to represent a chain of cons cells with a non-nil final cdr. For example, (rose violet . buttercup) is equivalent to (rose . (violet . buttercup)). The object looks like this: ___ ___ ___ ___ |___|___|--> |___|___|--> buttercup | | | | --> rose --> violet These diagrams make it evident that (rose . violet . buttercup) must have an invalid syntax since it would require that a cons cell have three parts rather than two. The list (rose violet) is equivalent to (rose . (violet)) and looks like this: ___ ___ ___ ___ |___|___|--> |___|___|--> nil | | | | --> rose --> violet Similarly, the three-element list (rose violet buttercup) is equivalent to (rose . (violet . (buttercup))). It looks like this: ___ ___ ___ ___ ___ ___ |___|___|--> |___|___|--> |___|___|--> nil | | | | | | --> rose --> violet --> buttercup ΓòÉΓòÉΓòÉ 5.3.4.2. Association List Type ΓòÉΓòÉΓòÉ An association list or alist is a specially-constructed list whose elements are cons cells. In each element, the car is considered a key, and the cdr is considered an associated value. (In some cases, the associated value is stored in the car of the cdr.) Association lists are often used to implement stacks, since new associations may easily be added to or removed from the front of the list. For example, (setq alist-of-colors '((rose . red) (lily . white) (buttercup . yellow))) sets the variable alist-of-colors to an alist of three elements. In the first element, rose is the key and red is the value. See Association Lists, for a further explanation of alists and for functions that work on alists. ΓòÉΓòÉΓòÉ 5.3.5. Array Type ΓòÉΓòÉΓòÉ An array is composed of an arbitrary number of other Lisp objects, arranged in a contiguous block of memory. Any element of an array may be accessed in constant time. In contrast, accessing an element of a list requires time proportional to the position of the element in the list. (Elements at the end of a list take longer to access than elements at the beginning of a list.) Emacs defines two types of array, strings and vectors. A string is an array of characters and a vector is an array of arbitrary objects. Both are one-dimensional. (Most other programming languages support multidimensional arrays, but we don't think they are essential in Emacs Lisp.) Each type of array has its own read syntax; see String Type, and Vector Type. An array may have any length up to the largest integer; but once created, it has a fixed size. The first element of an array has index zero, the second element has index 1, and so on. This is called zero-origin indexing. For example, an array of four elements has indices 0, 1, 2, and 3. The array type is contained in the sequence type and contains both strings and vectors. ΓòÉΓòÉΓòÉ 5.3.6. String Type ΓòÉΓòÉΓòÉ A string is an array of characters. Strings are used for many purposes in Emacs, as can be expected in a text editor; for example, as the names of Lisp symbols, as messages for the user, and to represent text extracted from buffers. Strings in Lisp are constants; evaluation of a string returns the same string. The read syntax for strings is a double-quote, an arbitrary number of characters, and another double-quote, "like this". The Lisp reader accepts the same formats for reading the characters of a string as it does for reading single characters (without the question mark that begins a character literal). You can enter a nonprinting character such as tab, C-a or M-C-A using the convenient escape sequences, like this: "\t, \C-a, \M-\C-a". You can include a double-quote in a string by preceding it with a backslash; thus, "\"" is a string containing just a single double-quote character. (See Character Type, for a description of the read syntax for characters.) In contrast with the C programming language, newlines are allowed in Emacs Lisp string literals. But an escaped newline---one that is preceded by `\'---does not become part of the string; i.e., the Lisp reader ignores an escaped newline in a string literal. "It is useful to include newlines in documentation strings, but the newline is \ ignored if escaped." => "It is useful to include newlines in documentation strings, but the newline is ignored if escaped." The printed representation of a string consists of a double-quote, the characters it contains, and another double-quote. However, any backslash or double-quote characters in the string are preceded with a backslash like this: "this \" is an embedded quote". See Strings and Characters, for functions that work on strings. ΓòÉΓòÉΓòÉ 5.3.7. Vector Type ΓòÉΓòÉΓòÉ A vector is a one-dimensional array of elements of any type. It takes a constant amount of time to access any element of a vector. (In a list, the access time of an element is proportional to the distance of the element from the beginning of the list.) The printed representation of a vector consists of a left square bracket, the elements, and a right square bracket. This is also the read syntax. Like numbers and strings, vectors are considered constants for evaluation. [1 "two" (three)] ; A vector of three elements. => [1 "two" (three)] See Vectors, for functions that work with vectors. ΓòÉΓòÉΓòÉ 5.3.8. Symbol Type ΓòÉΓòÉΓòÉ A symbol in GNU Emacs Lisp is an object with a name. The symbol name is the printed representation of the symbol. In ordinary use, the name is unique---no two symbols have the same name. A symbol may be used in programs as a variable, as a function name, or to hold a list of properties. Or it may serve only to be distinct from all other Lisp objects, so that its presence in a data structure may be recognized reliably. In a given context, usually only one of these uses is intended. A symbol name can contain any characters whatever. Most symbol names are written with letters, digits, and the punctuation characters `-+=*/'. Such names require no special punctuation; the characters of the name suffice as long as the name does not look like a number. (If it does, write a `\' at the beginning of the name to force interpretation as a symbol.) The characters `_~!@$%^&:<>{}' are less often used but also require no special punctuation. Any other characters may be included in a symbol's name by escaping them with a backslash. In contrast to its use in strings, however, a backslash in the name of a symbol quotes the single character that follows the backslash, without conversion. For example, in a string, `\t' represents a tab character; in the name of a symbol, however, `\t' merely quotes the letter t. To have a symbol with a tab character in its name, you must actually type an tab (preceded with a backslash). But you would hardly ever do such a thing. Common Lisp note: in Common Lisp, lower case letters are always ``folded'' to upper case, unless they are explicitly escaped. This is in contrast to Emacs Lisp, in which upper case and lower case letters are distinct. Here are several examples of symbol names. Note that the `+' in the fifth example is escaped to prevent it from being read as a number. This is not necessary in the last example because the rest of the name makes it invalid as a number. foo ; A symbol named `foo'. FOO ; A symbol named `FOO', different from `foo'. char-to-string ; A symbol named `char-to-string'. 1+ ; A symbol named `1+' ; (not `+1', which is an integer). \+1 ; A symbol named `+1' (not a very readable name). $*\ 1\ 2$ ; A symbol named `(* 1 2)' (a worse name). +-*/_~!@$%^&=:<>{} ; A symbol named `+-*/_~!@$%^&=:<>{}'. ; These characters need not be escaped. ΓòÉΓòÉΓòÉ 5.3.9. Lisp Function Type ΓòÉΓòÉΓòÉ Just as functions in other programming languages are executable, a Lisp function object is a piece of executable code. However, Lisp functions are primarily Lisp objects, and only secondarily the text which represents them. These Lisp objects are lambda expressions: lists whose first element is the symbol lambda ( see Lambda Expressions). In most programming languages, it is impossible to have a function without a name. In Lisp, a function has no intrinsic name. A lambda expression is also called an anonymous function ( see Anonymous Functions). A named function in Lisp is actually a symbol with a valid function in its function cell (see Defining Functions). Most of the time, functions are called when their names are written in Lisp expressions in Lisp programs. However, a function object found or constructed at run time can be called and passed arguments with the primitive functions funcall and apply. See Calling Functions. ΓòÉΓòÉΓòÉ 5.3.10. Lisp Macro Type ΓòÉΓòÉΓòÉ A Lisp macro is a user-defined construct that extends the Lisp language. It is represented as an object much like a function, but with different parameter-passing semantics. A Lisp macro has the form of a list whose first element is the symbol macro and whose cdr is a Lisp function object, including the lambda symbol. Lisp macro objects are usually defined with the built-in function defmacro, but any list that begins with macro is a macro as far as Emacs is concerned. See Macros, for an explanation of how to write a macro. ΓòÉΓòÉΓòÉ 5.3.11. Primitive Function Type ΓòÉΓòÉΓòÉ A primitive function is a function callable from Lisp but written in the C programming language. Primitive functions are also called subrs or built-in functions. (The word ``subr'' is derived from ``subroutine''.) Most primitive functions evaluate all their arguments when they are called. A primitive function that does not evaluate all its arguments is called a special form (see Special Forms). It does not matter to the caller of a function whether the function is primitive. However, this does matter if you are trying to substitute a function written in Lisp for a primitive of the same name. The reason is that the primitive function may be called directly from C code. When the redefined function is called from Lisp, the new definition will be used; but calls from C code may still use the old definition. The term function is used to refer to all Emacs functions, whether written in Lisp or C. See Lisp Function Type, for information about the functions written in Lisp. Primitive functions have no read syntax and print in hash notation with the name of the subroutine. (symbol-function 'car) ; Access the function cell of the symbol. => #<subr car> (subrp (symbol-function 'car)) ; Is this a primitive function? => t ; Yes. ΓòÉΓòÉΓòÉ 5.3.12. Autoload Type ΓòÉΓòÉΓòÉ An autoload object is a list whose first element is the symbol autoload. It is stored as the function definition of a symbol to say that a file of Lisp code should be loaded when necessary to find the true definition of that symbol. The autoload object contains the name of the file, plus some other information about the real definition. After the file has been loaded, the symbol should have a new function definition that is not an autoload object. The new definition is then called as if it had been there to begin with. From the user's point of view, the function call works as expected, using the function definition in the loaded file. An autoload object is usually created with the function autoload, which stores the object in the function cell of a symbol. See Autoload, for more details. ΓòÉΓòÉΓòÉ 5.4. Editing Types ΓòÉΓòÉΓòÉ The types in the previous section are common to many Lisp-like languages. But Emacs Lisp provides several additional data types for purposes connected with editing. ΓòÉΓòÉΓòÉ 5.4.1. Buffer Type ΓòÉΓòÉΓòÉ A buffer is an object that holds text that can be edited (see Buffers). Most buffers hold the contents of a disk file (see Files) so they can be edited, but some are used for other purposes. Most buffers are also meant to be seen by the user, and therefore displayed, at some time, in a window (see Windows). But a buffer need not be displayed in a window. The contents of a buffer are much like a string, but buffers are not used like strings in Emacs Lisp, and the available operations are different. For example, text can be inserted into a buffer very quickly, while ``inserting'' text into a string is accomplished by concatenation and the result is an entirely new string object. Each buffer has a designated position called point (see Positions). And one buffer is the current buffer. Most editing commands act on the contents of the current buffer in the neighborhood of point. Many other functions manipulate or test the characters in the current buffer and much of this manual is devoted to describing these functions (see Text). Several other data structures are associated with each buffer: o a local syntax table (see Syntax Tables); o a local keymap (see Keymaps); and, o a local variable binding list (see Variables). The local keymap and variable list contain entries which individually override global bindings or values. These are used to customize the behavior of programs in different buffers, without actually changing the programs. Buffers have no read syntax. They print in hash notation with the buffer name. (current-buffer) => #<buffer objects.texi> ΓòÉΓòÉΓòÉ 5.4.2. Window Type ΓòÉΓòÉΓòÉ A window describes the portion of the terminal screen that Emacs uses to display a buffer. Every window has one associated buffer, whose contents appear in the window. By contrast, a given buffer may appear in one window, no window, or several windows. Though many windows may exist simultaneously, one window is designated the selected window. This is the window where the cursor is (usually) displayed when Emacs is ready for a command. The selected window usually displays the current buffer, but this is not necessarily the case. Windows have no read syntax. They print in hash notation, giving the window number and the name of the buffer being displayed. The window numbers exist to identify windows uniquely, since the buffer displayed in any given window can change frequently. (selected-window) => #<window 1 on objects.texi> See Windows, for a description of the functions that work on windows. ΓòÉΓòÉΓòÉ 5.4.3. Window Configuration Type ΓòÉΓòÉΓòÉ A window configuration stores information about the positions and sizes of windows at the time the window configuration is created, so that the screen layout may be recreated later. Window configurations have no read syntax. They print as `#<window-configuration>'. See Window Configurations, for a description of several functions related to window configurations. ΓòÉΓòÉΓòÉ 5.4.4. Marker Type ΓòÉΓòÉΓòÉ A marker denotes a position in a specific buffer. Markers therefore have two components: one for the buffer, and one for the position. The position value is changed automatically as necessary as text is inserted into or deleted from the buffer. This is to ensure that the marker always points between the same two characters in the buffer. Markers have no read syntax. They print in hash notation, giving the current character position and the name of the buffer. (point-marker) => #<marker at 10779 in objects.texi> See Markers, for information on how to test, create, copy, and move markers. ΓòÉΓòÉΓòÉ 5.4.5. Process Type ΓòÉΓòÉΓòÉ The word process means a running program. Emacs itself runs in a process of this sort. However, in Emacs Lisp, a process is a Lisp object that designates a subprocess created by Emacs process. External subprocesses, such as shells, GDB, ftp, and compilers, may be used to extend the processing capability of Emacs. A process takes input from Emacs and returns output to Emacs for further manipulation. Both text and signals can be communicated between Emacs and a subprocess. Processes have no read syntax. They print in hash notation, giving the name of the process: (process-list) => (#<process shell>) See Processes, for information about functions that create, delete, return information about, send input or signals to, and receive output from processes. ΓòÉΓòÉΓòÉ 5.4.6. Stream Type ΓòÉΓòÉΓòÉ A stream is an object that can be used as a source or sink for characters---either to supply characters for input or to accept them as output. Many different types can be used this way: markers, buffers, strings, and functions. Most often, input streams (character sources) obtain characters from the keyboard, a buffer, or a file, and output streams (character sinks) send characters to a buffer, such as a `*Help*' buffer, or to the echo area. The object nil, in addition to its other meanings, may be used as a stream. It stands for the value of the variable standard-input or standard-output. Also, the object t as a stream specifies input using the minibuffer (see Minibuffers) or output in the echo area ( see The Echo Area). Streams have no special printed representation or read syntax, and print as whatever primitive type they are. See Streams, for a description of various functions related to streams, including various parsing and printing functions. ΓòÉΓòÉΓòÉ 5.4.7. Keymap Type ΓòÉΓòÉΓòÉ A keymap maps keys typed by the user to functions. This mapping controls how the user's command input is executed. Emacs defines two kinds of keymaps: full keymaps, which are vectors of 128 elements, and sparse keymaps, which are association lists whose first element is the symbol keymap. See Keymaps, for information about creating keymaps, handling prefix keys, local as well as global keymaps, and changing key bindings. ΓòÉΓòÉΓòÉ 5.4.8. Syntax Table Type ΓòÉΓòÉΓòÉ A syntax table is a vector of 256 integers. Each element of the vector defines how one character is interpreted when it appears in a buffer. For example, in C mode (see Major Modes), the `+' character is punctuation, but in Lisp mode it is a valid character in a symbol. These different interpretations are effected by changing the syntax table entry for `+', i.e., at index 43. Syntax tables are only used for scanning text in buffers, not for reading Lisp expressions. The table the Lisp interpreter uses to read expressions is built into the Emacs source code and cannot be changed; thus, to change the list delimiters to be `{' and `}' instead of `(' and `)' would be impossible. See Syntax Tables, for details about syntax classes and how to make and modify syntax tables. ΓòÉΓòÉΓòÉ 5.5. Type Predicates ΓòÉΓòÉΓòÉ The Emacs Lisp interpreter itself does not perform type checking on the actual arguments passed to functions when they are called. It could not do otherwise, since variables in Lisp are not declared to be of a certain type, as they are in other programming languages. It is therefore up to the individual function to test whether each actual argument belongs to a type that can be used by the function. All built-in functions do check the types of their actual arguments when appropriate and signal a wrong-type-argument error if an argument is of the wrong type. For example, here is what happens if you pass an argument to + which it cannot handle: (+ 2 'a) error--> Wrong type argument: integer-or-marker-p, a Many functions, called type predicates, are provided to test whether an object is a member of a given type. (Following a convention of long standing, the names of most Emacs Lisp predicates end in `p'.) Here is a table of predefined type predicates, in alphabetical order, with references to further information. atom see List-related Predicates arrayp see Array Functions bufferp see Buffer Basics char-or-string-p see Predicates for Strings consp see List-related Predicates integer-or-marker-p see Predicates on Markers integerp see Predicates on Numbers keymapp see Creating Keymaps listp see List-related Predicates markerp see Predicates on Markers natnump see Predicates on Numbers nlistp see List-related Predicates processp see Processes sequencep see Sequence Functions stringp see Predicates for Strings subrp see Function Cells symbolp see Symbols syntax-table-p see Syntax Tables user-variable-p see Defining Variables vectorp see Vectors windowp see Basic Windows ΓòÉΓòÉΓòÉ 5.6. Equality Predicates ΓòÉΓòÉΓòÉ Here we describe two functions that test for equality between any two objects. Other functions test equality between objects of specific types, e.g., strings. See the appropriate chapter describing the data type for these predicates. -- Function: eq object1 object2 This function returns t if object1 and object2 are the same object, nil otherwise. The ``same object'' means that a change in one will be reflected by the same change in the other. eq will be true if object1 and object2 are numbers with the same value. Also, since symbol names are normally unique, if the arguments are symbols with the same name, they are eq. For other types (e.g., lists, vectors, strings), two arguments with the same contents or elements are not necessarily eq to each other: they are eq only if they are the same object. (The make-symbol function returns an uninterned symbol that is not interned in the standard obarray. When uninterned symbols are in use, symbol names are no longer unique. Distinct symbols with the same name are not eq. See Creating Symbols.) (eq 'foo 'foo) => t (eq 456 456) => t (eq "asdf" "asdf") => nil (eq '(1 (2 (3))) '(1 (2 (3)))) => nil (eq [(1 2) 3] [(1 2) 3]) => nil (eq (point-marker) (point-marker)) => nil -- Function: equal object1 object2 This function returns t if object1 and object2 have equal components, nil otherwise. Whereas eq tests if its arguments are the same object, equal looks inside nonidentical arguments to see if their elements are the same. So, if two objects are eq, they are equal, but the converse is not always true. (equal 'foo 'foo) => t (equal 456 456) => t (equal "asdf" "asdf") => t (eq "asdf" "asdf") => nil (equal '(1 (2 (3))) '(1 (2 (3)))) => t (eq '(1 (2 (3))) '(1 (2 (3)))) => nil (equal [(1 2) 3] [(1 2) 3]) => t (eq [(1 2) 3] [(1 2) 3]) => nil (equal (point-marker) (point-marker)) => t (eq (point-marker) (point-marker)) => nil Comparison of strings is case-sensitive. (equal "asdf" "ASDF") => nil The test for equality is implemented recursively, and circular lists may therefore cause infinite recursion (leading to an error). ΓòÉΓòÉΓòÉ 6. Numbers ΓòÉΓòÉΓòÉ Integers are the only kind of number in version 18 Emacs Lisp. These are whole numbers such as -3, 0, 7, 13, and 511. In version 19, there is a compile time option to support floating point numbers, which are represented internally as the C type double. A floating point number is a number with a fractional part, such as -4.5, 0.0, or 2.71828. A floating point number can be expressed in an exponential notation as well: thus, 1.5e2 equals 150; in this example, `e2' stands for ten to the second power, and is multiplied by 1.5. ΓòÉΓòÉΓòÉ 6.1. Number Basics ΓòÉΓòÉΓòÉ The range of values for an integer depends on the machine. The range is -8388608 to 8388607 (24 bits; i.e., -2**23 to 2**23 - 1 ) on most machines, but on others it is -16777216 to 16777215 (25 bits), or -33554432 to 33554431 (26 bits). All of the examples shown below assume an integer has 24 bits. The Lisp reader reads numbers as a sequence of digits with an optional sign. 1 ; The integer 1. +1 ; Also the integer 1. -1 ; The integer -1. 16777217 ; Also the integer 1, due to overflow. 0 ; The number 0. -0 ; The number 0. 1. ; Invalid syntax. To understand how various functions work on integers, especially the bitwise operators (see Bitwise Operations), it is often helpful to view the numbers in their binary form. In 24 bit binary, the decimal integer 5 looks like this: 0000 0000 0000 0000 0000 0101 (We have inserted spaces between groups of 4 bits, and two spaces between groups of 8 bits, to make the binary integer easier to read.) The integer -1 looks like this: 1111 1111 1111 1111 1111 1111 -1 is represented as 24 ones. (This is called two's complement notation.) The negative integer, -5, is creating by subtracting 4 from -1. In binary, the decimal integer 4 is 100. Consequently, -5 looks like this: 1111 1111 1111 1111 1111 1011 In this implementation, the largest 24 bit binary integer is the decimal integer 8,388,607. In binary, this number looks like this: 0111 1111 1111 1111 1111 1111 Since the arithmetic functions do not check whether integers go outside their range, when you add 1 to 8,388,607, the value is negative integer -8,388,608: (+ 1 8388607) => -8388608 => 1000 0000 0000 0000 0000 0000 Many of the following functions accept markers for arguments as well as integers. (See Markers.) More precisely, the actual parameters to such functions may be either integers or markers, which is why we often give these parameters the name marker-or-int. When the actual parameter is a marker, the position value of the marker is used and the buffer of the marker is ignored. ΓòÉΓòÉΓòÉ 6.2. Type Predicates for Numbers ΓòÉΓòÉΓòÉ The functions in this section test whether the argument is a number or whether it is a certain sort of number. integerp and natnump can take any type of Lisp object as argument (the predicates would not be of much use otherwise); but the zerop predicate requires an integer as its argument. -- Function: integerp object This predicate tests whether its argument is an integer (a whole number) and returns t if so, nil otherwise. -- Function: natnump object The natnump predicate (whose name comes from the phrase ``natural-number-p'') tests to see whether its argument is a nonnegative integer, and returns t if so, nil otherwise. 0 is considered non-negative. Markers are not converted to integers, hence natnump of a marker is always nil. People have pointed out that this function is misnamed, because the term ``natural number'' is usually understood as excluding zero. We are open to suggestions for a better name to use in version 19. -- Function: zerop integer This predicate tests whether its argument is zero, and returns t if so, nil otherwise. These two forms are equivalent: (zerop x) == (= x 0). ΓòÉΓòÉΓòÉ 6.3. Comparison of Numbers ΓòÉΓòÉΓòÉ The integer type is implemented by storing the value in the ``pointer part'' of a Lisp object (which, on typical target machines, has 24 bits of pointer, 7 bits of type and 1 bit for the garbage collector). Because of this, the function eq will return t for two integers with the same value. See Equality Predicates. Common Lisp note: because of the way numbers are implemented in Common Lisp, you generally need to use `=' to test for equality between numbers. However, GNU Emacs Lisp does not need very large integers; as a consequence, it is possible to restrict them to the size of a single word, allowing eq to be used. -- Function: = marker-or-int1 marker-or-int2 This function tests whether its arguments are the same number, and returns t if so, nil otherwise. -- Function: /= marker-or-int1 marker-or-int2 This function tests whether its arguments are not the same number, and returns t if so, nil otherwise. -- Function: < marker-or-int1 marker-or-int2 This function tests whether its first argument is strictly less than its second argument. It returns t if so, nil otherwise. -- Function: <= marker-or-int1 marker-or-int2 This function tests whether its first argument is less than or equal to its second argument. It returns t if so, nil otherwise. -- Function: > marker-or-int1 marker-or-int2 This function tests whether its first argument is strictly greater than its second argument. It returns t if so, nil otherwise. -- Function: >= marker-or-int1 marker-or-int2 This function tests whether its first argument is greater than or equal to its second argument. It returns t if so, nil otherwise. -- Function: max marker-or-int &rest markers-or-ints This function returns the largest of its arguments. (max 20) => 20 (max 1 2) => 2 (max 1 3 2) => 3 -- Function: min marker-or-int &rest markers-or-ints This function returns the smallest of its arguments. ΓòÉΓòÉΓòÉ 6.4. Arithmetic Operations ΓòÉΓòÉΓòÉ Emacs Lisp provides the traditional four arithmetic operations: addition, subtraction, multiplication, and division. A remainder function supplements the (integer) division function. In addition, as a convenience, incrementing and decrementing functions are provided. It is important to note that in GNU Emacs Lisp, arithmetic functions do not check for overflow. Thus (1+ 8388607) may equal -8388608, depending on your hardware. -- Function: 1+ marker-or-int This function adds one to marker-or-int. -- Function: 1- marker-or-int This function subtracts one from marker-or-int. -- Function: + &rest markers-or-ints This function adds its arguments together. When given no arguments, + returns 0. It does not check for overflow. (+) => 0 (+ 1) => 1 (+ 1 2 3 4) => 10 -- Function: - &optional marker-or-int &rest other-markers-or-ints The - function serves two purposes: negation and subtraction. When - has a single argument, the value is the negative of the argument. When there are multiple arguments, each of the other-markers-or-ints is subtracted from marker-or-int, cumulatively. If there are no arguments, the result is 0. This function does not check for overflow. (- 10 1 2 3 4) => 0 (- 10) => -10 (-) => 0 -- Function: * &rest markers-or-integers This function multiplies its arguments together, and returns the product. When given no arguments, * returns 1. It does not check for overflow. (*) => 1 (* 1) => 1 (* 1 2 3 4) => 24 -- Function: / dividend divisor &rest divisors This function divides dividend by divisors and returns the quotient. If there are additional arguments divisors, then dividend is divided by each divisor in turn. Each argument may be an integer or a marker. The result is normally rounded towars zero after each division, but some machines may round differently with negative arguments. This is because the Lisp function / is implemented using the C division operator, which has the same possibility for machine-dependent rounding. In practice, all known machines round in the standard fashion. If you divide by 0, an arith-error error is signaled. (See Errors.) (/ 6 2) => 3 (/ 5 2) => 2 (/ 25 3 2) => 4 (/ -17 6) => -2 ; (Could be -3 on some machines.) -- Function: % dividend divisor This function returns the value of dividend modulo divisor; in other words, the integer remainder after division of dividend by divisor. The sign of the result is the sign of dividend. The sign of divisor is ignored. The arguments must be integers. For negative arguments, the value is in principle machine-dependent since the quotient is; but in practice, all known machines behave alike. An arith-error results if divisor is 0. (% 9 4) => 1 (% -9 4) => -1 (% 9 -4) => 1 (% -9 -4) => -1 For any two numbers dividend and divisor, (+ (% dividend divisor) (* (/ dividend divisor) divisor)) always equals dividend. ΓòÉΓòÉΓòÉ 6.5. Bitwise Operations on Integers ΓòÉΓòÉΓòÉ In a computer, an integer is represented as a binary number, a sequence of bits (digits which are either zero or one). A bitwise operation acts on the individual bits of such a sequence. For example, shifting moves the whole sequence left or right one or more places, reproducing the same pattern ``moved over''. The bitwise operations in Emacs Lisp apply only to integers. -- Function: lsh integer1 count lsh, which is an abbreviation for logical shift, shifts the bits in integer1 to the left count places, or to the right if count is negative. If count is negative, lsh shifts zeros into the most-significant bit, producing a positive result even if integer1 is negative. Contrast this with ash, below. Thus, the decimal number 5 is the binary number 00000101. Shifted once to the left, with a zero put in the one's place, the number becomes 00001010, decimal 10. Here are two examples of shifting the pattern of bits one place to the left. Since the contents of the rightmost place has been moved one place to the left, a value has to be inserted into the rightmost place. With lsh, a zero is placed into the rightmost place. (These examples show only the low-order eight bits of the binary pattern; the rest are all zero.) (lsh 5 1) => 10 00000101 => 00001010 ; Decimal 5 becomes decimal 10. (lsh 7 1) => 14 00000111 => 00001110 ; Decimal 7 becomes decimal 14. As the examples illustrate, shifting the pattern of bits one place to the left produces a number that is twice the value of the previous number. Note, however that functions do not check for overflow, and a returned value may be negative (and in any case, no more than a 24 bit value) when an integer is sufficiently left shifted. For example: (lsh 8388607 1) ; left shift => -2 In binary, in the 24 bit implementation, 0111 1111 1111 1111 1111 1111 ; Decimal 8,388,607 becomes 1111 1111 1111 1111 1111 1110 ; Decimal -2 Shifting the pattern of bits two places to the left produces results like this (with 8-bit binary numbers): (lsh 3 2) => 12 00000011 => 00001100 ; Decimal 3 becomes decimal 12. On the other hand, shifting the pattern of bits one place to the right looks like this: (lsh 6 -1) => 3 00000110 => 00000011 ; Decimal 6 becomes decimal 3. (lsh 5 -1) => 2 00000101 => 00000010 ; Decimal 5 becomes decimal 2. As the example illustrates, shifting the pattern of bits one place to the right divides the value of the binary number by two, rounding downward. -- Function: ash integer1 count ash (arithmetic shift) shifts the bits in integer1 to the left count places, or to the right if count is negative. ash gives the same results as lsh except when integer1 and count are both negative. In that case, ash puts a one in the leftmost position, while lsh puts a zero in the leftmost position. Thus, with ash, shifting the pattern of bits one place to the right looks like this: (ash -6 -1) => -3 ; Decimal -6 becomes decimal -3. 1111 1111 1111 1111 1111 1010 => 1111 1111 1111 1111 1111 1101 In contrast, shifting the pattern of bits one place to the right with lsh looks like this: (lsh -6 -1) => 8388605 ; Decimal -6 becomes decimal 8,388,605. 1111 1111 1111 1111 1111 1010 => 0111 1111 1111 1111 1111 1101 In this case, the 1 in the leftmost position is shifted one place to the right, and a zero is shifted into the leftmost position. Here are other examples: ; 24-bit binary values (lsh 5 2) ; 5 = 0000 0000 0000 0000 0000 0101 => 20 ; 20 = 0000 0000 0000 0000 0001 0100 (ash 5 2) => 20 (lsh -5 2) ; -5 = 1111 1111 1111 1111 1111 1011 => -20 ; -20 = 1111 1111 1111 1111 1110 1100 (ash -5 2) => -20 (lsh 5 -2) ; 5 = 0000 0000 0000 0000 0000 0101 => 1 ; 1 = 0000 0000 0000 0000 0000 0001 (ash 5 -2) => 1 (lsh -5 -2) ; -5 = 1111 1111 1111 1111 1111 1011 => 4194302 ; 0011 1111 1111 1111 1111 1110 (ash -5 -2) ; -5 = 1111 1111 1111 1111 1111 1011 => -2 ; -2 = 1111 1111 1111 1111 1111 1110 -- Function: logand &rest markers-or-ints This function returns the ``logical and'' of the arguments: the nth bit is set in the result if, and only if, the nth bit is set in all the arguments. (``Set'' means that the value of the bit is 1 rather than 0.) For example, using 4-bit binary numbers, the ``logical and'' of 13 and 12 is 12: 1101 combined with 1100 produces 1100. In both the binary numbers, the leftmost two bits are set (i.e., they are 1's), so the leftmost two bits of the returned value are set. However, for the rightmost two bits, each is zero in at least one of the arguments, so the rightmost two bits of the returned value are 0's. Therefore, (logand 13 12) => 12 If logand is not passed any argument, it returns a value of -1. This number is an identity element for logand because its binary representation consists entirely of ones. If logand is passed just one argument, it returns that argument. ; 24-bit binary values (logand 14 13) ; 14 = 0000 0000 0000 0000 0000 1110 ; 13 = 0000 0000 0000 0000 0000 1101 => 12 ; 12 = 0000 0000 0000 0000 0000 1100 (logand 14 13 4) ; 14 = 0000 0000 0000 0000 0000 1110 ; 13 = 0000 0000 0000 0000 0000 1101 ; 4 = 0000 0000 0000 0000 0000 0100 => 4 ; 4 = 0000 0000 0000 0000 0000 0100 (logand) => -1 ; -1 = 1111 1111 1111 1111 1111 1111 -- Function: logior &rest markers-or-ints This function returns the ``inclusive or'' of its arguments: the nth bit is set in the result if, and only if, the nth bit is set in at least one of the arguments. If there are no arguments, the result is zero, which is an identity element for this operation. If logior is passed just one argument, it returns that argument. ; 24-bit binary values (logior 12 5) ; 12 = 0000 0000 0000 0000 0000 1100 ; 5 = 0000 0000 0000 0000 0000 0101 => 13 ; 13 = 0000 0000 0000 0000 0000 1101 (logior 12 5 7) ; 12 = 0000 0000 0000 0000 0000 1100 ; 5 = 0000 0000 0000 0000 0000 0101 ; 7 = 0000 0000 0000 0000 0000 0111 => 15 ; 15 = 0000 0000 0000 0000 0000 1111 -- Function: logxor &rest markers-or-ints This function returns the ``exclusive or'' of its arguments: the nth bit is set in the result if, and only if, the nth bit is set in an odd number of the arguments. If there are no arguments, the result is 0. If logxor is passed just one argument, it returns that argument. ; 24-bit binary values (logxor 12 5) ; 12 = 0000 0000 0000 0000 0000 1100 ; 5 = 0000 0000 0000 0000 0000 0101 => 9 ; 9 = 0000 0000 0000 0000 0000 1001 (logxor 12 5 7) ; 12 = 0000 0000 0000 0000 0000 1100 ; 5 = 0000 0000 0000 0000 0000 0101 ; 7 = 0000 0000 0000 0000 0000 0111 => 14 ; 14 = 0000 0000 0000 0000 0000 1110 -- Function: lognot integer This function returns the logical complement of its argument: the nth bit is one in the result if, and only if, the nth bit is zero in integer, and vice-versa. (lognot 5) ; 5 = 0000 0000 0000 0000 0000 0101 => -6 ; -6 = 1111 1111 1111 1111 1111 1010 ΓòÉΓòÉΓòÉ 6.6. Random Numbers ΓòÉΓòÉΓòÉ -- Function: random &optional flag This function returns a pseudo-random number of type integer. When called more than once, this function returns a series of pseudo-random numbers. In a computer, a series of a pseudo-random numbers is generated in a deterministic fashion. The numbers are not truly random, but they have certain properties that mimic a random series. For example, all possible values occur equally often in a pseudo-random series. In Emacs, pseudo-random numbers are generated from a ``seed'' number. If the random function starts with the same seed, it generates the same sequence of numbers. Emacs always starts with the same seed value, so the sequence of values of random is actually the same in each Emacs run! For example, in one operating system, the first call to (random) after you start Emacs always returns -1457731, and the second one always returns -7692030. This is helpful for debugging. If you want different random numbers, execute (random t). This chooses a new seed based on the current time of day and on Emacs' process id number. On some machines, any integer representable in Lisp may be the result of random. On other machines, the result can never be larger than a certain maximum or less than a certain (negative) minimum. ΓòÉΓòÉΓòÉ 7. Strings and Characters ΓòÉΓòÉΓòÉ A string in Emacs Lisp is an array that contains an ordered sequence of characters. Strings are used as names of symbols, buffers, and files, to send messages to users, to hold text being copied between buffers, and for many other purposes. Because strings are so important, many functions are provided expressly for manipulating them. Emacs Lisp programs use strings more often than individual characters. ΓòÉΓòÉΓòÉ 7.1. Introduction to Strings and Characters ΓòÉΓòÉΓòÉ Characters are represented in Emacs Lisp as integers; whether an integer was intended as a character or not is determined only by how it is used. Strings in Emacs Lisp are arrays that contain an ordered sequence of characters. The length of a string (like any array) is fixed and independent of the string contents, and cannot be altered. Strings in Lisp are not terminated by a distinguished character code. (By contrast, strings in C are terminated by a character with ASCII code 0.) This means that any character, including the null character (ASCII code 0), is a valid element of a string. Since strings are considered arrays, you can operate on them with the general array functions. (See Sequences Arrays Vectors.) For example, you can access or change individual characters in a string using the functions aref and aset ( see Array Functions). Each character in a string is stored in a single byte. Therefore, numbers not in the range 0 to 255 are truncated when stored into a string. This means that a string takes up much less memory than a vector of the same length. See Text, for information about functions that display strings or copy them into buffers. See Character Type, and String Type, for information about the syntax of characters and strings. ΓòÉΓòÉΓòÉ 7.2. The Predicates for Strings ΓòÉΓòÉΓòÉ For more information about general sequence and array predicates, see Sequences Arrays Vectors, and Arrays. -- Function: stringp object This function returns t if object is a string, nil otherwise. -- Function: char-or-string-p object This function returns t if object is a string or a character (i.e., an integer), nil otherwise. ΓòÉΓòÉΓòÉ 7.3. Creating Strings ΓòÉΓòÉΓòÉ The following functions create strings, either from scratch, or by putting strings together, or by taking them apart. -- Function: make-string count character This function returns a string made up of count repetitions of character. If count is negative, an error is signaled. (make-string 5 ?x) => "xxxxx" (make-string 0 ?x) => "" Other functions to compare with this one include char-to-string (see String Conversion), make-vector (see Vectors), and make-list (see Building Lists). -- Function: substring string start &optional end This function returns a new string which consists of those characters from string in the range from (and including) the character at the index start up to (but excluding) the character at the index end. The first character is at index zero. (substring "abcdefg" 0 3) => "abc" Here the index for `a' is 0, the index for `b' is 1, and the index for `c' is 2. Thus, three letters, `abc', are copied from the full string. The index 3 marks the character position up to which the substring is copied. The character whose index is 3 is actually the fourth character in the string. A negative number counts from the end of the string, so that -1 signifies the index of the last character of the string. For example: (substring "abcdefg" -3 -1) => "ef" In this example, the index for `e' is -3, the index for `f' is -2, and the index for `g' is -1. Therefore, `e' and `f' are included, and `g' is excluded. When nil is used as an index, it falls after the last character in the string. Thus: (substring "abcdefg" -3 nil) => "efg" Omitting the argument end is equivalent to specifying nil. It follows that (substring string 0) returns a copy of all of string. (substring "abcdefg" 0) => "abcdefg" But we recommend copy-sequence for this purpose. A wrong-type-argument error is signaled if either start or end are non-integers. An args-out-of-range error is signaled if start indicates a character following end, or if either integer is out of range for string. Contrast this function with buffer-substring ( see Buffer Contents), which returns a string containing a portion of the text in the current buffer. The beginning of a string is at index 0, but the beginning of a buffer is at index 1. -- Function: concat &rest sequences This function returns a new string consisting of the characters in the arguments passed to it. The arguments may be strings, lists of numbers, or vectors of numbers; they are not themselves changed. If no arguments are passed to concat, it returns an empty string. (concat "abc" "-def") => "abc-def" (concat "abc" (list 120 (+ 256 121)) [122]) => "abcxyz" (concat "The " "quick brown " "fox.") => "The quick brown fox." (concat) => "" The second example above shows how characters stored in strings are taken modulo 256. In other words, each character in the string is stored in one byte. The concat function always constructs a new string that is not eq to any existing string. When an argument is an integer (not a sequence of integers), it is converted to a string of digits making up the decimal printed representation of the integer. This special case exists for compatibility with Mocklisp, and we don't recommend you take advantage of it. If you want to convert an integer in this way, use format (see Formatting Strings) or int-to-string ( see String Conversion). (concat 137) => "137" (concat 54 321) => "54321" For information about other concatenation functions, see mapconcat in Mapping Functions, vconcat in Vectors, and append in Building Lists. ΓòÉΓòÉΓòÉ 7.4. Comparison of Characters and Strings ΓòÉΓòÉΓòÉ -- Function: char-equal character1 character2 This function returns t if the arguments represent the same character, nil otherwise. This is done by comparing two integers modulo 256. (char-equal ?x ?x) => t (char-to-string (+ 256 ?x)) => "x" (char-equal ?x (+ 256 ?x)) => t -- Function: string= string1 string2 This function returns t if the characters of the two strings match exactly; case is significant. (string= "abc" "abc") => t (string= "abc" "ABC") => nil (string= "ab" "ABC") => nil -- Function: string-equal string1 string2 string-equal is another name for string=. -- Function: string< string1 string2 This function compares two strings a character at a time. First it scans both the strings at once to find the first pair of corresponding characters that do not match. If the lesser character of those two is the character from string1, then string1 is less, and this function returns t. If the lesser character is the one from string2, then string1 is greater, and this function returns nil. If the two strings match entirely, the value is nil. Pairs of characters are compared by their ASCII codes. Keep in mind that lower case letters have higher numeric values in the ASCII character set than their upper case counterparts; numbers and many punctuation characters have a lower numeric value than upper case letters. (string< "abc" "abd") => t (string< "abd" "abc") => nil (string< "123" "abc") => t When the strings have different lengths, and they match up to the length of string1, then the result is t. If they match up to the length of string2, the result is nil. A string without any characters in it is the smallest possible string. (string< "" "abc") => t (string< "ab" "abc") => t (string< "abc" "") => nil (string< "abc" "ab") => nil (string< "" "") => nil -- Function: string-lessp string1 string2 string-lessp is another name for string<. ΓòÉΓòÉΓòÉ 7.5. Conversion of Characters and Strings ΓòÉΓòÉΓòÉ Characters and strings may be converted into each other and into integers. format and prin1-to-string (see Output Functions) may also be used to convert Lisp objects into strings. read-from-string (see Input Functions) may be used to ``convert'' a string representation of a Lisp object into an object. See Documentation, for a description of the functions single-key-description and text-char-description, which return a string representing the Emacs standard notation of the argument character. These functions are used primarily for printing help messages. -- Function: char-to-string character This function returns a new string with a length of one character. The value of character, modulo 256, is used to initialize the element of the string. This function is similar to make-string with an integer argument of 1. (See Creating Strings.) This conversion can also be done with format using the `%c' format specification. (See Formatting Strings.) (char-to-string ?x) => "x" (char-to-string (+ 256 ?x)) => "x" (make-string 1 ?x) => "x" -- Function: string-to-char string This function returns the first character in string. If the string is empty, the function returns 0. The value is also 0 when the first character of string is the null character, ASCII code 0. (string-to-char "ABC") => 65 (string-to-char "xyz") => 120 (string-to-char "") => 0 (string-to-char "\000") => 0 This function may be eliminated in version 19 if it does not seem useful enough to retain. -- Function: int-to-string integer This function returns a string consisting of the digits of integer, base ten. When passed a positive integer as an argument, this function returns an unsigned string. When passed a negative integer, the function returns a string with a leading minus sign. (int-to-string 256) => "256" (int-to-string -23) => "-23" See also the function format in Formatting Strings. -- Function: string-to-int string This function returns the integer value of the characters in string, read as a number in base ten. The string is read starting from (and including) the first character, and it is read until a non-digit is encountered. If the first character is not a digit or a minus sign, this function returns 0. (string-to-int "256") => 256 (string-to-int "25 is a perfect square.") => 25 (string-to-int "X256") => 0 (string-to-int "-4") => -4 ΓòÉΓòÉΓòÉ 7.6. Formatting Strings ΓòÉΓòÉΓòÉ Formatting means constructing a string by substitution of computed values at various places in a constant string. This string controls how the other values are printed as well as where they appear; it is called a format string. Formatting is often useful for computing messages to be displayed. In fact, the functions message and error provide the same formatting feature described here; they differ from format only in how they use the result of formatting. -- Function: format string &rest objects This function returns a new string that is made by copying string and then replacing any format specification in the copy with encodings of the corresponding objects. The arguments objects are the computed values to be formatted. A format specification is a sequence of characters beginning with a `%'. Thus, if there is a `%d' in string, the format function replaces it with the printed representation of one of the values to be formatted (one of the arguments objects). For example: (format "The value of fill-column is %d." fill-column) => "The value of fill-column is 72." If string contains more than one format specification, the format specifications are matched in order with successive values from objects. Thus, the first format specification in string is matched with the first such value, the second format specification is matched with the second such value, and so on. Any extra format specifications (those for which there are no corresponding values) cause unpredictable behavior. Any extra values to be formatted will be ignored. Certain format specifications require values of particular types. However, no error is signaled if the value actually supplied fails to have the expected type. Instead, meaningless text is likely to be output. Here is a table of the characters that can follow `%' to make up a format specification: `s' Replace the specification with the printed representation of the object. If there is no corresponding object, the empty string is used. `o' Replace the specification with the base-eight representation of an integer. `d' Replace the specification with the base-ten representation of an integer. `x' Replace the specification with the base-sixteen representation of an integer. `c' Replace the specification with the character which is the value given. `%' A single `%' is placed in the string. This format specification is unusual in that it does not use a value. For example, (format "%% %d" 30) returns "% 30". Any other format character results in an `Invalid format operation' error. Here are several examples: (format "The name of this buffer is %s." (buffer-name)) => "The name of this buffer is strings.texi." (format "The buffer object prints as %s." (current-buffer)) => "The buffer object prints as #<buffer strings.texi>." (format "The octal value of 18 is %o, and the hex value is %x." 18 18) => "The octal value of 18 is 22, and the hex value is 12." All the specification characters allow an optional numeric prefix between the `%' and the character. The optional numeric prefix defines the minimum width for the object. If the printed representation of the object contains fewer characters than this, then it is padded. The padding is on the left if the prefix is positive (or starts with zero) and on the right if the prefix is negative. The padding character is normally a space, but if the numeric prefix starts with a zero, zeros are used for padding. (format "%06d will be padded on the left with zeros" 123) => "000123 will be padded on the left with zeros" (format "%-6d will be padded on the right" 123) => "123 will be padded on the right" No matter what the prefix, nothing in the printed representation will be truncated. This allows the programmer to specify minimum spacing without knowing how many characters there are in the object's printed representation. In the following three examples, `%7s' specifies a minimum width of 7. In the first case, the string inserted in place of `%7s' has only 3 letters, so 4 blank spaces are inserted for padding. In the second case, the string "specification" is 13 letters wide but is not truncated. In the third case, the padding is on the right. (This does not work in version 18, but does work in version 19.) (format "The word `%7s' actually has %d letters in it." "foo" (length "foo")) => "The word ` foo' actually has 3 letters in it." (format "The word `%7s' actually has %d letters in it." "specification" (length "specification")) => "The word `specification' actually has 13 letters in it." (format "The word `%-7s' actually has %d letters in it." "foo" (length "foo")) => "The word `foo ' actually has 3 letters in it." ;; `%-7s' fails to work in version 18, but does work in version 19. ;; In version 18, padding is not inserted. ΓòÉΓòÉΓòÉ 7.7. Character Case ΓòÉΓòÉΓòÉ The character case functions change the case of single characters or of the contents of strings. The functions convert only alphabetic characters (the letters `A' through `Z' and `a' through `z'); other characters are not altered. The functions do not modify the strings that are passed to them as arguments. The examples below use the characters `X' and `x' which have ASCII codes 88 and 120 respectively. -- Function: downcase string-or-char This function converts a character or a string to lower case. When the argument to downcase is a string, the function creates and returns a new string in which each letter in the argument that is upper case is converted to lower case. When the argument to downcase is a character, downcase returns the corresponding lower case character. This value is an integer. If the original character is lower case, or is not a letter, then the value equals the original character. (downcase "The cat in the hat") => "the cat in the hat" (downcase ?X) => 120 -- Function: upcase string-or-char This function converts a character or a string to upper case. When the argument to upcase is a string, the function creates and returns a new string in which each letter in the argument that is lower case is converted to upper case. When the argument to upcase is a character, upcase returns the corresponding upper case character. This value is an integer. If the original character is upper case, or is not a letter, then the value equals the original character. (upcase "The cat in the hat") => "THE CAT IN THE HAT" (upcase ?x) => 88 -- Function: capitalize string-or-char This function capitalizes strings or characters. If string-or-char is a string, the function creates and returns a new string, whose contents are a copy of string-or-char in which each word has been capitalized. This means that the first character of each word is converted to upper case, and the rest are converted to lower case. The definition of a word is any sequence of consecutive characters that are assigned to the word constituent category in the current syntax table (See Syntax Class Table). When the argument to capitalize is a character, capitalize has the same result as upcase. (capitalize "The cat in the hat") => "The Cat In The Hat" (capitalize "THE 77TH-HATTED CAT") => "The 77th-Hatted Cat" (capitalize ?x) => 88 ΓòÉΓòÉΓòÉ 8. Lists ΓòÉΓòÉΓòÉ A list represents a sequence of zero or more elements (which may be any Lisp objects). The important difference between lists and vectors is that two or more lists can share part of their structure; in addition, you can insert or delete elements in a list without copying the whole list. ΓòÉΓòÉΓòÉ 8.1. Lists and Cons Cells ΓòÉΓòÉΓòÉ Lists in Lisp are not a primitive data type; they are built up from cons cells. A cons cell is a data object which represents an ordered pair. It records two Lisp objects, one labeled as the car, and the other labeled as the cdr. (These names are traditional.) A list is made by chaining cons cells together, one cons cell per element. By convention, the cars of the cons cells are the elements of the list, and the cdrs are used to chain the list: the cdr of each cons cell is the following cons cell. The cdr of the last cons cell is nil. This asymmetry between the car and the cdr is entirely a matter of convention; at the level of cons cells, the car and cdr slots have the same characteristics. The symbol nil is considered a list as well as a symbol; it is the list with no elements. Therefore, the cdr of any nonempty list l is a list containing all the elements of l except the first. For convenience, the symbol nil is considered to have nil as its cdr (and also as its car). ΓòÉΓòÉΓòÉ 8.2. Lists as Linked Pairs of Boxes ΓòÉΓòÉΓòÉ A cons cell can be illustrated as a pair of boxes. The first box represents the car and the second box represents the cdr. Here is an illustration of the two-element list, (tulip lily), made from two cons cells: --------------- --------------- |car |cdr | |car |cdr | | tulip | o---------->| lily | nil | | | | | | | --------------- --------------- Each pair of boxes represents a cons cell. Each box ``refers to'', ``points to'' or ``contains'' a Lisp object. (These terms are synonymous.) The first box, which is the car of the first cons cell, contains the symbol tulip. The arrow from the cdr of the first cons cell to the second cons cell indicates that the cdr of the first cons cell points to the second cons cell. The same list can be illustrated in a different sort of box notation like this: ___ ___ ___ ___ |___|___|--> |___|___|--> nil | | | | --> tulip --> lily Here is a more complex illustration, this time of the three-element list, ((pine needles) oak maple), the first element of which is a two-element list: ___ ___ ___ ___ ___ ___ |___|___|--> |___|___|--> |___|___|--> nil | | | | | | | --> oak --> maple | | ___ ___ ___ ___ --> |___|___|--> |___|___|--> nil | | | | --> pine --> needles The same list is represented in the first box notation like this: --------------- --------------- --------------- |car |cdr | |car |cdr | |car |cdr | | o | o---------->| oak | o---------->| maple | nil | | | | | | | | | | | -- | ---------- --------------- --------------- | | | --------------- ----------------- | |car |cdr | |car |cdr | ------>| pine | o---------->| needles | nil | | | | | | | --------------- ----------------- See List Type, for the read and print syntax of lists, and for more ``box and arrow'' illustrations of lists. ΓòÉΓòÉΓòÉ 8.3. Predicates on Lists ΓòÉΓòÉΓòÉ The following predicates test whether a Lisp object is an atom, is a cons cell or is a list, or whether it is the distinguished object nil. (Many of these tests can be defined in terms of the others, but they are used so often that it is worth having all of them.) -- Function: consp object This function returns t if object is a cons cell, nil otherwise. nil is not a cons cell, although it is a list. -- Function: atom object This function returns t if object is an atom, nil otherwise. All objects except cons cells are atoms. The symbol nil is an atom and is also a list; it is the only Lisp object which is both. (atom object) == (not (consp object)) -- Function: listp object This function returns t if object is a cons cell or nil. Otherwise, it returns nil. (listp '(1)) => t (listp '()) => t -- Function: nlistp object This function is the opposite of listp: it returns t if object is not a list. Otherwise, it returns nil. (listp object) == (not (nlistp object)) -- Function: null object This function returns t if object is nil, and returns nil otherwise. This function is identical to not, but as a matter of clarity we use null when object is considered a list and not when it is considered a truth value (see not in Combining Conditions). (null '(1)) => nil (null '()) => t ΓòÉΓòÉΓòÉ 8.4. Accessing Elements of Lists ΓòÉΓòÉΓòÉ -- Function: car cons-cell This function returns the value pointed to by the first pointer of the cons cell cons-cell. Expressed another way, this function returns the car of cons-cell. As a special case, if cons-cell is nil, then car is defined to return nil; therefore, any list is a valid argument for car. An error is signaled if the argument is not a cons cell or nil. (car '(a b c)) => a (car '()) => nil -- Function: cdr cons-cell This function returns the value pointed to by the second pointer of the cons cell cons-cell. Expressed another way, this function returns the cdr of cons-cell. As a special case, if cons-cell is nil, then cdr is defined to return nil; therefore, any list is a valid argument for cdr. An error is signaled if the argument is not a cons cell or nil. (cdr '(a b c)) => (b c) (cdr '()) => nil -- Function: car-safe object This function lets you take the car of a cons cell while avoiding errors for other data types. It returns the car of object if object is a cons cell, nil otherwise. This is in contrast to car, which signals an error if object is not a list. (car-safe object) == (let ((x object)) (if (consp x) (car x) nil)) -- Function: cdr-safe object This function lets you take the cdr of a cons cell while avoiding errors for other data types. It returns the cdr of object if object is a cons cell, nil otherwise. This is in contrast to cdr, which signals an error if object is not a list. (cdr-safe object) == (let ((x object)) (if (consp x) (cdr x) nil)) -- Function: nth n list This function returns the nth element of list. Elements are numbered starting with zero, so the car of list is element number zero. If the length of list is n or less, the value is nil. If n is less than zero, then the first element is returned. (nth 2 '(1 2 3 4)) => 3 (nth 10 '(1 2 3 4)) => nil (nth -3 '(1 2 3 4)) => 1 (nth n x) == (car (nthcdr n x)) -- Function: nthcdr n list This function returns the nth cdr of list. In other words, it removes the first n links of list and returns what follows. If n is less than or equal to zero, then all of list is returned. If the length of list is n or less, the value is nil. (nthcdr 1 '(1 2 3 4)) => (2 3 4) (nthcdr 10 '(1 2 3 4)) => nil (nthcdr -3 '(1 2 3 4)) => (1 2 3 4) ΓòÉΓòÉΓòÉ 8.5. Building Cons Cells and Lists ΓòÉΓòÉΓòÉ Many functions build lists, as lists reside at the very heart of Lisp. cons is the fundamental list-building function; however, it is interesting to note that list is used more times in the source code for Emacs than cons. -- Function: cons object1 object2 This function is the fundamental function used to build new list structure. It creates a new cons cell, making object1 the car, and object2 the cdr. It then returns the new cons cell. The arguments object1 and object2 may be any Lisp objects, but most often object2 is a list. (cons 1 '(2)) => (1 2) (cons 1 '()) => (1) (cons 1 2) => (1 . 2) cons is often used to add a single element to the front of a list. This is called consing the element onto the list. For example: (setq list (cons newelt list)) -- Function: list &rest objects This function creates a list with objects as its elements. The resulting list is always nil-terminated. If no objects are given, the empty list is returned. (list 1 2 3 4 5) => (1 2 3 4 5) (list 1 2 '(3 4 5) 'foo) => (1 2 (3 4 5) foo) (list) => nil -- Function: make-list length object This function creates a list of length length, in which all the elements have the identical value object. Compare make-list with make-string (see Creating Strings). (make-list 3 'pigs) => (pigs pigs pigs) (make-list 0 'pigs) => nil -- Function: append &rest sequences This function returns a list containing all the elements of sequences. The sequences may be lists, vectors, strings, or integers. All arguments except the last one are copied, so none of them are altered. The final argument to append may be any object but it is typically a list. The final argument is not copied or converted; it becomes part of the structure of the new list. Here is an example: (setq trees '(pine oak)) => (pine oak) (setq more-trees (append '(maple birch) trees)) => (maple birch pine oak) trees => (pine oak) more-trees => (maple birch pine oak) (eq trees (cdr (cdr more-trees))) => t You can see what happens by looking at a box diagram. The variable trees is set to the list (pine oak) and then the variable more-trees is set to the list (maple birch pine oak). However, the variable trees continues to refer to the original list: more-trees trees | | | ___ ___ ___ ___ -> ___ ___ ___ ___ --> |___|___|--> |___|___|--> |___|___|--> |___|___|--> nil | | | | | | | | --> maple -->birch --> pine --> oak An empty sequence contributes nothing to the value returned by append. As a consequence of this, a final nil argument forces a copy of the previous argument. trees => (pine oak) (setq wood (append trees ())) => (pine oak) wood => (pine oak) (eq wood trees) => nil This once was the standard way to copy a list, before the function copy-sequence was invented. See Sequences Arrays Vectors. With the help of apply, we can append all the lists in a list of lists: (apply 'append '((a b c) nil (x y z) nil)) => (a b c x y z) If no sequences are given, nil is returned: (append) => nil In the special case where one of the sequences is an integer (not a sequence of integers), it is first converted to a string of digits making up the decimal print representation of the integer. This special case exists for compatibility with Mocklisp, and we don't recommend you take advantage of it. If you want to convert an integer in this way, use format (see Formatting Strings) or int-to-string (see String Conversion). (setq trees '(pine oak)) => (pine oak) (char-to-string ?\054) => "6" (setq longer-list (append trees 6 '(spruce))) => (pine oak 54 spruce) (setq x-list (append trees 6 6)) => (pine oak 54 . 6) See nconc in Rearrangement, for another way to join lists without copying. -- Function: reverse list This function creates a new list whose elements are the elements of list, but in reverse order. The original argument list is not altered. (setq x '(1 2 3 4)) => (1 2 3 4) (reverse x) => (4 3 2 1) x => (1 2 3 4) ΓòÉΓòÉΓòÉ 8.6. Modifying Existing List Structure ΓòÉΓòÉΓòÉ You can modify the car and cdr contents of a cons cell with the primitives setcar and setcdr. Common Lisp note: Common Lisp uses functions rplaca and rplacd to alter list structure; they change structure the same way as setcar and setcdr, but the Common Lisp functions return the cons cell while setcar and setcdr return the new car or cdr. ΓòÉΓòÉΓòÉ 8.6.1. Altering List Elements with setcar ΓòÉΓòÉΓòÉ Changing the car of a cons cell is done with setcar and replaces one element of a list with a different element. -- Function: setcar cons object This function stores object as the new car of cons, replacing its previous car. It returns the value object. For example: (setq x '(1 2)) => (1 2) (setcar x '4) => 4 x => (4 2) When a cons cell is part of the shared structure of several lists, storing a new car into the cons changes one element of each of these lists. Here is an example: ;; Create two lists that are partly shared. (setq x1 '(a b c)) => (a b c) (setq x2 (cons 'z (cdr x1))) => (z b c) ;; Replace the car of a shared link. (setcar (cdr x1) 'foo) => foo x1 ; Both lists are changed. => (a foo c) x2 => (z foo c) ;; Replace the car of a link that is not shared. (setcar x1 'baz) => baz x1 ; Only one list is changed. => (baz foo c) x2 => (z foo c) Here is a graphical depiction of the shared structure of the two lists x1 and x2, showing why replacing b changes them both: ___ ___ ___ ___ ___ ___ x1---> |___|___|----> |___|___|--> |___|___|--> nil | --> | | | | | | --> a | --> b --> c | ___ ___ | x2--> |___|___|-- | | --> z Here is an alternative form of box diagram, showing the same relationship: x1: --------------- --------------- --------------- |car |cdr | |car |cdr | |car |cdr | | a | o---------->| b | o---------->| c | nil | | | | -->| | | | | | --------------- | --------------- --------------- | x2: | --------------- | |car |cdr | | | z | o------- | | | --------------- ΓòÉΓòÉΓòÉ 8.6.2. Altering the CDR of a List ΓòÉΓòÉΓòÉ The lowest-level primitive for modifying a cdr is setcdr: -- Function: setcdr cons object This function stores object into the cdr of cons. The value returned is object, not cons. Here is an example of replacing the cdr of a list with a different list. All but the first element of the list are removed in favor of a different sequence of elements. The first element is unchanged, because it resides in the car of the list, and is not reached via the cdr. (setq x '(1 2 3)) => (1 2 3) (setcdr x '(4)) => (4) x => (1 4) You can delete elements from the middle of a list by altering the cdrs of the cons cells in the list. For example, here we delete the second element, b, from the list (a b c), by changing the cdr of the first cell: (setq x1 '(a b c)) => (a b c) (setcdr x1 '(c)) => (c) x1 => (a c) Here is the result in box notation: ----------------------- | | --------------- | --------------- | --------------- |car |cdr | | |car |cdr | -->|car |cdr | | a | o------- | b | o---------->| c | nil | | | | | | | | | | --------------- --------------- --------------- The second cons cell, which previously held the element b, still exists and its car is still b, but it no longer forms part of this list. It is equally easy to insert a new element by changing cdrs: (setq x1 '(a b c)) => (a b c) (setcdr x1 (cons 'd (cdr x1))) => (d b c) x1 => (a d b c) Here is this result in box notation: --------------- --------------- --------------- |car |cdr | |car |cdr | |car |cdr | | a | o | -->| b | o---------->| c | nil | | | | | | | | | | | | ---------- | -- | --------------- --------------- | | ------ ------- | | | --------------- | | |car |cdr | | -->| d | o------ | | | --------------- ΓòÉΓòÉΓòÉ 8.6.3. Functions that Rearrange Lists ΓòÉΓòÉΓòÉ Here are some functions that rearrange lists ``destructively'' by modifying the cdrs of their component cons cells. We call these functions ``destructive'' because the original lists passed as arguments to them are chewed up to produce a new list that is subsequently returned. -- Function: nconc &rest lists This function returns a list containing all the elements of lists. Unlike append (see Building Lists), the lists are not copied. Instead, the last cdr of each of the lists is changed to refer to the following list. The last of the lists is not altered. For example: (setq x '(1 2 3)) => (1 2 3) (nconc x '(4 5)) => (1 2 3 4 5) x => (1 2 3 4 5) Since the last argument of nconc is not itself modified, it is reasonable to use a constant list, such as '(4 5), as is done in the above example. For the same reason, the last argument need not be a list: (setq x '(1 2 3)) => (1 2 3) (nconc x 'z) => (1 2 3 . z) x => (1 2 3 . z) A common pitfall is to use a quoted constant list as a non-last argument to nconc. If you do this, your program will change each time you run it! Here is what happens: (defun add-foo (x) ; This function should add (nconc '(foo) x)) ; foo to the front of its arg. (symbol-function 'add-foo) => (lambda (x) (nconc (quote (foo)) x)) (setq xx (add-foo '(1 2))) ; It seems to work. => (foo 1 2) (setq xy (add-foo '(3 4))) ; What happened? => (foo 1 2 3 4) (eq xx xy) => t (symbol-function 'add-foo) => (lambda (x) (nconc (quote (foo 1 2 3 4) x))) -- Function: nreverse list This function reverses the order of the elements of list. Unlike reverse, nreverse alters its argument destructively by reversing the cdrs in the cons cells forming the list. The cons cell which used to be the last one in list becomes the first cell of the value. For example: (setq x '(1 2 3 4)) => (1 2 3 4) x => (1 2 3 4) (nreverse x) => (4 3 2 1) ;; The cell that was first is now last. x => (1) To avoid confusion, we usually store the result of nreverse back in the same variable which held the original list: (setq x (nreverse x)) Here is the nreverse of our favorite example, (a b c), presented graphically: Original list head: Reversed list: --------------- --------------- --------------- |car |cdr | |car |cdr | |car |cdr | | a | nil |<-- | b | o |<-- | c | o | | | | | | | | | | | | | | --------------- | ---------- | -- | ---------- | -- | | | | --------------- --------------- -- Function: sort list predicate This function sorts list stably, though destructively, and returns the sorted list. It compares elements using predicate. A stable sort is one in which elements with equal sort keys maintain their relative order before and after the sort. Stability is important when successive sorts are used to order elements according to different criteria. The argument predicate must be a function that accepts two arguments. It is called with two elements of list. To get an increasing order sort, the predicate should return t if the first element is ``less than'' the second, or nil if not. The destructive aspect of sort is that it rearranges the cons cells forming list by changing cdrs. A nondestructive sort function would create new cons cells to store the elements in their sorted order. If you wish to sort a list without destroying the original, copy it first with copy-sequence. The cars of the cons cells are not changed; the cons cell that originally contained the element a in list still has a in its car after sorting, but it now appears in a different position in the list due to the change of cdrs. For example: (setq nums '(1 3 2 6 5 4 0)) => (1 3 2 6 5 4 0) (sort nums '<) => (0 1 2 3 4 5 6) nums => (1 2 3 4 5 6) Note that the list in nums no longer contains 0; this is the same cons cell that it was before, but it is no longer the first one in the list. Don't assume a variable that formerly held the argument now holds the entire sorted list! Instead, save the result of sort and use that. Most often we store the result back into the variable that held the original list: (setq nums (sort nums '<)) See Sorting, for more functions that perform sorting. See documentation in Accessing Documentation, for a useful example of sort. See delq, in Sets And Lists, for another function that modifies cons cells. ΓòÉΓòÉΓòÉ 8.7. Using Lists as Sets ΓòÉΓòÉΓòÉ A list can represent an unordered mathematical set---simply consider a value an element of a set if it appears in the list, and ignore the order of the list. To form the union of two sets, use append (as long as you don't mind having duplicate elements). Two other useful functions for sets are memq and delq. Common Lisp note: Common Lisp has functions union (which avoids duplicate elements) and intersection for set operations, but GNU Emacs Lisp does not have them. You can write them in Lisp if you wish. -- Function: memq object list This function tests to see whether object is a member of list. If it is, memq returns a list starting with the first occurrence of object. Otherwise, it returns nil. The letter `q' in memq says that it uses eq to compare object against the elements of the list. For example: (memq 2 '(1 2 3 2 1)) => (2 3 2 1) (memq '(2) '((1) (2))) ; (2) and (2) are not eq. => nil -- Function: delq object list This function removes all elements eq to object from list. Elements at the front of the list are removed (when necessary) simply by advancing down the list and returning a sublist that starts after those elements: (delq 'a '(a b c)) == (cdr '(a b c)) When an element to be deleted appears in the middle of the list, removing it involves changing the cdrs (see Setcdr). (setq sample-list '(1 2 3 (4))) => (1 2 3 (4)) (delq 1 sample-list) => (2 3 (4)) sample-list => (1 2 3 (4)) (delq 2 sample-list) => (1 3 (4)) sample-list => (1 3 (4)) Note that (delq 2 sample-list) removes the second element of sample-list, but (delq 1 sample-list) does not remove the first element---it just returns a shorter list. Don't assume that a variable which formerly held the argument list now has fewer elements, or that it still holds the original list! Instead, save the result of delq and use that. Most often we store the result back into the variable that held the original list: (setq flowers (delq 'rose flowers)) In the following example, the (4) that delq attempts to match and the (4) in the sample-list are not eq: (delq '(4) sample-list) => (1 3 (4)) ΓòÉΓòÉΓòÉ 8.8. Association Lists ΓòÉΓòÉΓòÉ An association list, or alist for short, records a mapping from keys to values. It is a list of cons cells called associations: the car of each cell is the key, and the cdr is the associated value. (This usage of ``key'' is not related to the term ``key sequence''; it means any object which can be looked up in a table.) Here is an example of an alist. The key pine is associated with the value cones; the key oak is associated with acorns; and the key maple is associated with seeds. '((pine . cones) (oak . acorns) (maple . seeds)) The associated values in an alist may be any Lisp objects; so may the keys. For example, in the following alist, the symbol a is associated with the number 1, and the string "b" is associated with the list (2 3), which is the cdr of the alist element: ((a . 1) ("b" 2 3)) Sometimes it is better to design an alist to store the associated value in the car of the cdr of the element. Here is an example: '((rose red) (lily white) (buttercup yellow))) Here we regard red as the value associated with rose. One advantage of this method is that you can store other related information---even a list of other items---in the cdr of the cdr. One disadvantage is that you cannot use rassq (see below) to find the element containing a given value. When neither of these considerations is important, the choice is a matter of taste, as long as you are consistent about it for any given alist. Note that the same alist shown above could be regarded as having the associated value in the cdr of the element; the the value associated with rose would be the list (red). Association lists are often used to record information that you might otherwise keep on a stack, since new associations may be added easily to the front of the list. When searching an association list for an association with a given key, the first one found is returned, if there is more than one. In Emacs Lisp, it is not an error if an element of an association list is not a cons cell. The alist search functions simply ignore such elements. Many other versions of Lisp signal errors in such cases. Note that property lists are similar to association lists in several respects. A property list behaves like an association list in which each key can occur only once. See Property Lists, for a comparison of property lists and association lists. -- Function: assoc key alist This function returns the first association for key in alist. It compares key against the alist elements using equal (see Equality Predicates). It returns nil if no association in alist has a car equal to key. For example: (setq trees '((pine . cones) (oak . acorns) (maple . seeds))) => ((pine . cones) (oak . acorns) (maple . seeds)) (assoc 'oak trees) => (oak . acorns) (cdr (assoc 'oak trees)) => acorns (assoc 'birch trees) => nil Here is another example in which the keys and values are not symbols: (setq needles-per-cluster '((2 . ("Austrian Pine" "Red Pine")) (3 . "Pitch Pine") (5 . "White Pine"))) (cdr (assoc 3 needles-per-cluster)) => "Pitch Pine" (cdr (assoc 2 needles-per-cluster)) => ("Austrian Pine" "Red Pine") -- Function: assq key alist This function is like assoc in that it returns the first association for key in alist, but it makes the comparison using eq instead of equal. assq returns nil if no association in alist has a car eq to key. This function is used more often than assoc, since eq is faster than equal and most alists use symbols as keys. See Equality Predicates. (setq trees '((pine . cones) (oak . acorns) (maple . seeds))) (assq 'pine trees) => (pine . cones) On the other hand, assq is not usually useful in alists where the keys may not be symbols: (setq leaves '(("simple leaves" . oak) ("compound leaves" . horsechestnut))) (assq "simple leaves" leaves) => nil (assoc "simple leaves" leaves) => ("simple leaves" . oak) -- Function: rassq alist value This function returns the first association with value value in alist. It returns nil if no association in alist has a cdr eq to value. rassq is like assq except that the cdr of the alist associations is tested instead of the car. You can think of this as ``reverse assq'', finding the key for a given value. For example: (setq trees '((pine . cones) (oak . acorns) (maple . seeds))) (rassq 'acorns trees) => (oak . acorns) (rassq 'spores trees) => nil Note that rassq cannot be used to search for a value stored in the car of the cdr of an element: (setq colors '((rose red) (lily white) (buttercup yellow))) (rassq 'white colors) => nil In this case, the cdr of the association (lily white) is not the symbol white, but rather the list (white). This can be seen more clearly if the association is written in dotted pair notation: (lily white) == (lily . (white)) -- Function: copy-alist alist This function returns a two-level deep copy of alist: it creates a new copy of each association, so that you can alter the associations of the new alist without changing the old one. (setq needles-per-cluster '((2 . ("Austrian Pine" "Red Pine")) (3 . "Pitch Pine") (5 . "White Pine"))) => ((2 "Austrian Pine" "Red Pine") (3 . "Pitch Pine") (5 . "White Pine")) (setq copy (copy-alist needles-per-cluster)) => ((2 "Austrian Pine" "Red Pine") (3 . "Pitch Pine") (5 . "White Pine")) (eq needles-per-cluster copy) => nil (equal needles-per-cluster copy) => t (eq (car needles-per-cluster) (car copy)) => nil (cdr (car (cdr needles-per-cluster))) => "Pitch Pine" (eq (cdr (car (cdr needles-per-cluster))) (cdr (car (cdr copy)))) => t ΓòÉΓòÉΓòÉ 9. Sequences, Arrays, and Vectors ΓòÉΓòÉΓòÉ Recall that the sequence type is the union of three other Lisp types: lists, vectors, and strings. In other words, any list is a sequence, any vector is a sequence, and any string is a sequence. The common property that all sequences have is that each is an ordered collection of elements. An array is a single primitive object directly containing all its elements. Therefore, all the elements are accessible in constant time. The length of an existing array cannot be changed. Both strings and vectors are arrays. A list is a sequence of elements, but it is not a single primitive object; it is made of cons cells, one cell per element. Therefore, elements farther from the beginning of the list take longer to access, but it is possible to add elements to the list or remove elements. The elements of vectors and lists may be any Lisp objects. The elements of strings are all characters. The following diagram shows the relationship between these types: ___________________________________ | | | Sequence | | ______ ______________________ | | | | | | | | | List | | Array | | | | | | ________ _______ | | | |______| | | | | | | | | | | String | | Vector| | | | | |________| |_______| | | | |______________________| | |___________________________________| The Relationship between Sequences , Arrays , and Vectors ΓòÉΓòÉΓòÉ 9.1. Sequences ΓòÉΓòÉΓòÉ In Emacs Lisp, a sequence is either a list, a vector or a string. The common property that all sequences have is that each is an ordered collection of elements. This section describes functions that accept any kind of sequence. -- Function: sequencep object Returns t if object is a list, vector, or string, nil otherwise. -- Function: copy-sequence sequence Returns a copy of sequence. The copy is the same type of object as the original sequence, and it has the same elements in the same order. Storing a new element into the copy does not affect the original sequence, and vice versa. However, the elements of the new sequence are not copies; they are identical (eq) to the elements of the original. Therefore, changes made within these elements, as found via the copied sequence, are also visible in the original sequence. See also append in Building Lists, concat in Creating Strings, and vconcat in Vectors, for others ways to copy sequences. (setq bar '(1 2)) => (1 2) (setq x (vector 'foo bar)) => [foo (1 2)] (setq y (copy-sequence x)) => [foo (1 2)] (eq x y) => nil (equal x y) => t (eq (elt x 1) (elt y 1)) => t ;; Replacing an element of one sequence. (aset x 0 'quux) x => [quux (1 2)] y => [foo (1 2)] ;; Modifying the inside of a shared element. (setcar (aref x 1) 69) x => [quux (69 2)] y => [foo (69 2)] -- Function: length sequence Returns the number of elements in sequence. If sequence is a cons cell that is not a list (because the final cdr is not nil), a wrong-type-argument error is signaled. (length '(1 2 3)) => 3 (length nil) => 0 (length "foobar") => 6 (length [1 2 3]) => 3 -- Function: elt sequence index This function returns the element of sequence indexed by index. Legitimate values of index are integers ranging from 0 up to one less than the length of sequence; other values produce an args-out-of-range error. (elt [1 2 3 4] 2) => 3 (elt '(1 2 3 4) 2) => 3 (char-to-string (elt "1234" 2)) => "3" (elt [1 2 3 4] 4) error-->Args out of range: [1 2 3 4], 4 (elt [1 2 3 4] -1) error-->Args out of range: [1 2 3 4], -1 This function duplicates aref (see Array Functions) and nth (see List Elements), except that it works for any kind of sequence. ΓòÉΓòÉΓòÉ 9.2. Arrays ΓòÉΓòÉΓòÉ An array object refers directly to a number of other Lisp objects, called the elements of the array. Any element of an array may be accessed in constant time. In contrast, an element of a list requires access time that is proportional to the position of the element in the list. When you create an array, you must specify how many elements it has. The amount of space allocated depends on the number of elements. Therefore, it is impossible to change the size of an array once it is created. You cannot add or remove elements. However, you can replace an element with a different value. Emacs defines two types of array, both of which are one-dimensional: strings and vectors. A vector is a general array; its elements can be any Lisp objects. A string is a specialized array; its elements must be characters (i.e., integers between 0 and 255). Each type of array has its own read syntax. See String Type, and Vector Type. Both kinds of arrays share these characteristics: o The first element of an array has index zero, the second element has index 1, and so on. This is called zero-origin indexing. For example, an array of four elements has indices 0, 1, 2, and 3. o The elements of an array may be referenced or changed with the functions aref and aset, respectively (see Array Functions). In principle, if you wish to have an array of characters, you could use either a string or a vector. In practice, we always choose strings for such applications, for three reasons: o They occupy one-fourth the space of a vector of the same elements. o Strings are printed in a way that shows the contents more clearly as characters. o Many of the specialized editing and I/O facilities of Emacs accept only strings. For example, you cannot insert a vector of characters into a buffer the way you can insert a string. See Strings and Characters. ΓòÉΓòÉΓòÉ 9.3. Functions that Operate on Arrays ΓòÉΓòÉΓòÉ In this section, we describe the functions that accept both strings and vectors. -- Function: arrayp object This function returns t if object is an array (i.e., either a vector or a string). (arrayp [a]) => t (arrayp "asdf") => t -- Function: aref array index This function returns the indexth element of array. The first element is at index zero. (setq primes [2 3 5 7 11 13]) => [2 3 5 7 11 13] (aref primes 4) => 11 (elt primes 4) => 11 (aref "abcdefg" 1) => 98 ; `b' is ASCII code 98. See also the function elt, in Sequence Functions. -- Function: aset array index object This function sets the indexth element of array to be object. It returns object. (setq w [foo bar baz]) => [foo bar baz] (aset w 0 'fu) => fu w => [fu bar baz] (setq x "asdfasfd") => "asdfasfd" (aset x 3 ?Z) => 90 x => "asdZasfd" If array is a string and object is not a character, a wrong-type-argument error results. -- Function: fillarray array object This function fills the array array with pointers to object, replacing any previous values. It returns array. (setq a [a b c d e f g]) => [a b c d e f g] (fillarray a 0) => [0 0 0 0 0 0 0] a => [0 0 0 0 0 0 0] (setq s "When in the course") => "When in the course" (fillarray s ?-) => "------------------" If array is a string and object is not a character, a wrong-type-argument error results. The general sequence functions copy-sequence and length are often useful for objects known to be arrays. See Sequence Functions. ΓòÉΓòÉΓòÉ 9.4. Vectors ΓòÉΓòÉΓòÉ Arrays in Lisp, like arrays in most languages, are blocks of memory whose elements can be accessed in constant time. A vector is a general-purpose array; its elements can be any Lisp objects. (The other kind of array provided in Emacs Lisp is the string, whose elements must be characters.) The main uses of vectors in Emacs are as syntax tables (vectors of integers) and keymaps (vectors of commands). They are also used internally as part of the representation of a byte-compiled function; if you print such a function, you will see a vector in it. The indices of the elements of a vector are numbered starting with zero in Emacs Lisp. Vectors are printed with square brackets surrounding the elements in their order. Thus, a vector containing the symbols a, b and c is printed as [a b c]. You can write vectors in the same way in Lisp input. A vector, like a string or a number, is considered a constant for evaluation: the result of evaluating it is the same vector. The elements of the vector are not evaluated. See Self-Evaluating Forms. Here are examples of these principles: (setq avector [1 two '(three) "four" [five]]) => [1 two (quote (three)) "four" [five]] (eval avector) => [1 two (quote (three)) "four" [five]] (eq avector (eval avector)) => t Here are some functions that relate to vectors: -- Function: vectorp object This function returns t if object is a vector. (vectorp [a]) => t (vectorp "asdf") => nil -- Function: vector &rest objects This function creates and returns a vector whose elements are the arguments, objects. (vector 'foo 23 [bar baz] "rats") => [foo 23 [bar baz] "rats"] (vector) => [] -- Function: make-vector integer object This function returns a new vector consisting of integer elements, each initialized to object. (setq sleepy (make-vector 9 'Z)) => [Z Z Z Z Z Z Z Z Z] -- Function: vconcat &rest sequences This function returns a new vector containing all the elements of the sequences. The arguments sequences may be lists, vectors, or strings. If no sequences are given, an empty vector is returned. The value is a newly constructed vector that is not eq to any existing vector. (setq a (vconcat '(A B C) '(D E F))) => [A B C D E F] (eq a (vconcat a)) => nil (vconcat) => [] (vconcat [A B C] "aa" '(foo (6 7))) => [A B C 97 97 foo (6 7)] When an argument is an integer (not a sequence of integers), it is converted to a string of digits making up the decimal printed representation of the integer. This special case exists for compatibility with Mocklisp, and we don't recommend you take advantage of it. If you want to convert an integer in this way, use format (see Formatting Strings) or int-to-string ( see String Conversion). For other concatenation functions, see mapconcat in Mapping Functions, concat in Creating Strings, and append in Building Lists. The append function may be used to convert a vector into a list with the same elements (see Building Lists): (setq avector [1 two (quote (three)) "four" [five]]) => [1 two (quote (three)) "four" [five]] (append avector nil) => (1 two (quote (three)) "four" [five]) ΓòÉΓòÉΓòÉ 10. Symbols ΓòÉΓòÉΓòÉ A symbol is an object with a unique name. This chapter describes symbols, their components, and how they are created and interned. Property lists are also described. The uses of symbols as variables and as function names are described in separate chapters; see Variables, and Functions. You may test whether an arbitrary Lisp object is a symbol with symbolp: -- Function: symbolp object This function returns t if object is a symbol, nil otherwise. ΓòÉΓòÉΓòÉ 10.1. Symbol Components ΓòÉΓòÉΓòÉ Each symbol has four components (or ``cells''), each of which references another object: Print name The print name cell holds a string which names the symbol for reading and printing. See symbol-name in Creating Symbols. Value The value cell holds the current value of the symbol as a variable. When a symbol is used as a form, the value of the form is the contents of the symbol's value cell. See symbol-value in Accessing Variables. Function The function cell holds the function definition of the symbol. When a symbol is used as a function, its function definition is used in its place. This cell is also used by the editor command loop to record keymaps and keyboard macros. Because each symbol has separate value and function cells, variables and function names do not conflict. See symbol-function in Function Cells. Property list The property list cell holds the property list of the symbol. See symbol-plist in Property Lists. The print name cell always holds a string, and cannot be changed. The other three cells can be set individually to any specified Lisp object. The print name cell holds the string that is the name of the symbol. Since symbols are represented textually by their names, it is important not to have two symbols with the same name. The Lisp reader ensures this: every time it reads a symbol, it looks for an existing symbol with the specified name before it creates a new one. (In GNU Emacs Lisp, this is done with a hashing algorithm that uses an obarray; see Creating Symbols.) In normal usage, the function cell usually contains a function or macro, as that is what the Lisp interpreter expects to see there (see Evaluation). Keyboard macros (see Keyboard Macros), keymaps (see Keymaps) and autoload objects (see Autoloading) are also sometimes stored in the function cell of symbols. We often refer to ``the function foo'' when we really mean the function stored in the function cell of the symbol foo. The distinction will be made only when necessary. Similarly, the property list cell normally holds a correctly formatted property list (see Property Lists), as a number of functions will expect to see a property list there. The function cell or the value cell may be void, which means that the cell does not reference any object. (This is not the same thing as holding the symbol void, nor the same as holding the symbol nil.) Examining the value of a cell which is void results in an error, such as `Symbol's value as variable is void'. The four functions symbol-name, symbol-value, symbol-plist, and symbol-function return the contents of the four cells. Here as an example we show the contents of the four cells of the symbol buffer-file-name: (symbol-name 'buffer-file-name) => "buffer-file-name" (symbol-value 'buffer-file-name) => "/gnu/elisp/symbols.texi" (symbol-plist 'buffer-file-name) => (variable-documentation 29529) (symbol-function 'buffer-file-name) => #<subr buffer-file-name> Because this symbol is the variable which holds the name of the file being visited in the current buffer, the value cell contents we see are the name of the source file of this chapter of the Emacs Lisp Manual. The property list cell contains the list (variable-documentation 29529) which tells the documentation functions where to find documentation about buffer-file-name in the `DOC' file. (29529 is the offset from the beginning of the `DOC' file where the documentation for the function begins.) The function cell contains the function for returning the name of the file. Since buffer-file-name is a primitive function, its function definition has no read syntax and prints in hash notation ( see Primitive Function Type). A function definition written in Lisp will have a lambda expression (or byte-code) in this cell. ΓòÉΓòÉΓòÉ 10.2. Defining Symbols ΓòÉΓòÉΓòÉ A definition in Lisp is a special form that announces your intention to use a certain symbol in a particular way. In Emacs Lisp, you can define a symbol as a variable, or define it as a function (or macro), or both independently. A definition construct typically specifies a value or meaning for the symbol for one kind of use, plus documentation for its meaning when used in this way. Thus, when you define a symbol as a variable, you can supply an initial value for the variable, plus documentation for the variable. defvar and defconst are definitions that establish a symbol as a global variable. They are documented in detail in Defining Variables. defun defines a symbol as a function, creating a lambda expression and storing it in the function cell of the symbol. This lambda expression thus becomes the function definition of the symbol. (The term ``function definition'', meaning the contents of the function cell, is derived from the idea that defun gives the symbol its definition as a function.) See Functions. defmacro defines a symbol as a macro. It creates a macro object and stores it in the function cell of the symbol. Note that a given symbol can be a macro or a function, but not both at once, because both macro and function definitions are kept in the function cell, and that cell can hold only one Lisp object at any given time. See Macros. In GNU Emacs Lisp, a definition is not required in order to use a symbol as a variable or function. Thus, you can make a symbol a global variable with setq, whether you define it first or not. The real purpose of definitions is to guide programmers and programming tools. They inform programmers who read the code that certain symbols are intended to be used as variables, or as functions. In addition, utilities such as `etags' and `make-docfile' can recognize definitions, and add the appropriate information to tag tables and the `emacs/etc/DOC-version' file. See Accessing Documentation. ΓòÉΓòÉΓòÉ 10.3. Creating and Interning Symbols ΓòÉΓòÉΓòÉ To understand how symbols are created in GNU Emacs Lisp, it is necessary to know how Lisp reads them. It is essential to ensure that every time Lisp reads the same set of characters, it finds the same symbol. Failure to do so would be disastrous. When the Lisp reader encounters a symbol, it reads all the characters of the name. Then it ``hashes'' those characters to find an index in a table called an obarray. Hashing is an efficient method of looking something up. For example, instead of searching a telephone book cover to cover when looking up Jan Jones, you start with the J's and go from there. That is a simple version of hashing. Each element of the obarray is a bucket which holds all the symbols with a given hash code; to look for a given name, it is sufficient to look through all the symbols in the bucket for that name's hash code. If a symbol with the desired name is found, then it is used. If no such symbol is found, then a new symbol is created and added to the obarray bucket. Adding a symbol to an obarray is called interning it, and the symbol is then called an interned symbol. In Emacs Lisp, a symbol may be interned in only one obarray. Common Lisp note: in Common Lisp, a symbol may be interned in several obarrays at once. If a symbol is not in the obarray, then there is no way for Lisp to find it when its name is read. Such a symbol is called an uninterned symbol relative to the obarray. An uninterned symbol has all the other characteristics of symbols. It is possible, though uncommon, for two different symbols to have the same name in different obarrays; they are not eq or equal. In Emacs Lisp, an obarray is represented as a vector. Each element of the vector is a bucket; its value is either an interned symbol whose name hashes to that bucket, or 0 if the bucket is empty. Each interned symbol has an internal link (invisible to the user) to the next symbol in the bucket. Because these links are invisible, there is no way to scan the symbols in an obarray except using mapatoms (below). The order of symbols in a bucket is not significant. In an empty obarray, every element is 0, and you can create an obarray with (make-vector length 0). Prime numbers as lengths tend to result in good hashing; lengths one less than a power of two are also good. Most of the functions below take a name and sometimes an obarray as arguments. A wrong-type-argument error is signaled if the name is not a string, or if the obarray is not a vector. -- Function: symbol-name symbol This function returns the string that is symbol's name. For example: (symbol-name 'foo) => "foo" Changing the string by substituting characters, etc, will change the name of the symbol, but will fail to update the obarray, so don't do it! -- Function: make-symbol name This function returns a newly-allocated uninterned symbol whose name is name (which must be a string). Its value and function definition are void, and its property list is nil. In the example below, the value of sym is not eq to foo because it is a distinct uninterned symbol whose name is also `foo'. (setq sym (make-symbol "foo")) => foo (eq sym 'foo) => nil -- Function: intern name &optional obarray This function returns the interned symbol whose name is name. If there is no such symbol in the obarray, a new one is created, added to the obarray, and returned. If obarray is supplied, it specifies the obarray to use; otherwise, the value of the global variable obarray is used. (setq sym (intern "foo")) => foo (eq sym 'foo) => t -- Function: intern-soft name &optional obarray This function returns the symbol whose name is name, or nil if a symbol with that name is not found in the obarray. Therefore, you can use intern-soft to test whether a symbol with a given name is interned. If obarray is supplied, it specifies the obarray to use; otherwise the value of the global variable obarray is used. (intern-soft "frazzle") ; No such symbol exists. => nil (make-symbol "frazzle") ; Create an uninterned one. => frazzle (intern-soft "frazzle") ; That one cannot be found. => nil (setq sym (intern "frazzle")) ; Create an interned one. => frazzle (intern-soft "frazzle") ; That one can be found! => frazzle (eq sym 'frazzle) ; And it is the same one. => t -- Variable: obarray This variable is the standard obarray for use by intern and read. -- Function: mapatoms function &optional obarray This function applies function to every symbol in obarray. It returns nil. If obarray is not supplied, it defaults to the value of obarray, the standard obarray for ordinary symbols. (setq count 0) => 0 (defun count-syms (s) (setq count (1+ count))) => count-syms (mapatoms 'count-syms) => nil count => 1871 See documentation in Accessing Documentation, for another example using mapatoms. ΓòÉΓòÉΓòÉ 10.4. Property Lists ΓòÉΓòÉΓòÉ A property list (plist for short) is a list of paired elements stored in the property list cell of a symbol. Each of the pairs associates a property name (usually a symbol) with a property or value. Property lists are generally used to record information about a symbol, such as how to compile it, the name of the file where it was defined, or perhaps even the grammatical class of the symbol (representing a word) in a language understanding system. The property names and property values may be any Lisp objects, but the names are usually symbols. They are compared using eq. Here is an example of a property list, found on the symbol progn when the compiler is loaded: (lisp-indent-hook 0 byte-compile byte-compile-progn) Here lisp-indent-hook and byte-compile are property names, and the other two elements are the corresponding values. Association lists (see Association Lists) are very similar to property lists. In contrast to association lists, the order of the pairs in the property list is not significant since the property names must be distinct. Property lists are better than association lists when it is necessary to attach information to various Lisp function names or variables. If all the pairs are recorded in one association list, it will be necessary to search that entire list each time a function or variable is to be operated on. By contrast, if the information is recorded in the property lists of the function names or variables themselves, each search will scan only the length of one property list, which is usually short. For this reason, the documentation for a variable is recorded in a property named variable-documentation. The byte compiler likewise uses properties to record those functions needing special treatment. However, association lists have their own advantages. Depending on your application, it may be faster to add an association to the front of an association list than to update a property. All properties for a symbol are stored in the same property list, so there is a possibility of a conflict between different uses of a property name. (For this reason, it is a good idea to use property names that are probably unique, such as by including the name of the library in the property name.) An association list may be used like a stack where associations are pushed on the front of the list and later discarded; this is not possible with a property list. -- Function: symbol-plist symbol This function returns the property list of symbol. -- Function: setplist symbol plist This function sets symbol's property list to plist. Normally, plist should be a well-formed property list, but this is not enforced. (setplist 'foo '(a 1 b (2 3) c nil)) => (a 1 b (2 3) c nil) (symbol-plist 'foo) => (a 1 b (2 3) c nil) For symbols in special obarrays, which are not used for ordinary purposes, it may make sense to use the property list cell in a nonstandard fashion; in fact, the abbrev mechanism does so (see Abbrevs). -- Function: get symbol property This function finds the value of the property named property in symbol's property list. If there is no such property, nil is returned. Thus, there is no distinction between a value of nil and the absence of the property. The name property is compared with the existing property names using eq, so any object is a legitimate property. See put for an example. -- Function: put symbol property value This function puts value onto symbol's property list under the property name property, replacing any previous value. (put 'fly 'verb 'transitive) =>'transitive (put 'fly 'noun '(a buzzing little bug)) => (a buzzing little bug) (get 'fly 'verb) => transitive (symbol-plist 'fly) => (verb transitive noun (a buzzing little bug)) ΓòÉΓòÉΓòÉ 11. Evaluation ΓòÉΓòÉΓòÉ The evaluation of expressions in Emacs Lisp is performed by the Lisp interpreter---a program that receives a Lisp object as input and computes its value as an expression. The value is computed in a fashion that depends on the data type of the object, following rules described in this chapter. The interpreter runs automatically to evaluate portions of your program, but can also be called explicitly via the Lisp primitive function eval. ΓòÉΓòÉΓòÉ 11.1. Introduction to Evaluation ΓòÉΓòÉΓòÉ The Lisp interpreter, or evaluator, is the program which computes the value of an expression which is given to it. When a function written in Lisp is called, the evaluator computes the value of the function by evaluating the expressions in the function body. Thus, running any Lisp program really means running the Lisp interpreter. How the evaluator handles an object depends primarily on the data type of the object. A Lisp object which is intended for evaluation is called an expression or a form. The fact that expressions are data objects and not merely text is one of the fundamental differences between Lisp-like languages and typical programming languages. Any object can be evaluated, but in practice only numbers, symbols, lists and strings are evaluated very often. It is very common to read a Lisp expression and then evaluate the expression, but reading and evaluation are separate activities, and either can be performed alone. Reading per se does not evaluate anything; it converts the printed representation of a Lisp object to the object itself. It is up to the caller of read whether this object is a form to be evaluated, or serves some entirely different purpose. See Input Functions. Do not confuse evaluation with command key interpretation. The editor command loop translates keyboard input into a command (an interactively callable function) using the current keymaps, and then uses call-interactively to invoke the command. The execution of the command itself involves evaluation if the command is written in Lisp, but that is not a part of command key interpretation itself. See Command Loop. Evaluation is a recursive process. That is, evaluation of a form may cause eval to be called again in order to evaluate parts of the form. For example, evaluation of a function call first evaluates each argument of the function call, and then evaluates each form in the function body. Consider evaluation of the form (car x): the subform x must first be evaluated recursively, so that its value can be passed as an argument to the function car. The evaluation of forms takes place in a context called the environment, which consists of the current values and bindings of all Lisp variables. Whenever the form refers to a variable without creating a new binding for it, the value of the current binding is used. See Variables. Evaluation of a form may create new environments for recursive evaluation by binding variables (see Local Variables). These environments are temporary and will be gone by the time evaluation of the form is complete. The form may also make changes that persist; these changes are called side-effects. An example of a form that produces side-effects is (setq foo 1). Finally, evaluation of one particular function call, byte-code, invokes the byte-code interpreter on its arguments. Although the byte-code interpreter is not the same as the Lisp interpreter, it uses the same environment as the Lisp interpreter, and may on occasion invoke the Lisp interpreter. (See Byte Compilation.) The details of what evaluation means for each kind of form are described below (see Forms). ΓòÉΓòÉΓòÉ 11.2. Eval ΓòÉΓòÉΓòÉ Most often, forms are evaluated automatically, by virtue of their occurrence in a program being run. On rare occasions, you may need to write code that evaluates a form that is computed at run time, such as when the form is read from text being edited or found on a property list. On these occasions, use the eval function. The functions and variables described in this section evaluate forms, specify limits to the evaluation process, or record recently returned values. Evaluation is also performed by load (see Loading). -- Function: eval form This is the basic function for performing evaluation. It evaluates form in the current environment and returns the result. How the evaluation proceeds depends on the type of the object (see Forms). Since eval is a function, the argument expression that appears in a call to eval is evaluated twice: once as preparation before eval is called, and again by the eval function itself. Here is an example: (setq foo 'bar) => bar (setq bar 'baz) => baz ;; eval is called on the form bar, which is the value of foo (eval foo) => baz The number of currently active calls to eval is limited to max-lisp-eval-depth. -- Command: eval-current-buffer &optional stream This function evaluates the forms in the current buffer. It reads forms from the buffer and calls eval on them until the end of the buffer is reached, or until an error is signaled and not handled. If stream is supplied, the variable standard-output is bound to stream during the evaluation ( see Output Functions). eval-current-buffer always returns nil. -- Command: eval-region start end &optional stream This function evaluates the forms in the current buffer in the region defined by the positions start and end. It reads forms from the region and calls eval on them until the end of the region is reached, or until an error is signaled and not handled. If stream is supplied, standard-output is bound to it for the duration of the command. eval-region always returns nil. -- Variable: max-lisp-eval-depth This variable defines the maximum depth allowed in calls to eval, apply, and funcall before an error is signaled (with error message "Lisp nesting exceeds max-lisp-eval-depth"). eval is called recursively to evaluate the arguments of Lisp function calls and to evaluate bodies of functions. This limit, with the associated error when it is exceeded, is one way that Lisp avoids infinite recursion on an ill-defined function. The default value of this variable is 200. If you set it to a value less than 100, Lisp will reset it to 100 if the given value is reached. -- Variable: values The value of this variable is a list of values returned by all expressions which were read from buffers (including the minibuffer), evaluated, and printed. The elements are in order, most recent first. (setq x 1) => 1 (list 'A (1+ 2) auto-save-default) => (A 3 t) values => ((A 3 t) 1 ...) This variable is useful for referring back to values of forms recently evaluated. It is generally a bad idea to print the value of values itself, since this may be very long. Instead, examine particular elements, like this: ;; Refer to the most recent evaluation result. (nth 0 values) => (A 3 t) ;; That put a new element on, so all elements move back one. (nth 1 values) => (A 3 t) ;; This gets the element that was next-to-last before this example. (nth 3 values) => 1 ΓòÉΓòÉΓòÉ 11.3. Kinds of Forms ΓòÉΓòÉΓòÉ A Lisp object that is intended to be evaluated is called a form. How Emacs evaluates a form depends on its data type. Emacs has three different kinds of form that are evaluated differently: symbols, lists, and ``all other types''. All three kinds are described in this section, starting with ``all other types'' which are self-evaluating forms. ΓòÉΓòÉΓòÉ 11.3.1. Self-Evaluating Forms ΓòÉΓòÉΓòÉ A self-evaluating form is any form that is not a list or symbol. Self-evaluating forms evaluate to themselves: the result of evaluation is the same object that was evaluated. Thus, the number 25 evaluates to 25, and the string "foo" evaluates to the string "foo". Likewise, evaluation of a vector does not cause evaluation of the elements of the vector---it returns the same vector with its contents unchanged. '123 ; An object, shown without evaluation. => 123 123 ; Evaluated as usual---result is the same. => 123 (eval '123) ; Evaluated ``by hand''---result is the same. => 123 (eval (eval '123)) ; Evaluating twice changes nothing. => 123 It is common to write numbers, characters, strings, and even vectors in Lisp code, taking advantage of the fact that they self-evaluate. However, it is quite unusual to do this for types that lack a read syntax, because it is inconvenient and not very useful; however, it is possible to put them inside Lisp programs when they are constructed from subexpressions rather than read. Here is an example: ;; Build such an expression. (setq buffer (list 'print (current-buffer))) => (print #<buffer eval.texi>) ;; Evaluate it. (eval buffer) -| #<buffer eval.texi> => #<buffer eval.texi> ΓòÉΓòÉΓòÉ 11.3.2. Symbol Forms ΓòÉΓòÉΓòÉ When a symbol is evaluated, it is treated as a variable. The result is the variable's value, if it has one. If it has none (if its value cell is void), an error is signaled. For more information on the use of variables, see Variables. In the following example, the value of a symbol is set with setq. When the symbol is later evaluated, that value is returned. (setq a 123) => 123 (eval 'a) => 123 a => 123 The symbols nil and t are treated specially, so that the value of nil is always nil, and the value of t is always t. Thus, these two symbols act like self-evaluating forms, even though eval treats them like any other symbol. ΓòÉΓòÉΓòÉ 11.3.3. Classification of List Forms ΓòÉΓòÉΓòÉ A form that is a nonempty list is either a function call, a macro call, or a special form, according to its first element. These three kinds of forms are evaluated in different ways, described below. The rest of the list consists of arguments for the function, macro or special form. The first step in evaluating a nonempty list is to examine its first element. This element alone determines what kind of form the list is and how the rest of the list is to be processed. The first element is not evaluated, as it would be in some Lisp dialects including Scheme. If the first element of the list is a symbol, as it most commonly is, then the symbol's function cell is examined, and its contents are used instead of the original symbol. If the contents are another symbol, this process, called symbol function indirection, is repeated until a non-symbol is obtained. One possible consequence of this process is an infinite loop, in the event that a symbol's function cell refers to the same symbol. Or a symbol may have a void function cell, causing a void-function error. But if neither of these things happens, we eventually obtain a non-symbol, which ought to be a function or other suitable object. More precisely, we should now have a Lisp function (a lambda expression), a primitive function, a Lisp macro, a special form, or an autoload object. Each of these types is a case described in one of the following sections. If the object is not one of these types, the error invalid-function is signaled. The following example illustrates the symbol indirection process. We use fset to set the function cell of a symbol and symbol-function to get the function cell contents (see Function Cells). Specifically, we store the symbol car into the function cell of first, and the symbol first into the function cell of erste. ;; Build this function cell linkage: ;; ------------- ----- ------- ------- ;; | #<subr car> | <-- | car | <-- | first | <-- | erste | ;; ------------- ----- ------- ------- (symbol-function 'car) => #<subr car> (fset 'first 'car) => car (fset 'erste 'first) => first (erste '(1 2 3)) ; Call the function referenced by erste. => 1 By contrast, the following example calls a function without any symbol function indirection, because the first element is an anonymous Lisp function, not a symbol. ((lambda (arg) (erste arg)) '(1 2 3)) => 1 After that function is called, its body is evaluated; this does involve symbol function indirection when calling erste. ΓòÉΓòÉΓòÉ 11.3.4. Evaluation of Function Forms ΓòÉΓòÉΓòÉ If the first element of a list being evaluated is a Lisp function object or primitive function object, then that list is a function call. For example, here is a call to the function +: (+ 1 x) When a function call is evaluated, the first step is to evaluate the remaining elements of the list in the order they appear. The results are the actual argument values, one argument from each element. Then the function is called with this list of arguments, effectively using the function apply (see Calling Functions). If the function is written in Lisp, the arguments are used to bind the argument variables of the function (see Lambda Expressions); then the forms in the function body are evaluated in order, and the result of the last one is used as the value of the function call. ΓòÉΓòÉΓòÉ 11.3.5. Lisp Macro Evaluation ΓòÉΓòÉΓòÉ If the first element of a list being evaluated is a macro object, then the list is a macro call. When a macro call is evaluated, the elements of the rest of the list are not initially evaluated. Instead, these elements themselves are used as the arguments of the macro. The macro definition computes a replacement form, called the expansion of the macro, which is evaluated in place of the original form. The expansion may be any sort of form: a self-evaluating constant, a symbol or a list. If the expansion is itself a macro call, this process of expansion repeats until some other sort of form results. Normally, the argument expressions are not evaluated as part of computing the macro expansion, but instead appear as part of the expansion, so they are evaluated when the expansion is evaluated. For example, given a macro defined as follows: (defmacro cadr (x) (list 'car (list 'cdr x))) an expression such as (cadr (assq 'handler list)) is a macro call, and its expansion is: (car (cdr (assq 'handler list))) Note that the argument (assq 'handler list) appears in the expansion. See Macros, for a complete description of Emacs Lisp macros. ΓòÉΓòÉΓòÉ 11.3.6. Special Forms ΓòÉΓòÉΓòÉ A special form is a primitive function specially marked so that its arguments are not all evaluated. Special forms define control structures or perform variable bindings---things which functions cannot do. Each special form has its own rules for which arguments are evaluated and which are used without evaluation. Whether a particular argument is evaluated may depend on the results of evaluating other arguments. Here is a list, in alphabetical order, of all of the special forms in Emacs Lisp with a reference to where each is described. and see Combining Conditions catch see Catch and Throw cond see Conditionals condition-case see Errors defconst see Defining Variables defmacro see Defining Macros defun see Defining Functions defvar see Defining Variables function see Anonymous Functions if see Conditionals interactive see Interactive Call let see Local Variables let* see Local Variables or see Combining Conditions prog1 see Sequencing prog2 see Sequencing progn see Sequencing quote see Quoting save-excursion see Excursions save-restriction see Narrowing save-window-excursion see Window Configurations setq see Setting Variables setq-default see Creating Buffer-Local unwind-protect see Nonlocal Exits while see Iteration with-output-to-temp-buffer see Temporary Displays Common Lisp note: here are some comparisons of special forms in GNU Emacs Lisp and Common Lisp. setq, if, and catch are special forms in both Emacs Lisp and Common Lisp. defun is a special form in Emacs Lisp, but a macro in Common Lisp. save-excursion is a special form in Emacs Lisp, but doesn't exist in Common Lisp. throw is a special form in Common Lisp (because it must be able to throw multiple values), but it is a function in Emacs Lisp (which doesn't have multiple values). ΓòÉΓòÉΓòÉ 11.3.7. Autoloading ΓòÉΓòÉΓòÉ The autoload feature allows you to call a function or macro whose function definition has not yet been loaded into Emacs. When an autoload object appears as a symbol's function definition and that symbol is used as a function, Emacs will automatically install the real definition (plus other associated code) and then call that definition. (See Autoload.) ΓòÉΓòÉΓòÉ 11.4. Quoting ΓòÉΓòÉΓòÉ The special form quote returns its single argument ``unchanged''. -- Special Form: quote object This special form returns object, without evaluating it. This allows symbols and lists, which would normally be evaluated, to be included literally in a program. (It is not necessary to quote numbers, strings, and vectors since they are self-evaluating.) Use function instead of quote when quoting lambda expressions (see Anonymous Functions). Because quote is used so often in programs, a convenient read syntax is defined for it. An apostrophe character (`'') followed by a Lisp object (in read syntax) expands to a list whose first element is quote, and whose second element is the object. Thus, the read syntax 'x is an abbreviation for (quote x). Here are some examples of expressions that use quote: (quote (+ 1 2)) => (+ 1 2) (quote foo) => foo 'foo => foo ''foo => (quote foo) '(quote foo) => (quote foo) ['foo] => [(quote foo)] ΓòÉΓòÉΓòÉ 12. Control Structures ΓòÉΓòÉΓòÉ A Lisp program consists of expressions or forms (see Forms). We control the order of execution of the forms by enclosing them in control structures. Control structures are special forms which control when, whether, or how many times to execute the forms they contain. The simplest control structure is sequential execution: first form a, then form b, and so on. This is what happens when you write several forms in succession in the body of a function, or at top level in a file of Lisp code---the forms are executed in the order they are written. We call this textual order. For example, if a function body consists of two forms a and b, evaluation of the function evaluates first a and then b, and the function's value is the value of b. Naturally, Emacs Lisp has many kinds of control structures, including other varieties of sequencing, function calls, conditionals, iteration, and (controlled) jumps. The built-in control structures are special forms since their subforms are not necessarily evaluated. You can use macros to define your own control structure constructs (see Macros). ΓòÉΓòÉΓòÉ 12.1. Sequencing ΓòÉΓòÉΓòÉ Evaluating forms in the order they are written is the most common control structure. Sometimes this happens automatically, such as in a function body. Elsewhere you must use a control structure construct to do this: progn, the simplest control construct of Lisp. A progn special form looks like this: (progn a b c ...) and it says to execute the forms a, b, c and so on, in that order. These forms are called the body of the progn form. The value of the last form in the body becomes the value of the entire progn. When Lisp was young, progn was the only way to execute two or more forms in succession and use the value of the last of them. But programmers found they often needed to use a progn in the body of a function, where (at that time) only one form was allowed. So the body of a function was made into an ``implicit progn'': several forms are allowed just as in the body of an actual progn. Many other control structures likewise contain an implicit progn. As a result, progn is not used as often as it used to be. It is needed now most often inside of an unwind-protect, and, or or. -- Special Form: progn forms... This special form evaluates all of the forms, in textual order, returning the result of the final form. (progn (print "The first form") (print "The second form") (print "The third form")) -| "The first form" -| "The second form" -| "The third form" => "The third form" Two other control constructs likewise evaluate a series of forms but return a different value: -- Special Form: prog1 form1 forms... This special form evaluates form1 and all of the forms, in textual order, returning the result of form1. (prog1 (print "The first form") (print "The second form") (print "The third form")) -| "The first form" -| "The second form" -| "The third form" => "The first form" Here is a way to remove the first element from a list in the variable x, then return the value of that former element: (prog1 (car x) (setq x (cdr x))) -- Special Form: prog2 form1 form2 forms... This special form evaluates form1, form2, and all of the following forms, in textual order, returning the result of form2. (prog2 (print "The first form") (print "The second form") (print "The third form")) -| "The first form" -| "The second form" -| "The third form" => "The second form" ΓòÉΓòÉΓòÉ 12.2. Conditionals ΓòÉΓòÉΓòÉ Conditional control structures choose among alternatives. Emacs Lisp has two conditional forms: if, which is much the same as in other languages, and cond, which is a generalized case statement. -- Special Form: if condition then-form else-forms... if chooses between the then-form and the else-forms based on the value of condition. If the evaluated condition is non-nil, then-form is evaluated and the result returned. Otherwise, the else-forms are evaluated in textual order, and the value of the last one is returned. (The else part of if is an example of an implicit progn. See Sequencing.) If condition has the value nil, and no else-forms are given, if returns nil. if is a special form because the branch which is not selected is never evaluated---it is ignored. Thus, in the example below, true is not printed because print is never called. (if nil (print 'true) 'very-false) => very-false -- Special Form: cond clause... cond chooses among an arbitrary number of alternatives. Each clause in the cond must be a list. The car of this list is the condition; the remaining elements, if any, the body-forms. Thus, a clause looks like this: (condition body-forms...) cond tries the clauses in textual order, by evaluating the condition of each clause. If the value of condition is non-nil, the body-forms are evaluated, and the value of the last of body-forms becomes the value of the cond. The remaining clauses are ignored. If the value of condition is nil, the clause ``fails'', so the cond moves on to the following clause, trying its condition. If every condition evaluates to nil, so that every clause fails, cond returns nil. A clause may also look like this: (condition) Then, if condition is non-nil when tested, the value of condition becomes the value of the cond form. The following example has four clauses, which test for the cases where the value of x is a number, string, buffer and symbol, respectively: (cond ((numberp x) x) ((stringp x) x) ((bufferp x) (setq temporary-hack x) ; multiple body-forms (buffer-name x)) ; in one clause ((symbolp x) (symbol-value x))) Often we want the last clause to be executed whenever none of the previous clauses was successful. To do this, we use t as the condition of the last clause, like this: (t body-forms). The form t evaluates to t, which is never nil, so this clause never fails, provided the cond gets to it at all. For example, (cond ((eq a 1) 'foo) (t "default")) => "default" This expression is a cond which returns foo if the value of a is 1, and returns the string "default" otherwise. Both cond and if can usually be written in terms of the other. Therefore, the choice between them is a matter of taste and style. For example: (if a b c) == (cond (a b) (t c)) ΓòÉΓòÉΓòÉ 12.3. Constructs for Combining Conditions ΓòÉΓòÉΓòÉ This section describes three constructs that are often used together with if and cond to express complicated conditions. The constructs and and or can also be used individually as kinds of multiple conditional constructs. -- Function: not condition This function tests for the falsehood of condition. It returns t if condition is nil, and nil otherwise. The function not is identical to null, and we recommend using null if you are testing for an empty list. -- Special Form: and conditions... The and special form tests whether all the conditions are true. It works by evaluating the conditions one by one in the order written. If any of the conditions evaluates to nil, then the result of the and must be nil regardless of the remaining conditions; so the remaining conditions are ignored and the and returns right away. If all the conditions turn out non-nil, then the value of the last of them becomes the value of the and form. Here is an example. The first condition returns the integer 1, which is not nil. Similarly, the second condition returns the integer 2, which is not nil. The third condition is nil, so the remaining condition is never evaluated. (and (print 1) (print 2) nil (print 3)) -| 1 -| 2 => nil Here is a more realistic example of using and: (if (and (consp foo) (eq (car foo) 'x)) (message "foo is a list starting with x")) Note that (car foo) is not executed if (consp foo) returns nil, thus avoiding an error. and can be expressed in terms of either if or cond. For example: (and arg1 arg2 arg3) == (if arg1 (if arg2 arg3)) == (cond (arg1 (cond (arg2 arg3)))) -- Special Form: or conditions... The or special form tests whether at least one of the conditions is true. It works by evaluating all the conditions one by one in the order written. If any of the conditions evaluates to a non-nil value, then the result of the or must be non-nil; so the remaining conditions are ignored and the or returns right away. The value it returns is the non-nil value of the condition just evaluated. If all the conditions turn out nil, then the or expression returns nil. For example, this expression tests whether x is either 0 or nil: (or (eq x nil) (= x 0)) Like the and construct, or can be written in terms of cond. For example: (or arg1 arg2 arg3) == (cond (arg1) (arg2) (arg3)) You could almost write or in terms of if, but not quite: (if arg1 arg1 (if arg2 arg2 arg3)) This is not completely equivalent because it can evaluate arg1 or arg2 twice. By contrast, (or arg1 arg2 arg3) never evaluates any argument more than once. ΓòÉΓòÉΓòÉ 12.4. Iteration ΓòÉΓòÉΓòÉ Iteration means executing part of a program repetitively. For example, you might want to repeat some expressions once for each element of a list, or once for each integer from 0 to n. You can do this in Emacs Lisp with the special form while: -- Special Form: while condition forms... while first evaluates condition. If the result is non-nil, it evaluates forms in textual order. Then it reevaluates condition, and if the result is non-nil, it evaluates forms again. This process repeats until condition evaluates to nil. There is no limit on the number of iterations that may occur. The loop will continue until either condition evaluates to nil or until an error or throw jumps out of it (see Nonlocal Exits). The value of a while form is always nil. (setq num 0) => 0 (while (< num 4) (princ (format "Iteration %d." num)) (setq num (1+ num))) -| Iteration 0. -| Iteration 1. -| Iteration 2. -| Iteration 3. => nil If you would like to execute something on each iteration before the end-test, put it together with the end-test in a progn as the first argument of while, as shown here: (while (progn (forward-line 1) (not (looking-at "^$")))) This moves forward one line and continues moving by lines until an empty line is reached. ΓòÉΓòÉΓòÉ 12.5. Nonlocal Exits ΓòÉΓòÉΓòÉ A nonlocal exit is a transfer of control from one point in a program to another remote point. Nonlocal exits can occur in Emacs Lisp as a result of errors; you can also use them under explicit control. ΓòÉΓòÉΓòÉ 12.5.1. Explicit Nonlocal Exits: catch and throw ΓòÉΓòÉΓòÉ Most control constructs affect only the flow of control within the construct itself. The function throw is the sole exception: it performs a nonlocal exit on request. throw is used inside a catch, and jumps back to that catch. For example: (catch 'foo (progn ... (throw 'foo t) ...)) The throw transfers control straight back to the corresponding catch, which returns immediately. The code following the throw is not executed. The second argument of throw is used as the return value of the catch. The throw and the catch are matched through the first argument: throw searches for a catch whose first argument is eq to the one specified. Thus, in the above example, the throw specifies foo, and the catch specifies the same symbol, so that catch is applicable. If there is more than one applicable catch, the innermost one takes precedence. All Lisp constructs between the catch and the throw, including function calls, are exited automatically along with the catch. When binding constructs such as let or function calls are exited in this way, the bindings are unbound, just as they are when the binding construct is exited normally (see Local Variables). Likewise, the buffer and position saved by save-excursion (see Excursions) are restored, and so is the narrowing status saved by save-restriction and the window selection saved by save-window-excursion (see Window Configurations). Any cleanups established with the unwind-protect special form are executed if the unwind-protect is exited with a throw. The throw need not appear lexically within the catch that it jumps to. It can equally well be called from another function called within the catch. As long as the throw takes place chronologically after entry to the catch, and chronologically before exit from it, it has access to that catch. This is why throw can be used in commands such as exit-recursive-edit which throw back to the editor command loop (see Recursive Editing). Common Lisp note: most other versions of Lisp, including Common Lisp, have several ways of transferring control nonsequentially: return, return-from, and go, for example. Emacs Lisp has only throw. -- Special Form: catch tag body... catch establishes a return point for the throw function. The return point is distinguished from other such return points by tag, which may be any Lisp object. The argument tag is evaluated normally before the return point is established. With the return point in effect, the forms of the body are evaluated in textual order. If the forms execute normally, without error or nonlocal exit, the value of the last body form is returned from the catch. If a throw is done within body specifying the same value tag, the catch exits immediately; the value it returns is whatever was specified as the second argument of throw. -- Function: throw tag value The purpose of throw is to return from a return point previously established with catch. The argument tag is used to choose among the various existing return points; it must be eq to the value specified in the catch. If multiple return points match tag, the innermost one is used. The argument value is used as the value to return from that catch. If no return point is in effect with tag tag, then a no-catch error is signaled with data (tag value). ΓòÉΓòÉΓòÉ 12.5.2. Examples of catch and throw ΓòÉΓòÉΓòÉ One way to use catch and throw is to exit from a doubly nested loop. (In most languages, this would be done with a ``go to''.) Here we compute (foo i j) for i and j varying from 0 to 9: (defun search-foo () (catch 'loop (let ((i 0)) (while (< i 10) (let ((j 0)) (while (< j 10) (if (foo i j) (throw 'loop (list i j))) (setq j (1+ j)))) (setq i (1+ i)))))) If foo ever returns non-nil, we stop immediately and return a list of i and j. If foo always returns nil, the catch returns normally, and the value is nil, since that is the result of the while. Here are two tricky examples, slightly different, showing two return points at once. First, two return points with the same tag, hack: (defun catch2 (tag) (catch tag (throw 'hack 'yes))) => catch2 (catch 'hack (print (catch2 'hack)) 'no) -| yes => no Since both return points have tags that match the throw, it goes to the inner one, the one established in catch2. Therefore, catch2 returns normally with value yes, and this value is printed. Finally the second body form in the outer catch, which is 'no, is evaluated and returned from the outer catch. Now let's change the argument given to catch2: (defun catch2 (tag) (catch tag (throw 'hack 'yes))) => catch2 (catch 'hack (print (catch2 'quux)) 'no) => yes We still have two return points, but this time only the outer one has the tag hack; the inner one has the tag quux instead. Therefore, the throw returns the value yes from the outer return point. The function print is never called, and the body-form 'no is never evaluated. ΓòÉΓòÉΓòÉ 12.5.3. Errors ΓòÉΓòÉΓòÉ When Emacs Lisp attempts to evaluate a form that, for some reason, cannot be evaluated, it signals an error. When an error is signaled, Emacs's default reaction is to print an error message and terminate execution of the current command. This is the right thing to do in most cases, such as if you type C-f at the end of the buffer. In complicated programs, simple termination may not be what you want. For example, the program may have made temporary changes in data structures, or created temporary buffers which should be deleted before the program is finished. In such cases, you would use unwind-protect to establish cleanup expressions to be evaluated in case of error. Occasionally, you may wish the program to continue execution despite an error in a subroutine. In these cases, you would use condition-case to establish error handlers to recover control in case of error. Resist the temptation to use error handling to transfer control from one part of the program to another; use catch and throw. See Catch and Throw. ΓòÉΓòÉΓòÉ 12.5.3.1. How to Signal an Error ΓòÉΓòÉΓòÉ Most errors are signaled ``automatically'' within Lisp primitives which you call for other purposes, such as if you try to take the car of an integer or move forward a character at the end of the buffer; you can also signal errors explicitly with the functions error and signal. -- Function: error format-string &rest args This function signals an error with an error message constructed by applying format (see String Conversion) to format-string and args. Typical uses of error is shown in the following examples: (error "You have committed an error. Try something else.") error--> You have committed an error. Try something else. (error "You have committed %d errors. You don't learn fast." 10) error--> You have committed 10 errors. You don't learn fast. error works by calling signal with two arguments: the error symbol error, and a list containing the string returned by format. If you want to use a user-supplied string as an error message verbatim, don't just write (error string). If string contains `%', it will be interpreted as a format specifier, with undesirable results. Instead, use (error "%s" string). -- Function: signal error-symbol data This function signals an error named by error-symbol. The argument data is a list of additional Lisp objects relevant to the circumstances of the error. The argument error-symbol must be an error symbol---a symbol that has an error-conditions property whose value is a list of condition names. This is how different sorts of errors are classified. The number and significance of the objects in data depends on error-symbol. For example, with a wrong-type-arg error, there are two objects in the list: a predicate which describes the type that was expected, and the object which failed to fit that type. See Error Names, for a description of error symbols. Both error-symbol and data are available to any error handlers which handle the error: a list (error-symbol . data) is constructed to become the value of the local variable bound in the condition-case form (see Handling Errors). If the error is not handled, both of them are used in printing the error message. (signal 'wrong-number-of-arguments '(x y)) error--> Wrong number of arguments: x, y (signal 'no-such-error '("My unknown error condition.")) error--> peculiar error: "My unknown error condition." Common Lisp note: Emacs Lisp has nothing like the Common Lisp concept of continuable errors. ΓòÉΓòÉΓòÉ 12.5.3.2. How Emacs Processes Errors ΓòÉΓòÉΓòÉ When an error is signaled, Emacs searches for an active handler for the error. A handler is a specially marked place in the Lisp code of the current function or any of the functions by which it was called. If an applicable handler exists, its code is executed, and control resumes following the handler. The handler executes in the environment of the condition-case which established it; all functions called within that condition-case have already been exited, and the handler cannot return to them. If no applicable handler is in effect in your program, the current command is terminated and control returns to the editor command loop, because the command loop has an implicit handler for all kinds of errors. The command loop's handler uses the error symbol and associated data to print an error message. When an error is not handled explicitly, it may cause the Lisp debugger to be called. The debugger is enabled if the variable debug-on-error (see Error Debugging) is non-nil. Unlike error handlers, the debugger runs in the environment of the error, so that you can examine values of variables precisely as they were at the time of the error. ΓòÉΓòÉΓòÉ 12.5.3.3. Writing Code to Handle Errors ΓòÉΓòÉΓòÉ The usual effect of signaling an error is to terminate the command that is running and return immediately to the Emacs editor command loop. You can arrange to trap errors occurring in a part of your program by establishing an error handler with the special form condition-case. A simple example looks like this: (condition-case nil (delete-file filename) (error nil)) This deletes the file named filename, catching any error and returning nil if an error occurs. The second argument of condition-case is called the protected form. (In the example above, the protected form is a call to delete-file.) The error handlers go into effect when this form begins execution and are deactivated when this form returns. They remain in effect for all the intervening time. In particular, they are in effect during the execution of subroutines called by this form, and their subroutines, and so on. This is a good thing, since, strictly speaking, errors can be signaled only by Lisp primitives (including signal and error) called by the protected form, not by the protected form itself. The arguments after the protected form are handlers. Each handler lists one or more condition names (which are symbols) to specify which errors it will handle. The error symbol specified when an error is signaled also defines a list of condition names. A handler applies to an error if they have any condition names in common. In the example above, there is one handler, and it specifies one condition name, error, which covers all errors. The search for an applicable handler checks all the established handlers starting with the most recently established one. Thus, if two nested condition-case forms try to handle the same error, the inner of the two will actually handle it. When an error is handled, control returns to the handler, unbinding all variable bindings made by binding constructs that are exited and executing the cleanups of all unwind-protect forms that are exited by doing so. Then the body of the handler is executed. After this, execution continues by returning from the condition-case form. Because the protected form is exited completely before execution of the handler, the handler cannot resume execution at the point of the error, nor can it examine variable bindings that were made within the protected form. All it can do is clean up and proceed. Error signaling and handling have some resemblance to throw and catch, but they are entirely separate facilities. An error cannot be caught by a catch, and a throw cannot be handled by an error handler (though if there is no catch, throw will signal an error which can be handled). -- Special Form: condition-case var protected-form handlers... This special form establishes the error handlers handlers around the execution of protected-form. If protected-form executes without error, the value it returns becomes the value of the condition-case form; in this case, the condition-case has no effect. The condition-case form makes a difference when an error occurs during protected-form. Each of the handlers is a list of the form (conditions body...). conditions is a condition name to be handled, or a list of condition names; body is one or more Lisp expressions to be executed when this handler handles an error. Each error that occurs has an error symbol which describes what kind of error it is. The error-conditions property of this symbol is a list of condition names (see Error Names). Emacs searches all the active condition-case forms for a handler which specifies one or more of these names; the innermost matching condition-case handles the error. The handlers in this condition-case are tested in the order in which they appear. The body of the handler is then executed, and the condition-case returns normally, using the value of the last form in the body as the overall value. The argument var is a variable. condition-case does not bind this variable when executing the protected-form, only when it handles an error. At that time, var is bound locally to a list of the form (error-symbol . data), giving the particulars of the error. The handler can refer to this list to decide what to do. For example, if the error is for failure opening a file, the file name is the second element of data---the third element of var. If var is nil, that means no variable is bound. Then the error symbol and associated data are not made available to the handler. Here is an example of using condition-case to handle the error that results from dividing by zero. The handler prints out a warning message and returns a very large number. (defun safe-divide (dividend divisor) (condition-case err ;; Protected form. (/ dividend divisor) ;; The handler. (arith-error ; Condition. (princ (format "Arithmetic error: %s" err)) 1000000))) => safe-divide (safe-divide 5 0) -| Arithmetic error: (arith-error) => 1000000 The handler specifies condition name arith-error so that it will handle only division-by-zero errors. Other kinds of errors will not be handled, at least not by this condition-case. Thus, (safe-divide nil 3) error--> Wrong type argument: integer-or-marker-p, nil Here is a condition-case that catches all kinds of errors, including those signaled with error: (setq baz 34) => 34 (condition-case err (if (eq baz 35) t ;; This is a call to the function error. (error "Rats! The variable %s was %s, not 35." 'baz baz)) ;; This is the handler; it is not a form. (error (princ (format "The error was: %s" err)) 2)) -| The error was: (error "Rats! The variable baz was 34, not 35.") => 2 condition-case is often used to trap errors that are predictable, such as failure to open a file in a call to insert-file-contents. It is also used to trap errors that are totally unpredictable, such as when the program evaluates an expression read from the user. ΓòÉΓòÉΓòÉ 12.5.3.4. Error Symbols and Condition Names ΓòÉΓòÉΓòÉ When you signal an error, you specify an error symbol to specify the kind of error you have in mind. Each error has one and only one error symbol to categorize it. This is the finest classification of errors defined by the Lisp language. These narrow classifications are grouped into a hierarchy of wider classes called error conditions, identified by condition names. The narrowest such classes belong to the error symbols themselves: each error symbol is also a condition name. There are also condition names for more extensive classes, up to the condition name error which takes in all kinds of errors. Thus, each error has one or more condition names: error, the error symbol if that is distinct from error, and perhaps some intermediate classifications. In order for a symbol to be usable as an error symbol, it must have an error-conditions property which gives a list of condition names. This list defines the conditions which this kind of error belongs to. (The error symbol itself, and the symbol error, should always be members of this list.) Thus, the hierarchy of condition names is defined by the error-conditions properties of the error symbols. In addition to the error-conditions list, the error symbol should have an error-message property whose value is a string to be printed when that error is signaled but not handled. If the error-message property exists, but is not a string, the error message `peculiar error' is used. Here is how we define a new error symbol, new-error: (put 'new-error 'error-conditions '(error my-own-errors new-error)) => (error my-own-errors new-error) (put 'new-error 'error-message "A new error") => "A new error" This error has three condition names: new-error, the narrowest classification; my-own-errors, which we imagine is a wider classification; and error, which is the widest of all. Naturally, Emacs will never signal a new-error on its own; only an explicit call to signal (see Errors) in your code can do this: (signal 'new-error '(x y)) error--> A new error: x, y This error can be handled through any of the three condition names. This example handles new-error and any other errors in the class my-own-errors: (condition-case foo (bar nil t) (my-own-errors nil)) The significant way that errors are classified is by their condition names---the names used to match errors with handlers. An error symbol serves only as a convenient way to specify the intended error message and list of condition names. If signal were given a list of condition names rather than one error symbol, that would be cumbersome. By contrast, using only error symbols without condition names would seriously decrease the power of condition-case. Condition names make it possible to categorize errors at various levels of generality when you write an error handler. Using error symbols alone would eliminate all but the narrowest level of classification. See Standard Errors, for a list of all the standard error symbols and their conditions. ΓòÉΓòÉΓòÉ 12.5.4. Cleaning up from Nonlocal Exits ΓòÉΓòÉΓòÉ The unwind-protect construct is essential whenever you temporarily put a data structure in an inconsistent state; it permits you to ensure the data are consistent in the event of an error. -- Special Form: unwind-protect body cleanup-forms... unwind-protect executes the body with a guarantee that the cleanup-forms will be evaluated if control leaves body, no matter how that happens. The body may complete normally, or execute a throw out of the unwind-protect, or cause an error; in all cases, the cleanup-forms will be evaluated. Only the body is actually protected by the unwind-protect. If any of the cleanup-forms themselves exit nonlocally (e.g., via a throw or an error), it is not guaranteed that the rest of them will be executed. If the failure of one of the cleanup-forms has the potential to cause trouble, then it should be protected by another unwind-protect around that form. The number of currently active unwind-protect forms counts, together with the number of local variable bindings, against the limit max-specpdl-size (see Local Variables). For example, here we make an invisible buffer for temporary use, and make sure to kill it before finishing: (save-excursion (let ((buffer (get-buffer-create " *temp*"))) (set-buffer buffer) (unwind-protect body (kill-buffer buffer)))) You might think that we could just as well write (kill-buffer (current-buffer)) and dispense with the variable buffer. However, the way shown above is safer, if body happens to get an error after switching to a different buffer! (Alternatively, you could write another save-excursion around the body, to ensure that the temporary buffer becomes current in time to kill it.) Here is an actual example taken from the file `ftp.el'. It creates a process (see Processes) to try to establish a connection to a remote machine. As the function ftp-login is highly susceptible to numerous problems which the writer of the function cannot anticipate, it is protected with a form that guarantees deletion of the process in the event of failure. Otherwise, Emacs might fill up with useless subprocesses. (let ((win nil)) (unwind-protect (progn (setq process (ftp-setup-buffer host file)) (if (setq win (ftp-login process host user password)) (message "Logged in") (error "Ftp login failed"))) (or win (and process (delete-process process))))) This example actually has a small bug: if the user types C-g to quit, and the quit happens immediately after the function ftp-setup-buffer returns but before the variable process is set, the process will not be killed. There is no easy way to fix this bug, but at least it is very unlikely. ΓòÉΓòÉΓòÉ 13. Variables ΓòÉΓòÉΓòÉ A variable is a name used in a program to stand for a value. Nearly all programming languages have variables of some sort. In the text for a Lisp program, variables are written using the syntax for symbols. In Lisp, unlike most programming languages, programs are represented primarily as Lisp objects and only secondarily as text. The Lisp objects used for variables are symbols: the symbol name is the variable name, and the variable's value is stored in the value cell of the symbol. The use of a symbol as a variable is independent of whether the same symbol has a function definition. See Symbol Components. The textual form of a program is determined by its Lisp object representation; it is the read syntax for the Lisp object which constitutes the program. This is why a variable in a textual Lisp program is written as the read syntax for the symbol that represents the variable. ΓòÉΓòÉΓòÉ 13.1. Global Variables ΓòÉΓòÉΓòÉ The simplest way to use a variable is globally. This means that the variable has just one value at a time, and this value is in effect (at least for the moment) throughout the Lisp system. The value remains in effect until you specify a new one. When a new value replaces the old one, no trace of the old value remains in the variable. You specify a value for a symbol with setq. For example, (setq x '(a b)) gives the variable x the value (a b). Note that the first argument of setq, the name of the variable, is not evaluated, but the second argument, the desired value, is evaluated normally. Once the variable has a value, you can refer to it by using the symbol by itself as an expression. Thus, x => (a b) assuming the setq form shown above has already been executed. If you do another setq, the new value replaces the old one: x => (a b) (setq x 4) => 4 x => 4 ΓòÉΓòÉΓòÉ 13.2. Variables that Never Change ΓòÉΓòÉΓòÉ Emacs Lisp has two special symbols, nil and t, that always evaluate to themselves. These symbols cannot be rebound, nor can their value cells be changed. An attempt to change the value of nil or t signals a setting-constant error. nil == 'nil => nil (setq nil 500) error--> Attempt to set constant symbol: nil ΓòÉΓòÉΓòÉ 13.3. Local Variables ΓòÉΓòÉΓòÉ Global variables are given values that last until explicitly superseded with new values. Sometimes it is useful to create variable values that exist temporarily---only while within a certain part of the program. These values are called local, and the variables so used are called local variables. For example, when a function is called, its argument variables receive new local values which last until the function exits. Similarly, the let special form explicitly establishes new local values for specified variables; these last until exit from the let form. When a local value is established, the previous value (or lack of one) of the variable is saved away. When the life span of the local value is over, the previous value is restored. In the mean time, we say that the previous value is shadowed and not visible. Both global and local values may be shadowed. If you set a variable (such as with setq) while it is local, this replaces the local value; it does not alter the global value, or previous local values that are shadowed. To model this behavior, we speak of a local binding of the variable as well as a local value. The local binding is a conceptual place that holds a local value. Entry to a function, or a special form such as let, creates the local binding; exit from the function or from the let removes the local binding. As long as the local binding lasts, the variable's value is stored within it. Use of setq or set while there is a local binding stores a different value into the local binding; it does not create a new binding. We also speak of the global binding, which is where (conceptually) the global value is kept. A variable can have more than one local binding at a time (for example, if there are nested let forms that bind it). In such a case, the most recently created local binding that still exists is the current binding of the variable. (This is called dynamic scoping; see Variable Scoping.) If there are no local bindings, the variable's global binding is its current binding. We also call the current binding the most-local existing binding, for emphasis. Ordinary evaluation of a symbol always returns the value of its current binding. The special forms let and let* exist to create local bindings. -- Special Form: let (bindings...) forms... This function binds variables according to bindings and then evaluates all of the forms in textual order. The let-form returns the value of the last form in forms. Each of the bindings is either (i) a symbol, in which case that symbol is bound to nil; or (ii) a list of the form (symbol value-form), in which case symbol is bound to the result of evaluating value-form. If value-form is omitted, nil is used. All of the value-forms in bindings are evaluated in the order they appear and before any of the symbols are bound. Here is an example of this: Z is bound to the old value of Y, which is 2, not the new value, 1. (setq Y 2) => 2 (let ((Y 1) (Z Y)) (list Y Z)) => (1 2) -- Special Form: let* (bindings...) forms... This special form is like let, except that each symbol in bindings is bound as soon as its new value is computed, before the computation of the values of the following local bindings. Therefore, an expression in bindings may reasonably refer to the preceding symbols bound in this let* form. Compare the following example with the example above for let. (setq Y 2) => 2 (let* ((Y 1) (Z Y)) ; Use the just-established value of Y. (list Y Z)) => (1 1) Here is a complete list of the other facilities which create local bindings: o Function calls (see Functions). o Macro calls (see Macros). o condition-case (see Errors). -- Variable: max-specpdl-size This variable defines the limit on the number of local variable bindings and unwind-protect cleanups (see Nonlocal Exits) that are allowed before signaling an error (with data "Variable binding depth exceeds max-specpdl-size"). This limit, with the associated error when it is exceeded, is one way that Lisp avoids infinite recursion on an ill-defined function. The default value is 600. ΓòÉΓòÉΓòÉ 13.4. When a Variable is ``Void'' ΓòÉΓòÉΓòÉ If you have never given a symbol any value as a global variable, we say that that symbol's global value is void. In other words, the symbol's value cell does not have any Lisp object in it. If you try to evaluate the symbol, you get a void-variable error rather than a value. Note that a value of nil is not the same as void. The symbol nil is a Lisp object and can be the value of a variable just as any other object can be; but it is a value. A void variable does not have any value. After you have given a variable a value, you can make it void once more using makunbound. -- Function: makunbound symbol This function makes the current binding of symbol void. This causes any future attempt to use this symbol as a variable to signal the error void-variable, unless or until you set it again. makunbound returns symbol. (makunbound 'x) ; Make the global value of x void. => x x error--> Symbol's value as variable is void: x If symbol is locally bound, makunbound affects the most local existing binding. This is the only way a symbol can have a void local binding, since all the constructs that create local bindings create them with values. In this case, the voidness lasts at most as long as the binding does; when the binding is removed due to exit from the construct that made it, the previous or global binding is reexposed as usual, and the variable is no longer void unless the newly reexposed binding was void all along. (setq x 1) ; Put a value in the global binding. => 1 (let ((x 2)) ; Locally bind it. (makunbound 'x) ; Void the local binding. x) error--> Symbol's value as variable is void: x x ; The global binding is unchanged. => 1 (let ((x 2)) ; Locally bind it. (let ((x 3)) ; And again. (makunbound 'x) ; Void the innermost-local binding. x)) ; And refer: it's void. error--> Symbol's value as variable is void: x (let ((x 2)) (let ((x 3)) (makunbound 'x)) ; Void inner binding, then remove it. x) ; Now outer let binding is visible. => 2 A variable that has been made void with makunbound is indistinguishable from one that has never received a value and has always been void. You can use the function boundp to test whether a variable is currently void. -- Function: boundp variable boundp returns t if variable (a symbol) is not void; more precisely, if its current binding is not void. It returns nil otherwise. (boundp 'abracadabra) ; Starts out void. => nil (let ((abracadabra 5)) ; Locally bind it. (boundp 'abracadabra)) => t (boundp 'abracadabra) ; Still globally void. => nil (setq abracadabra 5) ; Make it globally nonvoid. => 5 (boundp 'abracadabra) => t ΓòÉΓòÉΓòÉ 13.5. Defining Global Variables ΓòÉΓòÉΓòÉ You may announce your intention to use a symbol as a global variable with a definition, using defconst or defvar. In Emacs Lisp, definitions serve three purposes. First, they inform the user who reads the code that certain symbols are intended to be used as variables. Second, they inform the Lisp system of these things, supplying a value and documentation. Third, they provide information to utilities such as etags and make-docfile, which create data bases of the functions and variables in a program. The difference between defconst and defvar is primarily a matter of intent, serving to inform human readers of whether programs will change the variable. Emacs Lisp does not restrict the ways in which a variable can be used based on defconst or defvar declarations. However, it also makes a difference for initialization: defconst unconditionally initializes the variable, while defvar initializes it only if it is void. One would expect user option variables to be defined with defconst, since programs do not change them. Unfortunately, this has bad results if the definition is in a library that is not preloaded: defconst would override any prior value when the library is loaded. Users would like to be able to set the option in their init files, and override the default value given in the definition. For this reason, user options must be defined with defvar. -- Special Form: defvar symbol [value [doc-string]] This special form informs a person reading your code that symbol will be used as a variable that the programs are likely to set or change. It is also used for all user option variables except in the preloaded parts of Emacs. Note that symbol is not evaluated; the symbol to be defined must appear explicitly in the defvar. If symbol already has a value (i.e., it is not void), value is not even evaluated, and symbol's value remains unchanged. If symbol is void and value is specified, it is evaluated and symbol is set to the result. (If value is not specified, the value of symbol is not changed in any case.) If the doc-string argument appears, it specifies the documentation for the variable. (This opportunity to specify documentation is one of the main benefits of defining the variable.) The documentation is stored in the symbol's variable-documentation property. The Emacs help functions (see Documentation) look for this property. If the first character of doc-string is `*', it means that this variable is considered to be a user option. This affects commands such as set-variable and edit-options. For example, this form defines foo but does not set its value: (defvar foo) => foo The following example sets the value of bar to 23, and gives it a documentation string: (defvar bar 23 "The normal weight of a bar.") => bar The following form changes the documentation string for bar, making it a user option, but does not change the value, since bar already has a value. (The addition (1+ 23) is not even performed.) (defvar bar (1+ 23) "*The normal weight of a bar.") => bar bar => 23 Here is an equivalent expression for the defvar special form: (defvar symbol value doc-string) == (progn (if (not (boundp 'symbol)) (setq symbol value)) (put 'symbol 'variable-documentation 'doc-string) 'symbol) The defvar form returns symbol, but it is normally used at top level in a file where its value does not matter. -- Special Form: defconst symbol [value [doc-string]] This special form informs a person reading your code that symbol has a global value, established here, that will not normally be changed or locally bound by the execution of the program. The user, however, may be welcome to change it. Note that symbol is not evaluated; the symbol to be defined must appear explicitly in the defconst. defconst always evaluates value and sets the global value of symbol to the result, provided value is given. *Note:* don't use defconst for user option variables in libraries that are not normally loaded. The user should be able to specify a value for such a variable in the `.emacs' file, so that it will be in effect if and when the library is loaded later. Here, pi is a constant that presumably ought not to be changed by anyone (attempts by the Indiana State Legislature notwithstanding). As the second form illustrates, however, this is only advisory. (defconst pi 3 "Pi to one place.") => pi (setq pi 4) => pi pi => 4 -- Function: user-variable-p variable This function returns t if variable is a user option, intended to be set by the user for customization, nil otherwise. (Variables other than user options exist for the internal purposes of Lisp programs, and users need not know about them.) User option variables are distinguished from other variables by the first character of the variable-documentation property. If the property exists and is a string, and its first character is `*', then the variable is a user option. Note that if the defconst and defvar special forms are used while the variable has a local binding, the local binding's value is set, and the global binding is not changed. This would be confusing. But the normal way to use these special forms is at top level in a file, where no local binding should be in effect. ΓòÉΓòÉΓòÉ 13.6. Accessing Variable Values ΓòÉΓòÉΓòÉ The usual way to reference a variable is to write the symbol which names it (see Symbol Forms). This requires you to specify the variable name when you write the program. Usually that is exactly what you want to do. Occasionally you need to choose at run time which variable to reference; then you can use symbol-value. -- Function: symbol-value symbol This function returns the value of symbol. This is the value in the innermost local binding of the symbol, or its global value if it has no local bindings. (setq abracadabra 5) => 5 (setq foo 9) => 9 ;; Here the symbol abracadabra ;; is the symbol whose value is examined. (let ((abracadabra 'foo)) (symbol-value 'abracadabra)) => foo ;; Here the value of abracadabra, ;; which is foo, ;; is the symbol whose value is examined. (let ((abracadabra 'foo)) (symbol-value abracadabra)) => 9 (symbol-value 'abracadabra) => 5 A void-variable error is signaled if symbol has neither a local binding nor a global value. ΓòÉΓòÉΓòÉ 13.7. How to Alter a Variable Value ΓòÉΓòÉΓòÉ The usual way to change the value of a variable is with the special form setq. When you need to compute the choice of variable at run time, use the function set. -- Special Form: setq [symbol form]... This special form is the most common method of changing a variable's value. Each symbol is given a new value, which is the result of evaluating the corresponding form. The most-local existing binding of the symbol is changed. The value of the setq form is the value of the last form. (setq x (1+ 2)) => 3 x ; x now has a global value. => 3 (let ((x 5)) (setq x 6) ; The local binding of x is set. x) => 6 x ; The global value is unchanged. => 3 Note that the first form is evaluated, then the first symbol is set, then the second form is evaluated, then the second symbol is set, and so on: (setq x 10 ; Notice that x is set y (1+ x)) ; before the value of y is computed. => 11 -- Function: set symbol value This function sets symbol's value to value, then returns value. Since set is a function, the expression written for symbol is evaluated to obtain the symbol to be set. The most-local existing binding of the variable is the binding that is set; shadowed bindings are not affected. If symbol is not actually a symbol, a wrong-type-argument error is signaled. (set one 1) error--> Symbol's value as variable is void: one (set 'one 1) => 1 (set 'two 'one) => one (set two 2) ; two evaluates to symbol one. => 2 one ; So it is one that was set. => 2 (let ((one 1)) ; This binding of one is set, (set 'one 3) ; not the global value. one) => 3 one => 2 Logically speaking, set is a more fundamental primitive that setq. Any use of setq can be trivially rewritten to use set; setq could even be defined as a macro, given the availability of set. However, set itself is rarely used; beginners hardly need to know about it. It is needed only when the choice of variable to be set is made at run time. For example, the command set-variable, which reads a variable name from the user and then sets the variable, needs to use set. Common Lisp note: in Common Lisp, set always changes the symbol's special value, ignoring any lexical bindings. In Emacs Lisp, all variables and all bindings are special, so set always affects the most local existing binding. ΓòÉΓòÉΓòÉ 13.8. Scoping Rules for Variable Bindings ΓòÉΓòÉΓòÉ A given symbol foo may have several local variable bindings, established at different places in the Lisp program, as well as a global binding. The most recently established binding takes precedence over the others. Local bindings in Emacs Lisp have indefinite scope and dynamic extent. Scope refers to where textually in the source code the binding can be accessed. Indefinite scope means that any part of the program can potentially access the variable binding. Extent refers to when, as the program is executing, the binding exists. Dynamic extent means that the binding lasts as long as the activation of the construct that established it. The combination of dynamic extent and indefinite scope is called dynamic scoping. By contrast, most programming languages use lexical scoping, in which references to a local variable must be textually within the function or block that binds the variable. Common Lisp note: variables declared ``special'' in Common Lisp are dynamically scoped like variables in Emacs Lisp. ΓòÉΓòÉΓòÉ 13.8.1. Scope ΓòÉΓòÉΓòÉ Emacs Lisp uses indefinite scope for local variable bindings. This means that any function anywhere in the program text might access a given binding of a variable. Consider the following function definitions: (defun binder (x) ; x is bound in binder. (foo 5)) ; foo is some other function. (defun user () ; x is used in user. (list x)) In a lexically scoped language, the binding of x from binder would never be accessible in user, because user is not textually contained within the function binder. However, in dynamically scoped Emacs Lisp, user may or may not refer to the binding of x established in binder, depending on circumstances: o If we call user directly without calling binder at all, then whatever binding of x is found, it cannot come from binder. o If we define foo as follows and call binder, then the binding made in binder will be seen in user: (defun foo (lose) (user)) o If we define foo as follows and call binder, then the binding made in binder will not be seen in user: (defun foo (x) (user)) Here, when foo is called by binder, it binds x. (The binding in foo is said to shadow the one made in binder.) Therefore, user will access the x bound by foo instead of the one bound by binder. ΓòÉΓòÉΓòÉ 13.8.2. Extent ΓòÉΓòÉΓòÉ Extent refers to the time during program execution that a variable name is valid. In Emacs Lisp, a variable is valid only while the form that bound it is executing. This is called dynamic extent. ``Local'' or ``automatic'' variables in most languages, including C and Pascal, have dynamic extent. One alternative to dynamic extent is indefinite extent. This means that a variable binding can live on past the exit from the form that made the binding. Common Lisp and Scheme, for example, support this, but Emacs Lisp does not. To illustrate this, the function below, make-add, returns a function that purports to add n to its own argument m. This would work in Common Lisp, but it does not work as intended in Emacs Lisp, because after the call to make-add exits, the variable n is no longer bound to the actual argument 2. (defun make-add (n) (function (lambda (m) (+ n m)))) ; Return a function. => make-add (fset 'add2 (make-add 2)) ; Define function add2 with (make-add 2). => (lambda (m) (+ n m)) (add2 4) ; Try to add 2 to 4. error--> Symbol's value as variable is void: n ΓòÉΓòÉΓòÉ 13.8.3. Implementation of Dynamic Scoping ΓòÉΓòÉΓòÉ A simple sample implementation (which is not how Emacs Lisp actually works) may help you understand dynamic binding. This technique is called deep binding and was used in early Lisp systems. Suppose there is a stack of bindings: variable-value pairs. At entry to a function or to a let form, we can push bindings on the stack for the arguments or local variables created there. We can pop those bindings from the stack at exit from the binding construct. We can find the value of a variable by searching the stack from top to bottom for a binding for that variable; the value from that binding is the value of the variable. To set the variable, we search for the current binding, then store the new value into that binding. As you can see, a function's bindings remain in effect as long as it continues execution, even during its calls to other functions. That is why we say the extent of the binding is dynamic. And any other function can refer to the bindings, if it uses the same variables while the bindings are in effect. That is why we say the scope is indefinite. The actual implementation of variable scoping in GNU Emacs Lisp uses a technique called shallow binding. Each variable has a standard place in which its current value is always found---the value cell of the symbol. In shallow binding, setting the variable works by storing a value in the value cell. When a new local binding is created, the local value is stored in the value cell, and the old value (belonging to a previous binding) is pushed on a stack. When a binding is eliminated, the old value is popped off the stack and stored in the value cell. We use shallow binding because it has the same results as deep binding, but runs faster, since there is never a need to search for a binding. ΓòÉΓòÉΓòÉ 13.8.4. Proper Use of Dynamic Scoping ΓòÉΓòÉΓòÉ Binding a variable in one function and using it in another is a powerful technique, but if used without restraint, it can make programs hard to understand. There are two clean ways to use this technique: o Use or bind the variable only in a few related functions, written close together in one file. Such a variable is used for communication within one program. You should write comments to inform other programmers that they can see all uses of the variable before them, and to advise them not to add uses elsewhere. o Give the variable a well-defined, documented meaning, and make all appropriate functions refer to it (but not bind it or set it) wherever that meaning is relevant. For example, the variable case-fold-search is defined as ``non-nil means ignore case when searching''; various search and replace functions refer to it directly or through their subroutines, but do not bind or set it. Then you can bind the variable in other programs, knowing reliably what the effect will be. ΓòÉΓòÉΓòÉ 13.9. Buffer-Local Variables ΓòÉΓòÉΓòÉ Global and local variable bindings are found in most programming languages in one form or another. Emacs also supports another, unusual kind of variable binding: buffer-local bindings, which apply only to one buffer. Emacs Lisp is meant for programming editing commands, and having different values for a variable in different buffers is an important customization method. ΓòÉΓòÉΓòÉ 13.9.1. Introduction to Buffer-Local Variables ΓòÉΓòÉΓòÉ A buffer-local variable has a buffer-local binding associated with a particular buffer. The binding is in effect when that buffer is current; otherwise, it is not in effect. If you set the variable while a buffer-local binding is in effect, the new value goes in that binding, so the global binding is unchanged; this means that the change is visible in that buffer alone. A variable may have buffer-local bindings in some buffers but not in others. The global binding is shared by all the buffers that don't have their own bindings. Thus, if you set the variable in a buffer that does not have a buffer-local binding for it, the new value is visible in all buffers except those with buffer-local bindings. (Here we are assuming that there are no let-style local bindings to complicate the issue.) The most common use of buffer-local bindings is for major modes to change variables that control the behavior of commands. For example, C mode and Lisp mode both set the variable paragraph-start to specify that only blank lines separate paragraphs. They do this by making the variable buffer-local in the buffer that is being put into C mode or Lisp mode, and then setting it to the new value for that mode. The usual way to make a buffer-local binding is with make-local-variable, which is what major mode commands use. This affects just the current buffer; all other buffers (including those yet to be created) continue to share the global value. A more powerful operation is to mark the variable as automatically buffer-local by calling make-variable-buffer-local. You can think of this as making the variable local in all buffers, even those yet to be created. More precisely, the effect is that setting the variable automatically makes the variable local to the current buffer if it is not already so. All buffers start out by sharing the global value of the variable as usual, but any setq creates a buffer-local binding for the current buffer. The new value is stored in the buffer-local binding, leaving the (default) global binding untouched. The global value can no longer be changed with setq; you need to use setq-default to do that. When a variable has local values in one or more buffers, you can get Emacs very confused by binding the variable with let, changing to a different current buffer in which a different binding is in effect, and then exiting the let. The best way to preserve your sanity is to avoid such situations. If you use save-excursion around each piece of code that changes to a different current buffer, you will not have this problem. Here is an example of incorrect code: (setq foo 'b) (set-buffer "a") (make-local-variable 'foo) (setq foo 'a) (let ((foo 'temp)) (set-buffer "b") ...) foo => 'a ; The old buffer-local value from buffer `a' ; is now the default value. (set-buffer "a") foo => 'temp ; The local value that should be gone ; is now the buffer-local value in buffer `a'. But save-excursion as shown here avoids the problem: (let ((foo 'temp)) (save-excursion (set-buffer "b") ...)) Local variables in a file you edit are also represented by buffer-local bindings for the buffer that holds the file within Emacs. See Auto Major Mode. ΓòÉΓòÉΓòÉ 13.9.2. Creating and Destroying Buffer-local Bindings ΓòÉΓòÉΓòÉ -- Command: make-local-variable symbol This function creates a buffer-local binding for symbol in the current buffer. Other buffers are not affected. The value returned is symbol. The buffer-local value of symbol starts out as the same value symbol previously had. ;; In buffer `b1': (setq foo 5) ; Affects all buffers. => 5 (make-local-variable 'foo) ; Now it is local in `b1'. => foo foo ; That did not change the value. => 5 (setq foo 6) ; Change the value in `b1'. => 6 foo => 6 ;; In buffer `b2', the value hasn't changed. (save-excursion (set-buffer "b2") foo) => 5 -- Command: make-variable-buffer-local symbol This function marks symbol automatically buffer-local, so that any attempt to set it will make it local to the current buffer at the time. The value returned is symbol. -- Function: buffer-local-variables &optional buffer This function tells you what the buffer-local variables are in buffer buffer. It returns an association list ( see Association Lists) in which each association contains one buffer-local variable and its value. If buffer is omitted, the current buffer is used. (setq lcl (buffer-local-variables)) => ((fill-column . 75) (case-fold-search . t) ... (mark-ring #<marker at 5454 in buffers.texi>) (require-final-newline . t)) Note that storing new values into the cdrs of the elements in this list will not change the local values of the variables. -- Command: kill-local-variable symbol This function deletes the buffer-local binding (if any) for symbol in the current buffer. As a result, the global (default) binding of symbol becomes visible in this buffer. Usually this results in a change in the value of symbol, since the global value is usually different from the buffer-local value just eliminated. It is possible to kill the local binding of a variable that automatically becomes local when set. This causes the variable to show its global value in the current buffer. However, if you set the variable again, this will once again create a local value. kill-local-variable returns symbol. -- Function: kill-all-local-variables This function eliminates all the buffer-local variable bindings of the current buffer. As a result, the buffer will see the default values of all variables. This function also resets certain other information pertaining to the buffer: its local keymap is set to nil, its syntax table is set to the value of standard-syntax-table, and its abbrev table is set to the value of fundamental-mode-abbrev-table. Every major mode command begins by calling this function, which has the effect of switching to Fundamental mode and erasing most of the effects of the previous major mode. kill-all-local-variables returns nil. ΓòÉΓòÉΓòÉ 13.9.3. The Default Value of a Buffer-Local Variable ΓòÉΓòÉΓòÉ The global value of a variable with buffer-local bindings is also called the default value, because it is the value that is in effect except when specifically overridden. The functions default-value and setq-default allow you to access and change the default value regardless of whether the current buffer has a buffer-local binding. For example, you could use setq-default to change the default setting of paragraph-start for most buffers; and this would work even when you are in a C or Lisp mode buffer which has a buffer-local value for this variable. -- Function: default-value symbol This function returns symbol's default value. This is the value that is seen in buffers that do not have their own values for this variable. If symbol is not buffer-local, this is equivalent to symbol-value (see Accessing Variables). -- Special Form: setq-default symbol value This sets the default value of symbol to value. symbol is not evaluated, but value is. The value of the setq-default form is value. If a symbol is not buffer-local for the current buffer, and is not marked automatically buffer-local, this has the same effect as setq. If symbol is buffer-local for the current buffer, then this changes the value that other buffers will see (as long as they don't have a buffer-local value), but not the value that the current buffer sees. ;; In buffer `foo': (make-local-variable 'local) => local (setq local 'value-in-foo) => value-in-foo (setq-default local 'new-default) => new-default local => value-in-foo (default-value 'local) => new-default ;; In (the new) buffer `bar': local => new-default (default-value 'local) => new-default (setq local 'another-default) => another-default (default-value 'local) => another-default ;; Back in buffer `foo': local => value-in-foo (default-value 'local) => another-default -- Function: set-default symbol value This function is like setq-default, except that symbol is evaluated. (set-default (car '(a b c)) 23) => 23 (default-value 'a) => 23 ΓòÉΓòÉΓòÉ 14. Functions ΓòÉΓòÉΓòÉ A Lisp program is composed mainly of Lisp functions. This chapter explains what functions are, how they accept arguments, and how to define them. ΓòÉΓòÉΓòÉ 14.1. What Is a Function? ΓòÉΓòÉΓòÉ In a general sense, a function is a rule for carrying on a computation given several values called arguments. The result of the computation is called the value of the function. The computation can also have side effects: lasting changes in the values of variables or the contents of data structures. Here are important terms for functions in Emacs Lisp and for other function-like objects. function In Emacs Lisp, a function is anything that can be applied to arguments in a Lisp program. In some cases, we will use it more specifically to mean a function written in Lisp. Special forms and macros are not functions. primitive A primitive is a function callable from Lisp that is written in C, such as car or append. These functions are also called built-in functions or subrs. (Special forms are also considered primitives.) Primitives provide the lowest-level interfaces to editing functions or operating system services, or in a few cases they perform important operations more quickly than a Lisp program could. Primitives can be modified or added only by changing the C sources and recompiling the editor. See Writing Emacs Primitives. lambda expression A lambda expression is a function written in Lisp. These are described in the following section. See Lambda Expressions. special form A special form is a primitive that is like a function but does not evaluate all of its arguments in the usual way. It may evaluate only some of the arguments, or may evaluate them in an unusual order, or several times. Many special forms are described in Control Structures. macro A macro is a construct defined in Lisp by the programmer. It differs from a function in that it translates a Lisp expression that you write into an equivalent expression to be evaluated instead of the original expression. See Macros, for how to define and use macros. command A command is an object that command-execute can invoke; it is a possible definition for a key sequence. Some functions are commands; a function written in Lisp is a command if it contains an interactive declaration (see Defining Commands). Such a function can be called from Lisp expressions like other functions; in this case, the fact that the function is a command makes no difference. Strings are commands also, even though they are not functions. A symbol is a command if its function definition is a command; such symbols can be invoked with M-x. The symbol is a function as well if the definition is a function. See Command Overview. keystroke command A keystroke command is a command that is bound to a key sequence (typically one to three keystrokes). The distinction is made here merely to avoid confusion with the meaning of ``command'' in non-Emacs editors; for programmers, the distinction is normally unimportant. -- Function: subrp object This function returns t if object is a built-in function (i.e. a Lisp primitive). (subrp 'message) ; message is a symbol, => nil ; not a subr object. (subrp (symbol-function 'message)) => t ΓòÉΓòÉΓòÉ 14.2. Lambda Expressions ΓòÉΓòÉΓòÉ A function written in Lisp is a list that looks like this: (lambda (arg-variables...) [documentation-string] [interactive-declaration] body-forms...) (Such a list is called a lambda expression, even though it is not an expression at all, for historical reasons.) ΓòÉΓòÉΓòÉ 14.2.1. Components of a Lambda Expression ΓòÉΓòÉΓòÉ A function written in Lisp (a ``lambda expression'') is a list that looks like this: (lambda (arg-variables...) [documentation-string] [interactive-declaration] body-forms...) The first element of a lambda expression is always the symbol lambda. This indicates that the list represents a function. The reason functions are defined to start with lambda is so that other lists, intended for other uses, will not accidentally be valid as functions. The second element is a list of argument variable names (symbols). This is called the lambda list. When a Lisp function is called, the argument values are matched up against the variables in the lambda list, which are given local bindings with the values provided. See Local Variables. The documentation string is an actual string that serves to describe the function for the Emacs help facilities. See Function Documentation. The interactive declaration is a list of the form (interactive code-string). This declares how to provide arguments if the function is used interactively. Functions with this declaration are called commands; they can be called using M-x or bound to a key. Functions not intended to be called in this way should not have interactive declarations. See Defining Commands, for how to write an interactive declaration. The rest of the elements are the body of the function: the Lisp code to do the work of the function (or, as a Lisp programmer would say, ``a list of Lisp forms to evaluate''). The value returned by the function is the value returned by the last element of the body. ΓòÉΓòÉΓòÉ 14.2.2. A Simple Lambda-Expression Example ΓòÉΓòÉΓòÉ Consider for example the following function: (lambda (a b c) (+ a b c)) We can call this function by writing it as the car of an expression, like this: ((lambda (a b c) (+ a b c)) 1 2 3) The body of this lambda expression is evaluated with the variable a bound to 1, b bound to 2, and c bound to 3. Evaluation of the body adds these three numbers, producing the result 6; therefore, this call to the function returns the value 6. Note that the arguments can be the results of other function calls, as in this example: ((lambda (a b c) (+ a b c)) 1 (* 2 3) (- 5 4)) Here all the arguments 1, (* 2 3), and (- 5 4) are evaluated, left to right. Then the lambda expression is applied to the argument values 1, 6 and 1 to produce the value 8. It is not often useful to write a lambda expression as the car of a form in this way. You can get the same result, of making local variables and giving them values, using the special form let (see Local Variables). And let is clearer and easier to use. In practice, lambda expressions are either stored as the function definitions of symbols, to produce named functions, or passed as arguments to other functions (see Anonymous Functions). However, calls to explicit lambda expressions were very useful in the old days of Lisp, before the special form let was invented. At that time, they were the only way to bind and initialize local variables. ΓòÉΓòÉΓòÉ 14.2.3. Advanced Features of Argument Lists ΓòÉΓòÉΓòÉ Our simple sample function, (lambda (a b c) (+ a b c)), specifies three argument variables, so it must be called with three arguments: if you try to call it with only two arguments or four arguments, you will get a wrong-number-of-arguments error. It is often convenient to write a function that allows certain arguments to be omitted. For example, the function substring accepts three arguments---a string, the start index and the end index---but the third argument defaults to the end of the string if you omit it. It is also convenient for certain functions to accept an indefinite number of arguments, as the functions and and + do. To specify optional arguments that may be omitted when a function is called, simply include the keyword &optional before the optional arguments. To specify a list of zero or more extra arguments, include the keyword &rest before one final argument. Thus, the complete syntax for an argument list is as follows: (required-vars... [&optional optional-vars...] [&rest rest-var]) The square brackets indicate that the &optional and &rest clauses, and the variables that follow them, are optional. A call to the function requires one actual argument for each of the required-vars. There may be actual arguments for zero or more of the optional-vars, and there cannot be any more actual arguments than these unless &rest exists. In that case, there may be any number of extra actual arguments. If actual arguments for the optional and rest variables are omitted, then they always default to nil. However, the body of the function is free to consider nil an abbreviation for some other meaningful value. This is what substring does; nil as the third argument means to use the length of the string supplied. There is no way for the function to distinguish between an explicit argument of nil and an omitted argument. Common Lisp note: Common Lisp allows the function to specify what default values will be used when an optional argument is omitted; GNU Emacs Lisp always uses nil. For example, an argument list that looks like this: (a b &optional c d &rest e) binds a and b to the first two actual arguments, which are required. If one or two more arguments are provided, c and d are bound to them respectively; any arguments after the first four are collected into a list and e is bound to that list. If there are only two arguments, c is nil; if two or three arguments, d is nil; if four arguments or fewer, e is nil. There is no way to have required arguments following optional ones---it would not make sense. To see why this must be so, suppose that c in the example were optional and d were required. If three actual arguments are given; then which variable would the third argument be for? Similarly, it makes no sense to have any more arguments (either required or optional) after a &rest argument. Here are some examples of argument lists and proper calls: ((lambda (n) (1+ n)) ; One required: 1) ; requires exactly one argument. => 2 ((lambda (n &optional n1) ; One required and one optional: (if n1 (+ n n1) (1+ n))) ; 1 or 2 arguments. 1 2) => 3 ((lambda (n &rest ns) ; One required and one rest: (+ n (apply '+ ns))) ; 1 or more arguments. 1 2 3 4 5) => 15 ΓòÉΓòÉΓòÉ 14.2.4. Documentation Strings of Functions ΓòÉΓòÉΓòÉ A lambda expression may optionally have a documentation string just after the lambda list. This string does not affect execution of the function; it is a kind of comment, but a systematized comment which actually appears inside the Lisp world and can be used by the Emacs help facilities. See Documentation, for how the documentation-string is accessed. It is a good idea to provide documentation strings for all commands, and for all other functions in your program that users of your program should know about; internal functions might as well have only comments, since comments don't take up any room when your program is loaded. The first line of the documentation string should stand on its own, because apropos displays just this first line. It should consist of one or two complete sentences that summarize the function's purpose. The start of the documentation string is usually indented, but since these spaces come before the starting double-quote, they are not part of the string. Some people make a practice of indenting any additional lines of the string so that the text lines up. This is a mistake. The indentation of the following lines is inside the string; what looks nice in the source code will look ugly when displayed by the help commands. You may wonder how the documentation string could be optional, since there are required components of the function that follow it (the body). Since evaluation of a string returns that string, without any side effects, it has no effect if it is not the last form in the body. Thus, in practice, there is no confusion between the first form of the body and the documentation string; if the only body form is a string then it serves both as the return value and as the documentation. ΓòÉΓòÉΓòÉ 14.3. Naming a Function ΓòÉΓòÉΓòÉ In most computer languages, every function has a name; the idea of a function without a name is nonsensical. In Lisp, a function in the strictest sense has no name. It is simply a list whose first element is lambda, or a primitive subr-object. However, a symbol can serve as the name of a function. This happens when you put the function in the symbol's function cell (see Symbol Components). Then the symbol itself becomes a valid, callable function, equivalent to the list or subr-object that its function cell refers to. The contents of the function cell are also called the symbol's function definition. In practice, nearly all functions are given names in this way and referred to through their names. For example, the symbol car works as a function and does what it does because the primitive subr-object #<subr car> is stored in its function cell. We give functions names because it is more convenient to refer to them by their names in other functions. For primitive subr-objects such as #<subr car>, names are the only way you can refer to them: there is no read syntax for such objects. For functions written in Lisp, the name is more convenient to use in a call than an explicit lambda expression. Also, a function with a name can refer to itself---it can be recursive. Writing the function's name in its own definition is much more convenient than making the function definition point to itself (something that is not impossible but that has various disadvantages in practice). Functions are often identified with the symbols used to name them. For example, we often speak of ``the function car'', not distinguishing between the symbol car and the primitive subr-object that is its function definition. For most purposes, there is no need to distinguish. Even so, keep in mind that a function need not have a unique name. While a given function object usually appears in the function cell of only one symbol, this is just a matter of convenience. It is very easy to store it in several symbols using fset; then each of the symbols is equally well a name for the same function. A symbol used as a function name may also be used as a variable; these two uses of a symbol are independent and do not conflict. ΓòÉΓòÉΓòÉ 14.4. Defining Named Functions ΓòÉΓòÉΓòÉ We usually give a name to a function when it is first created. This is called defining a function, and it is done with the defun special form. -- Special Form: defun name argument-list body-forms defun is the usual way to define new Lisp functions. It defines the symbol name as a function that looks like this: (lambda argument-list . body-forms) This lambda expression is stored in the function cell of name. The value returned by evaluating the defun form is name, but usually we ignore this value. As described previously (see Lambda Expressions), argument-list is a list of argument names and may include the keywords &optional and &rest. Also, the first two forms in body-forms may be a documentation string and an interactive declaration. Note that the same symbol name may also be used as a global variable, since the value cell is independent of the function cell. Here are some examples: (defun foo () 5) => foo (foo) => 5 (defun bar (a &optional b &rest c) (list a b c)) => bar (bar 1 2 3 4 5) => (1 2 (3 4 5)) (bar 1) => (1 nil nil) (bar) error--> Wrong number of arguments. (defun capitalize-backwards () "This function makes the last letter of a word upper case." (interactive) (backward-word 1) (forward-word 1) (backward-char 1) (capitalize-word 1)) => capitalize-backwards Be careful not to redefine existing functions unintentionally. defun will redefine even primitive functions such as car without any hesitation or notification. Redefining a function already defined is often done deliberately, and there is no way to distinguish deliberate redefinition from unintentional redefinition. ΓòÉΓòÉΓòÉ 14.5. Calling Functions ΓòÉΓòÉΓòÉ Defining functions is only half the battle. Functions don't do anything until you call them, i.e., tell them to run. This process is also known as invocation. The most common way of invoking a function is by evaluating a list. For example, evaluating the list (concat "a" "b") calls the function concat. See Evaluation, for a description of evaluation. When you write a list as an expression in your program, the function name is part of the program. This means that the choice of which function to call is made when you write the program. Usually that's just what you want. Occasionally you need to decide at run time which function to call. Then you can use the functions funcall and apply. -- Function: funcall function &rest arguments funcall calls function with arguments, and returns whatever function returns. Since funcall is a function, all of its arguments, including function, are evaluated before funcall is called. This means that you can use any expression to obtain the function to be called. It also means that funcall does not see the expressions you write for the arguments, only their values. These values are not evaluated a second time in the act of calling function; funcall enters the normal procedure for calling a function at the place where the arguments have already been evaluated. The argument function must be either a Lisp function or a primitive function. Special forms and macros are not allowed, because they make sense only when given the ``unevaluated'' argument expressions. funcall cannot provide these because, as we saw above, it never knows them in the first place. (setq f 'list) => list (funcall f 'x 'y 'z) => (x y z) (funcall f 'x 'y '(z)) => (x y (z)) (funcall 'and t nil) error--> Invalid function: #<subr and> Compare this example with that of apply. -- Function: apply function &rest arguments apply calls function with arguments, just like funcall but with one difference: the last of arguments is a list of arguments to give to function, rather than a single argument. We also say that this list is appended to the other arguments. apply returns the result of calling function. As with funcall, function must either be a Lisp function or a primitive function; special forms and macros do not make sense in apply. (setq f 'list) => list (apply f 'x 'y 'z) error--> Wrong type argument: listp, z (apply '+ 1 2 '(3 4)) => 10 (apply '+ '(1 2 3 4)) => 10 (apply 'append '((a b c) nil (x y z) nil)) => (a b c x y z) An interesting example of using apply is found in the description of mapcar; see the following section. It is common for Lisp functions to accept functions as arguments or find them in data structures (especially in hook variables and property lists) and call them using funcall or apply. Functions that accept function arguments are often called functionals. Sometimes, when you call such a function, it is useful to supply a no-op function as the argument. Here are two different kinds of no-op function: -- Function: identity arg This function returns arg and has no side effects. -- Function: ignore &rest args This function ignores any arguments and returns nil. ΓòÉΓòÉΓòÉ 14.6. Mapping Functions ΓòÉΓòÉΓòÉ A mapping function applies a given function to each element of a list or other collection. Emacs Lisp has three such functions; mapcar and mapconcat, which scan a list, are described here. For the third mapping function, mapatoms, see Creating Symbols. -- Function: mapcar function sequence mapcar applies function to each element of sequence in turn. The results are made into a nil-terminated list. The argument sequence may be a list, a vector or a string. The result is always a list. The length of the result is the same as the length of sequence. For example: (mapcar 'car '((a b) (c d) (e f))) => (a c e) (mapcar '1+ [1 2 3]) => (2 3 4) (mapcar 'char-to-string "abc") => ("a" "b" "c") ;; Call each function in my-hooks. (mapcar 'funcall my-hooks) (defun mapcar* (f &rest args) "Apply FUNCTION to successive cars of all ARGS, until one ends. Return the list of results." (if (not (memq 'nil args)) ; If no list is exhausted, (cons (apply f (mapcar 'car args)) ; Apply function to CARs. (apply 'mapcar* f ; Recurse for rest of elements. (mapcar 'cdr args))))) (mapcar* 'cons '(a b c) '(1 2 3 4)) => ((a . 1) (b . 2) (c . 3)) -- Function: mapconcat function sequence separator mapconcat applies function to each element of sequence: the results, which must be strings, are concatenated. Between each pair of result strings, mapconcat inserts the string separator. Usually separator contains a space or comma or other suitable punctuation. The argument function must be a function that can take one argument and returns a string. (mapconcat 'symbol-name '(The cat in the hat) " ") => "The cat in the hat" (mapconcat (function (lambda (x) (format "%c" (1+ x)))) "HAL-8000" "") => "IBM.9111" ΓòÉΓòÉΓòÉ 14.7. Anonymous Functions ΓòÉΓòÉΓòÉ In Lisp, a function is a list that starts with lambda (or alternatively a primitive subr-object); names are ``extra''. Although usually functions are defined with defun and given names at the same time, it is occasionally more concise to use an explicit lambda expression---an anonymous function. Such a list is valid wherever a function name is. Any method of creating such a list makes a valid function. Even this: (setq silly (append '(lambda (x)) (list (list '+ (* 3 4) 'x)))) => (lambda (x) (+ 12 x)) This computes a list that looks like (lambda (x) (+ 12 x)) and makes it the value (not the function definition!) of silly. Here is how we might call this function: (funcall silly 1) => 13 (It does not work to write (silly 1), because this function is not the function definition of silly. We have not given silly any function definition, just a value as a variable.) Most of the time, anonymous functions are constants that appear in your program. For example, you might want to pass one as an argument to the function mapcar, which applies any given function to each element of a list. Here we pass an anonymous function that multiplies a number by two: (defun double-each (list) (mapcar '(lambda (x) (* 2 x)) list)) => double-each (double-each '(2 11)) => (4 22) In such cases, we usually use the special form function instead of simple quotation to quote the anonymous function. -- Special Form: function function-object This special form returns function-object without evaluating it. In this, it is equivalent to quote. However, it serves as a note to the Emacs Lisp compiler that function-object is intended to be used only as a function, and therefore can safely be compiled. See Quoting, for comparison. Using function instead of quote makes a difference inside a function or macro that you are going to compile. For example: (defun double-each (list) (mapcar (function (lambda (x) (* 2 x))) list)) => double-each (double-each '(2 11)) => (4 22) If this definition of double-each is compiled, the anonymous function is compiled as well. By contrast, in the previous definition where ordinary quote is used, the argument passed to mapcar is the precise list shown: (lambda (arg) (+ arg 5)) The Lisp compiler cannot assume this list is a function, even though it looks like one, since it does not know what mapcar does with the list. Perhaps mapcar will check that the car of the third element is the symbol +! The advantage of function is that it tells the compiler to go ahead and compile the constant function. We sometimes write function instead of quote when quoting the name of a function, but this usage is just a sort of comment. (function symbol) == (quote symbol) == 'symbol See documentation in Accessing Documentation, for a realistic example using function and an anonymous function. ΓòÉΓòÉΓòÉ 14.8. Accessing Function Cell Contents ΓòÉΓòÉΓòÉ The function definition of a symbol is the object stored in the function cell of the symbol. The functions described here access, test, and set the function cell of symbols. -- Function: symbol-function symbol This returns the object in the function cell of symbol. If the symbol's function cell is void, a void-function error is signaled. This function does not check that the returned object is a legitimate function. (defun bar (n) (+ n 2)) => bar (symbol-function 'bar) => (lambda (n) (+ n 2)) (fset 'baz 'bar) => bar (symbol-function 'baz) => bar If you have never given a symbol any function definition, we say that that symbol's function cell is void. In other words, the function cell does not have any Lisp object in it. If you try to call such a symbol as a function, it signals a void-function error. Note that void is not the same as nil or the symbol void. The symbols nil and void are Lisp objects, and can be stored into a function cell just as any other object can be (and they can be valid functions if you define them in turn with defun); but nil or void is an object. A void function cell contains no object whatsoever. You can test the voidness of a symbol's function definition with fboundp. After you have given a symbol a function definition, you can make it void once more using fmakunbound. -- Function: fboundp symbol Returns t if the symbol has an object in its function cell, nil otherwise. It does not check that the object is a legitimate function. -- Function: fmakunbound symbol This function makes symbol's function cell void, so that a subsequent attempt to access this cell will cause a void-function error. (See also makunbound, in Local Variables.) (defun foo (x) x) => x (fmakunbound 'foo) => x (foo 1) error--> Symbol's function definition is void: foo -- Function: fset symbol object This function stores object in the function cell of symbol. The result is object. Normally object should be a function or the name of a function, but this is not checked. There are three normal uses of this function: o Copying one symbol's function definition to another. (In other words, making an alternate name for a function.) o Giving a symbol a function definition that is not a list and therefore cannot be made with defun. See Classifying Lists, for an example of this usage. o In constructs for defining or altering functions. If defun were not a primitive, it could be written in Lisp (as a macro) using fset. Here are examples of the first two uses: ;; Give first the same definition car has. (fset 'first (symbol-function 'car)) => #<subr car> (first '(1 2 3)) => 1 ;; Make the symbol car the function definition of xfirst. (fset 'xfirst 'car) => car (xfirst '(1 2 3)) => 1 (symbol-function 'xfirst) => car (symbol-function (symbol-function 'xfirst)) => #<subr car> ;; Define a named keyboard macro. (fset 'kill-two-lines "\^u2\^k") => "\^u2\^k" When writing a function that extends a previously defined function, the following idiom is often used: (fset 'old-foo (symbol-function 'foo)) (defun foo () "Just like old-foo, except more so." (old-foo) (more-so)) This does not work properly if foo has been defined to autoload. In such a case, when foo calls old-foo, Lisp will attempt to define old-foo by loading a file. Since this presumably defines foo rather than old-foo, it will not produce the proper results. The only way to avoid this problem is to make sure the file is loaded before moving aside the old definition of foo. ΓòÉΓòÉΓòÉ 14.9. Other Topics Related to Functions ΓòÉΓòÉΓòÉ Here is a table of several functions that do things related to function calling and function definitions. apply See Calling Functions. autoload See Autoload. call-interactively See Interactive Call. commandp See Interactive Call. documentation See Accessing Documentation. eval See Eval. funcall See Calling Functions. ignore See Calling Functions. interactive See Using Interactive. interactive-p See Interactive Call. mapatoms See Creating Symbols. mapcar See Mapping Functions. mapconcat See Mapping Functions. undefined See Key Lookup. ΓòÉΓòÉΓòÉ 15. Macros ΓòÉΓòÉΓòÉ Macros enable you to define new control constructs and other language features. A macro is defined much like a function, but instead of telling how to compute a value, it tells how to compute another Lisp expression which will in turn compute the value. We call this expression the expansion of the macro. Macros can do this because they operate on the unevaluated expressions for the arguments, not on the argument values as functions do. They can therefore construct an expansion containing these argument expressions or parts of them. ΓòÉΓòÉΓòÉ 15.1. A Simple Example of a Macro ΓòÉΓòÉΓòÉ Suppose we would like to define a Lisp construct to increment a variable value, much like the ++ operator in C. We would like to write (inc x) and have the effect of (setq x (1+ x)). Here's a macro definition that will do the job: (defmacro inc (var) (list 'setq var (list '1+ var))) When this is called with (inc x), the argument var has the value x---not the value of x. The body of the macro uses this to construct the expansion, which is (setq x (1+ x)). Once the macro definition returns this expansion, Lisp proceeds to evaluate it, thus incrementing x. ΓòÉΓòÉΓòÉ 15.2. Expansion of a Macro Call ΓòÉΓòÉΓòÉ A macro call looks just like a function call in that it is a list which starts with the name of the macro. The rest of the elements of the list are the arguments of the macro. Evaluation of the macro call begins like evaluation of a function call except for one crucial difference: the macro arguments are the actual expressions appearing in the macro call. They are not evaluated before they are given to the macro definition. By contrast, the arguments of a function are results of evaluating the elements of the function call list. Having obtained the arguments, Lisp invokes the macro definition just as a function is invoked. The argument variables of the macro are bound to the argument values from the macro call, or to a list of them in the case of a &rest argument. And the macro body executes and returns its value just as a function body does. The second crucial difference between macros and functions is that the value returned by the macro body is not the value of the macro call. Instead, it is an alternate expression for computing that value, also known as the expansion of the macro. The Lisp interpreter proceeds to evaluate the expansion as soon as it comes back from the macro. Since the expansion is evaluated in the normal manner, it may contain calls to other macros. It may even be a call to the same macro, though this is unusual. You can see the expansion of a given macro call by calling macroexpand: -- Function: macroexpand form &optional environment This function expands form, if it is a macro call. If the result is another macro call, it is expanded in turn, until something which is not a macro call results. That is the value returned by macroexpand. If form is not a macro call to begin with, it is returned as given. Note that macroexpand does not look at the subexpressions of form (although some macro definitions may do so). If they are macro calls themselves, macroexpand will not expand them. If environment is provided, it specifies an alist of macro definitions that shadow the currently defined macros. This is used by byte compilation. (defmacro inc (var) (list 'setq var (list '1+ var))) => inc (macroexpand '(inc r)) => (setq r (1+ r)) (defmacro inc2 (var1 var2) (list 'progn (list 'inc var1) (list 'inc var2))) => inc2 (macroexpand '(inc2 r s)) => (progn (inc r) (inc s)) ; inc not expanded here. ΓòÉΓòÉΓòÉ 15.3. Macros and Byte Compilation ΓòÉΓòÉΓòÉ You might ask why we take the trouble to compute an expansion for a macro and then evaluate the expansion. Why not have the macro body produce the desired results directly? The reason has to do with compilation. When a macro call appears in a Lisp program being compiled, the Lisp compiler calls the macro definition just as the interpreter would, and receives an expansion. But instead of evaluating this expansion, it compiles expansion as if it had appeared directly in the program. As a result, the compiled code produces the value and side effects intended for the macro, but executes at full compiled speed. This would not work if the macro body computed the value and side effects itself---they would be computed at compile time, which is not useful. In order for compilation of macro calls to work, the macros must be defined in Lisp when the calls to them are compiled. The compiler has a special feature to help you do this: if a file being compiled contains a defmacro form, the macro is defined temporarily for the rest of the compilation of that file. To use this feature, you must define the macro in the same file where it is used and before its first use. While byte-compiling a file, any require calls at top-level are executed. One way to ensure that necessary macro definitions are available during compilation is to require the file that defines them. See Features. ΓòÉΓòÉΓòÉ 15.4. Defining Macros ΓòÉΓòÉΓòÉ A Lisp macro is a list whose car is macro. Its cdr should be a function; expansion of the macro works by applying the function (with apply) to the list of unevaluated argument-expressions from the macro call. It is possible to use an anonymous Lisp macro just like an anonymous function, but this is never done, because it does not make sense to pass an anonymous macro to mapping functions such as mapcar. In practice, all Lisp macros have names, and they are usually defined with the special form defmacro. -- Special Form: defmacro name argument-list body-forms... defmacro defines the symbol name as a macro that looks like this: (macro lambda argument-list . body-forms) This macro object is stored in the function cell of name. The value returned by evaluating the defmacro form is name, but usually we ignore this value. The shape and meaning of argument-list is the same as in a function, and the keywords &rest and &optional may be used (see Argument List). Macros may have a documentation string, but any interactive declaration is ignored since macros cannot be called interactively. ΓòÉΓòÉΓòÉ 15.5. Backquote ΓòÉΓòÉΓòÉ It could prove rather awkward to write macros of significant size, simply due to the number of times the function list needs to be called. To make writing these forms easier, a macro ``' (often called backquote) exists. Backquote allows you to quote a list, but selectively evaluate elements of that list. In the simplest case, it is identical to the special form quote (see Quoting). For example, these two forms yield identical results: (` (a list of (+ 2 3) elements)) => (a list of (+ 2 3) elements) (quote (a list of (+ 2 3) elements)) => (a list of (+ 2 3) elements) By inserting a special marker, `,', inside of the argument to backquote, it is possible to evaluate desired portions of the argument: (list 'a 'list 'of (+ 2 3) 'elements) => (a list of 5 elements) (` (a list of (, (+ 2 3)) elements)) => (a list of 5 elements) It is also possible to have an evaluated list spliced into the resulting list by using the special marker `,@'. The elements of the spliced list become elements at the same level as the other elements of the resulting list. The equivalent code without using ` is often unreadable. Here are some examples: (setq some-list '(2 3)) => (2 3) (cons 1 (append some-list '(4) some-list)) => (1 2 3 4 2 3) (` (1 (,@ some-list) 4 (,@ some-list))) => (1 2 3 4 2 3) (setq list '(hack foo bar)) => (hack foo bar) (cons 'use (cons 'the (cons 'words (append (cdr list) '(as elements))))) => (use the words foo bar as elements) (` (use the words (,@ (cdr list)) as elements (,@ nil))) => (use the words foo bar as elements) The reason for (,@ nil) is to avoid a bug in Emacs version 18. The bug occurs when a call to ,@ is followed only by constant elements. Thus, (` (use the words (,@ (cdr list)) as elements)) would not work, though it really ought to. (,@ nil) avoids the problem by being a nonconstant element that does not affect the result. -- Macro: ` list This macro returns list as quote would, except that the list is copied each time this expression is evaluated, and any sublist of the form (, subexp) is replaced by the value of subexp. Any sublist of the form (,@ listexp) is replaced by evaluating listexp and splicing its elements into the containing list in place of this sublist. (A single sublist can in this way be replaced by any number of new elements in the containing list.) There are certain contexts in which `,' would not be recognized and should not be used: ;; Use of a `,' expression as the cdr of a list. (` (a . (, 1))) ; Not (a . 1) => (a \, 1) ;; Use of `,' in a vector. (` [a (, 1) c]) ; Not [a 1 c] error--> Wrong type argument ;; Use of a `,' as the entire argument of ``'. (` (, 2)) ; Not 2 => (\, 2) Common Lisp note: in Common Lisp, `,' and `,@' are implemented as reader macros, so they do not require parentheses. Emacs Lisp implements them as functions because reader macros are not supported (to save space). ΓòÉΓòÉΓòÉ 15.6. Common Problems Using Macros ΓòÉΓòÉΓòÉ The basic facts of macro expansion have all been described above, but there consequences are often counterintuitive. This section describes some important consequences that can lead to trouble, and rules to follow to avoid trouble. ΓòÉΓòÉΓòÉ 15.6.1. Evaluating Macro Arguments Too Many Times ΓòÉΓòÉΓòÉ When defining a macro you must pay attention to the number of times the arguments will be evaluated when the expansion is executed. The following macro (used to facilitate iteration) illustrates the problem. This macro allows us to write a simple ``for'' loop such as one might find in Pascal. (defmacro for (var from init to final do &rest body) "Execute a simple \"for\" loop, e.g., (for i from 1 to 10 do (print i))." (list 'let (list (list var init)) (cons 'while (cons (list '<= var final) (append body (list (list 'inc var))))))) => for (for i from 1 to 3 do (setq square (* i i)) (princ (format "\n%d %d" i square))) -> (let ((i 1)) (while (<= i 3) (setq square (* i i)) (princ (format "%d %d" i square)) (inc i))) -|1 1 -|2 4 -|3 9 => nil (The arguments from, to, and do in this macro are ``syntactic sugar''; they are entirely ignored. The idea is that you will write noise words (such as from, to, and do) in those positions in the macro call.) This macro suffers from the defect that final is evaluated on every iteration. If final is a constant, this is not a problem. If it is a more complex form, say (long-complex-calculation x), this can slow down the execution significantly. If final has side effects, executing it more than once is probably incorrect. A well-designed macro definition takes steps to avoid this problem by producing an expansion that evaluates the argument expressions exactly once unless repeated evaluation is part of the intended purpose of the macro. Here is a correct expansion for the for macro: (let ((i 1) (max 3)) (while (<= i max) (setq square (* i i)) (princ (format "%d %d" i square)) (inc i))) Here is a macro definition that creates this expansion: (defmacro for (var from init to final do &rest body) "Execute a simple for loop: (for i from 1 to 10 do (print i))." (` (let (((, var) (, init)) (max (, final))) (while (<= (, var) max) (,@ body) (inc (, var)))))) Unfortunately, this introduces another problem. Proceed to the following node. ΓòÉΓòÉΓòÉ 15.6.2. Local Variables in Macro Expansions ΓòÉΓòÉΓòÉ In the previous section, the definition of for was fixed as follows to make the expansion evaluate the macro arguments the proper number of times: (defmacro for (var from init to final do &rest body) "Execute a simple for loop: (for i from 1 to 10 do (print i))." (` (let (((, var) (, init)) (max (, final))) (while (<= (, var) max) (,@ body) (inc (, var)))))) The new definition of for has a new problem: it introduces a local variable named max which the user does not expect. This will cause trouble in examples such as the following: (let ((max 0)) (for x from 0 to 10 do (let ((this (frob x))) (if (< max this) (setq max this))))) The references to max inside the body of the for, which are supposed to refer to the user's binding of max, will instead access the binding made by for. The way to correct this is to use an uninterned symbol instead of max (see Creating Symbols). The uninterned symbol can be bound and referred to just like any other symbol, but since it is created by for, we know that it cannot appear in the user's program. Since it is not interned, there is no way the user can put it into the program later. It will not appear anywhere except where put by for. Here is a definition of for which works this way: (defmacro for (var from init to final do &rest body) "Execute a simple for loop: (for i from 1 to 10 do (print i))." (let ((tempvar (make-symbol "max"))) (` (let (((, var) (, init)) ((, tempvar) (, final))) (while (<= (, var) (, tempvar)) (,@ body) (inc (, var))))))) This creates an uninterned symbol named max and puts it in the expansion instead of the usual interned symbol max that appears in expressions ordinarily. ΓòÉΓòÉΓòÉ 15.6.3. Evaluating Macro Arguments in Expansion ΓòÉΓòÉΓòÉ Another problem can happen if you evaluate any of the macro argument expressions during the computation of the expansion, such as by calling eval (see Eval). If the argument is supposed to refer to the user's variables, you may have trouble if the user happens to use a variable with the same name as one of the macro arguments. Inside the macro body, the macro argument binding is the most local binding of this variable, so any references inside the form being evaluated will refer to it. Here is an example: (defmacro foo (a) (list 'setq (eval a) t)) => foo (setq x 'b) (foo x) -> (setq b t) => t ; and b has been set. ;; but (setq a 'b) (foo a) -> (setq 'b t) ; invalid! error--> Symbol's value is void: b It makes a difference whether the user types a or x, because a conflicts with the macro argument variable a. In general it is best to avoid calling eval in a macro definition at all. ΓòÉΓòÉΓòÉ 15.6.4. How Many Times is the Macro Expanded? ΓòÉΓòÉΓòÉ Occasionally problems result from the fact that a macro call is expanded each time it is evaluated in an interpreted function, but is expanded only once (during compilation) for a compiled function. If the macro definition has side effects, they will work differently depending on how many times the macro is expanded. In particular, constructing objects is a kind of side effect. If the macro is called once, then the objects are constructed only once. In other words, the same structure of objects is used each time the macro call is executed. In interpreted operation, the macro is reexpanded each time, producing a fresh collection of objects each time. Usually this does not matter---the objects have the same contents whether they are shared or not. But if the surrounding program does side effects on the objects, it makes a difference whether they are shared. Here is an example: (defmacro new-object () (list 'quote (cons nil nil))) (defun initialize (condition) (let ((object (new-object))) (if condition (setcar object condition)) object)) If initialize is interpreted, a new list (nil) is constructed each time initialize is called. Thus, no side-effect survives between calls. If initialize is compiled, then the macro new-object is expanded during compilation, producing a single ``constant'' (nil) that is reused and altered each time initialize is called. ΓòÉΓòÉΓòÉ 16. Loading ΓòÉΓòÉΓòÉ Loading a file of Lisp code means bringing its contents into the Lisp environment in the form of Lisp objects. Emacs finds and opens the file, reads the text, evaluates each form, and then closes the file. The load functions evaluate all the expressions in a file just as the eval-current-buffer function evaluates all the expressions in a buffer. The difference is that the load functions read and evaluate the text in the file as found on disk, not the text in an Emacs buffer. The loaded file must contain Lisp expressions, either as source code or, optionally, as byte-compiled code. Each form in the file is called a top-level form. There is no special format for the forms in a loadable file; any form in a file may equally well be typed directly into a buffer and evaluated there. (Indeed, most code is tested this way.) Most often, the forms are function definitions and variable definitions. A file containing Lisp code is often called a library. Thus, the ``Rmail library'' is a file containing code for Rmail mode. Similarly, a ``Lisp library directory'' is a directory of files containing Lisp code. ΓòÉΓòÉΓòÉ 16.1. How Programs Do Loading ΓòÉΓòÉΓòÉ There are several interface functions for loading. For example, the autoload function creates a Lisp object that loads a file when it is evaluated (see Autoload). require also causes files to be loaded (see Features). Ultimately, all these facilities call the load function to do the work. -- Function: load filename &optional missing-ok nomessage nosuffix This function finds and opens a file of Lisp code, evaluates all the forms in it, and closes the file. To find the file, load first looks for a file named `filename.elc', that is, for a file whose name has `.elc' appended. If such a file exists, it is loaded. But if there is no file by that name, then load looks for a file whose name has `.el' appended. If that file exists, it is loaded. Finally, if there is no file by either name, load looks for a file named filename with nothing appended, and loads it if it exists. (The load function is not clever about looking at filename. In the perverse case of a file named `foo.el.el', evaluation of (load "foo.el") will indeed find it.) If the optional argument nosuffix is non-nil, then the suffixes `.elc' and `.el' are not tried. In this case, the file name must be specified precisely. If filename is a relative file name, such as `foo.bar' or `baz/foo.bar', Emacs searches for the file using the variable load-path. Emacs does this by appending filename to each of the directories listed in load-path, and loading the first file it finds whose name matches. The current default directory is tried only if it is specified in load-path, where it is represented as nil. All three possible suffixes are tried in the first directory in load-path, then all three in the second directory in load-path, etc. Messages like `Loading foo...' and `Loading foo...done' are printed in the echo area while loading unless nomessage is non-nil. Any errors that are encountered while loading a file cause load to abort. If the load was done for the sake of autoload, certain kinds of top-level forms, those which define functions, are undone. The error file-error is signaled (with `Cannot open load file filename') if no file is found. No error is signaled if missing-ok is non-nil---then load just returns nil. load returns t if the file loads successfully. -- User Option: load-path The value of this variable is a list of directories to search when loading files with load. Each element is a string (which must be a directory name) or nil (which stands for the current working directory). The value of load-path is initialized from the environment variable EMACSLOADPATH, if it exists; otherwise it is set to the default specified in `emacs/src/paths.h' when Emacs is built. The syntax of EMACSLOADPATH is the same as that of PATH; fields are separated by `:', and `.' is used for the current default directory. Here is an example of how to set your EMACSLOADPATH variable from a csh `.login' file: setenv EMACSLOADPATH .:/user/liberte/emacs:/usr/local/lib/emacs/lisp Here is how to set it using sh: export EMACSLOADPATH EMACSLOADPATH=.:/user/liberte/emacs:/usr/local/lib/emacs/lisp Here is an example of code you can place in a `.emacs' file to add several directories to the front of your default load-path: (setq load-path (append (list nil "/user/liberte/emacs" "/usr/local/lisplib") load-path)) In this example, the path searches the current working directory first, followed by `/user/liberte/emacs' and `/usr/local/lisplib', which are then followed by the standard directories for Lisp code. When Emacs 18 is processing command options `-l' or `-load' which specify Lisp libraries to be loaded, it temporarily adds the current directory to the front of load-path so that files in the current directory can be specified easily. Emacs version 19 will also find such files in the current directory but without altering load-path. -- Variable: load-in-progress This variable is non-nil if Emacs is in the process of loading a file, and it is nil otherwise. This is how defun and provide determine whether a load is in progress, so that their effect can be undone if the load fails. To learn how load is used to build Emacs, see Building Emacs. ΓòÉΓòÉΓòÉ 16.2. Autoload ΓòÉΓòÉΓòÉ The autoload facility allows you to make a function or macro available but put off loading its actual definition. An attempt to call a symbol whose definition is an autoload object automatically reads the file to install the real definition and its other associated code, and then calls the real definition. To prepare a function or macro for autoloading, you must call autoload, specifying the function name and the name of the file to be loaded. This is usually done when Emacs is first built, by files such as `emacs/lisp/loaddefs.el'. The following example shows how doctor is prepared for autoloading in `loaddefs.el': (autoload 'doctor "doctor" "\ Switch to *doctor* buffer and start giving psychotherapy." t) The backslash and newline immediately following the double-quote are a convention used only in the preloaded Lisp files such as `loaddefs.el'; they cause the documentation string to be put in the `etc/DOC' file. (See Building Emacs.) In any other source file, you would write just this: (autoload 'doctor "doctor" "Switch to *doctor* buffer and start giving psychotherapy." t) Calling autoload creates an autoload object containing the name of the file and some other information, and makes this the definition of the specified symbol. When you later try to call that symbol as a function or macro, the file is loaded; the loading should redefine that symbol with its proper definition. After the file completes loading, the function or macro is called as if it had been there originally. If, at the end of loading the file, the desired Lisp function or macro has not been defined, then the error error is signaled (with data "Autoloading failed to define function function-name"). The autoloaded file may, of course, contain other definitions and may require or provide one or more features. If the file is not completely loaded (due to an error in the evaluation of the contents) any function definitions or provide calls that occurred during the load are undone. This is to ensure that the next attempt to call any function autoloading from this file will try again to load the file. If not for this, then some of the functions in the file might appear defined, but they may fail to work properly for the lack of certain subroutines defined later in the file and not loaded successfully. -- Function: autoload symbol filename &optional docstring interactive macro This function defines the function (or macro) named symbol so as to load automatically from filename. The string filename is a file name which will be passed to load when the function is called. The argument docstring is the documentation string for the function. Normally, this is the same string that is in the function definition itself. This makes it possible to look at the documentation without loading the real definition. If interactive is non-nil, then the function can be called interactively. This lets completion in M-x work without loading the function's real definition. The complete interactive specification need not be given here. If macro is non-nil, then the function is really a macro. If symbol already has a non-nil function definition that is not an autoload object, autoload does nothing and returns nil. If the function cell of symbol is void, or is already an autoload object, then it is set to an autoload object that looks like this: (autoload filename docstring interactive macro) For example, (symbol-function 'run-prolog) => (autoload "prolog" 169681 t nil) In this case, "prolog" is the name of the file to load, 169681 is the reference to the documentation string in the `emacs/etc/DOC' file (see Documentation Basics), t means the function is interactive, and nil that it is not a macro. ΓòÉΓòÉΓòÉ 16.3. Repeated Loading ΓòÉΓòÉΓòÉ You may load a file more than once in an Emacs session. For example, after you have rewritten and reinstalled a function definition by editing it in a buffer, you may wish to return to the original version; you can do this by reloading the file in which it is located. When you load or reload files, bear in mind that the load and load-library functions automatically load a byte-compiled file rather than a non-compiled file of similar name. If you rewrite a file that you intend to save and reinstall, remember to byte-compile it if necessary; otherwise you may find yourself inadvertently reloading the older, byte-compiled file instead of your newer, non-compiled file! When writing the forms in a library, keep in mind that the library might be loaded more than once. For example, the choice of defvar vs. defconst for defining a variable depends on whether it is desirable to reinitialize the variable if the library is reloaded: defconst does so, and defvar does not. (See Defining Variables.) The simplest way to add an element to an alist is like this: (setq minor-mode-alist (cons '(leif-mode " Leif") minor-mode-alist)) But this would add multiple elements if the library is reloaded. To avoid the problem, write this: (or (assq 'leif-mode minor-mode-alist) (setq minor-mode-alist (cons '(leif-mode " Leif") minor-mode-alist))) Occasionally you will want to test explicitly whether a library has already been loaded; you can do so as follows: (if (not (boundp 'foo-was-loaded)) execute-first-time-only) (setq foo-was-loaded t) ΓòÉΓòÉΓòÉ 16.4. Features ΓòÉΓòÉΓòÉ provide and require are an alternative to autoload for loading files automatically. They work in terms of named features. Autoloading is triggered by calling a specific function, but a feature is loaded the first time another program asks for it by name. The use of named features simplifies the task of determining whether required definitions have been defined. A feature name is a symbol that stands for a collection of functions, variables, etc. A program that needs the collection may ensure that they are defined by requiring the feature. If the file that contains the feature has not yet been loaded, then it will be loaded (or an error will be signaled if it cannot be loaded). The file thus loaded must provide the required feature or an error will be signaled. To require the presence of a feature, call require with the feature name as argument. require looks in the global variable features to see whether the desired feature has been provided already. If not, it loads the feature from the appropriate file. This file should call provide at the top-level to add the feature to features. Features are normally named after the files they are provided in so that require need not be given the file name. For example, in `emacs/lisp/prolog.el', the definition for run-prolog includes the following code: (interactive) (require 'shell) (switch-to-buffer (make-shell "prolog" "prolog")) (inferior-prolog-mode)) The expression (require 'shell) loads the file `shell.el' if it has not yet been loaded. This ensures that make-shell is defined. The `shell.el' file contains the following top-level expression: (provide 'shell) This adds shell to the global features list when the `shell' file is loaded, so that (require 'shell) will henceforth know that nothing needs to be done. When require is used at top-level in a file, it takes effect if you byte-compile that file (see Byte Compilation). This is in case the required package contains macros that the byte compiler must know about. Although top-level calls to require are evaluated during byte compilation, provide calls are not. Therefore, you can ensure that a file of definitions is loaded before it is byte-compiled by including a provide followed by a require for the same feature, as in the following example. (provide 'my-feature) ; Ignored by byte compiler, evaluated by load. (require 'my-feature) ; Evaluated by byte compiler. -- Function: provide feature This function announces that feature is now loaded, or being loaded, into the current Emacs session. This means that the facilities associated with feature are or will be available for other Lisp programs. The direct effect of calling provide is to add feature to the front of the list features if it is not already in the list. The argument feature must be a symbol. provide returns feature. features => (bar bish) (provide 'foo) => foo features => (foo bar bish) During autoloading, if the file is not completely loaded (due to an error in the evaluation of the contents) any function definitions or provide calls that occurred during the load are undone. See Autoload. -- Function: require feature &optional filename This function checks whether feature is present in the current Emacs session (using (featurep feature); see below). If it is not, then require loads filename with load. If filename is not supplied, then the name of the symbol feature is used as the file name to load. If feature is not provided after the file has been loaded, Emacs will signal the error error (with data `Required feature feature was not provided'). -- Function: featurep feature This function returns t if feature has been provided in the current Emacs session (i.e., feature is a member of features.) -- Variable: features The value of this variable is a list of symbols that are the features loaded in the current Emacs session. Each symbol was put in this list with a call to provide. The order of the elements in the features list is not significant. ΓòÉΓòÉΓòÉ 17. Byte Compilation ΓòÉΓòÉΓòÉ GNU Emacs Lisp has a compiler that translates functions written in Lisp into a special representation called byte-code that can be executed more efficiently. The compiler replaces Lisp function definitions with byte-code. When a byte-code function is called, its definition is evaluated by the byte-code interpreter. Because the byte-compiled code is evaluated by the byte-code interpreter, instead of being executed directly by the machine's hardware (as true compiled code is), byte-code is completely transportable from machine to machine without recompilation. It is not, however, as fast as true compiled code. See Compilation Errors, for how to investigate errors occurring in byte compilation. ΓòÉΓòÉΓòÉ 17.1. The Compilation Functions ΓòÉΓòÉΓòÉ An individual function or macro definition may be byte-compiled with the byte-compile function. A whole file may be byte-compiled with byte-compile-file and several files may be byte-compiled with byte-recompile-directory or batch-byte-compile. Only defun and defmacro forms in a file are byte-compiled; other top-level forms are not altered by byte compilation. Be careful when byte-compiling code that uses macros. Macro calls are expanded when they are compiled, so the macros must already be defined for proper compilation. For more details, see Compiling Macros. While byte-compiling a file, any require calls at top-level are executed. One way to ensure that necessary macro definitions are available during compilation is to require the file that defines them. See Features. A byte-compiled function is not as efficient as a primitive function written in C, but will run much faster than the version written in Lisp. For a rough comparison, consider the example below: (defun silly-loop (n) "Return time before and after N iterations of a loop." (let ((t1 (current-time-string))) (while (> (setq n (1- n)) 0)) (list t1 (current-time-string)))) => silly-loop (silly-loop 100000) => ("Thu Jan 12 20:18:38 1989" "Thu Jan 12 20:19:29 1989") ; 51 seconds (byte-compile 'silly-loop) => [Compiled code not shown] (silly-loop 100000) => ("Thu Jan 12 20:21:04 1989" "Thu Jan 12 20:21:17 1989") ; 13 seconds In this example, the interpreted code required 51 seconds to run, whereas the byte-compiled code required 13 seconds. These results are representative, but actual results will vary greatly. -- Function: byte-compile symbol This function byte-compiles the function definition of symbol, replacing the previous definition with the compiled one. The function definition of symbol must be the actual code for the function; i.e., the compiler will not follow indirection to another symbol. byte-compile does not compile macros. byte-compile returns the new, compiled definition of symbol. (defun factorial (integer) "Compute factorial of INTEGER." (if (= 1 integer) 1 (* integer (factorial (1- integer))))) => factorial (byte-compile 'factorial) => (lambda (integer) "Compute factorial of INTEGER." (byte-code "\301^HU\203 ^@\301\202^Q^@\302^H\303^HS!\"\207" [integer 1 * factorial] 4)) The string that is the first argument of byte-code is the actual byte-code. Each character in it is an instruction. The vector contains all the constants, variable names and function names used by the function, except for certain primitives that are coded as special instructions. The byte-compile function is not autoloaded as are byte-compile-file and byte-recompile-directory. -- Command: byte-compile-file filename This function compiles a file of Lisp code named filename into a file of byte-code. The output file's name is made by appending `c' to the end of filename. Compilation works by reading the input file one form at a time. If it is a definition of a function or macro, the compiled function or macro definition is written out. Other forms are copied out unchanged. All comments are discarded when the input file is read. This command returns t. When called interactively, it prompts for the file name. % ls -l push* -rw-r--r-- 1 lewis 791 Oct 5 20:31 push.el (byte-compile-file "~/emacs/push.el") => t % ls -l push* -rw-r--r-- 1 lewis 791 Oct 5 20:31 push.el -rw-rw-rw- 1 lewis 638 Oct 8 20:25 push.elc -- Command: byte-recompile-directory directory flag This function recompiles every `.el' file in directory that needs recompilation. A file needs recompilation if a `.elc' file exists but is older than the `.el' file. If a `.el' file exists, but there is no corresponding `.elc' file, then flag is examined. If it is nil, the file is ignored. If it is non-nil, the user is asked whether the file should be compiled. The returned value of this command is unpredictable. -- Function: batch-byte-compile This function runs byte-compile-file on the files remaining on the command line. This function must be used only in a batch execution of Emacs, as it kills Emacs on completion. Each file will be processed, even if an error occurs while compiling a previous file. (The file with the error will not, of course, produce any compiled code.) % emacs -batch -f batch-byte-compile *.el -- Function: byte-code code-string data-vector max-stack This is the function that actually interprets byte-code. A byte-compiled function is actually defined with a body that calls byte-code. Don't call this function yourself. Only the byte compiler knows how to generate valid calls to this function. ΓòÉΓòÉΓòÉ 17.2. Disassembled Byte-Code ΓòÉΓòÉΓòÉ People do not write byte-code; that job is left to the byte compiler. But we provide a disassembler to satisfy a cat-like curiosity. The disassembler converts the byte-compiled code into humanly readable form. The byte-code interpreter is implemented as a simple stack machine. Values get stored by being pushed onto the stack, and are popped off and manipulated, the results being pushed back onto the stack. When a function returns, the top of the stack is popped and returned as the value of the function. In addition to the stack, values used during byte-code execution can be stored in ordinary Lisp variables. Variable values can be pushed onto the stack, and variables can be set by popping the stack. -- Command: disassemble object &optional stream This function prints the disassembled code for object. If stream is supplied, then output goes there. Otherwise, the disassembled code is printed to the stream standard-output. The argument object can be a function name or a lambda expression. As a special exception, if this function is used interactively, it outputs to a buffer named `*Disassemble*'. Here are two examples of using the disassemble function. We have added explanatory comments to help you relate the byte-code to the Lisp source; these do not appear in the output of disassemble. (defun factorial (integer) "Compute factorial of an integer." (if (= 1 integer) 1 (* integer (factorial (1- integer))))) => factorial (factorial 4) => 24 (disassemble 'factorial) -| byte-code for factorial: doc: Compute factorial of an integer. args: (integer) 0 constant 1 ; Push 1 onto stack. 1 varref integer ; Get value of integer from the environment ; and push the value onto the stack. 2 eqlsign ; Pop top two values off stack, ; compare them, ; and push result onto stack. 3 goto-if-nil 10 ; Pop and test top of stack; ; if nil, go to 10, ; else continue. 6 constant 1 ; Push 1 onto top of stack. 7 goto 17 ; Go to 17 (in this case, 1 will be ; returned by the function). 10 constant * ; Push symbol * onto stack. 11 varref integer ; Push value of integer onto stack. 12 constant factorial ; Push factorial onto stack. 13 varref integer ; Push value of integer onto stack. 14 sub1 ; Pop integer, decrement value, ; push new value onto stack. ; Stack now contains: ; decremented value of integer ; factorial ; value of integer ; * 15 call 1 ; Call function factorial using ; the first (i.e., the top) element ; of the stack as the argument; ; push returned value onto stack. ; Stack now contains: ; result of result of recursive ; call to factorial ; value of integer ; * 16 call 2 ; Using the first two (i.e., the top two) ; elements of the stack as arguments, ; call the function *, ; pushing the result onto the stack. 17 return ; Return the top element of the stack. => nil The silly-loop function is somewhat more complex: (defun silly-loop (n) "Return time before and after N iterations of a loop." (let ((t1 (current-time-string))) (while (> (setq n (1- n)) 0)) (list t1 (current-time-string)))) => silly-loop (disassemble 'silly-loop) -| byte-code for silly-loop: doc: Return time before and after N iterations of a loop. args: (n) 0 constant current-time-string ; Push current-time-string ; onto top of stack. 1 call 0 ; Call current-time-string with no ; argument, pushing result onto stack. 2 varbind t1 ; Pop stack and bind t1 to popped value. 3 varref n ; Get value of n from the environment ; and push the value onto the stack. 4 sub1 ; Subtract 1 from top of stack. 5 dup ; Duplicate the top of the stack; ; i.e. copy the top of the stack ; and push the copy onto the stack. 6 varset n ; Pop the top of the stack, ; and bind n to the value. ; In effect, the sequence dup varset copies ; the top of the stack into the value of n ; without popping it. 7 constant 0 ; Push 0 onto stack. 8 gtr ; Pop top two values off stack, ; test if n is greater than 0 ; and push result onto stack. 9 goto-if-nil-else-pop 17 ; Goto 17 if n > 0 else pop top of stack and continue ; (this exits the while loop). 12 constant nil ; Push nil onto stack ; (this is the body of the loop). 13 discard ; Discard result of the body of the loop ; (a while loop is always evaluated ; for its side effects). 14 goto 3 ; Jump back to beginning of while loop. 17 discard ; Discard result of while loop ; by popping top of stack. 18 varref t1 ; Push value of t1 onto stack. 19 constant current-time-string ; Push current-time-string ; onto top of stack. 20 call 0 ; Call current-time-string again. 21 list2 ; Pop top two elements off stack, ; create a list of them, ; and push list onto stack. 22 unbind 1 ; Unbind t1 in local environment. 23 return ; Return value of the top of stack. => nil ΓòÉΓòÉΓòÉ 18. Debugging Lisp Programs ΓòÉΓòÉΓòÉ There are three ways to investigate a problem in an Emacs Lisp program, depending on what you are doing with the program when the problem appears. o If the problem occurs when you run the program, you can use the Lisp debugger to investigate what is happening during execution. o If the problem is syntactic, so that Lisp cannot even read the program, you can use the Emacs facilities for editing Lisp to localize it. o If the problem occurs when trying to compile the program with the byte compiler, you need to know how to examine the compiler's input buffer. Another useful debugging tool is a dribble file. When a dribble file is open, Emacs copies all keyboard input characters to that file. Afterward, you can examine the file to find out what input was used. See Terminal Input. For debugging problems in terminal descriptions, the open-termscript function can be useful. See Terminal Output. ΓòÉΓòÉΓòÉ 18.1. The Lisp Debugger ΓòÉΓòÉΓòÉ The Lisp debugger provides you with the ability to suspend evaluation of a form. While evaluation is suspended (a state that is commonly known as a break), you may examine the run time stack, examine the values of local or global variables, or change those values. Since a break is a recursive edit, all the usual editing facilities of Emacs are available; you can even run programs that will enter the debugger recursively. See Recursive Editing. ΓòÉΓòÉΓòÉ 18.1.1. Entering the Debugger When an Error Occurs ΓòÉΓòÉΓòÉ The most important time to enter the debugger is when a Lisp error happens. This allows you to investigate the immediate causes of the error. However, entry to the debugger is not a normal consequence of an error. Many commands frequently get Lisp errors when invoked in inappropriate contexts (such as C-f at the end of the buffer) and during ordinary editing it would be very unpleasant to enter the debugger each time this happens. If you want errors to enter the debugger, set the variable debug-on-error to non-nil. -- User Option: debug-on-error This variable determines whether the debugger is called when a error is signaled and not handled. If debug-on-error is non-nil, then the debugger is called when an error happens. Otherwise the debugger is not called for errors. ΓòÉΓòÉΓòÉ 18.2. Debugging Infinite Loops ΓòÉΓòÉΓòÉ When a program loops infinitely and fails to return, your first problem is to stop the loop. On most operating systems, you can do this with C-g, which causes quit. This works if you are not using X windows, and on Berkeley systems even if you are using X windows. On other inferior operating systems, C-g does not work when using X windows. This is because these systems do not allow Emacs to request a signal when input arrives from the X server. There is nothing Emacs can do about this. However, you can still stop a loop by sending Emacs the SIGINT signal. To do this, go to a shell in another window, use ps to find out the pid of the Emacs process, and then type `kill -INT pid'. Ordinary quitting gives no information about why the program was looping. To get more information, you can set the variable debug-on-quit to non-nil. Quitting with C-g is not considered an error, and debug-on-error has no effect on the handling of C-g. debug-on-quit has no effect on errors. Once you have the debugger running in the middle of the infinite loop, you can proceed from the debugger using the stepping commands. If you step through the entire loop, you will probably get enough information to solve the problem. -- User Option: debug-on-quit This variable determines whether the debugger is called when quit is signaled and not handled. If debug-on-quit is non-nil, then the debugger is called whenever you quit (that is, type C-g). If debug-on-quit is nil, then the debugger is not called when you quit. See Quitting. ΓòÉΓòÉΓòÉ 18.2.1. Entering the Debugger when Some Function is Called ΓòÉΓòÉΓòÉ To investigate a problem that happens in the middle of a program, one useful technique is to cause the debugger to be entered when a certain function is called. You can do this to the function in which the problem occurs, and then step through the function, or you can do this to a function called shortly before the problem, step quickly over the call to that function, and then step through its caller. -- Command: debug-on-entry function-name This function requests function-name to invoke the debugger each time it is called. It works by inserting the form (debug 'debug) into the function definition as the first form. Any function defined as Lisp code may be set to break on entry, regardless of whether it is interpreted code or compiled code. Even functions that are commands may be debugged---they will enter the debugger when called inside a function, or when called interactively. Primitive functions (i.e., those written in C) may not be debugged. When debug-on-entry is called interactively, it prompts for function-name in the minibuffer. If debug-on-entry is called more than once on the same function, the second call does nothing. debug-on-entry returns function-name. (defun fact (n) (if (zerop n) 1 (* n (fact (1- n))))) => fact (debug-on-entry 'fact) => fact (fact 3) => 6 ---------- Buffer: *Backtrace* ---------- Entering: * fact(3) eval-region(4870 4878 t) byte-code("...") eval-last-sexp(nil) (let ...) eval-insert-last-sexp(nil) * call-interactively(eval-insert-last-sexp) ---------- Buffer: *Backtrace* ---------- (symbol-function 'fact) => (lambda (n) (debug (quote debug)) (if (zerop n) 1 (* n (fact (1- n))))) -- Command: cancel-debug-on-entry function-name This function undoes the effect of debug-on-entry on function-name. When called interactively, it prompts for function-name in the minibuffer. If cancel-debug-on-entry is called more than once on the same function, the second call does nothing. cancel-debug-on-entry returns function-name. ΓòÉΓòÉΓòÉ 18.2.2. Explicit Entry to the Debugger ΓòÉΓòÉΓòÉ You can cause the debugger to be called at a certain point in your program by writing the expression (debug) at that point. To do this, visit the source file, insert the text `(debug)' at the proper place, and type C-M-x. Be sure to undo this insertion before you save the file! The place where you insert `(debug)' must be a place where an additional form can be evaluated and its value ignored. (If the value isn't ignored, it will alter the execution of the program!) Usually this means inside a progn or an implicit progn (see Sequencing). ΓòÉΓòÉΓòÉ 18.2.3. Using the Debugger ΓòÉΓòÉΓòÉ When the debugger is entered, it displays the previously selected buffer in one window and a buffer named `*Backtrace*' in another window. The backtrace buffer contains one line for each level of Lisp function execution currently going on. At the beginning of this buffer is a message describing the reason that the debugger was invoked (such as the error message and associated data, if it was invoked due to an error). The backtrace buffer is read-only and uses a special major mode, Debugger mode, in which letters are defined as debugger commands. The usual Emacs editing commands are available; thus, you can switch windows to examine the buffer that was being edited at the time of the error, switch buffers, visit files, or do any other sort of editing. However, the debugger is a recursive editing level (see Recursive Editing) and it is wise to go back to the backtrace buffer and exit the debugger (with the q command) when you you are finished with it. Exiting the debugger gets out of the recursive edit and kills the backtrace buffer. The contents of the backtrace buffer show you the functions that are executing and the arguments that were given to them. It also allows you to specify a stack frame by moving point to the line describing that frame. (A stack frame is the place where the Lisp interpreter records information about a particular invocation of a function. The frame whose line point is on is considered the current frame.) Some of the debugger commands operate on the current frame. The debugger itself should always be run byte-compiled, since it makes assumptions about how many stack frames are used for the debugger itself. These assumptions are false if the debugger is running interpreted. ΓòÉΓòÉΓòÉ 18.2.4. Debugger Commands ΓòÉΓòÉΓòÉ Inside the debugger (in Debugger mode), these special commands are available in addition to the usual cursor motion commands. (Keep in mind that all the usual facilities of Emacs, such as switching windows or buffers, are still available.) The most important use of debugger commands is for stepping through code, so that you can see how control flows. The debugger can step through the control structures of an interpreted function, but cannot do so in a byte-compiled function. If you would like to step through a byte-compiled function, replace it with an interpreted definition of the same function. (To do this, visit the source file for the function and type C-M-x on its definition.) c Exit the debugger and continue execution. When continuing is possible, it resumes execution of the program as if the debugger had never been entered (aside from the effect of any variables or data structures you may have changed while inside the debugger). Continuing is possible after entry to the debugger due to function entry or exit, explicit invocation, quitting or certain errors. Most errors cannot be continued; trying to continue an unsuitable error causes the same error to occur again. d Continue execution, but enter the debugger the next time any Lisp function is called. This allows you to step through the subexpressions of an expression, seeing what values the subexpressions compute, and what else they do. The stack frame made for the function call which enters the debugger in this way will be flagged automatically so that the debugger will be called again when the frame is exited. You can use the u command to cancel this flag. b Flag the current frame so that the debugger will be entered when the frame is exited. Frames flagged in this way are marked with stars in the backtrace buffer. u Don't enter the debugger when the current frame is exited. This cancels a b command on that frame. e Read a Lisp expression in the minibuffer, evaluate it, and print the value in the echo area. This is the same as the command M-ESC, except that e is not normally disabled like M-ESC. q Terminate the program being debugged; return to top-level Emacs command execution. If the debugger was entered due to a C-g but you really want to quit, and not debug, use the q command. r Return a value from the debugger. The value is computed by reading an expression with the minibuffer and evaluating it. The r command makes a difference when the debugger was invoked due to exit from a Lisp call frame (as requested with b); then the value specified in the r command is used as the value of that frame. The r also matters in certain cases of errors. For example, wrong-type-argument errors will use the debugger's return value instead of the invalid argument; no-catch errors will use the debugger value as a throw tag instead of the tag that was not found. If an error was signaled by calling the Lisp function signal, the debugger's return value is returned as the value of signal. ΓòÉΓòÉΓòÉ 18.2.5. Invoking the Debugger ΓòÉΓòÉΓòÉ Here we describe fully the function used to invoke the debugger. -- Function: debug &rest debugger-args This function enters the debugger. It switches buffers to a buffer named `*Backtrace*' (or `*Backtrace*<2>' if it is the second recursive entry to the debugger, etc.), and fills it with information about the stack of Lisp function calls. It then enters a recursive edit, leaving that buffer in Debugger mode and displayed in the selected window. Debugger mode provides a c command which operates by exiting the recursive edit, switching back to the previous buffer, and returning to whatever called debug. The r command also returns from debug. These are the only ways the function debug can return to its caller. If the first of the debugger-args passed to debug is nil (or if it is not one of the following special values), then the rest of the arguments to debug are printed at the top of the `*Backtrace*' buffer. This mechanism is used to display a message to the user. However, if the first argument passed to debug is one of the following special values, then it has special significance. Normally, these values are passed to debug only by the internals of Emacs and the debugger, and not by programmers calling debug. The special values are: lambda When the first argument is lambda, the debugger displays `Entering:' as a line of text at the top of the buffer. This means that a function is being entered when debug-on-next-call is non-nil. debug When the first argument is debug, the debugger displays `Entering:' just as in the lambda case. However, debug as the argument indicates that the reason for entering the debugger is that a function set to debug on entry is being entered. In addition, debug as the first argument directs the debugger to mark the function that called debug so that it will invoke the debugger when exited. (When lambda is the first argument, the debugger does not do this, because it has already been done by the interpreter.) t When the first argument is t, the debugger displays `Beginning evaluation of function call form:' as the top line in the buffer, to indicate that it was entered due to the evaluation of a list form at a time when debug-on-next-call is non-nil. exit When the first argument is exit, it indicates the exit of a stack frame previously marked to invoke the debugger on exit. The debugger displays `Return value:' on the top line of the buffer, followed by the value being returned from the frame. error When the first argument is error, the debugger indicates that it is being entered because an error or quit was signaled and not handled, by displaying `Signaling:' followed by the error signaled and any arguments to signal. For example, (let ((debug-on-error t)) (/ 1 0)) ---------- Buffer: *Backtrace* ---------- Signaling: (arith-error) /(1 0) ... If an error was signaled, presumably the variable debug-on-error is non-nil. If quit was signaled, then presumably the variable debug-on-quit is non-nil. nil Use nil as the first of the debugger-args when you want to enter the debugger explicitly. The rest of the debugger-args are printed on the top line of the buffer. You can use this feature to display messages---for example, to remind yourself of the conditions under which debug is called. ΓòÉΓòÉΓòÉ 18.2.6. Internals of the Debugger ΓòÉΓòÉΓòÉ This section describes functions and variables used internally by the debugger. -- Variable: debugger The value of this variable is the function to call to invoke the debugger. Its value must be a function of any number of arguments (or, more typically, the name of a function). Presumably this function will enter some kind of debugger. The default value of the variable is debug. The first argument that Lisp hands to the function indicates why it was called. The convention for arguments is detailed in the description of debug. -- Command: backtrace This function prints a trace of Lisp function calls currently active. This is the function used by debug to fill up the `*Backtrace*' buffer. It is written in C, since it must have access to the stack to determine which function calls are active. The return value is always nil. In the following example, backtrace is called explicitly in a Lisp expression. When the expression is evaluated, the backtrace is printed to the stream standard-output: in this case, to the buffer `backtrace-output'. Each line of the backtrace represents one function call. If the arguments of the function call are all known, they are displayed; if they are being computed, that fact is stated. The arguments of special forms are elided. (with-output-to-temp-buffer "backtrace-output" (let ((var 1)) (save-excursion (setq var (eval '(progn (1+ var) (list 'testing (backtrace)))))))) => nil ----------- Buffer: backtrace-output ------------ backtrace() (list ...computing arguments...) (progn ...) eval((progn (1+ var) (list (quote testing) (backtrace)))) (setq ...) (save-excursion ...) (let ...) (with-output-to-temp-buffer ...) eval-region(1973 2142 #<buffer *scratch*>) byte-code("... for eval-print-last-sexp ...") eval-print-last-sexp(nil) * call-interactively(eval-print-last-sexp) ----------- Buffer: backtrace-output ------------ -- User Option: stack-trace-on-error This variable controls whether Lisp automatically displays a backtrace buffer after every error that is not handled. A quit signal counts as an error for this variable. If it is non-nil then a backtrace is shown in a pop-up buffer named `*Backtrace*' on every error. If it is nil, then a backtrace is not shown. When a backtrace is shown, that buffer is not selected. If either debug-on-quit or debug-on-error is also non-nil, then a backtrace is shown in one buffer, and the debugger is popped up in another buffer with its own backtrace. We consider this feature to be obsolete and superseded by the debugger itself. -- Variable: debug-on-next-call This variable determines whether the debugger is called before the next eval, apply or funcall. It is automatically reset to nil when the debugger is entered. The d command in the debugger works by setting this variable. -- Function: backtrace-debug level flag This function sets the debug-on-exit flag of the eval frame level levels down to flag. If flag is non-nil, this will cause the debugger to be entered when that frame exits. The debug-on-exit flag is an entry in the stack frame of a function call. This flag is examined on every exit from a function. Normally, this function is only called by the debugger. ΓòÉΓòÉΓòÉ 18.3. Debugging Invalid Lisp Syntax ΓòÉΓòÉΓòÉ It is easy to make a syntax error in an Emacs Lisp program by omitting a parenthesis. The Lisp reader will detect an error, but cannot say where the real problem is. For example, if a close parenthesis is omitted, the reader will detect an imbalance at the end of the file, but it cannot tell anything about where the close parenthesis should have been. However, you can use the following techniques to figure out where. If the problem is not simply an imbalance of parentheses, a useful technique is to try C-M-e at the beginning of each defun, and see if it goes to the place where that defun appears to end. If it does not, there is a problem in that defun. However, unmatched parentheses are the most common syntax errors in Lisp, and we can give further advice for those cases. ΓòÉΓòÉΓòÉ 18.3.1. Excess Open Parentheses ΓòÉΓòÉΓòÉ The first step is to find the defun that is unbalanced. If there is an excess open parenthesis, the way to do this is to insert a close parenthesis at the end of the file and type C-M-b (backward-sexp). This will move you to the beginning of the defun that is unbalanced. (Then type C-SPC C-_ C-u C-SPC to set the mark there, undo the insertion of the close parenthesis, and finally return to the mark.) The next step is to determine precisely what is wrong. There is no way to be sure of this except to study the program, but often the existing indentation is a clue to where the parentheses should have been. The easiest way to use this clue is to reindent with C-M-q and see what moves. Before you do this, make sure the defun has enough close parentheses. Otherwise, C-M-q will get an error, or will reindent all the rest of the file until the end. So move to the end of the defun and insert a close parenthesis there. Don't use C-M-e to move there, since that too will fail to work until the defun is balanced. Then go to the beginning of the defun and type C-M-q. Usually all the lines from a certain point to the end of the function will shift to the right. There is probably a missing close parenthesis, or a superfluous open parenthesis, near that point. (However, don't assume this is true; study the code to make sure.) Once you have found the discrepancy, undo the C-M-q, since the old indentation is probably appropriate to the intended parentheses. After you think you have fixed the problem, use C-M-q again. It should not change anything, if the problem is really fixed. ΓòÉΓòÉΓòÉ 18.3.2. Excess Close Parentheses ΓòÉΓòÉΓòÉ To deal with an excess close parenthesis, first insert an open parenthesis at the beginning of the file and type C-M-f to find the end of the unbalanced defun. (Then type C-SPC C-_ C-u C-SPC to set the mark there, undo the insertion of the open parenthesis, and finally return to the mark.) Then find the actual matching close parenthesis by typing C-M-f at the beginning of the defun. This will leave you somewhere short of the place where the defun ought to end. It is possible that you will find a spurious close parenthesis in that vicinity. If you don't see a problem at that point, the next thing to do is to type C-M-q at the beginning of the defun. A range of lines will probably shift left; if so, the missing open parenthesis or spurious close parenthesis is probably near the first of those lines. (However, don't assume this is true; study the code to make sure.) Once you have found the discrepancy, undo the C-M-q, since the old indentation is probably appropriate to the intended parentheses. ΓòÉΓòÉΓòÉ 18.4. Debugging Problems in Compilation ΓòÉΓòÉΓòÉ When an error happens during byte compilation, it is normally due to an error in the program you are compiling. The compiler itself can't tell you where in the file the error occurred, so here is how to find out. What you should do is switch to the buffer `*Compiler Input*'. (Note that the buffer name starts with a space, so it will not show up in M-x list-buffers.) This buffer contains the program being compiled, and point shows how far the byte compiler was able to read. If the error was due to invalid Lisp syntax, point shows exactly where the invalid syntax was detected. The cause of the error is not necessarily near by! Use the techniques in the previous section to find the error. If the error was detected while compiling a form that had been read successfully, then point is located at the end of the form. In this case, it can't localize the error precisely, but can still show you which function to check. ΓòÉΓòÉΓòÉ 19. Reading and Printing Lisp Objects ΓòÉΓòÉΓòÉ Printing and reading are the operations of converting Lisp objects to textual form and vice versa. They use the printed representations and read syntax described in Types of Lisp Object. This chapter describes the Lisp functions for reading and printing. It also describes streams, which specify where to get the text (if reading) or where to put it (if printing). ΓòÉΓòÉΓòÉ 19.1. Introduction to Reading and Printing ΓòÉΓòÉΓòÉ Reading a Lisp object means parsing a Lisp expression in textual form and producing a corresponding Lisp object. This is how Lisp programs get into Lisp from files of Lisp code. We call the text the read syntax of the object. For example, reading the text `(a . 5)' returns a cons cell whose car is a and whose cdr is the number 5. Printing a Lisp object means producing text that represents that object---converting the object to its printed representation. Printing the cons cell described above produces the text `(a . 5)'. Reading and printing are usually inverse operations: printing the object that results from reading a given piece of text often produces the same text, and reading the text that results from printing an object usually produces a similar-looking object. For example, printing the symbol foo produces the text `foo', and reading that text returns the symbol foo. Printing a list whose elements are a and b produces the text `(a b)', and reading that text produces a list (but not the same list) with elements are a and b. However, these two operations are not precisely inverses. There are two kinds of exceptions: o Printing can produce text that cannot be read. For example, buffers, windows, subprocesses and markers print into text that starts with `#'; if you try to read this text, you get an error. There is no way to read those data types. o One object can have multiple textual representations. For example, `1' and `01' represent the same integer, and `(a b)' and `(a . (b))' represent the same list. Reading will accept any of the alternatives, but printing must choose one of them. ΓòÉΓòÉΓòÉ 19.2. Input Streams ΓòÉΓòÉΓòÉ Most of the Lisp functions for reading text take an input stream as an argument. The input stream specifies where or how to get the characters of the text to be read. Here are the possible types of input stream: buffer The input characters are read from buffer, starting with the character directly after point. Point advances as characters are read. marker The input characters are read from the buffer that marker is in, starting with the character directly after the marker. The marker position advances as characters are read. The value of point in the buffer has no effect when the stream is a marker. string The input characters are taken from string, starting at the first character in the string and using as many characters as required. function The input characters are generated by function, one character per call. In version 18, function is always called with no arguments and should return a character. t t used as a stream means that the input is read from the minibuffer. In fact, the minibuffer is invoked once and the text given by the user is made into a string that is then used as the input stream. nil nil used as a stream means that the value of standard-input should be used instead; that value is the default input stream, and must be a non-nil input stream. Here is an example of reading from a stream which is a buffer, showing where point is located before and after: ---------- Buffer: foo ---------- This-!- is the contents of foo. ---------- Buffer: foo ---------- (read (get-buffer "foo")) => is (read (get-buffer "foo")) => the ---------- Buffer: foo ---------- This is the -!-contents of foo. ---------- Buffer: foo ---------- Note that the first read skips a space at the beginning of the buffer. Reading skips any amount of whitespace preceding the significant text. Note also that the second read skips the space which terminates the symbol the. It has to read this space in order to know that no more letters follow. Here is an example of reading from a stream that is a marker, initialized to point at the beginning of the buffer shown. The value of the read is the symbol This. ---------- Buffer: foo ---------- This is the contents of foo. ---------- Buffer: foo ---------- (setq m (set-marker (make-marker) 1 (get-buffer "foo"))) => #<marker at 1 in foo> (read m) => This m => #<marker at 6 in foo> ;; After the first space. Here we read from the contents of a string: (read "(When in) the course") => (When in) The following example reads from the minibuffer, prompting with `Lisp expression: '. (That is always the prompt used when you read from the stream t.) The user's input is shown following the prompt. (read t) => 23 ---------- Buffer: Minibuffer ---------- Lisp expression: 23 RET Finally, here is an example of a stream that is a function, named useless-stream. Before we use the stream, we initialize the variable useless-list to a list of characters. Then each call to the function useless-stream obtains the next letter in the list: (setq useless-list (append "XY()" nil)) => (88 89 40 41) (defun useless-stream () (prog1 (car useless-list) (setq useless-list (cdr useless-list)))) => useless-stream Now we read using the stream thus constructed: (read 'useless-stream) => XY useless-list => (41) Note that the close parenthesis remains in the list. This is because the open parenthesis was read before the Lisp reader knew it had found the end of the symbol. A second attempt to read from the stream at this point would get an error due to the unmatched close parenthesis. -- Function: get-file-char This function is used internally as an input stream to read from the input file opened by the function load. Don't use this function yourself. ΓòÉΓòÉΓòÉ 19.3. Input Functions ΓòÉΓòÉΓòÉ This section describes the Lisp functions and variables that pertain to reading. In the functions below, stream stands for an input stream (see the previous section). If stream is nil or omitted, it defaults to the value of standard-input. An end-of-file error will result if an unterminated list or vector is found. -- Function: read &optional stream This function reads one textual Lisp expression from stream, returning it as a Lisp object. This is the basic Lisp input function. -- Function: read-from-string string &optional start end This function reads the first textual Lisp expression from the text in string. It returns a cons cell whose car is that expression, and whose cdr is an integer giving the position of the next remaining character in the string (i.e., the first one not read). If start is supplied, then reading begins at index start in the string (where the first character is at index 0). If end is also supplied, then reading stops at that index as if the rest of the string were not there. For example: (read-from-string "(setq x 55) (setq y 5)") => ((setq x 55) . 11) (read-from-string "\"A short string\"") => ("A short string" . 16) ;; Read starting at the first character. (read-from-string "(list 112)" 0) => ((list 112) . 10) ;; Read starting at the second character. (read-from-string "(list 112)" 1) => (list . 6) ;; Read starting at the seventh character, and stopping at the ninth. (read-from-string "(list 112)" 6 8) => (11 . 8) -- Variable: standard-input This variable holds the default input stream: the stream that read uses when the stream argument is nil. ΓòÉΓòÉΓòÉ 19.4. Output Streams ΓòÉΓòÉΓòÉ An output stream specifies what to do with the characters produced by printing. Most print functions accept an output stream as an optional argument. Here are the possible types of output stream: buffer The output characters are inserted into buffer at point. Point advances as characters are inserted. marker The output characters are inserted into the buffer that marker is in at the marker position. The position advances as characters are inserted. The value of point in the buffer has no effect when the stream is a marker. function The output characters are passed to function, which is responsible for storing them away. It is called with a single character as argument, as many times as there are characters to be output, and is free to do anything at all with the characters it receives. t The output characters are displayed in the echo area. nil nil specified as an output stream means that the value of standard-output should be used as the output stream; that value is the default output stream, and must be a non-nil output stream. Here is an example of a buffer used as an output stream. Point is initially located as shown immediately before the `h' in `the'. At the end, point is located directly before that same `h'. ---------- Buffer: foo ---------- This is t-!-he contents of foo. ---------- Buffer: foo ---------- (print "This is the output" (get-buffer "foo")) => "This is the output" ---------- Buffer: foo ---------- This is t "This is the output" -!-he contents of foo. ---------- Buffer: foo ---------- Now we show a use of a marker as an output stream. Initially, the marker points in buffer foo, between the `t' and the `h' in the word `the'. At the end, the marker has been advanced over the inserted text so that it still points before the same `h'. Note that the location of point, shown in the usual fashion, has no effect. ---------- Buffer: foo ---------- "This is the -!-output" ---------- Buffer: foo ---------- m => #<marker at 11 in foo> (print "More output for foo." marker) => "More output for foo." ---------- Buffer: foo ---------- "This is t "More output for foo." he -!-output" ---------- Buffer: foo ---------- m => #<marker at 35 in foo> The following example shows output to the echo area: (print "Echo Area output" t) => "Echo Area output" ---------- Echo Area ---------- "Echo Area output" ---------- Echo Area ---------- Finally, we show an output stream which is a function. The function eat-output takes each character that it is given and conses it onto the front of the list last-output (see Building Lists). At the end, the list contains all the characters output, but in reverse order. (setq last-output nil) => nil (defun eat-output (c) (setq last-output (cons c last-output))) => eat-output (print "This is the output" 'eat-output) => "This is the output" last-output => (10 34 116 117 112 116 117 111 32 101 104 116 32 115 105 32 115 105 104 84 34 10) Now we can put the output in the proper order by reversing the list: (concat (nreverse last-output)) => " \"This is the output\" " ΓòÉΓòÉΓòÉ 19.4.1. Output Functions ΓòÉΓòÉΓòÉ This section describes the Lisp functions and variables that pertain to printing. Some of the Emacs printing functions add quoting characters to the output when necessary so that it can be read properly. The quoting characters used are `\' and `"'; they are used to distinguish strings from symbols, and to prevent punctuation characters in strings and symbols from being taken as delimiters. See Printed Representation, for full details. You specify quoting or no quoting by the choice of printing function. If the text is to be read back into Lisp, then it is best to print with quoting characters to avoid ambiguity. Likewise, if the purpose is to describe a Lisp object clearly for a Lisp programmer. However, if the purpose of the output is to look nice for humans, then it is better to print without quoting. In the functions below, stream stands for an output stream. (See the previous section for a description of output streams.) If stream is nil or omitted, it defaults to the value of standard-output. -- Function: print object &optional stream The print is a convenient way of printing. It outputs the printed representation of object to stream, printing in addition one newline before object and another after it. Quoting characters are used. print returns object. For example: (progn (print 'The\ cat\ in) (print "the hat") (print " came back")) -| -| The\ cat\ in -| -| "the hat" -| -| " came back" -| => " came back" -- Function: prin1 object &optional stream This function outputs the printed representation of object to stream. It does not print any spaces or newlines to separate output as print does, but it does use quoting characters just like print. It returns object. (progn (prin1 'The\ cat\ in) (prin1 "the hat") (prin1 " came back")) -| The\ cat\ in"the hat"" came back" => " came back" -- Function: prin1-to-string object This function returns a string containing the text that prin1 would have printed for the same argument. (prin1-to-string 'foo) => "foo" (prin1-to-string (mark-marker)) => "#<marker at 2773 in strings.texi>" See format, in String Conversion, for other ways to obtain the printed representation of a Lisp object as a string. -- Function: princ object &optional stream This function outputs the printed representation of object to stream. It returns object. This function is intended to produce output that is readable by people, not by read, so quoting characters are not used and double-quotes are not printed around the contents of strings. It does not add any spacing between calls. (progn (princ 'The\ cat) (princ " in the \"hat\"")) -| The cat in the "hat" => " in the \"hat\"" -- Function: terpri &optional stream This function outputs a newline to stream. The name stands for ``terminate print''. -- Variable: standard-output The value of this variable is the default output stream, used when the stream argument is omitted or nil. -- Variable: print-escape-newlines If this variable is non-nil, then newline characters in strings are printed as `\n'. Normally they are printed as actual newlines. This variable affects the print functions prin1 and print; it does not affect princ in Emacs 18, but this may be changed. Here is an example using prin1: (prin1 "a\nb") -| "a -| b" => "a => b" (let ((print-escape-newlines t)) (prin1 "a\nb")) -| "a\nb" => "a => b" In the second expression, the local binding of print-escape-newlines is in effect during the call to prin1, but not during the printing of the result. -- Variable: print-length The value of this variable is the maximum number of elements of a list that will be printed. If the list being printed has more than this many elements, then it is abbreviated with an ellipsis. If the value is nil (the default), then there is no limit. (setq print-length 2) => 2 (print '(1 2 3 4 5)) -| (1 2 ...) => (1 2 ...) -- Function: write-char character &optional stream This function outputs character to stream. It returns character. ΓòÉΓòÉΓòÉ 20. Minibuffers ΓòÉΓòÉΓòÉ A minibuffer is a special buffer used by Emacs commands to read arguments more complicated than the single numeric prefix argument. These arguments include file names, buffer names, and command names (as in M-x). The minibuffer is displayed on the bottom line of the screen, in the same place as the echo area, but only while it is in use for reading an argument. ΓòÉΓòÉΓòÉ 20.1. Introduction to Minibuffers ΓòÉΓòÉΓòÉ In most ways, a minibuffer is a normal Emacs buffer. Most operations within a buffer, such as editing commands, work normally in a minibuffer. However, many operations for managing buffers do not apply to minibuffers. The name of a minibuffer always has the form ` *Minibuf-number', and it cannot be changed. There is a special window used only for minibuffers, and minibuffers cannot be displayed in any other window. This window is normally the single line at the bottom of the screen; it can be resized temporarily with the window sizing commands, but reverts to its normal size when the minibuffer is exited. A recursive minibuffer may be created when there is an active minibuffer and a command is invoked that requires input from a minibuffer. The first minibuffer is named `*Minibuf-0*'. Recursive minibuffers are named by incrementing the number at the end of the name. (The names begin with a space so that they won't show up in normal buffer lists.) Of several recursive minibuffers, the innermost (or most recently entered) is the active minibuffer, and is the only one that is displayed in a window. We usually call this ``the'' minibuffer. Recursive minibuffers may be allowed or disallowed by setting the variable enable-recursive-minibuffers. Like other buffers, a minibuffer may use any of several local keymaps (see Keymaps); these contain various exit commands and in some cases completion commands. See Completion. o minibuffer-local-map is for ordinary input (no completion). o minibuffer-local-ns-map is similar, except that SPC exits just like RET. This is used mainly for Mocklisp compatibility. o minibuffer-local-completion-map is for permissive completion. o minibuffer-local-must-match-map is for strict completion and for cautious completion. o repeat-complex-command-map is for use in C-x ESC. ΓòÉΓòÉΓòÉ 20.2. Reading Text Strings with the Minibuffer ΓòÉΓòÉΓòÉ The minibuffer is usually used to read text which is returned as a string, but can also be used to read a Lisp object in textual form. The most basic primitive for minibuffer input is read-from-minibuffer. -- Function: read-from-minibuffer prompt-string &optional initial keymap read This function is the most general way to get input through the minibuffer. By default, it accepts arbitrary text and returns it as a string; however, if read is non-nil, then it uses read to convert the text into a Lisp object ( see Input Functions). The first thing this function does is to activate a minibuffer and display it with prompt-string as the prompt. This value must be a string. Then, if initial is non-nil, it must be a string; its contents are inserted into the minibuffer as initial contents. The text thus inserted is treated as if the user had inserted it; the user can alter it with Emacs editing commands. If keymap is non-nil, that keymap is the local keymap to use while reading. If keymap is omitted or nil, the value of minibuffer-local-map is used as the keymap. Specifying a keymap is the most important way to customize minibuffer input for various applications including completion. When the user types a command to exit the minibuffer, the current minibuffer contents are usually made into a string which is the value of read-from-minibuffer. However, if read is non-nil, Emacs converts the result to a Lisp object and read-from-minibuffer returns that object, unevaluated. Suppose, for example, you are writing a search command and want to record the last search string and provide it as a default for the next search. Suppose that the previous search string is stored in the variable last-search-string. Here is how you can read a search string while providing the previous string as initial input to be edited: (read-from-minibuffer "Find string: " last-search-string) Assuming the value of last-search-string is `No', and the user wants to search for `Nope', the interaction looks like this: (setq last-search-string "No") (read-from-minibuffer "Find string: " last-search-string) ---------- Buffer: Minibuffer ---------- Find string: No-!- ---------- Buffer: Minibuffer ---------- ;; The user now types pe RET: => "Nope" -- Function: read-string prompt &optional initial This function reads a string from the minibuffer and returns it. The arguments prompt and initial are used as in read-from-minibuffer. This function is a simplified interface to read-from-minibuffer: (read-string prompt initial) == (read-from-minibuffer prompt initial nil nil) -- Variable: minibuffer-local-map This is the default local keymap for reading from the minibuffer. It is the keymap used by the minibuffer for local bindings in the function read-string. By default, it makes the following bindings: LFD exit-minibuffer RET exit-minibuffer C-g abort-recursive-edit -- Function: read-no-blanks-input prompt initial This function reads a string from the minibuffer, but does not allow whitespace characters as part of the input: instead, those characters terminate the input. The arguments prompt and initial are used as in read-from-minibuffer. This function is a simplified interface to read-from-minibuffer, and passes the value of minibuffer-local-ns-map as the keymap argument for that function. Since the keymap minibuffer-local-ns-map does not rebind C-q, it is possible to put a space into the string, by quoting it. (read-no-blanks-input prompt initial) == (read-from-minibuffer prompt initial minibuffer-local-ns-map) -- Variable: minibuffer-local-ns-map This built-in variable is the keymap used as the minibuffer local keymap in the function read-no-blanks-input. By default, it makes the following bindings: LFD exit-minibuffer SPC exit-minibuffer TAB exit-minibuffer RET exit-minibuffer C-g abort-recursive-edit ? self-insert-and-exit ΓòÉΓòÉΓòÉ 20.3. Reading Lisp Objects with the Minibuffer ΓòÉΓòÉΓòÉ This section describes functions for reading Lisp objects with the minibuffer. -- Function: read-minibuffer prompt &optional initial This function reads a Lisp object in the minibuffer and returns it, unevaluated. The arguments prompt and initial are used as in read-from-minibuffer; in particular, initial must be a string or nil. This function is a simplified interface to read-from-minibuffer: (read-minibuffer prompt initial) == (read-from-minibuffer prompt initial nil t) Here is an example in which we supply the string "(testing)" as initial input: (read-minibuffer "Enter an expression: " (format "%s" '(testing))) ;; Here is how the minibuffer is displayed: ---------- Buffer: Minibuffer ---------- Enter an expression: (testing)-!- ---------- Buffer: Minibuffer ---------- The user can type RET immediately to use the initial input as a default, or can edit the input. -- Function: eval-minibuffer prompt &optional initial This function reads a Lisp expression in the minibuffer, evaluates it, then returns the result. The arguments prompt and initial are used as in read-from-minibuffer. This function simply evaluates the result of a call to read-minibuffer: (eval-minibuffer prompt initial) == (eval (read-minibuffer prompt initial)) -- Function: edit-and-eval-command prompt form This function reads a Lisp expression in the minibuffer, and then evaluates it. The difference between this command and eval-minibuffer is that here the initial form is not optional and it is treated as a Lisp object to be converted to printed representation rather than as a string of text. It is printed with prin1, so if it is a string, double-quote characters (`"') will appear in the initial text. See Output Functions. The first thing edit-and-eval-command does is to activate the minibuffer with prompt as the prompt. The printed representation of form is then inserted in the minibuffer, and the user is allowed to edit. When the user exits the minibuffer, the edited text is read with read and then evaluated. The resulting value becomes the value of edit-and-eval-command. In the following example, we offer the user an expression with initial text which is a valid form already: (edit-and-eval-command "Please edit: " '(forward-word 1)) ;; After evaluating the preceding expression, ;; the following appears in the minibuffer: ---------- Buffer: Minibuffer ---------- Please edit: (forward-word 1)-!- ---------- Buffer: Minibuffer ---------- Typing RET right away would exit the minibuffer and evaluate the expression, thus moving point forward one word. edit-and-eval-command returns nil in this example. ΓòÉΓòÉΓòÉ 20.4. Completion ΓòÉΓòÉΓòÉ Completion is a feature that fills in the rest of a name starting from an abbreviation for it. Completion works by comparing the user's input against a list of valid names and determining how much of the name is determined uniquely by what the user has typed. For example, when you type C-x b (switch-to-buffer) and then type the first few letters of the name of the buffer to which you wish to switch, and then type TAB (minibuffer-complete), Emacs extends the name as far as it can. Standard Emacs commands offer completion for names of symbols, files, buffers, and processes; with the functions in this section, you can implement completion for other kinds of names. The try-completion function is the basic primitive for completion: it returns the longest determined completion of a given initial string, with a given set of strings to match against. The function completing-read provides a higher-level interface for completion. A call to completing-read specifies how to determine the list of valid names. The function then activates the minibuffer with a local keymap that binds a few keys to commands useful for completion. Other functions provide convenient simple interfaces for reading certain kinds of names with completion. ΓòÉΓòÉΓòÉ 20.4.1. Basic Completion Functions ΓòÉΓòÉΓòÉ -- Function: try-completion string alist-or-obarray &optional predicate This function returns the longest common substring of all possible completions of string in alist-or-obarray. If alist-or-obarray is an association list ( see Association Lists), the car of each cons cell in it is compared against string; if the beginning of the car equals string, the cons cell matches. If no cons cells match, try-completion returns nil. If only one cons cell matches, and the match is exact, then try-completion returns t. Otherwise, all matching strings are compared, and the longest initial sequence common to them is returned as a string. If alist-or-obarray is an obarray (see Creating Symbols), the names of all symbols in the obarray form the space of possible names. They are tested and used just like the cars of the elements of an association list. (The global variable obarray holds an obarray containing the names of all interned Lisp symbols.) If the argument predicate is non-nil, then it must be a function of one argument. It is used to test each possible match, and the match is accepted only if predicate returns non-nil. The argument given to predicate is either a cons cell from the alist (the car of which is a string) or else it is a symbol (not a symbol name) from the obarray. It is also possible to use a function as alist-or-obarray. Then the function is solely responsible for performing completion; try-completion returns whatever this function returns. The function is called with three arguments: string, predicate and nil. (The reason for the third argument is so that the same function can be used in all-completions and do the appropriate thing in either case.) See Programmed Completion. In the first of the following examples, the string `foo' is matched by three of the alist cars. All of the matches begin with the characters `fooba', so that is the result. In the second example, there is only one possible match, and it is exact, so the value is t. (try-completion "foo" '(("foobar1" 1) ("barfoo" 2) ("foobaz" 3) ("foobar2" 4))) => "fooba" (try-completion "foo" '(("barfoo" 2) ("foo" 3))) => t In the following example, numerous symbols begin with the characters `forw', and all of them begin with the word `forward'. In most of the symbols, this is followed with a `-', but not in all, so no more than `forward' can be completed. (try-completion "forw" obarray) => "forward" Finally, in the following example, only two of the three possible matches pass the predicate test (the string `foobaz' is too short). Both of those begin with the string `foobar'. (defun test (s) (> (length (car s)) 6)) => test (try-completion "foo" '(("foobar1" 1) ("barfoo" 2) ("foobaz" 3) ("foobar2" 4)) 'test) => "foobar" -- Function: all-completions string alist-or-obarray &optional predicate This function returns a list of all possible completions, instead of the longest substring they share. The parameters to this function are the same as to try-completion. If alist-or-obarray is a function, it is called with three arguments: string, predicate and t, and all-completions returns whatever the function returns. See Programmed Completion. Here is an example, using the same function test used in the example for try-completion: (defun test (s) (> (length (car s)) 6)) => test (all-completions "foo" '(("foobar1" 1) ("barfoo" 2) ("foobaz" 3) ("foobar2" 4)) (function test)) => ("foobar1" "foobar2") -- Variable: completion-ignore-case If the value of this variable is non-nil, Emacs does not consider case significant in completion. The two functions try-completion and all-completions have nothing in themselves to do with minibuffers. However, completion is most often used there, which is why it is described in this chapter. ΓòÉΓòÉΓòÉ 20.4.2. Programmed Completion ΓòÉΓòÉΓòÉ Sometimes it is not possible to create an alist or an obarray containing all the intended possible completions. In such a case, you can supply your own function to compute the completion of a given string. This is called programmed completion. To use this feature, pass the function as the alist-or-obarray argument to completing-read. This command will arrange to pass the function along to try-completion and all-completions, which will then let your function do all the work. The completion function should accept three arguments: o The string to be completed. o The predicate function to filter possible matches, or nil if none. Your function should call the predicale for each possible match and ignore the possible match if the predicate returns nil. o A flag specifying the type of operation. There are three flag values for three operations: o nil specifies try-completion. The completion function should return the completion of the specified string, or t if the string is an exact match already, or nil if the string matches no possibility. o t specifies all-completions. The completion function should return a list of all possible completions of the specified string. o lambda specifies a test for an exact match. The completion function should return t if the specified string is an exact match for some possibility; nil otherwise. Emacs uses programmed completion when completing file names. See File Name Completion. ΓòÉΓòÉΓòÉ 20.4.3. Completion and the Minibuffer ΓòÉΓòÉΓòÉ This section describes the basic interface for reading from the minibuffer with completion. -- Function: completing-read prompt alist-or-obarray &optional predicate require-match initial This function reads a string in the minibuffer, assisting the user by providing completion. It activates the minibuffer with prompt prompt, which must be a string. If initial is non-nil, completing-read inserts it into the minibuffer as part of the input. Then it allows the user to edit the input, providing several commands to attempt completion. The actual completion is done by passing alist-or-obarray and predicate to the function try-completion. This happens in certain commands bound in the local keymaps used for completion. If require-match is t, the user will not be allowed to exit unless the input completes to an element of alist-or-obarray. If require-match is neither nil nor t, then completing-read does not exit unless the input typed is itself an element of alist-or-obarray. To accomplish this, completing-read calls read-minibuffer with the keymap minibuffer-local-completion-map if require-match is nil, or else with the keymap minibuffer-local-must-match-map, if require-match is non-nil. Case is ignored when comparing the input against the possible matches if the built-in variable completion-ignore-case is non-nil. See Basic Completion. For example: (completing-read "Complete a foo: " '(("foobar1" 1) ("barfoo" 2) ("foobaz" 3) ("foobar2" 4)) nil t "fo") ;; After evaluating the preceding expression, ;; the following appears in the minibuffer: ---------- Buffer: Minibuffer ---------- Complete a foo: fo-!- ---------- Buffer: Minibuffer ---------- If the user then types DEL DEL b RET, completing-read returns barfoo. The completing-read function binds three variables to pass information to the commands which actually do completion. Here they are: minibuffer-completion-table This variable is bound to alist-or-obarray argument. It is passed to the try-completion function. minibuffer-completion-predicate This variable is bound to the predicate argument. It is passed to the try-completion function. minibuffer-completion-confirm This variable is bound to the require-match argument. It is used in the minibuffer-complete-and-exit function. ΓòÉΓòÉΓòÉ 20.4.4. Minibuffer Commands That Do Completion ΓòÉΓòÉΓòÉ This section describes the keymaps, commands and user options used in the minibuffer to do completion. -- Variable: minibuffer-local-completion-map completing-read uses this value as the local keymap when an exact match of one of the completions is not required. By default, this keymap makes the following bindings: ? minibuffer-completion-help SPC minibuffer-complete-word TAB minibuffer-complete LFD exit-minibuffer RET exit-minibuffer C-g abort-recursive-edit -- Variable: minibuffer-local-must-match-map completing-read uses this value as the local keymap when an exact match of one of the completions is required. Therefore, no keys are bound to exit-minibuffer, the command which exits the minibuffer unconditionally. By default, this keymap makes the following bindings: ? minibuffer-completion-help SPC minibuffer-complete-word TAB minibuffer-complete LFD minibuffer-complete-and-exit RET minibuffer-complete-and-exit C-g abort-recursive-edit -- Variable: minibuffer-completion-table The value of this variable is the alist or obarray used for completion in the minibuffer. This is the global variable that contains what completing-read passes to try-completion. It is used by all the minibuffer completion functions, such as minibuffer-complete-word. -- Variable: minibuffer-completion-predicate The value of this variable is the predicate that completing-read passes to try-completion. The variable is also used by the other minibuffer completion functions. -- Command: minibuffer-complete-word This function completes the minibuffer contents by at most a single word. Even if the minibuffer contents has only one completion, minibuffer-complete-word will not add any characters beyond the first character that is not a word constituent. See Syntax Tables. -- Command: minibuffer-complete This function completes the minibuffer contents as far as possible. -- Command: minibuffer-complete-and-exit This function completes the minibuffer contents, and exits if confirmation is not required, i.e., if minibuffer-completion-confirm is non-nil. If confirmation is required, it is given by repeating this command immediately. -- Variable: minibuffer-completion-confirm When the value of this variable is non-nil, Emacs asks for confirmation of a completion before exiting the minibuffer. The function minibuffer-complete-and-exit checks the value of this variable before it exits. -- Command: minibuffer-completion-help This function creates a list of the possible completions of the current minibuffer contents. It works by calling all-completions; the values of minibuffer-completion-table and minibuffer-completion-predicate are used as arguments. The list of completions is displayed as text in a buffer named `*Completions*'. -- Function: display-completion-list completions This function displays completions to the stream standard-output (usually a buffer). (See Streams, for more information about streams.) The argument completions is normally a list of completions just returned by all-completions, but it does not have to be. Each element may be a symbol or a string, either of which is simply printed, or a list of two strings, which is printed as if the strings were concatenated. This function is called by minibuffer-completion-help. -- User Option: completion-auto-help If this variable is non-nil, the completion commands automatically display a list of possible completions whenever nothing can be completed because the next character is not uniquely determined. ΓòÉΓòÉΓòÉ 20.4.5. High-Level Completion Functions ΓòÉΓòÉΓòÉ This section describes the higher-level convenient functions for reading certain sorts of names with completion. -- Function: read-buffer prompt &optional default existing This function reads the name of a buffer and returns it as a string. The argument default is the default name to use, the value to return if the user exits with an empty minibuffer. If non-nil, it should be a string. It is mentioned in the prompt, but is not inserted in the minibuffer as initial input. If existing is non-nil, then the name specified must be that of an existing buffer. The usual commands to exit the minibuffer will not exit if the text is not valid, and RET will do completion to attempt to find a valid name. (However, default is not checked for this; it is returned, whatever it is, if the user exits with the minibuffer empty.) In the following example, the user enters `minibuffer.t', and then types RET. The argument existing is t, and the only buffer name starting with the given input is `minibuffer.texi', so that name is the value. (read-buffer "Buffer name? " "foo" t) ;; After evaluating the preceding expression, ;; the following prompt appears, with an empty minibuffer: ---------- Buffer: Minibuffer ---------- Buffer name? (default foo) -!- ---------- Buffer: Minibuffer ---------- ;; The user types minibuffer.t RET. => "minibuffer.texi" -- Function: read-command prompt This function reads the name of a command and returns it as a Lisp symbol. The argument prompt is used as in read-from-minibuffer. Recall that a command is anything for which commandp returns t, and a command name is a symbol for which commandp returns t. See Interactive Call. (read-command "Command name? ") ;; After evaluating the preceding expression, ;; the following appears in the minibuffer: ---------- Buffer: Minibuffer ---------- Command name? ---------- Buffer: Minibuffer ---------- If the user types forward-c RET, then this function returns forward-char. The read-command function is a simplified interface to completing-read. It uses the commandp predicate to allow only commands to be entered, and it uses the variable obarray so as to be able to complete all extant Lisp symbols: (read-command prompt) == (intern (completing-read prompt obarray 'commandp t nil)) -- Function: read-variable prompt This function reads the name of a user variable and returns it as a symbol. (read-variable "Variable name? ") ;; After evaluating the preceding expression, ;; the following prompt appears with an empty minibuffer: ---------- Buffer: Minibuffer ---------- Variable name? -!- ---------- Buffer: Minibuffer ---------- If the user then types fill-p RET, read-variable will return fill-prefix. This function is similar to read-command, but uses the predicate user-variable-p instead of commandp: (read-variable prompt) == (intern (completing-read prompt obarray 'user-variable-p t nil)) ΓòÉΓòÉΓòÉ 20.4.6. Reading File Names ΓòÉΓòÉΓòÉ Here is another high-level completion function, designed for reading a file name. It provides special features including automatic insertion of the default directory. -- Function: read-file-name prompt &optional directory default existing This function reads a file name in the minibuffer, prompting with prompt and providing completion. If default is non-nil, then the value of default will be returned by the function if the user just types RET. If existing is non-nil, then the name must refer to an existing file; then RET performs completion to make the name valid if possible, and then refuses to exit if it is not valid. If the value of existing is neither nil nor t, then RET also requires confirmation after completion. The argument directory specifies the directory to use for completion of relative file names. Usually it is inserted in the minibuffer as initial input as well. It defaults to the current buffer's default directory. Here is an example: (read-file-name "The file is ") ;; After evaluating the preceding expression, ;; the following appears in the minibuffer: ---------- Buffer: Minibuffer ---------- The file is /gp/gnu/elisp/-!- ---------- Buffer: Minibuffer ---------- Typing manual TAB results in the following: ---------- Buffer: Minibuffer ---------- The file is /gp/gnu/elisp/manual.texi-!- ---------- Buffer: Minibuffer ---------- If the user types RET, read-file-name returns "/gp/gnu/elisp/manual.texi". -- User Option: insert-default-directory This variable is used by read-file-name. The value of this variable controls whether read-file-name starts by placing the name of the default directory in the minibuffer. If the value of this variable is nil, then read-file-name does not place any initial input in the minibuffer. In that case, the default directory is still used for completion of relative file names, but is not displayed. For example: ;; Here the minibuffer starts out containing the default directory. (let ((insert-default-directory t)) (read-file-name "The file is ")) ---------- Buffer: Minibuffer ---------- The file is ~lewis/manual/-!- ---------- Buffer: Minibuffer ---------- ;; Here the minibuffer is empty and only the prompt appears on its line. (let ((insert-default-directory nil)) (read-file-name "The file is ")) ---------- Buffer: Minibuffer ---------- The file is -!- ---------- Buffer: Minibuffer ---------- ΓòÉΓòÉΓòÉ 20.4.7. Lisp Symbol Completion ΓòÉΓòÉΓòÉ If you type a part of a symbol, and then type M-TAB (lisp-complete-symbol), it will attempt to fill in as much more of the symbol name as it can. Not only does this save typing, but it can help you with the name of a symbol that you have partially forgotten. -- Command: lisp-complete-symbol This function performs completion on the symbol name preceding point. The name is completed against the symbols in the global variable obarray, and characters from the completion are inserted into the buffer, making the name longer. If there is more than one completion, a list of all possible completions is placed in the `*Help*' buffer. The bell rings if there is no possible completion in obarray. If an open parenthesis immediately precedes the name, only symbols with function definitions are considered. (By reducing the number of alternatives, this may succeed in completing more characters.) Otherwise, symbols with either a function definition, a value, or at least one property are considered. lisp-complete-symbol returns t if the symbol had an exact, and unique, match; otherwise, it returns nil. In the following example, the user has already inserted `(forwa' into the buffer `foo.el'. The command lisp-complete-symbol then completes the name to `(forward-'. ---------- Buffer: foo.el ---------- (forwa-!- ---------- Buffer: foo.el ---------- (lisp-complete-symbol) => nil ---------- Buffer: foo.el ---------- (forward--!- ---------- Buffer: foo.el ---------- ΓòÉΓòÉΓòÉ 20.5. Yes-or-No Queries ΓòÉΓòÉΓòÉ This section describes functions used to ask the user a yes-or-no question. The function y-or-n-p can be answered with a single character; it is useful for questions where an inadvertent wrong answer will not have serious consequences. yes-or-no-p is suitable for more momentous questions, since it requires three or four characters to answer. Strictly speaking, yes-or-no-p uses the minibuffer and y-or-n-p does not; but it seems best to describe them together. -- Function: y-or-n-p prompt This function asks the user a question, expecting input in the echo area. It returns t if the user types y, nil if the user types n. This function also accepts SPC to mean yes and DEL to mean no. The answer is a single character, with no RET needed to terminate it. Upper and lower case are equivalent. ``Asking the question'' means printing prompt in the echo area, followed by the string `(y or n) '. If the input is not one of the expected answers (y, n, SPC, or DEL), the function responds `Please answer y or n.', and repeats the request. This function does not actually use the minibuffer, since it does not allow editing of the answer. It actually uses the echo area ( see The Echo Area), which uses the same screen space as the minibuffer. The cursor moves to the echo area while the question is being asked. In the following example, the user first types q, which is invalid. At the next prompt the user types n. (y-or-n-p "Do you need a lift? ") ;; After evaluating the preceding expression, ;; the following prompt appears in the echo area: ---------- Echo area ---------- Do you need a lift? (y or n) ---------- Echo area ---------- ;; If the user then types q, the following appears: ---------- Echo area ---------- Please answer y or n. Do you need a lift? (y or n) ---------- Echo area ---------- ;; When the user types a valid answer, it is displayed after the question: ---------- Echo area ---------- Do you need a lift? (y or n) y ---------- Echo area ---------- Note that we show successive lines of echo area messages here. Only one will appear on the screen at a time. -- Function: yes-or-no-p prompt This function asks the user a question, expecting input in minibuffer. It returns t if the user enters `yes', nil if the user types `no'. The user must type RET to finalize the response. Upper and lower case are equivalent. yes-or-no-p starts by displaying prompt in the echo area, followed by `(yes or no) '. The user must type one of the expected responses; otherwise, the function responds `Please answer yes or no.', waits about two seconds and repeats the request. yes-or-no-p requires more work from the user than y-or-n-p and is appropriate for more crucial decisions. Here is an example: (yes-or-no-p "Do you really want to remove your entire directory? ") ;; After evaluating the preceding expression, ;; the following prompt appears with an empty minibuffer: ---------- Buffer: minibuffer ---------- Do you really want to remove your entire directory? (yes or no) ---------- Buffer: minibuffer ---------- If the user first types y RET, which is invalid because this function demands the entire word `yes', it responds by displaying these prompts, with a brief pause between them: ---------- Buffer: minibuffer ---------- Please answer yes or no. Do you really want to remove your entire directory? (yes or no) ---------- Buffer: minibuffer ---------- ΓòÉΓòÉΓòÉ 20.6. Minibuffer Miscellany ΓòÉΓòÉΓòÉ Some basic minibuffer functions and variables are described in this section. -- Command: exit-minibuffer This function exits the active minibuffer. It is normally bound to keys in minibuffer local keymaps. -- Command: self-insert-and-exit This function exits the active minibuffer after inserting the last character typed on the keyboard (found in last-command-char; see Command Loop Info). -- Variable: minibuffer-help-form The current value of this variable is used to rebind help-form locally inside the minibuffer (see Help Functions). -- Function: minibuffer-window This function returns the window that is used for the minibuffer. There is one and only one minibuffer window in Emacs 18; this window always exists and cannot be deleted. -- Variable: minibuffer-scroll-window If the value of this variable is non-nil, it should be a window object. When the function scroll-other-window is called in the minibuffer, it will scroll the minibuffer-scroll-window window. Finally, some functions and variables deal with recursive minibuffers (see Recursive Editing): -- Function: minibuffer-depth This function returns the current depth of activations of the minibuffer, a nonnegative integer. If no minibuffers are active, it returns zero. -- User Option: enable-recursive-minibuffers If this variable is non-nil, you can invoke commands (such as find-file) which use minibuffers even while in the minibuffer window. Such invocation produces a recursive editing level for a new minibuffer. The outer-level minibuffer is invisible while you are editing the inner one. This variable only affects invoking the minibuffer while the minibuffer window is selected. If you switch windows while in the minibuffer, you can always invoke minibuffer commands while some other window is selected. ΓòÉΓòÉΓòÉ 21. Command Loop ΓòÉΓòÉΓòÉ When you run Emacs, it enters the editor command loop almost immediately. This loop reads key sequences, executes their definitions, and displays the results. In this chapter, we describe how these things are done, and the subroutines that allow Lisp programs to do them. ΓòÉΓòÉΓòÉ 21.1. Command Loop Overview ΓòÉΓòÉΓòÉ The first thing the command loop must do is read a key sequence, which is a sequence of characters that translates into a command. It does this by calling the function read-key-sequence. Your Lisp code can also call this function (see Keyboard Input). Lisp programs can also do input at a lower level with read-char or discard pending input with discard-input. The key sequence is translated into a command through the keymaps of the current buffer. See Keymaps, for information on how this is done. The result should be a keyboard macro or an interactively callable function. If the key is M-x, then it reads the name of another command, which is used instead. This is done by the command execute-extended-command (see Interactive Call). Once the command is read, it must be executed, which includes reading arguments to be given to it. This is done by calling command-execute (see Interactive Call). For commands written in Lisp, the interactive specification says how to read the arguments. This may use the prefix argument ( see Prefix Command Arguments) or may read with prompting in the minibuffer (see Minibuffers). For example, the command find-file has an interactive specification which says to read a file name using the minibuffer. The command's function body does not use the minibuffer; if you call this command from Lisp code as a function, you must supply the file name string as an ordinary Lisp function argument. If the command is a string (i.e., a keyboard macro) then the function execute-kbd-macro is used to execute it. You can call this function yourself (see Keyboard Macros). If a command runs away, typing C-g will terminate its execution immediately. This is called quitting (see Quitting). ΓòÉΓòÉΓòÉ 21.2. Defining Commands ΓòÉΓòÉΓòÉ A Lisp function becomes a command when its body contains, at top level, a form which calls the special form interactive. This form does nothing when actually executed, but its presence serves as a flag to indicate that interactive calling is permitted. Its argument controls the reading of arguments for an interactive call. ΓòÉΓòÉΓòÉ 21.2.1. Using interactive ΓòÉΓòÉΓòÉ This section describes how to write the interactive form that makes a Lisp function an interactively-callable command. -- Special Form: interactive arg-descriptor This special form declares that the function in which it appears is a command, and that it may therefore be called interactively (via M-x or by entering a key sequence bound to it). The argument arg-descriptor declares the way the arguments to the command are to be computed when the command is called interactively. A command may be called from Lisp programs like any other function, but then the arguments are supplied by the caller and arg-descriptor has no effect. The interactive form has its effect because the command loop (actually, its subroutine call-interactively) scans through the function definition looking for it, before calling the function. Once the function is called, all its body forms including the interactive form are executed, but at this time interactive simply returns nil without even evaluating its argument. There are three possibilities for the argument arg-descriptor: o It may be omitted or nil; then the command is called with no arguments. This leads quickly to an error if the command requires one or more arguments. o It may be a Lisp expression that is not a string; then it should be a form that is evaluated to get a list of arguments to pass to the command. o It may be a string; then its contents should consist of a code character followed by a prompt (if required for that code character). The prompt ends either with the end of the string or with a newline. Here is a simple example: (interactive "bFrobnicate buffer: ") The code letter `b' says to read the name of an existing buffer, with completion. The buffer name will be the sole argument passed to the command. The rest of the string is a prompt. If there is a newline character in the string, it terminates the prompt. If the string does not end there, then the rest of the string should contain another code character and prompt, specifying another argument. Any number of arguments may be specified in this way. If the first character in the string is `*', then an error is signaled if the buffer is read-only. Otherwise, the following character is the first code character. ΓòÉΓòÉΓòÉ 21.2.2. Code Characters for interactive ΓòÉΓòÉΓòÉ The code character descriptions below contain a number of key words, defined here as follows: Completion Provide completion. TAB, SPC, and RET perform name completion because the argument is read using completing-read (see Completion). ? displays a list of possible completions. Existing Require the name of an existing object. An invalid name is not accepted; the commands to exit the minibuffer do not exit if the current input is not valid. Default A default value of some sort is used if the user enters no text in the minibuffer. The default depends on the code character. Prompt A prompt immediately follows the code character. The prompt ends either with the end of the string or with a newline. No I/O This code letter computes an argument without reading any input. Therefore, it does not use a prompt string, and any prompt string you supply is ignored. Here are the code character descriptions for use with interactive: `a' A function name (i.e., a symbol which is fboundp). Existing, Completion, Prompt. `b' The name of an existing buffer. By default, uses the name of the current buffer (see Buffers). Existing, Completion, Default, Prompt. `B' A buffer name. The buffer need not exist. By default, uses the name of a recently used buffer other than the current buffer. Completion, Prompt. `c' A character. The cursor does not move into the echo area. Prompt. `C' A command name (i.e., a symbol satisfying commandp). Existing, Completion, Prompt. `d' The position of point as a number (see Point). No I/O. `D' A directory name. The default is the current default directory of the current buffer, default-directory (see System Environment). Existing, Completion, Default, Prompt. `f' A file name of an existing file (see File Names). The default directory is default-directory. Existing, Completion, Default, Prompt. `F' A file name. The file need not exist. Completion, Default, Prompt. `k' A key sequence (see Keymap Terms). This keeps reading characters until a command (or undefined command) is found in the current key maps. The key sequence argument is represented as a string. The cursor does not move into the echo area. Prompt. This kind of input is used by commands such as describe-key and global-set-key. `m' The position of the mark as a number. No I/O. `n' A number read with the minibuffer. If the input is not a number, the user is asked to try again. The prefix argument, if any, is not used. Prompt. `N' The raw prefix argument. If the prefix argument is nil, then a number is read as with n. Requires a number. Prompt. `p' The numeric prefix argument. (Note that this `p' is lower case.) No I/O. `P' The raw prefix argument. (Note that this `P' is upper case.) See Prefix Command Arguments. No I/O. `r' Point and the mark, as two numeric arguments, smallest first. This is the only code letter that specifies two successive arguments rather than one. No I/O. `s' Arbitrary text, read in the minibuffer and returned as a string (see Text from Minibuffer). Terminate the input with either LFD or RET. (C-q may be used to include either of these characters in the input.) Prompt. `S' An interned symbol whose name is read in the minibuffer. Any whitespace character terminates the input. (Use C-q to include whitespace in the string.) Other characters that normally terminate a symbol (e.g., parentheses and brackets) do not do so here. Prompt. `v' A variable declared to be a user option (i.e., satisfying user-variable-p). See High-Level Completion. Existing, Completion, Prompt. `x' A Lisp object specified in printed representation, terminated with a LFD or RET. The object is not evaluated. See Object from Minibuffer. Prompt. `X' A Lisp form is read as with x, but then evaluated so that its value becomes the argument for the command. Prompt. ΓòÉΓòÉΓòÉ 21.2.3. Examples of Using interactive ΓòÉΓòÉΓòÉ Here are some examples of interactive: (defun foo1 () ; foo1 takes no arguments, (interactive) ; just moves forward two words. (forward-word 2)) => foo1 (defun foo2 (n) ; foo2 takes one argument, (interactive "p") ; which is the numeric prefix. (forward-word (* 2 n))) => foo2 (defun foo3 (n) ; foo3 takes one argument, (interactive "nCount:") ; which is read with the Minibuffer. (forward-word (* 2 n))) => foo3 (defun three-b (b1 b2 b3) "Select three existing buffers (prompting for them in the Minibuffer). Put them into three windows, selecting the last one." (interactive "bBuffer1:\nbBuffer2:\nbBuffer3:") (delete-other-windows) (split-window (selected-window) 8) (switch-to-buffer b1) (other-window 1) (split-window (selected-window) 8) (switch-to-buffer b2) (other-window 1) (switch-to-buffer b3)) => three-b (three-b "*scratch*" "declarations.texi" "*mail*") => nil ΓòÉΓòÉΓòÉ 21.3. Interactive Call ΓòÉΓòÉΓòÉ After the command loop has translated a key sequence into a definition, it invokes that definition using the function command-execute. If the definition is a function that is a command, command-execute calls call-interactively, which reads the arguments and calls the command. You can also call these functions yourself. -- Function: commandp object Returns t if object is suitable for calling interactively; that is, if object is a command. Otherwise, returns nil. The interactively callable objects include strings (treated as keyboard macros), lambda expressions that contain a top-level call to interactive, autoload objects that are declared as interactive (non-nil fourth argument to autoload), and some of the primitive functions. A symbol is commandp if its function definition is commandp. Keys and keymaps are not commands. Rather, they are used to look up commands (see Keymaps). See documentation in Accessing Documentation, for a realistic example of using commandp. -- Function: call-interactively command &optional record-flag This function calls the interactively callable function command, reading arguments according to its interactive calling specifications. An error is signaled if command cannot be called interactively (i.e., it is not a command). Note that strings are not accepted, even though they are considered commands. If record-flag is non-nil, then this command and its arguments are unconditionally added to the list command-history. Otherwise, the command is added only if it uses the minibuffer to read an argument. See Command History. -- Function: command-execute command &optional record-flag This function executes command as an editing command. The argument command must satisfy the commandp predicate; i.e., it must be an interactively callable function or a string. A string as command is executed with execute-kbd-macro. A function is passed to call-interactively, along with the optional record-flag. A symbol is handled by using its function definition in its place. A symbol with an autoload definition counts as a command if it was declared to stand for an interactively callable function. Such a definition is handled by loading the specified library and then rechecking the definition of the symbol. -- Command: execute-extended-command prefix-argument This primitive function reads a command name from the minibuffer using completing-read (see Completion). Then it uses command-execute to call the specified command. Whatever that command returns becomes the value of execute-extended-command. If the command asks for a prefix argument, the value prefix-argument is supplied. If execute-extended-command is called interactively, the current raw prefix argument is used for prefix-argument, and thus passed on to whatever command is run. execute-extended-command is the normal definition of M-x, so it uses the string `M-x ' as a prompt. (It would be better to take the prompt from the characters used to invoke execute-extended-command, but that is painful to implement.) A description of the value of the prefix argument, if any, also becomes part of the prompt. (execute-extended-command 1) ---------- Buffer: Minibuffer ---------- M-x forward-word RET ---------- Buffer: Minibuffer ---------- => t -- Function: interactive-p This function returns t if the containing function (the one that called interactive-p) was called interactively, with call-interactively. (It makes no difference whether call-interactively was called from Lisp or directly from the editor command loop.) Note that if the containing function was called by Lisp evaluation (or with apply or funcall), then it was not called interactively. The usual application of interactive-p is for deciding whether to print an informative message. As a special exception, interactive-p returns nil whenever a keyboard macro is being run. This is to suppress the informative messages and speed execution of the macro. For example: (defun foo () (interactive) (and (interactive-p) (message "foo"))) => foo (defun bar () (interactive) (setq foobar (list (foo) (interactive-p)))) => bar ;; Type M-x foo. -| foo ;; Type M-x bar. ;; This does not print anything. foobar => (nil t) ΓòÉΓòÉΓòÉ 21.4. Information from the Command Loop ΓòÉΓòÉΓòÉ The editor command loop sets several Lisp variables to keep status records for itself and for commands that are run. -- Variable: last-command This variable records the name of the previous command executed by the command loop (the one before the current command). Normally the value is a symbol with a function definition, but this is not guaranteed. The value is set by copying the value of this-command when a command returns to the command loop, except when the command specifies a prefix argument for the following command. -- Variable: this-command This variable records the name of the command now being executed by editor command loop. Like last-command, it is normally a symbol with a function definition. This variable is set by the command loop just before the command is run, and its value is copied into last-command when the command finishes (unless the command specifies a prefix argument for the following command). Some commands change the value of this variable during their execution, simply as a flag for whatever command runs next. In particular, the functions that kill text set this-command to kill-region so that any kill commands immediately following will know to append the killed text to the previous kill. -- Function: this-command-keys This function returns a string containing the key sequence that invoked the present command, plus any previous commands that generated the prefix argument for this command. (this-command-keys) ;; Now type C-u C-x C-e. => "^U^X^E" -- Variable: last-command-char This variable is set to the last character that was typed on the terminal and was part of a command. The principal use of this variable is in self-insert-command, which uses it to decide which character to insert. last-command-char ;; Now type C-u C-x C-e. => 5 The value is 5 because that is the ASCII code for C-e. -- Variable: echo-keystrokes This variable determines how much time should elapse before command characters are echoed. Its value must be an integer, which specifies the number of seconds to wait before echoing. If the user types a prefix key (say C-x) and then delays this many seconds before continuing, the key C-x is echoed in the echo area. Any subsequent characters in the same command will be echoed as well. If the value is zero, then command input is not echoed. ΓòÉΓòÉΓòÉ 21.5. Keyboard Input ΓòÉΓòÉΓòÉ The editor command loop reads keyboard input using read-key-sequence, which uses read-char. These and other functions for keyboard input are also available for use in Lisp programs. See also momentary-string-display in Temporary Displays, and sit-for in Waiting. See Terminal Input, for functions and variables for controlling terminal input modes and debugging terminal input. -- Function: read-char This function reads a character from the command input (either from direct keyboard input or from an executing keyboard macro), and returns it. No message is displayed to indicate that keyboard input is expected. If you want to display a message, call message first. If cursor-in-echo-area is non-nil, then the cursor moves to the echo area, to the end of any message displayed there. Otherwise the cursor does not move. See The Echo Area. In the first example, the user types 1 (which is ASCII code 49). The second example shows a keyboard macro definition that calls read-char from the minibuffer. read-char reads the keyboard macro's very next character, which is 1. The value of this function is displayed in the echo area by the command eval-expression. (read-char) => 49 (symbol-function 'foo) => "^[^[(read-char)^M1" (execute-kbd-macro foo) -| 49 => nil -- Function: read-quoted-char &optional prompt This function is like read-char, except that if the first character read is an octal digit (0-7), it reads up to two more octal digits (but stopping if a non-octal digit is found) and returns the character represented by those digits as an octal number. Quitting is suppressed when the first character is read, so that the user can enter a C-g. See Quitting. If prompt is supplied, it specifies a string for prompting the user. The prompt string is always printed in the echo area and followed by a single `-'. In the following example, the user types in the octal number 177 (which is 127 in decimal). (read-quoted-char "What character") ---------- Echo Area ---------- What character-177 ---------- Echo Area ---------- => 127 -- Function: read-key-sequence prompt This function reads a key sequence and returns it as a string. It keeps reading characters until it has accumulated a full key sequence; that is, enough characters to specify a non-prefix command using the current local and global keymaps. read-key-sequence is used by the command loop to read command input. If an input character is an upper case letter and has no definition, but the lower case equivalent is defined, then the character is converted to lower case. Note that lookup-key does not perform case conversion in this way. Quitting is suppressed inside read-key-sequence. In other words, a C-g typed while reading with this function is treated like any other character, and quit-flag is not set. See Quitting. The argument prompt is either a string to be displayed in the echo area as a prompt, or nil, meaning that no prompt is displayed. In the example below, the prompt `?' is displayed in the echo area, and the user types C-x C-f. (read-key-sequence "?") ---------- Echo Area ---------- ?C-x C-f ---------- Echo Area ---------- => "^X^F" -- Variable: unread-command-char This variable holds a character waiting to be read as the next input from the command input stream, or to the integer -1 if no character is waiting. The variable is used because in some cases an input function reads a character and then decides not to use it. Storing the character in this variable causes it to be processed normally by the command loop or when read-char is next called. For example, the function that governs prefix arguments reads any number of digits. When it finds a non-digit character, it must unread the character so that it becomes input for the next command. Likewise, incremental search uses this feature to unread a control character used to terminate the search. -- Function: input-pending-p This function determines whether command input is currently available. It returns immediately, with value t if there is input, nil otherwise. On rare occasions it may return t when no input is available. -- Variable: last-input-char This variable records the last terminal input character read, whether as part of a command or explicitly by a Lisp program. In the example below, a character is read (the character 1, ASCII code 49). It becomes the value of last-input-char, while C-e (from the C-x C-e command used to evaluate this expression) remains the value of last-command-char. (progn (print (read-char)) (print last-command-char) last-input-char) -| 49 -| 5 => 49 -- Function: discard-input This function discards the contents of the terminal input buffer and cancels any keyboard macro that might be in the process of definition. It returns nil. In the following example, the user may type a number of characters right after starting the evaluation of the form. After the sleep-for finishes sleeping, any characters that have been typed are discarded. (progn (sleep-for 2) (discard-input)) => nil ΓòÉΓòÉΓòÉ 21.6. Quitting ΓòÉΓòÉΓòÉ Typing C-g while the command loop has run a Lisp function causes Emacs to quit whatever it is doing. This means that control returns to the innermost active command loop. Typing C-g while the command loop is waiting for keyboard input does not cause a quit; it acts as an ordinary input character. In the simplest case, you cannot tell the difference, because C-g normally runs the command keyboard-quit, whose effect is to quit. However, when C-g follows a prefix key, the result is an undefined key. The effect is to cancel the prefix key as well as any prefix argument. In the minibuffer, C-g has a different definition: it aborts out of the minibuffer. This means, in effect, that it exits the minibuffer and then quits. (Simply quitting would return to the command loop within the minibuffer.) The reason why C-g does not quit directly when the command reader is reading input is so that its meaning can be redefined in the minibuffer in this way. C-g following a prefix key is not redefined in the minibuffer, and it has its normal effect of canceling the prefix key and prefix argument. This too would not be possible if C-g quit directly. C-g causes a quit by setting the variable quit-flag to a non-nil value. Emacs checks this variable at appropriate times and quits if it is not nil. Setting quit-flag non-nil in any way thus causes a quit. At the level of C code, quits cannot happen just anywhere; only at the special places which check quit-flag. The reason for this is that quitting at other places might leave an inconsistency in Emacs's internal state. Because quitting is delayed until a safe place, quitting cannot make Emacs crash. Certain functions such as read-key-sequence or read-quoted-char prevent quitting entirely even though they wait for input. Instead of quitting, C-g serves as the requested input. In the case of read-key-sequence, this serves to bring about the special behavior of C-g in the command loop. In the case of read-quoted-char, this is so that C-q can be used to quote a C-g. You can prevent quitting for a portion of a Lisp function by binding the variable inhibit-quit to a non-nil value. Then, although C-g still sets quit-flag to t as usual, the usual result of this---a quit---is prevented. Eventually, inhibit-quit will become nil again, such as when its binding is unwound at the end of a let form. At that time, if quit-flag is still non-nil, the requested quit happens immediately. This behavior is ideal for a ``critical section'', where you wish to make sure that quitting does not happen within that part of the program. In some functions (such as read-quoted-char), C-g is handled in a special way which does not involve quitting. This is done by reading the input with inhibit-quit bound to t and setting quit-flag to nil before inhibit-quit becomes nil again. This excerpt from the definition of read-quoted-char shows how this is done; it also shows that normal quitting is permitted after the first character of input. (defun read-quoted-char (&optional prompt) "...documentation..." (let ((count 0) (code 0) char) (while (< count 3) (let ((inhibit-quit (zerop count)) (help-form nil)) (and prompt (message "%s-" prompt)) (setq char (read-char)) (if inhibit-quit (setq quit-flag nil))) ...) (logand 255 code))) -- Variable: quit-flag If this variable is non-nil, then Emacs quits immediately, unless inhibit-quit is non-nil. Typing C-g sets quit-flag non-nil, regardless of inhibit-quit. -- Variable: inhibit-quit This variable determines whether Emacs should quit when quit-flag is set to a value other than nil. If inhibit-quit is non-nil, then quit-flag has no special effect. -- Command: keyboard-quit This function signals the quit condition with (signal 'quit nil). This is the same thing that quitting does. (See signal in Errors.) You can specify a character other than C-g to use for quitting. See the function set-input-mode in Terminal Input. ΓòÉΓòÉΓòÉ 21.7. Prefix Command Arguments ΓòÉΓòÉΓòÉ Most Emacs commands can use a prefix argument, a number specified before the command itself. (Don't confuse prefix arguments with prefix keys.) The prefix argument is represented by a value that is always available (though it may be nil, meaning there is no prefix argument). Each command may use the prefix argument or ignore it. There are two representations of the prefix argument: raw and numeric. The editor command loop uses the raw representation internally, and so do the Lisp variables that store the information, but commands can request either representation. Here are the possible values of a raw prefix argument: o nil, meaning there is no prefix argument. Its numeric value is 1, but numerous commands make a distinction between nil and the integer 1. o An integer, which stands for itself. o A list of one element, which is an integer. This form of prefix argument results from one or a succession of C-u's with no digits. The numeric value is the integer in the list, but some commands make a distinction between such a list and an integer alone. o The symbol -. This indicates that M-- or C-u - was typed, without following digits. The equivalent numeric value is -1, but some commands make a distinction between the integer -1 and the symbol -. The various possibilities may be illustrated by calling the following function with various prefixes: (defun print-prefix (arg) "Print the value of the raw prefix arg at point." (interactive "P") (message "%s" arg)) Here are the results of calling print-prefix with various raw prefix arguments: M-x print-prefix -| nil C-u M-x print-prefix -| (4) C-u C-u M-x print-prefix -| (16) C-u 3 M-x print-prefix -| 3 M-3 M-x print-prefix -| 3 ; (Same as C-u 3.) C-u - M-x print-prefix -| - M- - M-x print-prefix -| - ; (Same as C-u -.) C-u -7 M-x print-prefix -| -7 M- -7 M-x print-prefix -| -7 ; (Same as C-u -7.) There are two variables used to store the prefix argument: prefix-arg and current-prefix-arg. Commands such as universal-argument that set up prefix arguments for other commands store them in prefix-arg. In contrast, current-prefix-arg conveys the prefix argument to the current command, so setting it has no effect on the prefix arguments for future commands. Normally, commands specify which representation to use for the prefix argument, either numeric or raw, in the interactive declaration. (See Interactive Call.) Alternatively, functions may look at the value of the prefix argument directly in the variable current-prefix-arg, but this is less clean. Don't call universal-argument, digit-argument, or negative-argument unless you intend to let the user enter the prefix argument for the next command. -- Command: universal-argument This command reads input and specifies a prefix argument for the following command. Don't call this command yourself unless you know what you are doing. -- Command: digit-argument arg This command adds to the prefix argument for the following command. The argument arg is the raw prefix argument as it was before this command; it is used to compute the updated prefix argument. Don't call this command yourself unless you know what you are doing. -- Command: negative-argument arg This command adds to the numeric argument for the next command. The argument arg is the raw prefix argument as it was before this command; its value is negated to form the new prefix argument. Don't call this command yourself unless you know what you are doing. -- Function: prefix-numeric-value arg This function returns the numeric meaning of a valid raw prefix argument value, arg. The argument may be a symbol, a number, or a list. If it is nil, the value 1 is returned; if it is any other symbol, the value -1 is returned. If it is a number, that number is returned; if it is a list, the car of that list (which should be a number) is returned. -- Variable: current-prefix-arg This variable is the value of the raw prefix argument for the current command. Commands may examine it directly, but the usual way to access it is with (interactive "P"). -- Variable: prefix-arg The value of this variable is the raw prefix argument for the next editing command. Commands that specify prefix arguments for the following command work by setting this variable. ΓòÉΓòÉΓòÉ 21.8. Recursive Editing ΓòÉΓòÉΓòÉ The Emacs command loop is entered automatically when Emacs starts up. This top-level invocation of the command loop is never exited until the Emacs is killed. Lisp programs can also invoke the command loop. Since this makes more than one activation of the command loop, we call it recursive editing. A recursive editing level has the effect of suspending whatever command invoked it and permitting the user to do arbitrary editing before resuming that command. The commands available during recursive editing are the same ones available in the top-level editing loop and defined in the keymaps. Only a few special commands exit the recursive editing level; the others return to the recursive editing level when finished. (The special commands for exiting are always available, but do nothing when recursive editing is not in progress.) All command loops, including recursive ones, set up all-purpose error handlers so that an error in a command run from the command loop will not exit the loop. Minibuffer input is a special kind of recursive editing. It has a few special wrinkles, such as enabling display of the minibuffer and the minibuffer window, but fewer than you might suppose. Certain keys behave differently in the minibuffer, but that is only because of the minibuffer's local map; if you switch windows, you get the usual Emacs commands. To invoke a recursive editing level, call the function recursive-edit. This function contains the command loop; it also contains a call to catch with tag exit, which makes it possible to exit the recursive editing level by throwing to exit (see Catch and Throw). If you throw a value other than t, then recursive-edit returns normally to the function that called it. The command C-M-c (exit-recursive-edit) does this. Throwing a t value causes recursive-edit to quit, so that control returns to the command loop one level up. This is called aborting, and is done by C-] (abort-recursive-edit). Most applications should not use recursive editing, except as part of using the minibuffer. Usually it is more convenient for the user if you change the major mode of the current buffer temporarily to a special major mode, which has a command to go back to the previous mode. (This technique is used by the w command in Rmail.) Or, if you wish to give the user different text to edit ``recursively'', create and select a new buffer in a special mode. In this mode, define a command to complete the processing and go back to the previous buffer. (The m command in Rmail does this.) Recursive edits are useful in debugging. You can insert a call to debug into a function definition as a sort of breakpoint, so that you can look around when the function gets there. debug invokes a recursive edit but also provides the other features of the debugger. Recursive editing levels are also used when you type C-r in query-replace or use C-x q (kbd-macro-query). -- Function: recursive-edit This function invokes the editor command loop. It is called automatically by the initialization of Emacs, to let the user begin editing. When called from a Lisp program, it enters a recursive editing level. In the following example, the function simple-rec first advances point one word, then enters a recursive edit, printing out a message in the echo area. The user can then do any editing desired, and then type C-M-c to exit and continue executing simple-rec. (defun simple-rec () (forward-word 1) (message "Recursive edit in progress.") (recursive-edit) (forward-word 1)) => simple-rec (simple-rec) => nil -- Command: exit-recursive-edit This function exits from the innermost recursive edit (including minibuffer input). Its definition is effectively (throw 'exit nil). -- Command: abort-recursive-edit This function aborts the command that requested the innermost recursive edit (including minibuffer input), by signaling quit after exiting the recursive edit. Its definition is effectively (throw 'exit t). See Quitting. -- Command: top-level This function exits all recursive editing levels; it does not return a value, as it jumps completely out of any computation directly back to the main command loop. -- Function: recursion-depth This function returns the current depth of recursive edits. When no recursive edit is active, it returns 0. ΓòÉΓòÉΓòÉ 21.9. Disabling Commands ΓòÉΓòÉΓòÉ Disabling a command marks the command as requiring user confirmation before it can be executed. Disabling is used for commands which might be confusing to beginning users, to prevent them from using the commands by accident. The low-level mechanism for disabling a command is to put a non-nil disabled property on the Lisp symbol for the command. These properties are normally set up by the user's `.emacs' file with Lisp expressions such as this: (put 'upcase-region 'disabled t) For a few commands, these properties are present by default and may be removed by the `.emacs' file. If the value of the disabled property is a string, that string is included in the message printed when the command is used: (put 'delete-region 'disabled "Text deleted this way cannot be yanked back!\n") See Disabling, for the details on what happens when a disabled command is invoked interactively. Disabling a command has no effect on calling it as a function from Lisp programs. -- Command: enable-command command Allow command to be executed without special confirmation from now on. The user's `.emacs' file is optionally altered so that this will apply to future sessions. -- Command: disable-command command Require special confirmation to execute command from now on. The user's `.emacs' file is optionally altered so that this will apply to future sessions. -- Variable: disabled-command-hook The value of this variable is a function to be called instead of any command that is disabled (i.e., that has a non-nil disabled property). By default, the value of disabled-command-hook is a function defined to ask the user whether to proceed. ΓòÉΓòÉΓòÉ 21.10. Command History ΓòÉΓòÉΓòÉ The command loop keeps a history of the complex commands that have been executed, to make it convenient to repeat these commands. A complex command is one for which the interactive argument reading uses the minibuffer. This includes any M-x command, any M-ESC command, and any command whose interactive specification reads an argument from the minibuffer. Explicit use of the minibuffer during the execution of the command itself does not cause the command to be considered complex. -- Variable: command-history This variable's value is a list of recent complex commands, each represented as a form to evaluate. It continues to accumulate all complex commands for the duration of the editing session, but all but the first (most recent) thirty elements are deleted when a garbage collection takes place (see Garbage Collection). command-history => ((switch-to-buffer "chistory.texi") (describe-key "^X^[") (visit-tags-table "~/emacs/src/") (find-tag "repeat-complex-command")) There are a number of commands and even two entire modes devoted to facilitating the editing and recall of previous commands. The commands repeat-complex-command, and list-command-history are described in the user manual ( see Repetition). -- Variable: repeat-complex-command-map The value of this variable is a sparse keymap used by the minibuffer inside of read-complex-command. ΓòÉΓòÉΓòÉ 21.11. Keyboard Macros ΓòÉΓòÉΓòÉ A keyboard macro is a canned sequence of keystrokes that can be considered a command and made the definition of a key. Don't confuse keyboard macros with Lisp macros (see Macros). -- Function: execute-kbd-macro macro &optional count This function executes macro as a string of editor commands. If macro is a string, then the characters in that string are executed exactly as if they had been typed as command input. If macro is a symbol, then its function definition is used in place of macro. If that is another symbol, this process repeats. Eventually the result should be a string. If the result is neither a symbol nor a string, an error is signaled. The argument count is a repeat count; macro is executed that many times. If count is omitted or nil, macro is executed once. If it is 0, macro is executed over and over until it encounters an error or a failing search. -- Variable: last-kbd-macro This variable is the definition of the most recently defined keyboard macro. Its value is a string or nil. -- Variable: executing-macro This variable contains the string that defines the keyboard macro that is currently executing. It is nil if no macro is currently executing. -- Variable: defining-kbd-macro This variable indicates whether a keyboard macro is being defined. It is set to t by start-kbd-macro, and nil by end-kbd-macro. It is not hard to use this variable to make a command behave differently when run from a keyboard macro (perhaps indirectly by calling interactive-p). However, do not set this variable yourself. The user-level commands for defining, running and editing keyboard macros include call-last-kbd-macro, insert-kbd-macro, start-kbd-macro, end-kbd-macro, kbd-macro-query, and name-last-kbd-macro. They are described in the user's manual (see Keyboard Macros). ΓòÉΓòÉΓòÉ 22. Keymaps ΓòÉΓòÉΓòÉ The bindings between keyboard input and commands are recorded in data structures called keymaps. Each binding in a keymap associates (or binds) an individual character either with another keymap or with a command. When a character is bound to a keymap, that keymap is used to look up the next character typed; this continues until a command is found. This process is called key lookup. ΓòÉΓòÉΓòÉ 22.1. Keymaps: Terminology ΓòÉΓòÉΓòÉ A keymap is a table mapping characters to definitions (which can be any Lisp objects, though only certain types are meaningful for execution by the command loop). Given a character and a keymap, Emacs can get the character's definition. A sequence of keyboard input characters that form a unit is called a key sequence, or key for short. A sequence of one character is always a key sequence, and so are some multicharacter sequences. A keymap determines a binding or definition for any key sequence. If the key sequence is a single character, its binding is the definition of the character in the keymap. The binding of a multicharacter key sequence is found by an iterative process: the binding of the first character is found, and must be a keymap; then the second character's binding is found in that keymap, and so on until all the characters in the key sequence are used up. If the binding of a key sequence is a keymap, we call the key sequence a prefix key. Otherwise, we call it a complete key (because no more characters can be added to it). If the binding is nil, we call the key undefined. Examples of prefix keys are C-c, C-x, and C-x 4. Examples of defined complete keys are X, RET, and C-x 4 C-f. Examples of undefined complete keys are C-x C-g, and C-c 3. See Prefix Keys, for more details. The rule for finding the binding of a key sequence assumes that the intermediate bindings (found for the characters before the last) are all keymaps; if this is not so, the sequence of characters does not form a unit---it is not really a key sequence. In other words, removing one or more characters from the end of any key must always yield a prefix key. For example, C-f C-f is not a key; C-f is not a prefix key, so a longer sequence starting with C-f cannot be a key. Note that the set of possible multicharacter key sequences depends on the bindings for prefix keys; therefore, it can be different for different keymaps, and can change when bindings are changed. However, a one-character sequence is always a key sequence, because it does not depend on any prefix keys for its validity. At any time, two primary keymaps are in use for finding key bindings: the global map, which is shared by all buffers, and the local keymap, which is usually associated with a specific major mode. The local keymap bindings shadow (i.e., take precedence over) the corresponding global bindings. See Global and Local Keymaps, for details. ΓòÉΓòÉΓòÉ 22.2. Creating Keymaps ΓòÉΓòÉΓòÉ A keymap can be represented as one of two kinds of Lisp object: a vector or a list. A full keymap is a vector of length 128. The binding for a character in such a keymap is found by indexing into the vector with the character as the index. A sparse keymap is a list whose car is the symbol keymap, and whose remaining elements are cons cells of the form (char . binding). It is called a sparse keymap because it stores only the entries which are significant. Use a sparse keymap when you expect only a few entries. (define-key automatically creates sparse keymaps for intermediate keymaps.) Keymaps record directly only character codes less than 128; they are unable to handle directly the META characters, whose codes are from 128 to 255. Instead, META characters are regarded for purposes of key lookup as sequences of two characters, the first of which is ESC (the usual value of meta-prefix-char). Thus, the key M-a is really represented as ESC a, and its global binding is found at the slot for `a' in esc-map. Here as an example is the local keymap for Lisp mode, a sparse keymap. It defines C-c C-l as the run-lisp command, M-C-q as indent-sexp, and M-C-x as lisp-send-defun. lisp-mode-map => (keymap (9 . lisp-indent-line) ; TAB (127 . backward-delete-char-untabify) ; DEL (3 keymap (12 . run-lisp)) ; C-c C-l (27 keymap (17 . indent-sexp) ; M-C-q, treated as ESC C-q (24 . lisp-send-defun))) ; M-C-x, treated as ESC C-x -- Function: keymapp object This function returns t if object is a keymap, nil otherwise. A keymap is either a vector of length 128, or a list with the form (keymap pairs...), where pairs stands for a series of associations, cons cells of the form (char . binding). (keymapp '(keymap)) => t (keymapp (current-global-map)) => t -- Function: make-keymap This function creates and returns a new full keymap (i.e., a vector of length 128). All entries in the keymap are nil, which means that no characters are defined. (make-keymap) => [nil nil nil ... nil nil] -- Function: make-sparse-keymap This function creates and returns a new sparse keymap with no entries. In this keymap, no characters are defined. (make-sparse-keymap) => (keymap) -- Function: copy-keymap keymap This function returns a copy of keymap. Any keymaps which appear directly as bindings in keymap are also copied recursively, and so on to any number of levels. However, recursive copying does not take place when the definition of a character is a symbol whose function definition is a keymap; the same symbol appears in the new copy. (setq map (copy-keymap (current-local-map))) => (keymap (27 keymap ; (This implements META characters.) (83 . center-paragraph) (115 . center-line)) (9 . tab-to-tab-stop)) (eq map (current-local-map)) => nil (equal map (current-local-map)) => t ΓòÉΓòÉΓòÉ 22.3. Key Lookup ΓòÉΓòÉΓòÉ Key lookup is the process of finding the binding of a key sequence from a given keymap. Actual execution of the definition is not part of key lookup. When the key sequence consists of multiple characters, the characters are handled sequentially: the binding of the first character is found, and must be a keymap; then the second character's binding is found in that keymap, and so on until all the characters in the key sequence are used up. (The binding thus found for the last character may or may not be a keymap.) Thus, the process of key lookup is defined in terms of a simpler process for looking up a single character in a keymap. How this is done depends on the type of object associated with the character in that keymap. The value directly associated with a character in a keymap is called a keymap entry. While any Lisp object may be stored in a keymap entry, only a few types of object make sense for key lookup. Here is a list of them, and what they mean: nil nil means that the characters used so far in the lookup form an undefined key. When a sparse keymap fails to mention a character, that is equivalent to an entry of nil. keymap The characters used so far in the lookup form a prefix key. The next character of the key sequence is looked up in keymap, which may be full or sparse. command The characters used so far in the lookup form a complete key, and command is its definition. string string represents a keyboard macro. The characters used so far in the lookup form a complete key, and string is its definition. (See Keyboard Macros, for more information.) list The meaning of a list depends on the types of the elements of the list. o If the car of list is the symbol keymap, then the list is a sparse keymap, and is treated as a keymap (see above). o If the car of list is lambda, then the list is a lambda expression. This is presumed to be a command, and is treated as such (see above). o If the car of list is a keymap and the cdr is a character, then this entry is an indirection to a slot in the other keymap. When an indirect entry is found in key lookup, it is immediately replaced by the entry in the specified keymap for the specified character. This permits you to define one key as an alias for another key. For example, an entry whose car is the keymap called esc-map and whose cdr is 32 (the code for space) means, ``Use the global definition of Meta-SPC, whatever that may be.'' symbol The function definition of symbol is used in place of symbol. If that too is a symbol, then this process is repeated, any number of times. Ultimately this should lead to a definition which is a keymap, a command or a string. A list is allowed if it is a keymap or a command, but indirect entries are not understood when found via symbols. Note that keymaps and strings are not valid functions, so a symbol with a keymap or string as its function definition is likewise not valid as a function. It is, however, valid as a key binding. If the definition is a string, then the symbol is also valid as an argument to command-execute (see Interactive Call). The symbol undefined is worth special mention: it means to treat the key as undefined. Strictly speaking, the key is defined, and its definition is the symbol undefined, but that command does the same thing that is done automatically for an undefined key: it rings the bell (by calling ding) but does not signal an error. undefined is used in local keymaps to override a global key binding and make the key undefined locally. A local binding of nil would fail to do this because it would not override the global binding. anything else If any other type of object is found, the characters used so far in the lookup form a complete key, and the object is its definition. In short, a keymap entry may be a keymap, a command, a string, a symbol which leads to one of them, or an indirection or nil. Here is an example of a sparse keymap with two characters bound to commands and one bound to another keymap. This map is the normal value of emacs-lisp-mode-map. Note that 9 is the code for TAB, 127 for DEL, 27 for ESC, 17 for C-q and 24 for C-x. (keymap (9 . lisp-indent-line) (127 . backward-delete-char-untabify) (27 keymap (17 . indent-sexp) (24 . eval-defun))) ΓòÉΓòÉΓòÉ 22.4. Functions for Key Lookup ΓòÉΓòÉΓòÉ Here are the functions and variables pertaining to key lookup. -- Function: lookup-key keymap key This function returns the definition of key in keymap. If the string key is not a valid key sequence according to the prefix keys specified in keymap (which means it is ``too long'' and has extra characters at the end), then the value is a number, the number of characters at the front of key that compose a complete key. All the other functions described in this chapter that look up keys use lookup-key. (lookup-key (current-global-map) "\C-x\C-f") => find-file (lookup-key (current-global-map) "\C-x\C-f12345") => 2 If key contains a meta-character, that character is implicitly replaced by a two-character sequence: the value of meta-prefix-char, followed by the corresponding non-meta character. Thus, the first example below is handled by conversion into the second example. (lookup-key (current-global-map) "\M-f") => forward-word (lookup-key (current-global-map) "\ef") => forward-word This function does not perform automatic downcasing like that of read-key-sequence (see Keyboard Input). -- Command: undefined Used in keymaps to undefine keys. It calls ding, but does not cause an error. -- Variable: meta-prefix-char This variable is the meta-prefix character code. It is used when translating a meta-character to a two-character sequence so it can be looked up in a keymap. For useful results, the value should be a prefix character (see Prefix Keys). The default value is 27, which is the ASCII code for ESC. As long as the value of meta-prefix-char remains 27, key lookup translates M-b into ESC b, which is normally defined as the backward-word command. However, if you set meta-prefix-char to 24, the code for C-x, then Emacs will translate M-b into C-x b, and call the switch-to-buffer command. meta-prefix-char ; The default value. => 27 (key-binding "\M-b") => backward-word ?\C-x ; The print representation => 24 ; of a character. (setq meta-prefix-char 24) => 24 (key-binding "\M-b") => switch-to-buffer ; Now, typing M-b is ; like typing C-x b. (setq meta-prefix-char 27) ; Avoid confusion! => 27 ; Restore the default value! ΓòÉΓòÉΓòÉ 22.5. Prefix Keys ΓòÉΓòÉΓòÉ A prefix key has an associated keymap which defines what to do with key sequences that start with the prefix key. For example, ctl-x-map is the keymap used for characters following the prefix key C-x. Here is a list of the standard prefix keys of Emacs and their keymaps: o ctl-x-map is the variable name for the map used for characters that follow C-x. This map is also the function definition of Control-X-prefix. o ctl-x-4-map used is for characters that follow C-x 4. o esc-map is used for characters that follow ESC. Thus, the global definitions of all Meta characters are actually found here. This map is also the function definition of ESC-prefix. o help-map is used for characters that follow C-h. o mode-specific-map is for characters that follow C-c. This map is not actually mode specific; its name was chosen to be informative for the user in C-h b (display-bindings), where it describes the main use of the C-c prefix key. The binding of a prefix key is the keymap to use for looking up the characters that follow the prefix key. (It may instead be a symbol whose function definition is a keymap. The effect is the same, but the symbol serves as a name for the prefix key.) Thus, the binding of C-x is the symbol Control-X-prefix, whose function definition is the keymap for C-x commands. This keymap is also the value of ctl-x-map. Prefix key definitions of this sort can appear in either the global map or a local map. The definitions of C-c, C-x, C-h and ESC as prefix keys appear in the global map, so these prefix keys are always available. Major modes can locally redefine a key as a prefix by putting a prefix key definition for it in the local map. If a key is defined as a prefix in both the local map and the global, the two definitions are effectively merged: the commands defined in the local map's prefix definition take priority; those not defined there are taken from the global map. In the following example, C-p is made a prefix key in the local keymap (so that C-p is identical to C-x). Then the binding for C-p C-f is the function find-file, just like C-x C-f. The key sequence C-p 6 is not found in either the local map or global map. (use-local-map (make-sparse-keymap)) => nil (local-set-key "\C-p" ctl-x-map) => nil (key-binding "\C-p\C-f") => find-file (key-binding "\C-p6") => nil -- Function: define-prefix-command symbol This function defines symbol as a prefix command: it creates a full keymap and stores it as symbol's function definition. Storing the symbol as the binding of a key makes the key a prefix key which has a name. This function returns symbol. It is convenient to store the keymap as the value of a variable as well. In version 19, this function stores the keymap in both the function definition and value of symbol. However, in version 18, only the function definition of symbol is set, not the value. ΓòÉΓòÉΓòÉ 22.6. Global and Local Keymaps ΓòÉΓòÉΓòÉ The global keymap holds the bindings of keys that are defined regardless of the current buffer, such as C-f. The variable global-map holds this keymap. Each buffer may have another keymap, its local keymap, which may contain new or overriding definitions for keys. Each buffer records which local keymap is used with it. Both the global and local keymaps are used to determine what command to execute when a key is entered. The key lookup proceeds as described earlier (see Key Lookup), but Emacs first searches for the key in the local map; if Emacs does not find a local definition, Emacs then searches the global map. Since every buffer that uses the same major mode normally uses the very same local keymap, it may appear as if the keymap is local to the mode. A change to the local keymap of a buffer (using local-set-key, for example) will be seen also in the other buffers that share that keymap. The local keymaps that are used for Lisp mode, C mode, and several other major modes exist even if they have not yet been used. These local maps are the values of the variables lisp-mode-map, c-mode-map, and so on. For most other modes, which are less frequently used, the local keymap is constructed only when the mode is used for the first time in a session. The minibuffer has local keymaps, too; they contain various completion and exit commands. See Minibuffers. See Standard Keymaps, for a list of standard keymaps. -- Variable: global-map This variable contains the default global keymap that maps Emacs keyboard input to commands. Normally this keymap is the global keymap. The default global keymap is a full keymap that binds self-insert-command to all of the printing characters. -- Function: current-global-map This function returns the current global keymap. Normally, this is the same as the value of the global-map. (current-global-map) => [set-mark-command beginning-of-line ... delete-backward-char] -- Function: current-local-map This function returns the current buffer's local keymap, or nil if it has none. In the following example, the keymap for the `*scratch*' buffer (using Lisp Interaction mode) is a sparse keymap in which the entry for ESC, ASCII code 27, is another sparse keymap. (current-local-map) => (keymap (10 . eval-print-last-sexp) (9 . lisp-indent-line) (127 . backward-delete-char-untabify) (27 keymap (24 . eval-defun) (17 . indent-sexp))) -- Function: use-global-map keymap This function makes keymap the new current global keymap. The keymap map must be a full keymap (a vector of length 128). It returns nil. It is very unusual to change the global keymap. -- Function: use-local-map keymap This function makes keymap the new current local keymap of the current buffer. If keymap is nil, then there will be no local keymap. It returns nil. Most major modes use this function. -- Function: key-binding key This function returns the definition for key in the current keymaps trying the current buffer's local map and then the global map. The result is nil if key is undefined in the keymaps. An error is signaled if key is not a string. (key-binding "\C-x\C-f") => find-file -- Function: local-key-binding key This function returns the definition for key in the current local keymap, or nil if it is undefined there. -- Function: global-key-binding key This function returns the definition for command key in the current global keymap, or nil if it is undefined there. ΓòÉΓòÉΓòÉ 22.7. Changing Key Bindings ΓòÉΓòÉΓòÉ The way to rebind a key is to change its entry in a keymap. You can change the global keymap, so that the change is effective in all buffers (except those that override the global definition with a local one). Or you can change the current buffer's local map, which usually affects all buffers using the same major mode. The global-set-key and local-set-key functions are convenient interfaces for these operations. Or you can change bindings in any map by specifying it explicitly in define-key. People often use global-set-key in their `.emacs' file for simple customization. For example, (global-set-key "\C-x\C-\\" 'next-line) redefines C-x C-\ to move down a line. In writing the string for the key sequence to rebind, it is useful to use the special escape sequences for control and meta characters (see String Type). In a string, the syntax `\C-' means that the following character is a control character and `\M-' means that the following character is a META character. Thus, the string "\M-x" is read as containing a single M-x, "\C-f" is read as containing a single C-f, and "\M-\C-x" and "\C-\M-x" are both read as containing a single C-M-x. For the functions below, an error is signaled if keymap is not a keymap or if key is not a string representing a key sequence. -- Function: define-key keymap key definition This function sets the binding for key in keymap. (If key is more than one character long, the change is actually made in another keymap reached from keymap.) The argument definition can be any Lisp object, but only certain types are meaningful. (For a list of meaningful types, see Key Lookup.) The value returned by define-key is definition. Every prefix of key must be a prefix key (i.e., bound to a keymap) or undefined; otherwise an error is signaled (with data (error "Key sequence key uses invalid prefix characters")). If some prefix of key is undefined, then define-key defines it as a prefix key so that the rest of key may be defined as specified. In the following example, a sparse keymap is created and a number of bindings are added to it. (setq map (make-sparse-keymap)) => (keymap) (define-key map "\C-f" 'forward-char) => forward-char map => (keymap (6 . forward-char)) ;; Build sparse submap for C-x and bind f in that. (define-key map "\C-xf" 'forward-word) => forward-word map => (keymap (24 keymap ; C-x (102 . forward-word)) ; f (6 . forward-char)) ; C-f ;; Bind C-p to the ctl-x-map. (define-key map "\C-p" ctl-x-map) => [nil ... find-file ... backward-kill-sentence] ; ctl-x-map ;; Bind C-f to foo in the ctl-x-map. (define-key map "\C-p\C-f" 'foo) => 'foo map => (keymap ; Note foo in ctl-x-map. (16 . [nil ... foo ... backward-kill-sentence]) (24 keymap (102 . forward-word)) (6 . forward-char)) Note that storing a new binding for C-p C-f actually works by changing an entry in ctl-x-map, and this has the effect of changing the bindings of both C-p C-f and C-x C-f in the default global map. -- Function: substitute-key-definition olddef newdef keymap This function replaces olddef with newdef for any keys in keymap that were bound to olddef. In other words, olddef is replaced with newdef wherever it appears. It returns nil. Prefix keymaps that appear within keymap are not checked recursively for keys bound to olddef; they are not changed at all. Perhaps it would be better to check nested keymaps recursively. (setq map '(keymap (?1 . olddef-1) (?2 . olddef-2) (?3 . olddef-1))) => (keymap (49 . olddef-1) (50 . olddef-2) (51 . olddef-1)) (substitute-key-definition 'olddef-1 'newdef map) => nil map => (keymap (49 . newdef) (50 . olddef-2) (51 . newdef)) ;; The following will redefine C-x C-f, if you do it in an ;; Emacs with standard bindings. (substitute-key-definition 'find-file 'find-file-read-only (current-global-map)) -- Function: suppress-keymap keymap &optional nodigits This function changes the contents of the full keymap keymap by replacing the self-insertion commands for numbers with the digit-argument function, unless nodigits is non-nil, and by replacing the functions for the rest of the printing characters with undefined. This means that ordinary insertion of text is impossible in a buffer with a local keymap on which suppress-keymap has been called. suppress-keymap returns nil. The suppress-keymap function does not make it impossible to modify a buffer, as it does not suppress commands such as yank and quote-insert. To prevent any modification of a buffer, make it read-only (see Read Only Buffers). Since this function modifies keymap, you would normally use it on a newly created keymap. Operating on an existing keymap that is used for some other purpose is likely to cause trouble; for example, suppressing global-map would make it impossible to use most of Emacs. Most often, suppress-keymap is used for initializing local keymaps of modes such as Rmail and Dired where insertion of text is not desirable and the buffer is read-only. Here is an example taken from the file `emacs/lisp/dired.el', showing how the local keymap for Dired mode is set up: ... (setq dired-mode-map (make-keymap)) (suppress-keymap dired-mode-map) (define-key dired-mode-map "r" 'dired-rename-file) (define-key dired-mode-map "\C-d" 'dired-flag-file-deleted) (define-key dired-mode-map "d" 'dired-flag-file-deleted) (define-key dired-mode-map "v" 'dired-view-file) (define-key dired-mode-map "e" 'dired-find-file) (define-key dired-mode-map "f" 'dired-find-file) ... ΓòÉΓòÉΓòÉ 22.8. Commands for Binding Keys ΓòÉΓòÉΓòÉ This section describes some convenient interactive interfaces for changing key bindings. They work by calling define-key. -- Command: global-set-key key definition This function sets the binding of key in the current global map to definition. (global-set-key key definition) == (define-key (current-global-map) key definition) -- Command: global-unset-key key This function removes the binding of key from the current global map. One use of this function is in preparation for defining a longer key which uses it implicitly as a prefix---which would not be allowed otherwise. For example: (global-unset-key "\C-l") => nil (global-set-key "\C-l\C-l" 'redraw-display) => nil This function is implemented simply using define-key: (global-unset-key key) == (define-key (current-global-map) key nil) -- Command: local-set-key key definition This function sets the binding of key in the current local keymap to definition. (local-set-key key definition) == (define-key (current-local-map) key definition) -- Command: local-unset-key key This function removes the binding of key from the current local map. (local-unset-key key) == (define-key (current-local-map) key nil) ΓòÉΓòÉΓòÉ 22.9. Scanning Keymaps ΓòÉΓòÉΓòÉ This section describes functions used to scan all the current keymaps for the sake of printing help information. -- Function: accessible-keymaps keymap This function returns a list of all the keymaps that can be accessed (via prefix keys) from keymap. The list returned is an association list with elements of the form (key . map), where key is a prefix whose definition in keymap is map. The elements of the alist are ordered so that the key increases in length. The first element is always ("" . keymap), because the specified keymap is accessible from itself with a prefix of no characters. In the example below, the returned alist indicates that the key ESC, which is displayed as "^[", is a prefix key whose definition is the sparse keymap (keymap (83 . center-paragraph) (115 . foo)). (accessible-keymaps (current-local-map)) =>(("" keymap (27 keymap ; Note this keymap for ESC is repeated below. (83 . center-paragraph) (115 . center-line)) (9 . tab-to-tab-stop)) ("^[" keymap (83 . center-paragraph) (115 . foo))) In the following example, C-h is a prefix key that uses a sparse keymap starting (118 . describe-variable) .... Another prefix, C-x 4, uses the full keymap beginning [nil ...] (which happens to be ctl-x-4-map). (accessible-keymaps (current-global-map)) => (("" . [set-mark-command beginning-of-line ... delete-backward-char]) ("^C" keymap (13 . x-flush-mouse-queue)) ("^H" keymap (118 . describe-variable) ... (8 . help-for-help)) ("^X" . [x-flush-mouse-queue ... backward-kill-sentence]) ("^[" . [mark-sexp backward-sexp ... backward-kill-word]) ("^X4" . [nil ... find-file-other-window nil ... nil nil])) -- Function: where-is-internal command &optional keymap firstonly This function returns a list of key sequences (of any length) that are bound to command in keymap and the global keymap. The argument command can be any object; it is compared with all keymap entries using eq. If keymap is not supplied, then the global map alone is used. If firstonly is non-nil, then the value is a single string representing the first key sequence found, rather than a list of all possible key sequences. This function is used by where-is ( see Help). (where-is-internal 'describe-function) => ("\^hf" "\^hd") -- Command: describe-bindings This function creates a listing of all defined keys, and their definitions. The listing is put in a buffer named `*Help*', which is then displayed in a window. A meta character is shown as ESC followed by the corresponding non-meta character. Control characters are indicated with C-. When several consecutive characters have the same definition, they are shown together, as `firstchar..lastchar'. In this instance, you need to know the ASCII codes to understand which characters this means. For example, in the default global map, the characters `SPC .. ~' are described by a single line. SPC is ASCII 32, ~ is ASCII 126, and the characters between them include all the normal printing characters, (e.g., letters, digits, punctuation, etc.); all these characters are bound to self-insert-command. ΓòÉΓòÉΓòÉ 23. Major and Minor Modes ΓòÉΓòÉΓòÉ A mode is a set of definitions that customize Emacs and can be turned on and off while you edit. There are two varieties of modes: major modes, which are mutually exclusive and used for editing particular kinds of text, and minor modes, which provide features that may be enabled individually. This chapter covers both major and minor modes, the way they are indicated in the mode line, and how they run hooks supplied by the user. Related topics such as keymaps and syntax tables are covered in separate chapters. (See Keymaps, and Syntax Tables.) ΓòÉΓòÉΓòÉ 23.1. Major Modes ΓòÉΓòÉΓòÉ Major modes specialize Emacs for editing particular kinds of text. Each buffer has only one major mode at a time. The least specialized major mode is called Fundamental mode. This mode has no mode-specific definitions or variable settings, so each Emacs command behaves in its default manner, and each option is in its default state. All other major modes redefine various keys and options. For example, Lisp Interaction mode provides special key bindings for LFD (eval-print-last-sexp), TAB (lisp-indent-line), and other keys. When you need to write several editing commands to help you perform a specialized editing task, creating a new major mode is usually a good idea. In practice, writing a major mode is easy (in sharp contrast to writing a minor mode, which is often difficult). If the new mode is similar to an old one, it is often unwise to modify the old one to serve two purposes, since it may become harder to use and maintain. Instead, copy and rename an existing major mode definition and alter it for its new function. For example, Rmail Edit mode, which is in `emacs/lisp/rmailedit.el', is a major mode that is very similar to Text mode except that it provides three additional commands. Its definition is distinct from that of Text mode, but was derived from it. Rmail Edit mode is an example of a case where one piece of text is put temporarily into a different major mode so it can be edited in a different way (with ordinary Emacs commands rather than Rmail). In such cases, the temporary major mode usually has a command to switch back to the buffer's usual mode (Rmail mode, in this case). You might be tempted to present the temporary redefinitions inside a recursive edit and restore the usual ones when the user exits; but this is a bad idea because it constrains the user's options when it is done in more than one buffer: recursive edits must be exited most-recently-entered first. Using alternative major modes avoids this limitation. See Recursive Editing. The standard GNU Emacs Lisp library directory contains the code for several major modes, in files including `text-mode.el', `texinfo.el', `lisp-mode.el', `c-mode.el', and `rmail.el'. You can look at these libraries to see how modes are written. Text mode is perhaps the simplest major mode aside from Fundamental mode. Rmail mode is a rather complicated, full-featured mode. ΓòÉΓòÉΓòÉ 23.1.1. Major Mode Conventions ΓòÉΓòÉΓòÉ The code for existing major modes follows various coding conventions, including conventions for local keymap and syntax table initialization, global names, and hooks. Please keep these conventions in mind when you create a new major mode: o Define a command whose name ends in `-mode', with no arguments, that switches to the new mode in the current buffer. This command should set up the keymap, syntax table, and local variables in an existing buffer without changing the buffer's text. o Write a documentation string for this command which describes the special commands available in this mode. C-h m (describe-mode) will print this. The documentation string may include the special documentation substrings, `\[command]', `\{keymap}', and `\<keymap>', that enable the documentation to adapt automatically to the user's own key bindings. The describe-mode function replaces these special documentation substrings with their current meanings. See Accessing Documentation. o The major mode command should set the variable major-mode to the major mode command symbol. This is how describe-mode discovers which documentation to print. o The major mode command should set the variable mode-name to the ``pretty'' name of the mode, as a string. This appears in the mode line. o Since all global names are in the same name space, all the global variables, constants, and functions that are part of the mode should have names that start with the major mode name (or with an abbreviation of it if the name is long). o The major mode should usually have its own keymap, which is used as the local keymap in all buffers in that mode. The major mode function should call use-local-map to install this local map. See Global and Local Keymaps, for more information. This keymap should be kept in a global variable named modename-mode-map. This variable is usually set up when the library that defines the mode is loaded. Use defvar to set the variable, so that it is not reinitialized if it already has a value. (Such reinitialization could discard customizations made by the user.) o The mode may have its own syntax table or may share one with other related modes. If it has its own syntax table, it should store this in a variable named modename-mode-syntax-table. The reasons for this are the same as for using a keymap variable. See Syntax Tables. o The mode may have its own abbrev table or may share one with other related modes. If it has its own abbrev table, it should store this in a variable named modename-mode-abbrev-table. See Abbrev Tables. o To give a variable a buffer-local binding, use make-local-variable in the major mode command, not make-variable-buffer-local. The latter function would make the variable local to every buffer in which it is subsequently set, which would affect buffers that do not use this mode. It is undesirable for a mode to have such global effects. See Buffer-Local Variables. o If hooks are appropriate for the mode, the major mode command should run the hooks after completing all other initialization so the user may further customize any of the settings. See Hooks. o If this mode is appropriate only for specially-prepared text, then the major mode command symbol should have a property named mode-class with value special, put on as follows: (put 'funny-mode 'mode-class 'special) This tells Emacs that new buffers created while the current buffer has Funny mode should not inherit Funny mode. Modes such as Dired, Rmail, and Buffer List use this feature. o If it is desirable that Emacs use the new mode by default after visiting files with certain recognizable names, add an element to auto-mode-alist to select the mode for those file names. If you define the mode command to autoload, you should add this element at the same time. Otherwise, it is sufficient to add the element in the file that contains the mode definition. See Auto Major Mode. o In the documentation, you should provide a sample autoload form and a sample auto-mode-alist addition that users can include in their `.emacs' files. o The top level forms in the file defining the mode should be written so that they may be evaluated more than once without adverse consequences. Even if you never load the file more than once, someone else will. ΓòÉΓòÉΓòÉ 23.1.2. Major Mode Examples ΓòÉΓòÉΓòÉ Text mode is perhaps the simplest mode besides Fundamental mode. Here are excerpts from `text-mode.el' that illustrate many of the conventions listed above: ;; Create mode-specific tables. (defvar text-mode-syntax-table nil "Syntax table used while in text mode.") (if text-mode-syntax-table () ; Do not change the table if it is already set up. (setq text-mode-syntax-table (make-syntax-table)) (set-syntax-table text-mode-syntax-table) (modify-syntax-entry ?\" ". " text-mode-syntax-table) (modify-syntax-entry ?\\ ". " text-mode-syntax-table) (modify-syntax-entry ?' "w " text-mode-syntax-table)) (defvar text-mode-abbrev-table nil "Abbrev table used while in text mode.") (define-abbrev-table 'text-mode-abbrev-table ()) (defvar text-mode-map nil "") ; Create a mode-specific keymap. (if text-mode-map () ; Do not change the keymap if it is already set up. (setq text-mode-map (make-sparse-keymap)) (define-key text-mode-map "\t" 'tab-to-tab-stop) (define-key text-mode-map "\es" 'center-line) (define-key text-mode-map "\eS" 'center-paragraph)) Here is the complete major mode function definition for Text mode: (defun text-mode () "Major mode for editing text intended for humans to read. Special commands: \\{text-mode-map} Turning on text-mode calls the value of the variable text-mode-hook, if that value is non-nil." (interactive) (kill-all-local-variables) (use-local-map text-mode-map) ; This provides the local keymap. (setq mode-name "Text") ; This name goes into the mode line. (setq major-mode 'text-mode) ; This is how describe-mode ; finds the doc string to print. (setq local-abbrev-table text-mode-abbrev-table) (set-syntax-table text-mode-syntax-table) (run-hooks 'text-mode-hook)) ; Finally, this permits the user to ; customize the mode with a hook. The three Lisp modes (Lisp mode, Emacs Lisp mode, and Lisp Interaction mode) have more features than Text mode and the code is correspondingly more complicated. Here are excerpts from `lisp-mode.el' that illustrate how these modes are written. ;; Create mode-specific table variables. (defvar lisp-mode-syntax-table nil "") (defvar emacs-lisp-mode-syntax-table nil "") (defvar lisp-mode-abbrev-table nil "") (if (not emacs-lisp-mode-syntax-table) ; Do not change the table ; if it is already set. (let ((i 0)) (setq emacs-lisp-mode-syntax-table (make-syntax-table)) ;; Set syntax of chars up to 0 to class of chars that are ;; part of symbol names but not words. ;; (The number 0 is 48 in the ascii character set.) (while (< i ?0) (modify-syntax-entry i "_ " emacs-lisp-mode-syntax-table) (setq i (1+ i))) ... ;; Set the syntax for other characters. (modify-syntax-entry ? " " emacs-lisp-mode-syntax-table) (modify-syntax-entry ?\t " " emacs-lisp-mode-syntax-table) ... (modify-syntax-entry ?$ "() " emacs-lisp-mode-syntax-table) (modify-syntax-entry ?$ ")( " emacs-lisp-mode-syntax-table) ...)) ;; Create an abbrev table for lisp-mode. (define-abbrev-table 'lisp-mode-abbrev-table ()) Much code is shared among the three Lisp modes; the code is all in one library. The following function sets various variables; it is called by each of the major Lisp mode functions: (defun lisp-mode-variables (lisp-syntax) ;; The lisp-syntax argument is nil in Emacs Lisp mode, ;; and t in the other two Lisp modes. (cond (lisp-syntax (if (not lisp-mode-syntax-table) ;; The Emacs Lisp mode syntax table always exists, but ;; the Lisp Mode syntax table is created the first time a ;; mode that needs it is called. This is to save space. (progn (setq lisp-mode-syntax-table (copy-syntax-table emacs-lisp-mode-syntax-table)) ;; Change some entries for Lisp mode. (modify-syntax-entry ?\| "\" " lisp-mode-syntax-table) (modify-syntax-entry ?\[ "_ " lisp-mode-syntax-table) (modify-syntax-entry ?\] "_ " lisp-mode-syntax-table))) (set-syntax-table lisp-mode-syntax-table))) (setq local-abbrev-table lisp-mode-abbrev-table) ...) Functions such as forward-word use the value of the paragraph-start variable. Since Lisp code is different from ordinary text, the paragraph-start variable needs to be set specially to handle Lisp. Also, comments are indented in a special fashion in Lisp and the Lisp modes need their own mode-specific comment-indent-hook. The code to set these variables is the rest of lisp-mode-variables. (make-local-variable 'paragraph-start) (setq paragraph-start (concat "^$\\|" page-delimiter)) ... (make-local-variable 'comment-indent-hook) (setq comment-indent-hook 'lisp-comment-indent)) Each of the different Lisp modes has a slightly different keymap. For example, Lisp mode binds C-c C-l to run-lisp, but the other Lisp modes do not. However, all Lisp modes have some commands in common. The following function adds these common commands to a given keymap. (defun lisp-mode-commands (map) (define-key map "\e\C-q" 'indent-sexp) (define-key map "\177" 'backward-delete-char-untabify) (define-key map "\t" 'lisp-indent-line)) Here is an example of using lisp-mode-commands to initialize a keymap, as part of the code for Emacs Lisp mode. First defvar is used to declare a mode-specific keymap variable. Then boundp tests whether the emacs-lisp-mode-map variable has a value (is not void). If the variable does have a value, we do not change it. This lets the user customize the keymap if he or she so wishes. Otherwise, we initialize it to a new sparse keymap and install the default key bindings. (defvar emacs-lisp-mode-map () "") (if emacs-lisp-mode-map () (setq emacs-lisp-mode-map (make-sparse-keymap)) (define-key emacs-lisp-mode-map "\e\C-x" 'eval-defun) (lisp-mode-commands emacs-lisp-mode-map)) Finally, here is the complete major mode function definition for Emacs Lisp mode. (defun emacs-lisp-mode () "Major mode for editing Lisp code to run in Emacs. Commands: Delete converts tabs to spaces as it moves back. Blank lines separate paragraphs. Semicolons start comments. \\{emacs-lisp-mode-map} Entry to this mode calls the value of emacs-lisp-mode-hook if that value is non-nil." (interactive) (kill-all-local-variables) (use-local-map emacs-lisp-mode-map) ; This provides the local keymap. (set-syntax-table emacs-lisp-mode-syntax-table) (setq major-mode 'emacs-lisp-mode) ; This is how describe-mode ; finds out what to describe. (setq mode-name "Emacs-Lisp") ; This goes into the mode line. (lisp-mode-variables nil) ; This define various variables. (run-hooks 'emacs-lisp-mode-hook)) ; This permits the user to use a ; hook to customize the mode. ΓòÉΓòÉΓòÉ 23.1.3. How Emacs Chooses a Major Mode Automatically ΓòÉΓòÉΓòÉ Based on information in the file name or in the file itself, Emacs automatically selects a major mode for the new buffer when a file is visited. -- Command: fundamental-mode Fundamental mode is a major mode that is not specialized for anything in particular. Other major modes are defined in effect by comparison with this one---their definitions say what to change, starting from Fundamental mode. The fundamental-mode function does not run any hooks, so it is not readily customizable. -- Command: normal-mode &optional find-file This function establishes the proper major mode and local variable bindings for the current buffer. First it calls set-auto-mode, then it runs hack-local-variables to parse, and bind or evaluate as appropriate, any local variables. If the find-file argument to normal-mode is non-nil, normal-mode assumes that the find-file function is calling it. In this case, if inhibit-local-variables is non-nil, it asks for confirmation before processing a local variables list. If you run normal-mode yourself, the argument find-file is normally nil, so confirmation is not requested. normal-mode uses condition-case around the call to the major mode function, so errors are caught and reported as a `File mode specification error', followed by the original error message. -- Function: set-auto-mode This function selects the major mode that is appropriate for the current buffer. It may base its decision on the value of the `-*-' line, on the visited file name (using auto-mode-alist), or on the value of a local variable). However, this function does not look for the `mode:' local variable near the end of a file; the hack-local-variables function does that. See How Major Modes are Chosen. -- User Option: default-major-mode This variable holds the default major mode for new buffers. The standard value is fundamental-mode. If the value of default-major-mode is nil, Emacs uses the (previously) current buffer's major mode for major mode of a new buffer. However, if the major mode symbol has a mode-class property with value special, then it is not used for new buffers; Fundamental mode is used instead. The modes that have this property are those such as Dired and Rmail that are useful only with text that has been specially prepared. -- Variable: initial-major-mode The value of this variable determines the major mode of the initial `*scratch*' buffer. The value should be a symbol that is a major mode command name. The default value is lisp-interaction-mode. -- Variable: auto-mode-alist This variable contains an association list of file name patterns (regular expressions; see Regular Expressions) and corresponding major mode functions. Usually, the file name patterns test for suffixes, such as `.el' and `.c', but this need not be the case. Each element of the alist looks like (regexp . mode-function). For example, (("^/tmp/fol/" . text-mode) ("\\.texinfo$" . texinfo-mode) ("\\.texi$" . texinfo-mode) ("\\.el$" . emacs-lisp-mode) ("\\.c$" . c-mode) ("\\.h$" . c-mode) ...) When you visit a file whose full path name matches a regexp, set-auto-mode calls the corresponding mode-function. This feature enables Emacs to select the proper major mode for most files. Here is an example of how to prepend several pattern pairs to an existing auto-mode-alist. (You might use this sort of expression in your `.emacs' file.) (setq auto-mode-alist (append '(("/\\.[^/]*$" . fundamental-mode) ; Filename starts with a dot. ("[^\\./]*$" . fundamental-mode) ; Filename has no dot. ("\\.C$" . c++-mode)) auto-mode-alist)) -- Function: hack-local-variables &optional force This function parses, and binds or evaluates as appropriate, any local variables for the current buffer. If the variable inhibit-local-variables is non-nil, and force is nil, then the user is asked for confirmation if the buffer does contain local variable specifications. A non-nil value of force is passed by normal-mode when it is called explicitly by the user. See Local Variables in Files, for the syntax of the local variables section of a file. -- User Option: inhibit-local-variables When this variable is non-nil, hack-local-variables asks the user for confirmation before obeying a file's local-variables list. ΓòÉΓòÉΓòÉ 23.1.4. Getting Help about a Major Mode ΓòÉΓòÉΓòÉ The describe-mode function is used to provide information about major modes. It is normally called with C-h m. The describe-mode function uses the value of major-mode, which is why every major mode function needs to set the major-mode variable. -- Command: describe-mode This function displays the documentation of the current major mode. The describe-mode function calls the documentation function using the value of major-mode as an argument. Thus, it displays the documentation string of the major mode function. (See Accessing Documentation.) -- Variable: major-mode This variable holds the symbol for the current buffer's major mode. This symbol should be the name of the function that is called to initialize the mode. The describe-mode function uses the documentation string of this symbol as the documentation of the major mode. ΓòÉΓòÉΓòÉ 23.2. Minor Modes ΓòÉΓòÉΓòÉ A minor mode provides features that users may enable or disable independently of the choice of major mode. Minor modes can be enabled individually or in combination. Minor modes would be better named ``Generally available, optional feature modes'' except that such a name is unwieldy. A minor mode is not usually a modification of single major mode. For example, auto-fill-mode and auto-justify-mode may be used in any major mode that permits text insertion. To be general, a minor mode must be effectively independent of the things major modes do. A minor mode is often much more difficult to implement than a major mode. One reason is that you should be able to deactivate a minor mode and restore the environment of the major mode to the state it was in before the minor mode was activated. Often the biggest problem in implementing a minor mode is finding a way to insert the necessary hook into the rest of Emacs. For example, some minor modes, such as Auto Fill mode, change how text is inserted into a buffer. Without auto-fill-hook, Auto Fill mode would be implementable only at great pain and great cost in editing efficiency. ΓòÉΓòÉΓòÉ 23.2.1. Conventions for Writing Minor Modes ΓòÉΓòÉΓòÉ There are conventions for writing minor modes just as there are for major modes. Several of the major mode conventions apply to minor modes as well: those regarding the name of the mode initialization function, the names of global symbols, and the use of keymaps and other tables. In addition, there are several conventions that are specific to minor modes. o The minor mode should be represented by a symbol whose name ends in `-mode'. This symbol should be both a command to turn the mode on or off and a variable which records whether the mode is on. This variable is used in conjunction with the minor-mode-alist to display the minor mode name in the mode line. It may also be directly responsible for controlling the features of the minor mode. If you want the minor mode to be enabled separately in each buffer, make the variable buffer-local. o A minor mode command should accept one optional argument. If the argument is nil, the function should toggle the mode (turn it on if it is off, and off if it is on). Otherwise, the function should turn the mode on if the argument is a positive integer, a symbol, or a list whose car is a positive integer; it should turn the mode off otherwise. (This convention has not been implemented with full consistency in Emacs version 18.) Here is an example taken from the definition of overwrite-mode: (setq overwrite-mode (if (null arg) (not overwrite-mode) (> (prefix-numeric-value arg) 0))) o Add an element to minor-mode-alist for each minor mode (see Mode Line Variables). This element should be a list of the following form: (mode-variable string) Here mode-variable is the variable that indicates the enablement of the minor mode, and string is a short string, starting with a space, to represent the mode in the mode line. These strings must be short so that there is room for several of them at once. When you add an element to minor-mode-alist, use assq to check for an existing element, to avoid duplication. For example: (or (assq 'leif-mode minor-mode-alist) (setq minor-mode-alist (cons '(leif-mode " Leif") minor-mode-alist))) o If the minor mode adds new key bindings to the local keymap, you should be able to restore the keymap to its original value when you deactivate the minor mode. ΓòÉΓòÉΓòÉ 23.2.2. Limitations of Minor Modes ΓòÉΓòÉΓòÉ It is very difficult to write a minor mode that responds to arbitrary self-inserting characters. The principal problem is that self-insert-command, the command to insert the last key typed, is a primitive function written in C. It does not call any hooks, except in special cases. Unfortunately, you cannot simply substitute your own definition of self-insert-command for the existing one, as you can with most functions. This is a consequence of the way the editor command loop works: it checks whether a key is bound to self-insert-command and, if so, it calls the primitive self-insert-command function directly. It does not check to see whether you have written another version of the function to substitute for it. This is done for speed. (In general, if you substitute a Lisp function for a primitive, the C code within Emacs will continue to call the original primitive, but Lisp code will call your substitute Lisp function.) Instead of attempting to replace the function definition for self-insert-command, you could rebind certain keys that call self-insert-command. This can be made to work as long as no two minor modes try to rebind the same key. In a minor mode that is global (affects all buffers), do the rebinding in the global map. In a minor mode that is local to individual buffers, you will need to copy the local keymap (since it is usually shared with all the other buffers in the same major mode) and then make the change. ΓòÉΓòÉΓòÉ 23.3. Mode Line Format ΓòÉΓòÉΓòÉ Each Emacs window includes a mode line which displays status information about the buffer displayed in the window. The mode line contains information about the buffer such as its name, associated file, depth of recursive editing, and the major and minor modes of the buffer. This section describes how the contents of the mode line are controlled. It is in the chapter on modes because much of the information displayed in the mode line relates to the enabled major and minor modes. mode-line-format is a buffer-local variable that holds a template used to display the mode line of the current buffer. All windows for the same buffer use the same mode-line-format and the mode lines will appear the same (except perhaps for the percentage of the file scrolled off the top). The mode line of a window is normally updated whenever a different buffer is shown in the window, or when the buffer's modified-status changes from nil to t or vice-versa. If you modify any of the variables referenced by mode-line-format, you may want to force an update of the mode line so as to display the new information. You can do this with the following expression: (set-buffer-modified-p (buffer-modified-p)) The mode line is usually displayed in inverse video; see mode-line-inverse-video in Screen Attributes. ΓòÉΓòÉΓòÉ 23.3.1. The Data Structure of the Mode Line ΓòÉΓòÉΓòÉ The mode line contents are controlled by a data structure of lists, strings, symbols and numbers kept in the buffer-local variable mode-line-format. The data structure is called a mode line construct, and it is built in recursive fashion out of simpler mode line constructs. -- Variable: mode-line-format The value of this variable is a mode line construct with overall responsibility for the mode line format. The value of this variable controls which other variables are used to form the mode line text, and where they appear. A mode line construct may be as simple as a fixed string of text, but it usually specifies how to use other variables to construct the text. Many of these variables are themselves defined to have mode line constructs as their values. The default value of mode-line-format incorporates the values of variables such as mode-name and minor-mode-alist. Because of this, very few modes need to alter mode-line-format. For most purposes, it is sufficient to alter the variables referenced by mode-line-format. A mode line construct may be a list, cons cell, symbol, or string. If the value is a list, each element may be a list, a cons cell, a symbol, or a string. string A string as a mode line construct is displayed verbatim in the mode line except for %-constructs. Decimal digits after the % specify the field width for space filling on the right (i.e., the data is left justified). See %-Constructs. symbol A symbol as a mode line construct stands for its value. The value of symbol is used in place of symbol unless symbol is t or nil, or is void, in which case symbol is ignored. There is one exception: if the value of symbol is a string, it is processed verbatim in that the %-constructs are not recognized. (string rest...) or (list rest...) A list whose first element is a string or list, means to concatenate all the elements. This is the most common form of mode line construct. (symbol then else) A list whose first element is a symbol is a conditional. Its meaning depends on the value of symbol. If the value is non-nil, the second element of the list (then) is processed recursively as a mode line element. But if the value of symbol is nil, the third element of the list (if there is one) is processed recursively. (width rest...) A list whose first element is an integer specifies truncation or padding of the results of rest. The remaining elements rest are processed recursively as mode line constructs and concatenated together. Then the result is space filled (if width is positive) or truncated (to -width columns, if width is negative) on the right. For example, the usual way to show what percentage of a buffer is above the top of the window is to use a list like this: (-3 . "%p"). If you do alter mode-line-format itself, the new value should use all the same variables that are used by the default value, rather than duplicating their contents or displaying the information in another fashion. This permits customizations made by the user, by libraries (such as display-time) or by major modes via changes to those variables remain effective. Here is an example of a mode-line-format that might be useful for shell-mode since it contains the hostname and default directory. (setq mode-line-format (list "" 'mode-line-modified "%b--" (getenv "HOST") ; One element is not constant. ":" 'default-directory " " 'global-mode-string " %[(" 'mode-name 'minor-mode-alist "%n" 'mode-line-process ")%]----" '(-3 . "%p") "-%-")) ΓòÉΓòÉΓòÉ 23.3.2. Variables Used in the Mode Line ΓòÉΓòÉΓòÉ This section describes variables incorporated by the standard value of mode-line-format into the text of the mode line. There is nothing inherently special about these variables; any other variables could have the same effects on the mode line if mode-line-format were changed to use them. -- Variable: mode-line-modified This variable holds the mode-line construct for displaying whether the current buffer is modified. The default value mode-line-modified is ("--%1*%1*-"). This means that the mode line displays `--**-' if the buffer is modified, `-----' if the buffer is not modified, and `--%%-' if the buffer is read only. Changing this variable does not force an update of the mode line. -- Variable: mode-line-buffer-identification This variable identifies the buffer being displayed in the window. Its default value is `Emacs: %17b', which means that it displays `Emacs:' followed by the buffer name. You may want to change this in modes such as Rmail that do not behave like a ``normal'' Emacs. -- Variable: global-mode-string This variable holds a string that is displayed in the mode line for the use of display-time. The `%M' construct substitutes the value of global-mode-string, but this is obsolete, since the variable is included directly in the mode line. -- Variable: mode-name This buffer-local variable holds the ``pretty'' name of the current buffer's major mode. Each major mode should set this variable so that the mode name will appear in the mode line. -- Variable: minor-mode-alist This variable holds an association list whose elements specify how the mode line should indicate that a minor mode is active. Each element of the minor-mode-alist should be a two-element list: (minor-mode-variable mode-line-string) The string mode-line-string is included in the mode line when the value of minor-mode-variable is non-nil and not otherwise. These strings should begin with spaces so that they don't run together. Conventionally, the minor-mode-variable for a specific mode is set to a non-nil value when that minor mode is activated. The default value of minor-mode-alist is: minor-mode-alist => ((abbrev-mode " Abbrev") (overwrite-mode " Ovwrt") (auto-fill-hook " Fill") (defining-kbd-macro " Def")) (Note that in version 19, auto-fill-hook will be renamed to auto-fill-function.) minor-mode-alist is not buffer-local. The variables mentioned in the alist should be buffer-local if the minor mode can be enabled separately in each buffer. -- Variable: mode-line-process This buffer-local variable contains the mode line information on process status in modes used for communicating with subprocesses. It is displayed immediately following the major mode name, with no intervening space. For example, its value in the `*shell*' buffer is (": %s"), which allows the shell to display its status along with the major mode as: `(Shell: run)'. Normally this variable is nil. -- Variable: default-mode-line-format This variable holds the default mode-line-format for buffers that do not override it. This is the same as (default-value 'mode-line-format). The default value of default-mode-line-format is: ("" mode-line-modified mode-line-buffer-identification " " global-mode-string " %[(" mode-name minor-mode-alist "%n" mode-line-process ")%]----" (-3 . "%p") "-%-") ΓòÉΓòÉΓòÉ 23.3.3. %-Constructs in the Mode Line ΓòÉΓòÉΓòÉ The following table lists the recognized %-constructs and what they mean. %b the current buffer name, using the buffer-name function. %f the visited file name, using the buffer-file-name function. %* `%' if the buffer is read only (see buffer-read-only); `*' if the buffer is modified (see buffer-modified-p); `-' otherwise. %s the status of the subprocess belonging to the current buffer, using process-status. %p the percent of the buffer above the top of window, or `Top', `Bottom' or `All'. %n `Narrow' when narrowing is in effect; nothing otherwise (see narrow-to-region in Narrowing). %[ an indication of the depth of recursive editing levels (not counting minibuffer levels): one `[' for each editing level. %] one `]' for each recursive editing level (not counting minibuffer levels). %% the character `%'---this is how to include a literal `%' in a string in which %-constructs are allowed. %- dashes sufficient to fill the remainder of the mode line. The following two %-constructs are still supported but are obsolete since use of the mode-name and global-mode-string variables will produce the same results. %m the value of mode-name. %M the value of global-mode-string. Currently, only display-time modifies global-mode-string. ΓòÉΓòÉΓòÉ 23.4. Hooks ΓòÉΓòÉΓòÉ A hook is a variable whose value is a hook function (or a list of hook functions) to be called by parts of Emacs on certain defined occasions. The purpose of hooks is to facilitate customization, and the value of a hook is most often set up in the `.emacs' file, but it may be changed by programs. The function or functions used in a hook may be any of the valid kinds of functions that funcall accepts (see What Is a Function). Most modes run hooks as the last step of initialization. This makes it easy for a user to customize the behavior of the mode, by overriding the local variable assignments already made by the mode. But hooks may also be used in other contexts. For example, the functions named by find-file-not-found-hooks are called whenever a file is not found by find-file. For example, you can put the following expression in your `.emacs' file if you want to turn on Auto Fill mode when in Lisp Interaction mode: (setq lisp-interaction-mode-hook 'turn-on-auto-fill) The next example shows how to use a hook to customize the way Emacs formats C code. (People often have strong personal preferences for one format compared to another.) Here the hook function is an anonymous lambda expression. (setq c-mode-hook (function (lambda () (setq c-indent-level 4 c-argdecl-indent 0 c-label-offset -4 c-continued-statement-indent 0 c-brace-offset 0 comment-column 40)))) (setq c++-mode-hook c-mode-hook) Finally, here is an example of how to use the Text mode hook to provide a customized mode line for buffers in Text mode, displaying the default directory in addition to the standard components of the mode line. (This may cause the mode line to run out of space if you have very long path names or display the time and load.) (setq text-mode-hook (function (lambda () (setq mode-line-format '(mode-line-modified "Emacs: %14b" " " default-directory " " global-mode-string "%[(" mode-name minor-mode-alist "%n" mode-line-process ") %]---" (-3 . "%p") "-%-"))))) See Standard Hooks, for a list of standard hook variables. Most hook variables are initially void. This has no effect on examples such as the previous ones, where the hook variable is set without reference to any previous value. However, if you want to add an element to a hook variable which you use as a list of functions, you need to make sure the variable is not void. Here is how to do it using defvar: (defvar foo-hook nil) (or (memq 'my-hook foo-hook) (setq foo-hook (cons 'my-hook foo-hook))) At the appropriate time, Emacs uses the run-hooks function to run the hooks you have specified. -- Function: run-hooks &rest hookvar This function takes one or more hook names as arguments and runs each one in turn. Each hookvar argument should be a symbol that is a hook variable. These arguments are processed in the order specified. If a hook variable has a non-nil value, that value may be a function or a list of functions. If the value is a function (either a lambda expression or a symbol with a function definition), it is called. If it is a list, the elements are called, in order. The hook functions are called with no arguments. For example: (run-hooks 'emacs-lisp-mode-hook) Major mode functions use this function to call any hooks defined by the user. ΓòÉΓòÉΓòÉ 24. Documentation ΓòÉΓòÉΓòÉ GNU Emacs Lisp has convenient on-line help facilities, most of which derive their information from the documentation strings associated with functions and variables. This chapter describes how to write good documentation strings for your Lisp programs, as well as how to write programs to access documentation. Note that the documentation strings for Emacs are not the same thing as the Emacs manual. Manuals have their own source files, written in the Texinfo language; documentation strings are specified in the definitions of the functions and variables they apply to. A collection of documentation strings is not sufficient as a manual because a good manual is not organized in that fashion; it is organized in terms of topics of discussion. ΓòÉΓòÉΓòÉ 24.1. Documentation Basics ΓòÉΓòÉΓòÉ A documentation string is written using the Lisp syntax for strings, with double-quote characters surrounding the text of the string. This is because it really is a Lisp string object. The string serves as documentation when it is written in the proper place in the definition of a function or variable. In a function definition, the documentation string follows the argument list. In a variable definition, the documentation string follows the initial value of the variable. When you write a documentation string, make the first line a complete sentence (or two complete sentences) since some commands, such as apropos, print only the first line of a multi-line documentation string. Also, you should not indent the second line of a documentation string, if you have one, because that looks odd when you use C-h f (describe-function) or C-h v (describe-variable). Documentation strings may contain several special substrings, which stand for key bindings to be looked up in the current keymaps when the documentation is displayed. This allows documentation strings to refer to the keys for related commands and be accurate even when a user rearranges the key bindings. (See Accessing Documentation.) Within the Lisp world, a documentation string is kept with the function or variable that it describes: o The documentation for a function is stored in the function definition itself (see Lambda Expressions). The function documentation knows how to extract it. o The documentation for a variable is stored on the variable's property list under the property name variable-documentation. The function documentation-property knows how to extract it. However, to save space, the documentation for preloaded functions and variables (including primitive functions and autoloaded functions) are stored in the `emacs/etc/DOC-version' file. Both the documentation and the documentation-property functions know how to access `emacs/etc/DOC-version', and the process is transparent to the user. In this case, the documentation string is replaced with an integer offset into the `emacs/etc/DOC-version' file. Keeping the documentation strings out of the Emacs core image saves a significant amount of space. See Building Emacs. For information on the uses of documentation strings, see where-is-internal and describe-bindings in Global and Local Keymaps. Also, see Help. The `emacs/etc' directory contains two utilities for printing the `emacs/etc/DOC-version' file in hardcopy. These are `sorted-doc.c' and `digest-doc.c'. ΓòÉΓòÉΓòÉ 24.2. Access to Documentation Strings ΓòÉΓòÉΓòÉ -- Function: documentation-property symbol property This function returns the documentation string that is recorded symbol's property list under property property. This uses the function get, but does more than that: it also retrieves the string from the file `emacs/etc/DOC-version' if necessary, and runs substitute-command-keys to substitute the actual (current) key bindings. (documentation-property 'command-line-processed 'variable-documentation) => "t once command line has been processed" (symbol-plist 'command-line-processed) => (variable-documentation 188902) -- Function: documentation function This function returns the documentation string of function. If the documentation string is stored in the `emacs/etc/DOC-version' file, this function will access it there. In addition, documentation runs substitute-command-keys on the resulting string, so the value contains the actual (current) key bindings. The function documentation signals a void-function error unless function has a function definition. However, function does not need to have a documentation string. If there is no documentation string, documentation returns nil. Here is an example of using documentation and documentation-property to display the documentation strings for several symbols in a `*Help*' buffer. (defun describe-symbols (pattern) "Describe the Emacs Lisp symbols matching PATTERN. All symbols that have PATTERN in their name are described in the *Help* buffer." (interactive "sDescribe symbols matching: ") (let ((describe-func (function (lambda (s) ;; Print description of symbol. (if (fboundp s) ; It is a function. (princ (format "%s\t%s\n%s\n\n" s (if (commandp s) (concat "Command: " (or (mapconcat 'key-description (where-is-internal s) " "))) "Function") (or (documentation s) "not documented")))) (if (boundp s) ; It is a variable. (princ (format "%s\t%s\n%s\n\n" s (if (user-variable-p s) "Option " "Variable") (or (documentation-property s 'variable-documentation) "not documented"))))))) sym-list) ;; Build a list of symbols that match pattern. (mapatoms (function (lambda (sym) (if (string-match pattern (symbol-name sym)) (setq sym-list (cons sym sym-list)))))) ;; Display the data. (with-output-to-temp-buffer "*Help*" (mapcar describe-func (sort sym-list 'string<)) (print-help-return-message)))) The describe-symbols function works like apropos, but provides more information. (describe-symbols "goal") ---------- Buffer: *Help* ---------- goal-column Option *Semipermanent goal column for vertical motion, as set by C-x C-n, or nil. set-goal-column Command: C-x C-n Set the current horizontal position as a goal for C-n and C-p. Those commands will move to this position in the line moved to rather than trying to keep the same horizontal position. With a non-nil argument, clears out the goal column so that C-n and C-p resume vertical motion. temporary-goal-column Variable Current goal column for vertical motion. It is the column where point was at the start of current run of vertical motion commands. ---------- Buffer: *Help* ---------- -- Function: Snarf-documentation filename This function is used only during Emacs initialization, just before the runnable Emacs is dumped. It finds the file offsets of the documentation strings stored in the file filename, and records them in the in-core function definitions and variable property lists in place of the actual strings. See Building Emacs. The file filename is found in the `emacs/etc' directory (usually filename is "DOC-version"). When the dumped Emacs is later executed, the same file is found in the exec-directory (see Subprocess Creation). ΓòÉΓòÉΓòÉ 24.3. Substituting Key Bindings in Documentation ΓòÉΓòÉΓòÉ -- Function: substitute-command-keys string This function returns string with certain special substrings replaced by the actual (current) key bindings are listed. This permits the documentation to be displayed with accurate information about key bindings. (The key bindings may be changed by the user between the time Emacs is built and the time that the documentation is asked for.) This table lists the forms of the special substrings and what they are replaced with: \[command] is replaced either by a keystroke sequence that will invoke command, or by `M-x command' if command is not bound to any key sequence. \{mapvar} is replaced by a summary (made by describe-bindings) of the value of mapvar, taken as a keymap. \<mapvar> makes this call to substitute-command-keys use the value of mapvar as the keymap for future `\[command]' substrings. This special string does not produce any replacement text itself; it only affects the replacements done later. *Note:* each `\' must be doubled when written in a string in Emacs Lisp. Here are examples of the special substrings: (substitute-command-keys "To abort recursive edit, type: \\[abort-recursive-edit]") => "To abort recursive edit, type: C-]" (substitute-command-keys "The keys that are defined for the minibuffer here are: \\{minibuffer-local-must-match-map}") => "The keys that are defined for the minibuffer here are: ? minibuffer-completion-help SPC minibuffer-complete-word TAB minibuffer-complete LFD minibuffer-complete-and-exit RET minibuffer-complete-and-exit C-g abort-recursive-edit " (substitute-command-keys "To abort a recursive edit from the minibuffer, type\ \\<minibuffer-local-must-match-map>\\[abort-recursive-edit].") => "To abort a recursive edit from the minibuffer, type C-g." ΓòÉΓòÉΓòÉ 24.4. Describing Characters for Help Messages ΓòÉΓòÉΓòÉ These functions convert characters or strings to textual descriptions. These descriptions are useful for including arbitrary text characters or key sequences in messages, because they convert non-printing characters to sequences of printing characters. The description of a printing character is the character itself. -- Function: key-description string This function returns a string containing the Emacs standard notation for the keyboard characters in string. See the examples for single-key-description. -- Function: single-key-description character This function returns a string describing character in the standard Emacs notation for keyboard input. A normal printing character is represented by itself, but a control character turns into a string starting with `C-', a meta character turns into a string starting with `M-', and space, linefeed, etc. are transformed to `SPC', `LFD', etc. (single-key-description ?\C-x) => "C-x" (key-description "\C-x \M-y \n \t \r \f123") => "C-x SPC M-y SPC LFD SPC TAB SPC RET SPC C-l 1 2 3" -- Function: text-char-description character This function returns a string describing character in the standard Emacs notation for characters that appear in text---like single-key-description, except that that control characters are represented with a leading caret (which is how control characters in Emacs buffers are usually displayed). (text-char-description ?\C-c) => "^C" (text-char-description ?\M-m) => "M-m" (text-char-description ?\C-\M-m) => "M-^M" ΓòÉΓòÉΓòÉ 24.5. Help Functions ΓòÉΓòÉΓòÉ Emacs provides a variety of on-line help functions, all accessible to the user as subcommands of the prefix C-h. For more information about them, see Help. Here we describe some program-level interfaces to the same information. -- Command: apropos regexp &optional predicate noprint This function finds all symbols whose names contain a match for the regular expression regexp, and returns a list of them. Normally it displays the symbols in a buffer named `*Help*', each with a one-line description. If noprint is non-nil, it does not display them, but just returns the list. If predicate is non-nil, it should be a function to be called on each symbol that has matched regexp. Only symbols for which predicate returns a non-nil value are listed or displayed. When you call apropos interactively, it prompts for regexp in the minibuffer. In the first of the following examples, apropos finds all the symbols with names containing `exec'. They are returned but not displayed. In the second example, it finds and returns only those symbols that are also commands; in addition, they are then displayed in the `*Help*' buffer. (apropos "exec" nil t) => (Buffer-menu-execute command-execute exec-directory exec-path execute-extended-command execute-kbd-macro executing-kbd-macro executing-macro) (apropos "exec" 'commandp) => (Buffer-menu-execute execute-extended-command) ---------- Buffer: *Help* ---------- Buffer-menu-execute Function: Save and/or delete buffers marked with M-x Buffer-menu-save or M-x Buffer-menu-delete commands. execute-extended-command ESC x Function: Read function name, then read its arguments and call it. ---------- Buffer: *Help* ---------- The command C-h a (command-apropos) calls apropos, but specifies a predicate to restrict the output to symbols that are commands. The call to apropos looks like this: (apropos string 'commandp) -- Command: help-command This command is not a function, but rather a symbol which is equivalent to the keymap called help-map. It is defined in `help.el' as follows: (define-key global-map "\C-h" 'help-command) (fset 'help-command help-map) -- Variable: help-map The value of this variable is a local keymap for characters following the Help key, C-h. -- Function: print-help-return-message &optional function This function builds a string which is a message explaining how to restore the previous state of the windows after a help command. After building the message, it applies function to it if function is non-nil. Otherwise it calls message to display it in the echo area. This function expects to be called inside a with-output-to-temp-buffer special form, and expects standard-output to have the value bound by that special form. For an example of its use, see the example in the section describing the documentation function (see Accessing Documentation). The constructed message will have one of the forms shown below. ---------- Echo Area ---------- Type C-x 1 to remove help window. ---------- Echo Area ---------- ---------- Echo Area ---------- Type C-x 4 b RET to restore old contents of help window. ---------- Echo Area ---------- -- Variable: help-char The value of this variable is the character that Emacs recognizes as meaning Help. When Emacs reads this character (which is usually 8, the value of C-h), Emacs evaluates (eval help-form), and displays the result if it is a string. If help-form's value is nil, this character is read normally. -- Variable: help-form The value of this variable is a form to execute when the character help-char is read. If the form returns a string, that string is displayed. If help-form is nil, then the help character is not recognized. Entry to the minibuffer binds this variable to the value of minibuffer-help-form. The following two functions are found in the library `helper'. They are for modes that want to provide help without relinquishing control, such as the ``electric'' modes. You must load that library with (require 'helper) in order to use them. Their names begin with `Helper' to distinguish them from the ordinary help functions. -- Command: Helper-describe-bindings This command pops up a window displaying a help buffer containing a listing of all of the key bindings from both the local and global keymaps. It works by calling describe-bindings. -- Command: Helper-help This command provides help for the current mode. It prompts the user in the minibuffer with the message `Help (Type ? for further options)', and then provides assistance in finding out what the key bindings are, and what the mode is intended for. It returns nil. This can be customized by changing the map Helper-help-map. ΓòÉΓòÉΓòÉ 25. Files ΓòÉΓòÉΓòÉ In Emacs, you can find, create, view, save, and otherwise work with files and file directories. This chapter describes most of the file-related functions of Emacs Lisp, but a few others are described in Buffers, and those related to backups and auto-saving are described in Backups and Auto-Saving. ΓòÉΓòÉΓòÉ 25.1. Visiting Files ΓòÉΓòÉΓòÉ Visiting a file means reading a file into a buffer. Once this is done, we say that the buffer is visiting that file, and call the file ``the visited file'' of the buffer. A file and a buffer are two different things. A file is information recorded permanently in the computer (unless you delete it). A buffer, on the other hand, is information inside of Emacs that will vanish at the end of the editing session (or when you kill the buffer). Usually, a buffer contains information that you have copied from a file; then we say the buffer is visiting that file. The copy in the buffer is what you modify with editing commands. Such changes to the buffer do not change the file; therefore, to make the changes permanent, you must save the buffer, which means copying the altered buffer contents back into the file. In spite of the distinction between files and buffers, people often refer to a file when they mean a buffer and vice-versa. Indeed, we say, ``I am editing a file,'' rather than, ``I am editing a buffer which I will soon save as a file of the same name.'' Humans do not usually need to make the distinction explicit. When dealing with a computer program, however, it is good to keep the distinction in mind. ΓòÉΓòÉΓòÉ 25.1.1. Functions for Visiting Files ΓòÉΓòÉΓòÉ This section describes the functions normally used to visit files. For historical reasons, these functions have names starting with `find-' rather than `visit-'. See Buffer File Name, for functions and variables that access the visited file name of a buffer or that find an existing buffer by its visited file name. -- Command: find-file filename This function reads the file filename into a buffer and displays that buffer in the selected window so that the user can edit it. The body of the find-file function is very simple and looks like this: (switch-to-buffer (find-file-noselect filename)) (See switch-to-buffer in Displaying Buffers.) When find-file is called interactively, it prompts for filename in the minibuffer. -- Function: find-file-noselect filename This function is the guts of all the file-visiting functions. It reads a file into a buffer and returns the buffer. You may then make the buffer current or display it in a window if you wish, but this function does not do so. If no buffer is currently visiting filename, then one is created and the file is visited. If filename does not exist, the buffer is left empty, and find-file-noselect displays the message `New file' in the echo area. If a buffer is already visiting filename, then find-file-noselect uses that buffer rather than creating a new one. However, it does verify that the file has not changed since it was last visited or saved in that buffer. If the file has changed, then this function asks the user whether to reread the changed file. If the user says `yes', any changes previously made in the buffer will be lost. The find-file-noselect function calls after-find-file after the file is read in (see Subroutines of Visiting). The after-find-file function sets the buffer major mode, parses local variables, warns the user if there exists an auto-save file more recent than the file just visited, and finishes by running the functions in find-file-hooks. The find-file-noselect function returns the buffer that is visiting the file filename. (find-file-noselect "/etc/fstab") => #<buffer fstab> -- Command: find-alternate-file filename This function reads the file filename into a buffer and selects it, killing the buffer current at the time the command is run. It is useful if you have visited the wrong file by mistake, so that you can get rid of the buffer that you did not want to create, at the same time as you visit the file you intended. When this function is called interactively, it prompts for filename. -- Command: find-file-other-window filename This function visits the file filename and displays its buffer in a window other than the selected window. If there are two or more windows on the screen, then the window that is not selected is used. If there is only one window, it is split. The function returns nil. When this function is called interactively, it prompts for filename. -- Command: find-file-read-only filename This function visits the file named filename and selects its buffer, just like find-file, but it marks the buffer as read-only. See Read Only Buffers, for related functions and variables. When this function is called interactively, it prompts for filename. -- Command: view-file filename This function views filename in View mode, returning to the previous buffer when done. View mode is a mode that allows you to skim rapidly through the file but does not let you modify it. After loading the file, view-file calls the value of view-hook if that is non-nil. When this function is called interactively, it prompts for filename. -- Variable: find-file-hooks The value of this variable is a list of functions to be called after a file is visited. The file's local-variables specification (if any) will have been processed before the hooks are run. The buffer visiting the file is current when the hook functions are run. -- Variable: find-file-not-found-hooks The value of this variable is a list of functions to be called when find-file or find-file-noselect is passed a nonexistent filename. These functions are called as soon as the error is detected. buffer-file-name is already set up. The functions are called in the order given, until one of them returns non-nil. ΓòÉΓòÉΓòÉ 25.1.2. Subroutines of Visiting ΓòÉΓòÉΓòÉ The find-file-noselect function uses the create-file-buffer and after-find-file functions as subroutines. Sometimes it is useful to call them directly. -- Function: create-file-buffer filename This function creates a suitably named buffer for visiting filename, and returns it. The string filename (sans directory) is used unchanged if that name is free; otherwise, a string such as `<2>' is appended to get an unused name. See also Creating Buffers. *Note:* create-file-buffer does not associate the new buffer with a file and does not make it the current buffer. (create-file-buffer "foo") => #<buffer foo> (create-file-buffer "foo") => #<buffer foo<2>> (create-file-buffer "foo") => #<buffer foo<3>> This function is used by find-file-noselect. -- Function: after-find-file &optional error warn This function is called by find-file-noselect and by the default revert function (see Reverting). It sets the buffer major mode, and parses local variables (see Auto Major Mode). If there was an error in opening the file, the calling function should pass error a non-nil value. In that case, after-find-file issues a warning: `(New File)'. Note that, for serious errors, you would not even call after-find-file. Only ``file not found'' errors get here with a non-nil error. If warn is non-nil, then this function issues a warning if an auto-save file exists and is more recent than the visited file. The last thing after-find-file does is call all the functions in find-file-hooks. ΓòÉΓòÉΓòÉ 25.2. Saving Buffers ΓòÉΓòÉΓòÉ When you edit a file in Emacs, you are actually working on a buffer that is visiting that file---that is, the contents of the file are copied into the buffer and the copy is what you edit. Changes to the buffer do not change the file until you save the buffer, which means copying the contents of the buffer into the file. -- Command: save-buffer &optional backup-option This function saves the contents of the current buffer in its visited file if the buffer has been modified since it was last visited or saved. Otherwise it does nothing. save-buffer is responsible for making backup files. Normally, backup-option is nil, and save-buffer makes a backup file only if this is the first save or if the buffer was previously modified. Other values for backup-option request the making of backup files in other circumstances: o With an argument of 4 or 64, reflecting 1 or 3 C-u's, the save-buffer function marks this version of the file to be backed up when the buffer is next saved. o With an argument of 16 or 64, reflecting 2 or 3 C-u's, the save-buffer function unconditionally backs up the previous version of the file before saving it. -- Command: save-buffers-kill-emacs &optional kill-silently-p This function offers to save each buffer that needs to be saved, and then kills the Emacs job. With a non-nil value of the optional kill-silently-p argument, it unconditionally and silently saves all the file-visiting buffers, and then kills the job. The save-buffers-kill-emacs function is defined as follows: (defun save-buffers-kill-emacs (&optional arg) "Offer to save each buffer, then kill this Emacs fork..." (interactive "P") (save-some-buffers arg t) (kill-emacs)) -- Command: save-some-buffers &optional save-silently-p exiting This command saves some modified file-visiting buffers. Normally it asks the user about each buffer. If the function is called with a non-nil value of the optional save-silently-p argument, this function saves all the file-visiting buffers without querying the user. The optional exiting argument, if non-nil, requests this function to offer also to save certain other buffers that are not visiting files. These are buffers that have a non-nil local value of buffer-offer-save. (A user who says yes to saving one of these will be asked to specify a file name to use.) The save-buffers-kill-emacs function passes a non-nil value for this argument. -- Variable: buffer-offer-save When this variable is non-nil in a buffer, Emacs offers to save the buffer on exit even if the buffer is not visiting a file. The variable is automatically local in all buffers. Normally, Mail mode (used for editing outgoing mail) sets this to t. -- Command: write-file filename This function writes the current buffer into file filename, makes the buffer visit that file, and marks it not modified. The buffer is renamed to correspond to filename unless that name is already in use. -- Variable: write-file-hooks The value of this variable is a list of functions to be called before writing out a buffer to its visited file. If one of them returns non-nil, the file is considered already written and the rest of the functions are not called, nor is the usual code for writing the file executed. If the file-precious-flag variable is nil, the file is moved to the backup file before any of the hooks are called. If none of the hooks actually write the file, but one does return non-nil, the file will not exist, although the backup will. Here is an example showing how to add an element to write-file-hooks but avoid adding it twice: (or (memq 'my-write-file-hook write-file-hooks) (setq write-file-hooks (cons 'my-write-file-hook write-file-hooks))) -- Variable: file-precious-flag If this variable is non-nil, then save-buffer protects against I/O errors while saving by renaming the original file to a temporary name before writing the new contents of the file. If the new contents are successfully written, the renamed original file is deleted. Otherwise, it is renamed back to the original name. This procedure prevents problems such as a lack of disk space from resulting in an invalid file. Some modes set this non-nil locally in particular buffers. -- User Option: require-final-newline This variable determines whether files may be written out that do not end with a newline. If the value of the variable is t, then Emacs silently puts a newline at the end of the file whenever the buffer being saved does not already end in one. If the value of the variable is non-nil, but not t, then Emacs asks the user whether to add a newline each time the case arises. If the value of the variable is nil, then Emacs doesn't add newlines at all. nil is the default value, but a few major modes change it to t. ΓòÉΓòÉΓòÉ 25.3. Reading from Files ΓòÉΓòÉΓòÉ You can copy a file directly from the disk and insert it into a buffer using the insert-file-contents function, or its interactive variant, insert-file. -- Command: insert-file filename This function inserts the contents of file filename into the current buffer after point, and sets the mark at the end of the inserted text. An error is signaled if filename is not the name of a file that can be read. This function is for interactive use and does little more than call insert-file-contents. -- Function: insert-file-contents filename &optional visit This function inserts the contents of file filename into the current buffer after point. It returns a list of the absolute file name and the length of the data inserted. An error is signaled if filename is not the name of a file that can be read. If visit is non-nil, it also marks the buffer as unmodified and sets up various fields in the buffer so that it is visiting the file filename: these include the buffer's visited file name and its last save file modtime. This feature is used by find-file-noselect and you should probably not use it yourself. ΓòÉΓòÉΓòÉ 25.4. Writing to Files ΓòÉΓòÉΓòÉ You can write the contents of a buffer, or part of a buffer, directly to a file on disk using the append-to-file and write-region functions. Don't use these functions to write to files that are being visited; that could cause confusion in the mechanisms for visiting. -- Command: append-to-file start end filename This function appends the contents of the region delimited by start and end in the current buffer to the end of file filename. If that file does not exist, it is created. This function returns nil. An error is signaled if filename specifies a nonwritable file, or a nonexistent file in a directory where files cannot be created. -- Command: write-region start end filename &optional append visit This function writes the region (of the current buffer) delimited by start and end into the file specified by filename. If append is non-nil, then the region is appended to the existing file contents (if any). If visit is t, then Emacs establishes an association between the buffer and the file: the buffer is then visiting that file. It also sets the last file modification time for the current buffer to filename's modtime, and marks the buffer as not modified. This feature is used by write-file and you should probably not use it yourself. Normally, this function displays a message `Wrote file filename' in the echo area. If visit is neither t nor nil, then this message is inhibited. This feature is useful for programs that use files for internal purposes, files which the user does not need to know about. ΓòÉΓòÉΓòÉ 25.5. File Locks ΓòÉΓòÉΓòÉ When two users edit the same file at the same time, they are likely to interfere with each other. Emacs tries to prevent this situation from arising by recording a file lock when a file is being modified. Emacs can then detect the first attempt to modify a buffer visiting a file that is locked by another Emacs job, and ask the user what to do. File locks do not work properly when multiple machines can share filesystems, such as with NFS. Perhaps a better file locking system will be implemented in the future. When file locks do not work, it is possible for two users to make changes simultaneously, but Emacs will still be able to warn the user who saves second. Also, the detection of modification of a buffer visiting a file changed on disk catches some cases of simultaneous editing; see Modification Time. -- Function: file-locked-p filename This function returns nil if the file filename is not locked by this Emacs process. It returns t if it is locked by this Emacs, and it returns the name of the user who has locked it if it is locked by someone else. (file-locked-p "foo") => nil -- Function: lock-buffer &optional filename This function locks the file filename, if the current buffer is modified. The argument filename defaults to the current buffer's visited file. Nothing is done if the current buffer is not visiting a file, or is not modified. -- Function: unlock-buffer This function unlocks the file being visited in the current buffer, if the buffer is modified. If the buffer is not modified, then the file should not be locked, so this function does nothing. It also does nothing if the current buffer is not visiting a file. -- Function: ask-user-about-lock file other-user This function is called when the user tries to modify file, but it is locked by another user name other-user. The value it returns controls what Emacs will do with the file: o A value of t tells Emacs to grab the lock on the file. Then this user may edit the file and other-user loses the lock. o A value of nil tells Emacs to ignore the lock and let this user edit the file anyway. o This function may instead signal a file-locked error, in which case the change to the buffer which the user was about to make does not take place. The error message for this error looks like this: error--> File is locked: file other-user where file is the name of the file and other-user is the name of the user who has locked the file. The default definition of this function asks the user to choose what to do. If you wish, you can replace the ask-user-about-lock function with your own version that decides in another way. The code for its usual definition is in `userlock.el'. ΓòÉΓòÉΓòÉ 25.6. Information about Files ΓòÉΓòÉΓòÉ The functions described in this section are similar in as much as they all operate on strings which are interpreted as file names. All have names that begin with the word `file'. These functions all return information about actual files or directories, so their arguments must all exist as actual files or directories unless otherwise noted. Most of the file-oriented functions take a single argument, filename, which must be a string. The file name is expanded using expand-file-name, so `~' is handled correctly, as are relative file names (including `../'). Environment variable substitutions, such as `$HOME', are not recognized by these functions. ΓòÉΓòÉΓòÉ 25.6.1. Testing Accessibility ΓòÉΓòÉΓòÉ These functions test for permission to access a file in specific ways. -- Function: file-exists-p filename This function returns t if a file named filename appears to exist. This does not mean you can necessarily read the file, only that you can find out its attributes. (On Unix, this is true if the file exists and you have execute permission on the containing directories, regardless of the protection of the file itself.) If the file does not exist, or if fascist access control policies prevent you from finding the attributes of the file, this function returns nil. -- Function: file-readable-p filename This function returns t if a file named filename exists and you can read it. It returns nil otherwise. (file-readable-p "files.texi") => t (file-exists-p "/usr/spool/mqueue") => t (file-readable-p "/usr/spool/mqueue") => nil -- Function: file-writable-p filename This function returns t if filename can be written or created by you. It is writable if the file exists and you can write it. It is creatable if the file does not exist, but the specified directory does exist and you can write in that directory. file-writable-p returns nil otherwise. In the third example below, `foo' is not writable because the parent directory does not exist, even though the user could create it. (file-writable-p "~rms/foo") => t (file-writable-p "/foo") => nil (file-writable-p "~rms/no-such-dir/foo") => nil -- Function: file-newer-than-file-p filename1 filename2 This functions returns t if the file filename1 is newer than file filename2. If filename1 does not exist, it returns nil. If filename2 does not exist, it returns t. In the following example, assume that the file `aug-19' was written on the 19th, and `aug-20' was written on the 20th. The file `no-file' doesn't exist at all. (file-newer-than-file-p "aug-19" "aug-20") => nil (file-newer-than-file-p "aug-20" "aug-19") => t (file-newer-than-file-p "aug-19" "no-file") => t (file-newer-than-file-p "no-file" "aug-19") => nil ΓòÉΓòÉΓòÉ 25.6.2. Distinguishing Kinds of Files ΓòÉΓòÉΓòÉ This section describes how to distinguish directories and symbolic links from ordinary files. -- Function: file-symlink-p filename If filename is a symbolic link, the file-symlink-p function returns the file name to which it is linked. This may be the name of a text file, a directory, or even another symbolic link, or of no file at all. If filename is not a symbolic link (or there is no such file), file-symlink-p returns nil. (file-symlink-p "foo") => nil (file-symlink-p "sym-link") => "foo" (file-symlink-p "sym-link2") => "sym-link" (file-symlink-p "/bin") => "/pub/bin" -- Function: file-directory-p filename This function returns t if filename is the name of an existing directory, nil otherwise. (file-directory-p "~rms") => t (file-directory-p "~rms/lewis/files.texi") => nil (file-directory-p "~rms/lewis/no-such-file") => nil (file-directory-p "$HOME") => nil (file-directory-p (substitute-in-file-name "$HOME")) => t ΓòÉΓòÉΓòÉ 25.6.3. Other Information about Files ΓòÉΓòÉΓòÉ This section describes the functions for getting detailed information about a file, other than its contents. This information includes the mode bits that control access permission, the owner and group numbers, the number of names, the inode number, the size, and the times of access and modification. -- Function: file-modes filename This function returns the mode bits of filename, as an integer. The mode bits are also called the file permissions, and they specify access control in the usual Unix fashion. If the low-order bit is 1, then the file is executable by all users, if the second lowest-order bit is 1, then the file is writable by all users, etc. The highest value returnable is 4095 (7777 octal), meaning that everyone has read, write, and execute permission, that the suid bit is set for both others and group, and that the sticky bit is set. (file-modes "~/junk/diffs") => 492 ; decimal integer (format "%o" 492) => 754 ; convert to octal (set-file-modes "~/junk/diffs" 438) => nil (format "%o" 438) => 666 ; convert to octal % ls -l diffs -rw-rw-rw- 1 lewis 0 3063 Oct 30 16:00 diffs -- Function: file-nlinks filename This functions returns the number of names (i.e., hard links) that file filename has. If the file does not exist, then this function returns nil. Note that symbolic links have no effect on this function, because they are not considered to be names of the files they link to. % ls -l foo* -rw-rw-rw- 2 rms 4 Aug 19 01:27 foo -rw-rw-rw- 2 rms 4 Aug 19 01:27 foo1 (file-nlinks "foo") => 2 (file-nlinks "doesnt-exist") => nil -- Function: file-attributes filename This function returns a list of attributes of file filename. If the specified file cannot be opened, it returns nil. The elements of the list, in order, are: 1. t for a directory, a string for a symbolic link (the name linked to), or nil for a text file. 2. The number of names the file has. Alternate names, also known as hard links, can be created with add-name-to-file ( see Changing File Attributes). 3. The file's uid. 4. The file's gid. 5. The time of last access, as a list of two integers. The first integer has the high-order 16 bits of time, the second has the low 16 bits. 6. The time of last modification as a list of two integers (as above). 7. The time of last status change as a list of two integers (as above). 8. The size of the file in bytes. 9. The file's modes, as a string of ten letters or dashes as in `ls -l'. 10. t if the file's gid would change if file were deleted and recreated; nil otherwise. 11. The file's inode number. For example, here are the file attributes for `files.texi': (file-attributes "files.texi") => (nil 1 2235 75 (8489 20284) (8489 20284) (8489 20285) 14906 "-rw-rw-rw-" nil 20920) and here is how the result is interpreted: nil is neither a directory nor a symbolic link. 1 has only one name (the name `files.texi' in the current default directory). 2235 is owned by the user with uid 2235. 75 is in the group with gid 75. (8489 20284) was last accessed on Aug 19 00:09. Unfortunately, you cannot convert this number into a time string in Emacs. (8489 20284) was last accessed on Aug 19 00:09. (8489 20285) last had its inode changed on Aug 19 00:09. 14906 is 14906 characters long. "-rw-rw-rw-" has a mode of read and write access for the owner, group, and world. nil would retain the same gid if it were recreated. 20920 has an inode number of 20920. ΓòÉΓòÉΓòÉ 25.7. Contents of Directories ΓòÉΓòÉΓòÉ A directory is a kind of file that contains other files entered under various names. Directories are a feature of the file system. Emacs can list the names of the files in a directory as a Lisp list, or display the names in a buffer using the ls shell command. In the latter case, it can optionally display information about each file, depending on the value of switches passed to the ls command. -- Function: directory-files directory &optional full-name match-regexp This function returns a list of the names of the files in the directory directory. If full-name is non-nil, the function returns the files' absolute file names. Otherwise, it returns just the names relative to the specified directory. If match-regexp is non-nil, function returns only those file names that contain that regular expression---the other file names are discarded from the list. (directory-files "~lewis") => ("#foo#" "#foo.el#" "." ".." "dired-mods.el" "files.texi" "files.texi.~1~") An error is signaled if directory is not the name of a directory that can be read. -- Function: file-name-all-versions file dirname This function returns a list of all versions of the file named file in directory dirname. -- Command: list-directory filespec &optional verbose This command displays a list of files in or matching filespec. It calls the shell command ls, passing it options under the control of verbose. The argument filespec may be either a directory name or a file name containing optional wildcards. Wildcards are processed by the shell. The options passed to ls are the value of list-directory-verbose-switches if verbose is non-nil; list-directory-brief-switches otherwise. Interactively, the raw prefix argument is used for verbose. -- Variable: list-directory-brief-switches This variable contains switches for list-directory to pass to ls for a short or ``brief'' listing. The default value is "-CF". -- Variable: list-directory-verbose-switches This variable contains switches for list-directory to pass to ls for a verbose or ``long'' listing. The default value is "-l". ΓòÉΓòÉΓòÉ 25.8. Changing File Names and Attributes ΓòÉΓòÉΓòÉ The functions in this section rename, copy, delete, link, and set the modes of files. In the functions that have an argument newname, if a file by the name of newname already exists, the actions taken depend on the value of the argument ok-if-already-exists: o A file-already-exists error is signaled if ok-if-already-exists is nil. o Confirmation is requested if ok-if-already-exists is a number. o No confirmation is requested if ok-if-already-exists is any other value, in which case the old file is removed. -- Function: add-name-to-file oldname newname &optional ok-if-already-exists This function gives the file named oldname the additional name newname. This means that newname will be a new ``hard link'' to oldname. In the first part of the following example, we list two files, `foo' and `foo3'. % ls -l fo* -rw-rw-rw- 1 rms 29 Aug 18 20:32 foo -rw-rw-rw- 1 rms 24 Aug 18 20:31 foo3 Then we evaluate the form (add-name-to-file "~/lewis/foo" "~/lewis/foo2"). Again we list the files. This shows two names, `foo' and `foo2'. (add-name-to-file "~/lewis/foo1" "~/lewis/foo2") => nil % ls -l fo* -rw-rw-rw- 2 rms 29 Aug 18 20:32 foo -rw-rw-rw- 2 rms 29 Aug 18 20:32 foo2 -rw-rw-rw- 1 rms 24 Aug 18 20:31 foo3 Finally, we evaluate (add-name-to-file "~/lewis/foo" "~/lewis/foo3" t), and list the files again. Now there are three names for one file: `foo', `foo2', and `foo3'. The old contents of `foo3' are lost. (add-name-to-file "~/lewis/foo1" "~/lewis/foo3") => nil % ls -l fo* -rw-rw-rw- 3 rms 29 Aug 18 20:32 foo -rw-rw-rw- 3 rms 29 Aug 18 20:32 foo2 -rw-rw-rw- 3 rms 29 Aug 18 20:32 foo3 This function is meaningless on VMS, where multiple names for one file are not allowed. See also file-nlinks in Information about Files. -- Command: rename-file filename newname &optional ok-if-already-exists This command renames the file filename as newname. If filename has additional names aside from filename, it continues to have those names. In fact, adding the name newname with add-name-to-file and then deleting filename has the same effect as renaming, aside from momentary intermediate states. In an interactive call, this function prompts for filename and newname in the minibuffer; also, it requests confirmation if newname already exists. -- Command: copy-file oldname newname &optional ok-if-exists time This command copies the file oldname to newname. An error is signaled if oldname does not exist. If time is non-nil, then this functions gives the new file the same last-modified time that the old one has. (This works on only some operating systems.) In an interactive call, this function prompts for filename and newname in the minibuffer; also, it requests confirmation if newname already exists. -- Command: delete-file filename This command deletes the file filename, like the shell command `rm filename'. If the file has multiple names, it continues to exist under the other names. A suitable kind of file-error error is signaled if the file does not exist, or is not deletable. (In Unix, a file is deletable if its directory is writable.) -- Command: make-symbolic-link filename newname &optional ok-if-exists This command makes a symbolic link to filename, named newname. This is like the shell command `ln -s filename newname'. In an interactive call, filename and newname are read in the minibuffer, and ok-if-exists is set to the numeric prefix argument. -- Function: set-file-modes filename mode This function sets mode bits of filename to mode (which must be a integer). Only the 12 low bits of mode are used. ΓòÉΓòÉΓòÉ 25.9. File Names ΓòÉΓòÉΓòÉ Files are generally referred to by their names, in Emacs as elsewhere. File names in Emacs are represented as strings. The functions that operate on a file all expect a file name argument. In addition to operating on files themselves, Emacs Lisp programs often need to operate on the names; i.e., to take them apart and to use part of a name to construct related file names. This section describes how to manipulate file names. The functions in this section do not actually access files, so they can operate on file names that do not refer to an existing file or directory. On VMS, all these functions understand both VMS file name syntax and Unix syntax. This is so that all the standard Lisp libraries can specify file names in Unix syntax and work properly on VMS without change. ΓòÉΓòÉΓòÉ 25.9.1. File Name Components ΓòÉΓòÉΓòÉ The operating system groups files into directories. To specify a file, you must specify the directory, and the file's name in that directory. Therefore, a file name in Emacs is considered to have two main parts: the directory name part, and the nondirectory part (or file name within the directory). Either part may be empty. Concatenating these two parts reproduces the original file name. On Unix, the directory part is everything up to and including the last slash; the nondirectory part is the rest. The rules in VMS syntax are complicated. For some purposes, the nondirectory part is further subdivided into the name and the version number. On Unix, only backup files have version numbers in their names; on VMS, every file has a version number, but most of the time the file name actually used in Emacs omits the version number. Version numbers are found mostly in directory lists. -- Function: file-name-directory filename This function returns the directory part of filename (or nil if filename does not include a directory part). On Unix, the function returns a string ending in a slash. On VMS, it returns a string ending in one of the three characters `:', `]', or `>'. (file-name-directory "lewis/foo") ; Unix example => "lewis/" (file-name-directory "foo") ; Unix example => nil (file-name-directory "[X]FOO.TMP") ; VMS example => "[X]" -- Function: file-name-nondirectory filename This function returns the nondirectory part of filename. (file-name-nondirectory "lewis/foo") => "foo" (file-name-nondirectory "foo") => "foo" ;; The following example is accurate only on VMS. (file-name-nondirectory "[X]FOO.TMP") => "FOO.TMP" -- Function: file-name-sans-versions filename This function returns filename without any file version numbers, backup version numbers, or trailing tildes. (file-name-sans-versions "~rms/foo.~1~") => "~rms/foo" (file-name-sans-versions "~rms/foo~") => "~rms/foo" (file-name-sans-versions "~rms/foo") => "~rms/foo" ;; The following example applies to VMS only. (file-name-sans-versions "foo;23") => "foo" ΓòÉΓòÉΓòÉ 25.9.2. Directory Names ΓòÉΓòÉΓòÉ A directory name is the name of a directory. A directory is a kind of file, and it has a file name, which is related to the directory name but not identical to it. (This is not quite the same as the usual Unix terminology.) These two different names for the same entity are related by a syntactic transformation. On Unix, this is simple: a directory name ends in a slash, whereas the directory's name as a file lacks that slash. On VMS, the relationship is more complicated. The difference between a directory name and its name as a file is subtle but crucial. When an Emacs variable or function argument is described as being a directory name, a file name of a directory is not acceptable. All of these functions take a single argument, filename, which must be a string. Environment variable substitutions such as `$HOME', and the symbols `~', and `..', are not expanded. Use expand-file-name or substitute-in-file-name for that (see File Name Expansion). -- Function: file-name-as-directory filename This function returns a string representing filename in a form that the operating system will interpret as the name of a directory. In Unix, this means that a slash is appended to the string. On VMS, the function converts a string of the form `[X]Y.DIR.1' to the form `[X.Y]'. (file-name-as-directory "~rms/lewis") => "~rms/lewis/" -- Function: directory-file-name dirname This function returns a string representing dirname in a form that the operating system will interpret as the name of a file. On Unix, this means that a final slash is removed from the string. On VMS, the function converts a string of the form `[X.Y]' to `[X]Y.DIR.1'. (directory-file-name "~lewis/") => "~lewis" ΓòÉΓòÉΓòÉ 25.9.3. Absolute and Relative File Names ΓòÉΓòÉΓòÉ All the directories in the file system form a tree starting at the root directory. A file name can specify all the directory names starting from the root of the tree; then it is called an absolute file name. Or it can specify the position of the file in the tree relative to a default directory; then it is called an relative file name. On Unix, an absolute file name starts with a slash or a tilde (`~'), and a relative one does not. The rules on VMS are complicated. -- Function: file-name-absolute-p filename This function returns t if file filename is an absolute file name, nil otherwise. On VMS, this function understands both Unix syntax and VMS syntax. (file-name-absolute-p "~rms/foo") => t (file-name-absolute-p "rms/foo") => nil (file-name-absolute-p "$HOME") => nil (file-name-absolute-p "/user/rms/foo") => t ΓòÉΓòÉΓòÉ 25.9.4. Functions that Expand Filenames ΓòÉΓòÉΓòÉ Expansion of a file name means converting a relative file name to an absolute one. Since this is done relative to a default directory, you must specify the default directory name as well as the file name to be expanded. Expansion also canonicalizes file names by eliminating redundancies such as `./' and `name/../'. -- Function: expand-file-name filename &optional directory This function converts filename to an absolute file name. If directory is supplied, it is the directory to start with if filename is relative. Otherwise, the current buffer's value of default-directory is used. For example: (expand-file-name "foo") => "/xcssun/users/rms/lewis/foo" (expand-file-name "../foo") => "/xcssun/users/rms/foo" (expand-file-name "foo" "/usr/spool") => "/usr/spool/foo" (expand-file-name "$HOME/foo") => "/xcssun/users/rms/lewis/$HOME/foo" Filenames containing `.' or `..' are simplified to their canonical form: (expand-file-name "bar/../foo") => "/xcssun/users/rms/lewis/foo" `~/' is expanded into the user's home directory. A `/' or `~' following a `/' is taken to be the start of an absolute file name that overrides what precedes it, so everything before that `/' or `~' is deleted. For example: (expand-file-name "/a1/gnu//usr/local/lib/emacs/etc/MACHINES") => "/usr/local/lib/emacs/etc/MACHINES" (expand-file-name "/a1/gnu/~/foo") => "/xcssun/users/rms/foo" In both cases, `/a1/gnu/' has been discarded because an absolute file name follows it. Note that expand-file-name does not expand environment variables; that is done only by substitute-in-file-name. -- Variable: default-directory The value of this buffer-local variable is the default directory for the current buffer. It is local in every buffer. expand-file-name uses the default directory when its second argument is nil. On Unix systems, the value is always a string ending with a slash. default-directory => "/user/lewis/manual/" -- Function: substitute-in-file-name filename This function replaces environment variables names in filename with the values to which they are set by the operating system. Following standard Unix shell syntax, `$' is the prefix to substitute an environment variable value. The environment variable name is the series of alphanumeric characters (including underscores) that follow the `$'. If the character following the `$' is a `{', then the variable name is everything up to the matching `}'. Here we assume that the environment variable HOME, which holds the user's home directory name, has the value `/xcssun/users/rms'. (substitute-in-file-name "$HOME/foo") => "/xcssun/users/rms/foo" If a `~' or a `/' appears following a `/', after substitution, everything before the following `/' is discarded: (substitute-in-file-name "bar/~/foo") => "~/foo" (substitute-in-file-name "/usr/local/$HOME/foo") => "/xcssun/users/rms/foo" On VMS, `$' substitution is not done, so this function does nothing on VMS except discard superfluous initial components as shown above. ΓòÉΓòÉΓòÉ 25.9.5. Generating Unique File Names ΓòÉΓòÉΓòÉ Some programs need to write temporary files. Here is the usual way to construct a name for such a file: (concat "/tmp/" (make-temp-name name-of-application)) The directory `/tmp/' is chosen because that is the standard place on Unix for temporary files. The task of make-temp-name is to prevent two different users or two different jobs from trying to use the same name. -- Function: make-temp-name string This function generates string that can be used as a unique name. The name will start with the prefix string, and finish with a number that is different in each Emacs job. (make-temp-name "foo") => "foo021304" (make-temp-name "foo") => "foo021304" To prevent conflicts among different application libraries run in the same Emacs, each application should have its own string. The number added to the end of the name distinguishes between the same application running in different Emacses. ΓòÉΓòÉΓòÉ 25.9.6. File Name Completion ΓòÉΓòÉΓòÉ This section describes low-level subroutines for completing a file name. For other completion functions, see Completion. -- Function: file-name-all-completions partial-filename directory This function returns a list of all possible completions for a file whose name starts with partial-filename in directory directory. The order of the completions is the order of the files in the directory, which is unpredictable and conveys no useful information. The argument partial-filename must be a file name containing no directory part and no slash. The current buffer's default directory is prepended to directory, if directory is not an absolute file name. In the following example, suppose that the current default directory, `~rms/lewis', has five files whose names begin with `f': `foo', `file~', `file.c', `file.c.~1~', and `file.c.~2~'. (file-name-all-completions "f" "") => ("foo" "file~" "file.c.~2~" "file.c.~1~" "file.c") (file-name-all-completions "fo" "") => ("foo") -- Function: file-name-completion filename directory This function completes the file name filename in directory directory. It returns the longest prefix common to all file names in directory directory that start with filename. If only one match exists and filename matches it exactly, the function returns t. The function returns nil if directory directory contains no name starting with filename. In the following example, suppose that the current default directory has five files whose names begin with `f': `foo', `file~', `file.c', `file.c.~1~', and `file.c.~2~'. (file-name-completion "fi" "") => "file" (file-name-completion "file.c.~1" "") => "file.c.~1~" (file-name-completion "file.c.~1~" "") => t (file-name-completion "file.c.~3" "") => nil -- User Option: completion-ignored-extensions file-name-completion usually ignores file names that end in any string in this list. It does not ignore them when all the possible completions end in one of these suffixes or when a buffer showing all possible completions is displayed. A typical value might look like this: completion-ignored-extensions => (".o" ".elc" "~" ".dvi") ΓòÉΓòÉΓòÉ 26. Backups and Auto-Saving ΓòÉΓòÉΓòÉ Backup files and auto-save files are two methods by which Emacs tries to protect the user from the consequences of crashes or of the user's own errors. Auto-saving preserves the text from earlier in the current editing session; backup files preserve file contents prior to the current session. ΓòÉΓòÉΓòÉ 26.1. Backup Files ΓòÉΓòÉΓòÉ A backup file is a copy of the old contents of a file you are editing. Emacs makes a backup file the first time you save a buffer into its visited file. Normally, this means that the backup file contains the contents of the file as it was before the current editing session. The contents of the backup file normally remain unchanged once it exists. Backups are usually made by renaming the visited file to a new name. Optionally, you can specify that backup files should be made by copying the visited file. This choice makes a difference for files with multiple names; it also can affect whether the edited file remains owned by the original owner or becomes owned by the user editing it. By default, Emacs makes a single backup file for each file edited. You can alternatively request numbered backups; then each new backup file gets a new name. You can delete old numbered backups when you don't want them any more, or Emacs can delete them automatically. ΓòÉΓòÉΓòÉ 26.1.1. Making Backup Files ΓòÉΓòÉΓòÉ -- Function: backup-buffer This function makes a backup of the file visited by the current buffer, if appropriate. It is called by save-buffer before saving the buffer the first time. -- Variable: buffer-backed-up This buffer-local variable indicates whether this buffer's file has been backed up on account of this buffer. If it is non-nil, then the backup file has been written. Otherwise, the file should be backed up when it is next saved (if backup files are enabled). -- Variable: make-backup-files This variable determines whether or not backup files will be created. If it is non-nil, then Emacs will create a backup of each file when it is saved for the first time. The following example shows how to change the make-backup-files variable only in the `RMAIL' buffer and not elsewhere. Setting it nil stops Emacs from making backups of the `RMAIL' file, which may save disk space. (You would put this code in your `.emacs' file.) (setq rmail-mode-hook (function (lambda () (make-local-variable 'make-backup-files) (setq make-backup-files nil))) ΓòÉΓòÉΓòÉ 26.1.2. Backup by Renaming or by Copying? ΓòÉΓòÉΓòÉ There are two ways that Emacs can make a backup file: o Emacs can rename the original file so that it becomes a backup file, and then write the buffer being saved into a new file. In this case, any other names (i.e., hard links) of the original file will now refer to the backup file. The new file will be owned by the user doing the editing, and its group will be the default for the user or the directory. o Emacs can copy the original file into a backup file, and then overwrite the original file with new contents. In this case, any other names (i.e., hard links) of the original file will still refer to the current version of the file. The file's owner and group will be unchanged. The first method, renaming, is the default. The variable backup-by-copying, if non-nil, says to use the second method, which is to copy the original file and overwrite it with the new buffer contents. The variable file-precious-flag, if non-nil, also has this effect (as a sideline of its main significance). See Saving Buffers. The variables backup-by-copying-when-linked and backup-by-copying-when-mismatch, if non-nil, cause the second method to be used in certain special cases. They have no effect on the treatment of files that don't fall into the special cases. -- Variable: backup-by-copying This variable controls whether to make backup files by copying. If it is non-nil, then Emacs always copies the current contents of the file into the backup file before writing the buffer to be saved to the file. (In many circumstances, this has the same effect as file-precious-flag.) -- Variable: backup-by-copying-when-linked This variable controls whether to make backups by copying for files with multiple names (hard links). If it is non-nil, then Emacs will use copying to create backups for those files. This variable is significant only if backup-by-copying is nil, since copying is always used when that variable is non-nil. -- Variable: backup-by-copying-when-mismatch This variable controls whether to make backups by copying when renaming would cause either the owner or the group of the file to change. If it is non-nil then Emacs will create backups by copying in such cases. The value has no effect when renaming would not result in changing the owner or group of the file; that is, for files which are owned by the user and whose group matches the default for a new file created there by the user. This variable is significant only if backup-by-copying is nil, since copying is always used when that variable is non-nil. ΓòÉΓòÉΓòÉ 26.1.3. Making and Deleting Numbered Backup Files ΓòÉΓòÉΓòÉ If a file's name is `foo', the names of its numbered backup versions are `foo.~v~', for various integers v, like this: `foo.~1~', `foo.~2~', `foo.~3~', ..., `foo.~259~', and so on. -- User Option: version-control This variable controls whether to make a single non-numbered backup file or multiple numbered backups. nil Make numbered backups if the visited file already has numbered backups; otherwise, do not. never Do not make numbered backups. anything else Do make numbered backups. The use of numbered backups ultimately leads to a large number of backup versions, which must then be deleted. Emacs can do this automatically. -- User Option: kept-new-versions The value of this variable is the number of oldest versions to keep when a new numbered backup is made. The newly made backup is included in the count. The default value is 2. -- User Option: kept-old-versions The value of this variable is the number of oldest versions to keep when a new numbered backup is made. The default value is 2. -- User Option: dired-kept-versions This variable plays a role in Dired's dired-clean-directory (.) command like that played by kept-old-versions when a backup file is made. The default value is 2. If there are backups numbered 1, 2, 3, 5, and 7, and both of these variables have the value 2, then the backups numbered 1 and 2 will be kept as old versions and those numbered 5 and 7 will be kept as new versions; backup version 3 will be deleted. The function find-backup-file-name is responsible for determining which backup versions to delete, but does not delete them itself. -- User Option: trim-versions-without-asking If this variable is non-nil, then excess backup versions are deleted silently. Otherwise, the user is asked whether to delete them. ΓòÉΓòÉΓòÉ 26.1.4. Naming Backup Files ΓòÉΓòÉΓòÉ The functions in this section are documented mainly because you can customize the naming conventions for backup files by redefining them. -- Function: backup-file-name-p filename This function returns a non-nil value if filename is a possible name for a backup file. A file with the name filename need not exist; the function just checks the name. (backup-file-name-p "foo") => nil (backup-file-name-p "foo~") => 3 The standard definition of this function is as follows: (defun backup-file-name-p (file) "Return non-nil if FILE is a backup file name (numeric or not)..." (string-match "~$" file)) Thus, the function returns a non-nil value if the file name ends with a `~' This simple expression is placed in a separate function to make it easy to redefine for customization. -- Function: make-backup-file-name filename This function returns a string which is the name to use for a non-numbered backup file for file filename. On Unix, this is just filename with a tilde appended. The standard definition of this function is as follows: (defun make-backup-file-name (file) "Create the non-numeric backup file name for FILE..." (concat file "~")) You can change the backup file naming convention by redefining this function. In the following example, make-backup-file-name is redefined to prepend a `.' as well as to append a tilde. (defun make-backup-file-name (filename) (concat "." filename "~")) (make-backup-file-name "backups.texi") => ".backups.texi~" If you do redefine make-backup-file-name, be sure to redefine backup-file-name-p and find-backup-file-name as well. -- Function: find-backup-file-name filename This function computes the file name for a new backup file for filename. It may also propose certain existing backup files for deletion. find-backup-file-name returns a list whose car is the name for the new backup file and whose cdr is a list of backup files whose deletion is proposed. Two variables called kept-old-versions and kept-new-versions determine which old backup versions will be kept (by excluding them from the list of backup files ripe for deletion). See Numbered Backups. In this example, `~rms/foo.~5~' is the name to use for the new backup file, and `~rms/foo.~3~' is an ``excess'' version that the caller should consider deleting now. (find-backup-file-name "~rms/foo") => ("~rms/foo.~5~" "~rms/foo.~3~") ΓòÉΓòÉΓòÉ 26.2. Auto-Saving ΓòÉΓòÉΓòÉ Emacs periodically saves all files that you are visiting; this is called auto-saving. Auto-saving prevents you from losing more than a limited amount of work if the system crashes. By default, auto-saves happen every 300 keystrokes. See Auto-Saving: Protection Against Disasters, for information on auto-save for users. Here we describe the functions used to implement auto-saving and the variables that control them. -- Variable: buffer-auto-save-file-name This buffer-local variable is the name of the file used for auto-saving the current buffer. It is nil if the buffer should not be auto-saved. buffer-auto-save-file-name => "/xcssun/users/rms/lewis/#files.texi#" -- Command: auto-save-mode arg When used interactively without an argument, this command is a toggle switch: it turns on auto-saving of the current buffer if it is off, and vice-versa. With an argument arg, the command turns auto-saving on if the value of arg is t, a nonempty list, or a positive integer. Otherwise, it turns auto-saving off. -- Function: auto-save-file-name-p filename This function returns a non-nil value if filename is a string that could be the name of an auto-save file. It works based on knowledge of the naming convention for auto-save files: a name that begins and ends with hash marks (`#') is a possible auto-save file name. The argument filename should not contain a directory part. (make-auto-save-file-name) => "/xcssun/users/rms/lewis/#files.texi#" (auto-save-file-name-p "#files.texi#") => 0 (auto-save-file-name-p "files.texi") => nil The standard definition of this function is as follows: (defun auto-save-file-name-p (filename) "Return non-nil if FILENAME can be yielded by..." (string-match "^#.*#$" filename)) This function exists so that you can customize it if you wish to change the naming convention for auto-save files. If you redefine it, be sure to redefine make-auto-save-file-name correspondingly. -- Function: make-auto-save-file-name This function returns the file name to use for auto-saving the current buffer. This is just the file name with hash marks (`#') appended and prepended to it. This function does not look at the variable auto-save-visited-file-name; that should be checked before this function is called. (make-auto-save-file-name) => "/xcssun/users/rms/lewis/#backup.texi#" The standard definition of this function is as follows: (defun make-auto-save-file-name () "Return file name to use for auto-saves of current buffer..." (if buffer-file-name (concat (file-name-directory buffer-file-name) "#" (file-name-nondirectory buffer-file-name) "#") (expand-file-name (concat "#%" (buffer-name) "#")))) This exists as a separate function so that you can redefine it to customize the naming convention for auto-save files. Be sure to change auto-save-file-name-p in a corresponding way. -- Variable: auto-save-visited-file-name If this variable is non-nil, Emacs will auto-save buffers in the files they are visiting. That is, the auto-save is done in the same file which you are editing. Normally, this variable is nil, so auto-save files have distinct names that are created by make-auto-save-file-name. When you change the value of this variable, the value does not take effect until the next time auto-save mode is reenabled in any given buffer. If auto-save mode is already enabled, auto-saves continue to go in the same file name until auto-save-mode is called again. -- Function: recent-auto-save-p This function returns t if the current buffer has been auto-saved since the last time it was read in or saved. -- Function: set-buffer-auto-saved This function marks the current buffer as auto-saved. The buffer will not be auto-saved again until the buffer text is changed again. The function returns nil. -- User Option: auto-save-interval The value of this variable is the number of characters that Emacs reads from the keyboard between auto-saves. Each time this many more characters are read, auto-saving is done for all buffers in which it is enabled. -- User Option: auto-save-default If this variable is non-nil, buffers that are visiting files have auto-saving enabled by default. Otherwise, they do not. -- Command: do-auto-save &optional no-message This function auto-saves all buffers that need to be auto-saved. This is all buffers for which auto-saving is enabled and that have been changed since the last time they were auto-saved. Normally, if any buffers are auto-saved, a message `Auto-saving...' is displayed in the echo area while auto-saving is going on. However, if no-message is non-nil, the message is inhibited. -- Function: delete-auto-save-file-if-necessary This function deletes the auto-save file for the current buffer if variable delete-auto-save-files is non-nil. It is called every time a buffer is saved. -- Variable: delete-auto-save-files This variable is used by the function delete-auto-save-file-if-necessary. If it is non-nil, Emacs will delete auto-save files when a true save is done (in the visited file). This saves on disk space and unclutters your directory. -- Function: rename-auto-save-file This function adjusts the current buffer's auto-save file name if the visited file name has changed. It also renames an existing auto-save file. If the visited file name has not changed, this function does nothing. ΓòÉΓòÉΓòÉ 26.3. Reverting ΓòÉΓòÉΓòÉ If you have made extensive changes to a file and then change your mind about them, you can get rid of them by reading in the previous version of the file with the revert-buffer command. See Reverting a Buffer. -- Command: revert-buffer &optional no-auto-save-offer-p noconfirm This command replaces the buffer text with the text of the visited file on disk. This action undoes all changes since the file was visited or saved. When the value of the no-auto-save-offer-p argument is nil, and the latest auto-save file is more recent than the visited file, revert-buffer asks the user whether to use that instead. Otherwise, it always uses the latest backup file. This argument is the numeric prefix argument when the function is called interactively. When the value of the noconfirm argument is non-nil, revert-buffer does not ask for confirmation for the reversion action. This means that the buffer contents are deleted and replaced by the text from the file on the disk, with no further opportunities for the user to prevent it. Since reverting works by deleting the entire text of the buffer and inserting the file contents, all the buffer's markers are relocated to point at the beginning of the buffer. This is not ``correct'', but then, there is no way to determine what would be correct. It is not possible to determine, from the text before and after, which characters after reversion correspond to which characters before. If the value of the revert-buffer-function variable is non-nil, it is called as a function with no arguments to do the work. -- Variable: revert-buffer-function The value of this variable is the function to use to revert this buffer; but if the value of this variable is nil, then the revert-buffer function carries out its default action. Modes such as Dired mode, in which the text being edited does not consist of a file's contents but can be regenerated in some other fashion, give this variable a buffer-local value that is a function to regenerate the contents. -- Command: recover-file filename This function visits filename, but gets the contents from its last auto-save file. This is useful after the system has crashed, to resume editing the same file without losing all the work done in the previous session. An error is signaled if there is no auto-save file for filename, or if filename is newer than its auto-save file. If filename does not exist, but its auto-save file does, then the auto-save file is read as usual. This last situation may occur if you visited a nonexistent file and never actually saved it. ΓòÉΓòÉΓòÉ 27. Buffers ΓòÉΓòÉΓòÉ A buffer is a Lisp object containing text to be edited. Buffers are used to hold the contents of files that are being visited; there may also be buffers which are not visiting files. While several buffers may exist at one time, exactly one buffer is designated the current buffer at any time. Most editing commands act on the contents of the current buffer. Each buffer, including the current buffer, may or may not be displayed in any windows. ΓòÉΓòÉΓòÉ 27.1. Buffer Basics ΓòÉΓòÉΓòÉ A buffer is a Lisp object containing text to be edited. Buffers are used to hold the contents of files that are being visited; there may also be buffers which are not visiting files. While several buffers may exist at one time, exactly one buffer is designated the current buffer at any time. Most editing commands act on the contents of the current buffer. Each buffer, including the current buffer, may or may not be displayed in any windows. Buffers in Emacs editing are objects which have distinct names and hold text that can be edited. Buffers appear to Lisp programs as a special data type. The contents of a buffer may be viewed as an extendible string; insertions and deletions may occur in any part of the buffer. See Text. A Lisp buffer object contains numerous pieces of information. Some of this information is directly accessible to the programmer through variables, while other information is only accessible through special-purpose functions. For example, the width of a tab character is directly accessible through a variable, while the value of point is accessible only through a primitive function. Buffer-specific information that is directly accessible is stored in buffer-local variable bindings, which are variable values that are effective only in a particular buffer. This feature allows each buffer to override the values of certain variables. Most major modes override variables such as fill-column or comment-column in this way. For more information about buffer-local variables and functions related to them, see Buffer-Local Variables. For functions and variables related to visiting files in buffers, see Visiting Files and Saving Buffers. For functions and variables related to the display of buffers in windows, see Buffers and Windows. -- Function: bufferp object This function returns t if object is a buffer, nil otherwise. ΓòÉΓòÉΓòÉ 27.2. Buffer Names ΓòÉΓòÉΓòÉ Each buffer has a unique name, which is a string. The buffer name may be used in place of the buffer object in many functions that operate on buffers. Buffers that are generally ephemeral and uninteresting to the user have names starting with a space, which prevents them from being listed by the list-buffers or buffer-menu commands. Many of the following functions accept either a buffer or a buffer name (a string) as an argument. Any argument called buffer-or-name is of this sort, and an error is signaled if it is neither a string nor a buffer. Any argument called buffer is required to be an actual buffer object, not a name. -- Function: buffer-name &optional buffer This function returns the name of buffer as a string. If buffer is not supplied, it defaults to the current buffer. If buffer-name returns nil, it means that buffer has been killed. See Killing Buffers. (buffer-name) => "buffers.texi" (setq foo (get-buffer "temp")) => #<buffer temp> (kill-buffer foo) => nil (buffer-name foo) => nil foo => #<killed buffer> -- Command: rename-buffer newname This function renames the current buffer to newname. An error is signaled if newname is not a string, or if there is already a buffer with that name. The function returns nil. One application of this command is to rename the `*shell*' buffer to some other name, thus making it possible to create a second shell buffer under the name `*shell*'. -- Function: get-buffer buffer-or-name This function returns the buffer specified by buffer-or-name. If buffer-or-name is a string and there is no buffer with that name, the value is nil. If buffer-or-name is a buffer, it is returned as given. (That is not very useful, so the argument is usually a name.) For example: (setq b (get-buffer "lewis")) => #<buffer lewis> (get-buffer b) => #<buffer lewis> (get-buffer "Frazzle-nots") => nil ΓòÉΓòÉΓòÉ 27.3. Buffer File Name ΓòÉΓòÉΓòÉ The buffer file name is the name of the file that is visited in that buffer. When a buffer is not visiting a file, its buffer file name is nil. Most of the time, the buffer name is the same as the nondirectory part of the buffer file name, but the buffer file name and the buffer name are distinct and can be set independently. See Visiting Files. -- Function: buffer-file-name &optional buffer This function returns the absolute file name of the file that buffer is visiting. If buffer is not visiting any file, buffer-file-name returns nil. If buffer is not supplied, it defaults to the current buffer. (buffer-file-name (other-buffer)) => "/usr/user/lewis/manual/files.texi" -- Variable: buffer-file-name This buffer-local variable contains the name of the file being visited in the current buffer, or nil if it is not visiting a file. buffer-file-name => "/usr/user/lewis/manual/buffers.texi" It is risky to change this variable's value without doing various other things. See the definition of set-visited-file-name in `files.el'; some of the things done there, such as changing the buffer name, are not necessary, but others are essential to avoid confusing Emacs. -- Function: get-file-buffer filename This function returns the buffer visiting file filename. If there is no such buffer, it returns nil. The argument filename, which must be a string, is expanded ( see File Name Expansion), then compared against the visited file names of all live buffers. (get-file-buffer "buffers.texi") => #<buffer buffers.texi> In unusual circumstances, there can be more than one buffer visiting the same file name. In such cases, this function returns the first such buffer in the buffer list. -- Command: set-visited-file-name filename If filename is a non-empty string, this function changes the name of the file visited in current buffer to filename. (If the buffer had no visited file, this gives it one.) The next time the buffer is saved it will go in the newly-specified file. The buffer is always marked as modified, since it does not (as far as Emacs knows) match the contents of filename, even if it matched the former visited file. If filename is nil or the empty string, that stands for ``no visited file''. In this case, set-visited-file-name marks the buffer as having no visited file. When set-visited-file-name is called interactively, it prompts for filename in the minibuffer. See also clear-visited-file-modtime and verify-visited-file-modtime in Buffer Modification. ΓòÉΓòÉΓòÉ 27.4. Buffer Modification ΓòÉΓòÉΓòÉ Emacs keeps a flag called the modified flag for each buffer, to record whether you have changed the text of the buffer. This flag is set to t whenever you alter the contents of the buffer, and cleared to nil when you save it. Thus, the flag shows whether there are unsaved changes. The flag value is normally shown in the mode line (see Mode Line Variables), and controls saving ( see Saving Buffers) and auto-saving (see Auto-Saving). Some Lisp programs set the flag explicitly. For example, the Lisp function set-visited-file-name sets the flag to t, because the text does not match the newly-visited file, even if it is unchanged from the file formerly visited. The functions that modify the contents of buffers are described in Text. -- Function: buffer-modified-p &optional buffer This function returns t if the buffer buffer has been modified since it was last read in from a file or saved, or nil otherwise. If buffer is not supplied, the current buffer is tested. -- Function: set-buffer-modified-p flag This function marks the current buffer as modified if flag is non-nil, or as unmodified if the flag is nil. Another effect of calling this function is to cause unconditional redisplay of the mode line for the current buffer. In fact, the standard way to force redisplay of the mode line is as follows: (set-buffer-modified-p (buffer-modified-p)) -- Command: not-modified This command marks the current buffer as unmodified, and not needing to be saved. Don't use this function in programs, since it prints a message; use set-buffer-modified-p (above) instead. ΓòÉΓòÉΓòÉ 27.5. Comparison of Modification Time ΓòÉΓòÉΓòÉ Suppose that you visit a file and make changes in its buffer, and meanwhile the file itself is changed on disk. At this point, saving the buffer would overwrite the changes in the file. Occasionally this may be what you want, but usually it would lose valuable information. Emacs therefore checks the file's modification time using the functions described below before saving the file. -- Function: verify-visited-file-modtime buffer This function compares Emacs's record of the modification time for the file that the buffer is visiting against the actual modification time of the file as recorded by the operating system. The two will be the same unless some other process has written the file since Emacs visited or saved it. The function returns t if the last actual modification time and Emacs's recorded modification time are the same, nil otherwise. -- Function: clear-visited-file-modtime This function clears out the record of the last modification time of the file being visited by the current buffer. As a result, the next attempt to save this buffer will not complain of a discrepancy in file modification times. This function is called in set-visited-file-name and other exceptional places where the usual test to avoid overwriting a changed file should not be done. -- Function: ask-user-about-supersession-threat fn This function is used to ask a user how to proceed after an attempt to modify an obsolete buffer. An obsolete buffer is an unmodified buffer for which the associated file on disk is newer than the last save-time of the buffer. This means some other program has probably altered the file. This function is called automatically by Emacs on the proper occasions. It exists so you can customize Emacs by redefining it. See the file `userlock.el' for the standard definition. Depending on the user's answer, the function may return normally, in which case the modification of the buffer proceeds, or it may signal a file-supersession error with data (fn), in which case the proposed buffer modification is not allowed. See also the file locking mechanism in File Locks. ΓòÉΓòÉΓòÉ 27.6. Read-Only Buffers ΓòÉΓòÉΓòÉ A buffer may be designated as read-only. This means that the buffer's contents may not be modified, although you may change your view of the contents by scrolling, narrowing, or widening, etc. Read-only buffers are used in two kinds of situations: o A buffer visiting a file is made read-only if the file is write-protected. Here, the purpose is to show the user that editing the buffer with the aim of saving it in the file may be futile or undesirable. The user who wants to change the buffer text despite this can do so after clearing the read-only flag with the function toggle-read-only. o Modes such as Dired and Rmail make buffers read-only when altering the contents with the usual editing commands is probably a mistake. The special commands of the mode in question bind buffer-read-only to nil (with let) around the places where they change the text. -- Variable: buffer-read-only This buffer-local variable specifies whether the buffer is read-only. The buffer is read-only if this variable is non-nil. -- Command: toggle-read-only This command changes whether the current buffer is read-only. It is intended for interactive use; don't use it in programs. At any given point in a program, you should know whether you want the read-only flag on or off; so you can set buffer-read-only explicitly to the proper value, t or nil. -- Function: barf-if-buffer-read-only This function signals a buffer-read-only error if the current buffer is read-only. See Interactive Call, for another way to signal an error if the current buffer is read-only. ΓòÉΓòÉΓòÉ 27.7. The Buffer List ΓòÉΓòÉΓòÉ The buffer list is a list of all buffers that have not been killed. The order of the buffers in the list is based primarily on how recently each buffer has been displayed in the selected window. Several functions, notably other-buffer, make use of this ordering. -- Function: buffer-list This function returns a list of all buffers, including those whose names begin with a space. The elements are actual buffers, not their names. (buffer-list) => (#<buffer buffers.texi> #<buffer *Minibuf-1*> #<buffer buffer.c> #<buffer *Help*> #<buffer TAGS>) ;; Note that the name of the minibuffer begins with a space! (mapcar (function buffer-name) (buffer-list)) => ("buffers.texi" " *Minibuf-1*" "buffer.c" "*Help*" "TAGS") This list is a copy of a list used inside Emacs; modifying it has no effect on the buffers. -- Function: other-buffer &optional buffer-or-name This function returns the first buffer in the buffer list other than buffer-or-name. Usually this is the buffer most recently shown in the selected window, aside from buffer-or-name. Buffers are moved to the front of the list when they are selected and to the end when they are buried. Buffers whose names start with a space are not even considered. If buffer-or-name is not supplied (or if it is not a buffer), then other-buffer returns the first buffer on the buffer list that is not visible in any window. If no suitable buffer exists, the buffer `*scratch*' is returned (and created, if necessary). -- Command: list-buffers &optional files-only This function displays a listing of the names of existing buffers. It clears the buffer `*Buffer List*', then inserts the listing into that buffer and displays it in a window. list-buffers is intended for interactive use, and is described fully in The GNU Emacs Manual. It returns nil. -- Command: bury-buffer &optional buffer-or-name This function puts buffer-or-name at the end of the buffer list without changing the order of any of the other buffers on the list. This buffer therefore becomes the least desirable candidate for other-buffer to return, and appears last in the list displayed by list-buffers. If buffer-or-name is the current buffer, then it is replaced in the selected window by the buffer chosen using other-buffer. If the buffer is displayed in a window other than the selected one, it remains there. If buffer-or-name is not supplied, it defaults to the current buffer. This is what happens in an interactive call. ΓòÉΓòÉΓòÉ 27.8. Creating Buffers ΓòÉΓòÉΓòÉ This section describes the two primitives for creating buffers. get-buffer-create creates a buffer if it finds no existing buffer; generate-new-buffer always creates a new buffer, and gives it a unique name. Two other functions to create buffers are with-output-to-temp-buffer (see Temporary Displays) and create-file-buffer (see Visiting Files). -- Function: get-buffer-create name This function returns a buffer named name. If such a buffer already exists, it is returned. If such a buffer does not exist, one is created and returned. The buffer does not become the current buffer---this function does not change which buffer is current. An error is signaled if name is not a string. (get-buffer-create "foo") => #<buffer foo> The major mode for the new buffer is chosen according to the value of default-major-mode. See Auto Major Mode. -- Function: generate-new-buffer name This function returns a newly created, empty buffer. If there is no buffer named name, then that is the name of the new buffer. If there is a buffer with that name, then suffixes of the form `<n>' are added to name, where n stands for successive integers starting with 2. New suffixes are tried until an unused name is found. An error is signaled if name is not a string. (generate-new-buffer "bar") => #<buffer bar> (generate-new-buffer "bar") => #<buffer bar<2>> (generate-new-buffer "bar") => #<buffer bar<3>> The major mode for the new buffer is chosen according to the value of default-major-mode. See Auto Major Mode. ΓòÉΓòÉΓòÉ 27.9. Killing Buffers ΓòÉΓòÉΓòÉ Killing a buffer makes its name unknown to Emacs and makes its space available for other use. The buffer object for the buffer which has been killed remains in existence as long as anything refers to it, but it is specially marked so that you cannot make it current or display it. Killed buffers retain their identity, however; two distinct buffers, when killed, remain distinct according to eq. The buffer-name of a killed buffer is nil. You can use this feature to test whether a buffer has been killed: (defun killed-buffer-p (buffer) "Return t if BUFFER is killed." (not (buffer-name buffer))) -- Command: kill-buffer buffer-or-name This function kills the buffer buffer-or-name, freeing all its memory for use as space for other buffers. (In Emacs version 18, the memory is not returned to the operating system.) It returns nil. Any processes that have this buffer as the process-buffer are sent the SIGHUP signal, which normally causes them to terminate. (The usual meaning of SIGHUP is that a dialup line has been disconnected.) See Deleting Processes. If the buffer is visiting a file when kill-buffer is called and the buffer has not been saved since it was last modified, the user is asked to confirm before the buffer is killed. This is done even if kill-buffer is not called interactively. To prevent the request for confirmation, clear the modified flag before calling kill-buffer. See Buffer Modification. (kill-buffer "foo.unchanged") => nil (kill-buffer "foo.changed") ---------- Buffer: Minibuffer ---------- Buffer foo.changed modified; kill anyway? (yes or no) yes ---------- Buffer: Minibuffer ---------- => nil ΓòÉΓòÉΓòÉ 27.10. The Current Buffer ΓòÉΓòÉΓòÉ There are in general many buffers in an Emacs session. At any time, one of them is designated as the current buffer. This is the buffer in which most editing takes place, because most of the primitives for examining or changing text in a buffer operate implicitly on the current buffer (see Text). Normally the buffer that is displayed on the screen in the selected window is the current buffer, but this is not always so: a Lisp program can designate any buffer as current temporarily in order to operate on its contents, without changing what is displayed on the screen. The way to designate a current buffer in a Lisp program is by calling set-buffer. The specified buffer remains current until a new one is designated. When an editing command returns to the editor command loop, the command loop designates the buffer displayed in the selected window as current, to prevent confusion: the buffer that the cursor is in, when Emacs reads a command, is the one to which the command will apply. (See Command Loop.) Therefore, set-buffer is not usable for switching visibly to a different buffer so that the user can edit it. For this, you must use the functions described in Displaying Buffers. However, Lisp functions that change to a different current buffer should not rely on the command loop to set it back afterwards. Editing commands written in Emacs Lisp can be called from other programs as well as from the command loop. It is convenient for the caller if the subroutine does not change which buffer is current (unless, of course, that is the subroutine's purpose). Therefore, you should normally use set-buffer within a save-excursion that will restore the current buffer when your program is done (see Excursions). Here is an example, the code for the command append-to-buffer (with the documentation string abridged): (defun append-to-buffer (buffer start end) "Append to specified buffer the text of the region..." (interactive "BAppend to buffer: \nr") (let ((oldbuf (current-buffer))) (save-excursion (set-buffer (get-buffer-create buffer)) (insert-buffer-substring oldbuf start end)))) In this function, a local variable is bound to the current buffer, and then save-excursion records the values of point, the mark, and the original buffer. Next, set-buffer makes another buffer current. Finally, insert-buffer-substring copies the string from the original current buffer to the new current buffer. If the buffer appended to happens to be displayed in some window, then the next redisplay will show how its text has changed. Otherwise, you will not see the change immediately on the screen. The buffer becomes current temporarily during the execution of the command, but this does not cause it to be displayed. -- Function: current-buffer This function returns the current buffer. (current-buffer) => #<buffer buffers.texi> -- Function: set-buffer buffer-or-name This function makes buffer-or-name the current buffer. However, it does not display the buffer in the currently selected window or in any other window. This means that the user cannot necessarily see the buffer, but Lisp programs can in any case work on it. This function returns the buffer identified by buffer-or-name. An error is signaled if buffer-or-name does not identify an existing buffer. ΓòÉΓòÉΓòÉ 28. Windows ΓòÉΓòÉΓòÉ This chapter describes most of the functions and variables related to Emacs windows. See Emacs Display, for information on how text is displayed in windows. ΓòÉΓòÉΓòÉ 28.1. Basic Concepts of Emacs Windows ΓòÉΓòÉΓòÉ A window is the physical area of the screen in which a buffer is displayed. The term is also used to refer to a Lisp object which represents that screen area in Emacs Lisp. It should be clear from the context which is meant. There is always at least one window displayed on the screen, and there is exactly one window that we call the selected window. The cursor is in the selected window. The selected window's buffer is usually the current buffer (except when set-buffer has been used.) See Current Buffer. For all intents, a window only exists while it is displayed on the terminal. Once removed from the display, the window is effectively deleted and should not be used, even though there may still be references to it from other Lisp objects. (See Deleting Windows.) Each window has the following attributes: o window height o window width o window edges with respect to the screen o the buffer it displays o position within the buffer at the upper left of the window o the amount of horizontal scrolling, in columns o point o the mark o how recently the window was selected Applications use multiple windows for a variety of reasons, but most often to give different views of the same information. In Rmail, for example, you can move through a summary buffer in one window while the other window shows messages one at a time as they are reached. Use of the word ``window'' to refer to a view of a buffer was established long ago in Emacs. The metaphor was inspired by how you look out a house window---at part (or sometimes all) of an overall view. You see part (or sometimes all) of a buffer through an Emacs window. In Emacs, each window may look on a different view, like different windows of a house. The term ``window'' as used in this manual means something different from the term as used in a window system like X Windows. In this manual, the term ``window'' refers to the nonoverlapping subdivisions of the Emacs display. If Emacs is displaying on a window system, the Emacs display may itself be one X window among many on the screen. But Emacs version 18 knows nothing of this. For those familiar with windowing systems, Emacs's windows are rectangles tiled onto the rectangle of the screen, and every portion of the screen is part of some window, except (sometimes) the minibuffer area. This limitation helps avoid wasting the historically scarce resource of screen space. It also works well with character-only terminals. Because of the way in which Emacs creates new windows and resizes them, you can't create every conceivable tiling on an Emacs screen. See Splitting Windows. Also, see Size of Window. See Emacs Display, for information on how the contents of the window's buffer are displayed in the window. -- Function: windowp object This function returns t if object is a window. ΓòÉΓòÉΓòÉ 28.2. Splitting Windows ΓòÉΓòÉΓòÉ The functions described here are the primitives used to split a window into two windows. Two higher level functions sometimes split a window, but not always: pop-to-buffer and display-buffer (see Displaying Buffers). The functions described here do not accept a buffer as an argument. They let the two ``halves'' of the split window display the same buffer previously visible in the window that was split. -- Function: one-window-p &optional no-mini This function returns non-nil if there is only one window. The argument no-mini, if non-nil, means don't count the minibuffer even if it is active; otherwise, the minibuffer window is included, if active, in the total number of windows which is compared against one. -- Command: split-window &optional window size horizontal This function splits window into two windows. The original window window remains the selected window, but occupies only part of its former screen area. The rest is occupied by a newly created window which is returned as the value of this function. If horizontal is non-nil, then window splits side by side, keeping the leftmost size columns and giving the rest of the columns to the new window. Otherwise, it splits into halves one above the other, keeping the upper size lines and giving the rest of the lines to the new window. The original window is therefore the right-hand or upper of the two, and the new window is the left-hand or lower. If window is omitted or nil, then the selected window is split. If size is omitted or nil, then window is divided evenly into two parts. (If there is an odd line, it is allocated to the new window.) When split-window is called interactively, all its arguments are nil. The following example starts with one window on a screen that is 50 lines high by 80 columns wide; then the window is split. (setq w (selected-window)) => #<window 8 on windows.texi> (window-edges) ; Edges in order: left--top--right--bottom => (0 0 80 50) (setq w2 (split-window w 15)) ; Returns window created => #<window 28 on windows.texi> (window-edges w2) => (0 15 80 50) ; Bottom window; top is line 15 (window-edges w) => (0 0 80 15) ; Top window The screen looks like this: __________ | | line 0 | w | |__________| | | line 15 | w2 | |__________| line 50 column 0 column 80 Next, the top window is split horizontally: (setq w3 (split-window w 35 t)) => #<window 32 on windows.texi> (window-edges w3) => (35 0 80 15) ; Left edge at column 35 (window-edges w) => (0 0 35 15) ; Right edge at column 35 (window-edges w2) => (0 15 80 50) ; Bottom window unchanged Now, the screen looks like this: column 35 __________ | | | line 0 | w | w3 | |___|______| | | line 15 | w2 | |__________| line 50 column 0 column 80 -- Command: split-window-vertically size This function splits the selected window into two windows, one above the other, leaving the selected window with size lines. This function is simply an interface to split-windows. Here is the complete function definition for it: (defun split-window-vertically (&optional arg) "Split selected window into two windows, one above the other..." (interactive "P") (split-window nil (and arg (prefix-numeric-value arg)))) -- Command: split-window-horizontally size This function splits the selected window into two windows side-by-side, leaving the selected window with size columns. This function is simply an interface to split-windows. Here is the complete definition for split-window-horizontally (except for part of the documentation string): (defun split-window-horizontally (&optional arg) "Split selected window into two windows side by side..." (interactive "P") (split-window nil (and arg (prefix-numeric-value arg)) t)) ΓòÉΓòÉΓòÉ 28.3. Deleting Windows ΓòÉΓòÉΓòÉ A deleted window no longer appears on the screen. In Emacs version 18, the space it took up on the screen is divided proportionally among all siblings; in version 19, the space is given to one adjacent sibling. -- Command: delete-window &optional window This function removes window from the display. If window is omitted, then the selected window is deleted. An error is signaled if there is only one window when delete-window is called. *Warning:* erroneous information or fatal errors may result from using a deleted window. Use (window-point window) to test whether a window has been deleted; it yields nil for a deleted window. This function returns nil. When delete-window is called interactively, window defaults to the selected window. -- Command: delete-other-windows &optional window This function makes window the only window on the screen by deleting all the other windows. If window is omitted or nil, then the selected window is used by default. The result is nil. -- Command: delete-windows-on buffer This function deletes all windows showing buffer. If there are no windows showing buffer, then this function does nothing. If all windows are showing buffer (including the case where there is only one window), then the screen reverts to having a single window showing the buffer chosen by other-buffer. See The Buffer List. If there are several windows showing different buffers, then those showing buffer are removed, and the others are expanded to fill the void. The result is nil. ΓòÉΓòÉΓòÉ 28.4. Selecting Windows ΓòÉΓòÉΓòÉ When a window is selected, the buffer in the window becomes the current buffer, and the cursor will appear in it. -- Function: selected-window This function returns the selected window. This is the window in which the cursor appears and to which many commands apply. -- Function: select-window window This function makes window the selected window. The cursor then appears in window (on redisplay). The buffer being displayed in window is immediately designated the current buffer. The return value is window. (setq w (next-window)) (select-window w) => #<window 65 on windows.texi> The following functions choose one of the windows on the screen, offering various criteria for the choice. -- Function: get-lru-window This function returns the window least recently ``used'' (that is, selected). The selected window is always the most recently used window. The selected window can be the least recently used window if it is the only window. A newly created window becomes the least recently used window until it is selected. The minibuffer window is not considered a candidate. -- Function: get-largest-window This function returns the window with the largest area (height times width). If there are no side-by-side windows, then this is the window with the most lines. The minibuffer window is not considered a candidate. If there are two windows of the same size, then the function returns the window which is first in the cyclic ordering of windows (see following section), starting from the selected window. ΓòÉΓòÉΓòÉ 28.5. Cycling Ordering of Windows ΓòÉΓòÉΓòÉ When you use the command C-x o (other-window) to select the next window, it moves through all the windows on the screen in a specific cyclic order. For any given configuration of windows, this order never varies. It is called the cyclic ordering of windows. This ordering generally goes from top to bottom, and from left to right. But it may go down first or go right first, depending on the order in which the screen, or the windows within the screen, were split. If the screen was first split vertically (into windows one above each other), and then the subwindows were split horizontally, then the ordering is left to right in the top, and then left to right in the next lower part of the screen, and so on. If the screen was first split horizontally, the ordering is top to bottom in the left part, and so on. In general, within each set of siblings at any level in the window tree, the order is left to right, or top to bottom. -- Function: next-window window &optional minibuf This function returns the window following window in the cyclic ordering of windows. This is the window which C-x o would select if done when window is selected. If window is the only window visible, then this function returns window. The value of the argument minibuf determines whether the minibuffer is included in the window order. Normally, when minibuf is nil, the minibuffer is included if it is currently active; this is the behavior of C-x o. If minibuf is t, then the cyclic ordering includes the minibuffer window even if it is not active. If minibuf is neither t nor nil, then the minibuffer window is not included even if it is active. (The minibuffer window is active while the minibuffer is in use. See Minibuffers.) This example shows two windows, which both happen to be displaying the same buffer: (selected-window) => #<window 56 on windows.texi> (next-window (selected-window)) => #<window 52 on windows.texi> (next-window (next-window (selected-window))) => #<window 56 on windows.texi> -- Function: previous-window window This function returns the window preceding window in the cyclic ordering of windows. -- Command: other-window count This function selects the countth next window in the cyclic order. If count is negative, then it selects the -countth preceding window. It returns nil. In an interactive call, count is the numeric prefix argument. ΓòÉΓòÉΓòÉ 28.6. Buffers and Windows ΓòÉΓòÉΓòÉ This section describes low-level functions to examine windows or to show buffers in windows in a precisely controlled fashion. See Displaying Buffers, for related functions that find a window to use and specify a buffer for it. The functions described there are easier to use than these, but they employ heuristics in choosing or creating a window; use these functions when you need complete control. -- Function: set-window-buffer window buffer-or-name This function makes window display buffer-or-name as its contents. It returns nil. (set-window-buffer (selected-window) "foo") => nil -- Function: window-buffer &optional window This function returns the buffer that window is displaying. If window is omitted, then this function returns the buffer for the selected window. (window-buffer) => #<buffer windows.texi> -- Function: get-buffer-window buffer-or-name This function returns a window currently displaying buffer-or-name, or nil if there is none. If there are several such windows, then the function returns the first one in the cyclic ordering of windows, starting from the selected window. See Cyclic Window Ordering. -- Command: replace-buffer-in-windows buffer This function replaces buffer with some other buffer in all windows displaying it. The other buffer used is chosen with other-buffer. In the usual applications of this function, you don't care which other buffer is used; you just want to make sure that buffer is off the screen. This function returns nil. ΓòÉΓòÉΓòÉ 28.7. Displaying Buffers in Windows ΓòÉΓòÉΓòÉ In this section we describe convenient functions that choose a window automatically and use it to display a specified buffer. These functions can also split an existing window in certain circumstances. We also describe variables that parameterize the heuristics used for choosing a window. See Buffers and Windows, for low-level functions that give you more precise control. Do not use the functions in this section in order to make a buffer current so that a Lisp program can access or modify it; they are too drastic for that purpose, since they change the display of buffers on the screen, which is gratuitous and will surprise the user. Instead, use set-buffer (see Current Buffer) and save-excursion (see Excursions), which designate buffers as current for programmed access without affecting the display of buffers in windows. -- Command: switch-to-buffer buffer-or-name &optional norecord This function makes buffer-or-name the current buffer, and also displays the buffer in the selected window. This means that a human can see the buffer and subsequent keyboard commands will apply to it. Contrast this with set-buffer, which makes buffer-or-name the current buffer but does not display it in the selected window. See Current Buffer. If buffer-or-name does not identify an existing buffer, then a new buffer by that name is created. Normally the specified buffer is put at the front of the buffer list. This affects the operation of other-buffer. However, if norecord is non-nil, this is not done. See The Buffer List. The switch-to-buffer function is often used interactively, as the binding of C-x b. It is also used frequently in programs. It always returns nil. -- Command: switch-to-buffer-other-window buffer-or-name This function makes buffer-or-name the current buffer and displays it in a window not currently selected. It then selects that window. The handling of the buffer is the same as in switch-to-buffer. The previously selected window is absolutely never used to display the buffer. If it is the only window, then it is split to make a distinct window for this purpose. If the selected window is already displaying the buffer, then it continues to do so, but another window is nonetheless found to display it in as well. -- Function: pop-to-buffer buffer-or-name &optional other-window This function makes buffer-or-name the current buffer and switches to it in some window, preferably not the window previously selected. The ``popped-to'' window becomes the selected window. If the variable pop-up-windows is non-nil, windows may be split to create a new window that is different from the original window. If other-window is non-nil, pop-to-buffer finds or creates another window even if buffer-or-name is already visible in the selected window. Thus buffer-or-name could end up displayed in two windows. On the other hand, if buffer-or-name is already displayed in the selected window and other-window is nil, then the selected window is considered sufficient display for buffer-or-name, so that nothing needs to be done. If buffer-or-name is a string that does not name an existing buffer, a buffer by that name is created. An example use of this function is found at the end of Filter Functions. -- Function: display-buffer buffer-or-name &optional not-this-window This function makes buffer-or-name appear in some window, like pop-to-buffer, but it does not select that window and does not make the buffer current. The identity of the selected window is unaltered by this function. If not-this-window is non-nil, it means that the specified buffer should be displayed in a window other than the selected one, even if it is already on display in the selected window. This can cause the buffer to appear in two windows at once. Otherwise, if buffer-or-name is already being displayed in any window, that is good enough, so this function does nothing. If the variable pop-up-windows is non-nil, windows can be split to display the buffer. If there are multiple windows, display-buffer will split the largest window if it has more than the number of lines specified by the variable split-height-threshold. display-buffer returns the window chosen to display buffer-or-name. -- User Option: pop-up-windows This variable controls whether display-buffer makes new windows. If it is non-nil and there is only one window on the screen, then that window is split. If it is nil, then display-buffer does not split the single window, but rather replaces its buffer. This variable also affects pop-to-buffer, which uses display-buffer as a subroutine. -- User Option: split-height-threshold This variable determines when display-buffer may split a window, if there are multiple windows. display-buffer splits the largest window if it has at least this many lines. If there is only one window, it is split regardless of this value, provided pop-up-windows is non-nil. ΓòÉΓòÉΓòÉ 28.8. Window Point ΓòÉΓòÉΓòÉ Each window has its own value of point, independent of the value of point in other windows displaying the same buffer. This makes it useful to have multiple windows showing one buffer. o The window point is established when a window is first created; it is initialized from the buffer's point, or from the window point of another window opened on the buffer if such a window exists. o Selecting a window sets the value of point in its buffer to the window's value of point. Conversely, deselecting a window copies the buffer's value of point into the window. Thus, when you switch between windows that display a given buffer, the point value for the selected window is in effect in the buffer, while the point values for the other windows are stored in those windows. o As long as the selected window displays the current buffer, the window's point and the buffer's point always move together; they remain equal. o See Positions, for more details on positions. As far as the user is concerned, point is where the cursor is, and when the user switches to another buffer, the cursor jumps to the position of point in that buffer. -- Function: window-point window This function returns the current position of point in window. For a nonselected window, this is the value point would have (in that window's buffer) if that window were selected. When window is the selected window and its buffer is also the current buffer, the value returned is the same as point in that buffer. Strictly speaking, it would be more correct to return the ``top-level'' value of point, outside of any save-excursion forms. But that value is hard to find. -- Function: set-window-point window position This function positions point in window at position position in window's buffer. ΓòÉΓòÉΓòÉ 28.9. The Display-Start Position ΓòÉΓòÉΓòÉ Each window contains a marker used to keep track of a buffer position which specifies where in the buffer display should start. This position is called the display-start position of the window. The character after this position is the one that appears at the upper left corner of the window. It is usually, but not inevitably, at the beginning of a text line. -- Function: window-start &optional window This function returns the display-start position of window window. If window is nil, the selected window is used. (window-start) => 7058 For a more complicated example of use, see the description of count-lines in Text Lines. -- Function: set-window-start window position &optional noforce This function sets the display-start position of window to position in window's buffer. The display routines insist that the position of point be visible when a buffer is displayed. Normally, they change the display-start position (that is, scroll the window) whenever necessary to make point visible. However, if you specify the start position with this function with nil for noforce, it means you want display to start at position even if that would put the location of point off the screen. What the display routines do in this case is move point instead, to the left margin on the middle line in the window. For example, if point is 1 and you attempt to set the start of the window to 2, then the position of point would be ``above'' the top of the window. The display routines would automatically move point if it is still 1 when redisplay occurs. Here is an example: ;; Here is what `foo' looks like before executing ;; the set-window-start expression. ---------- Buffer: foo ---------- -!-This is the contents of buffer foo. 2 3 4 5 6 ---------- Buffer: foo ---------- (set-window-start (selected-window) (1+ (window-start))) ;; Here is what `foo' looks like after executing ;; the set-window-start expression. ---------- Buffer: foo ---------- his is the contents of buffer foo. 2 3 -!-4 5 6 ---------- Buffer: foo ---------- => 2 However, when noforce is non-nil, set-window-start does nothing if the specified start position would make point invisible. This function returns position, regardless of whether the noforce option caused that position to be overruled. -- Function: pos-visible-in-window-p &optional position window This function returns t if position is within the range of text currently visible on the screen in window. It returns nil if position is scrolled vertically out of view. The argument position defaults to the current position of point; window, to the selected window. Here is an example: (or (pos-visible-in-window-p (point) (selected-window)) (recenter 0)) The pos-visible-in-window-p function considers only vertical scrolling. It returns t if position is out of view only because window has been scrolled horizontally. See Horizontal Scrolling. ΓòÉΓòÉΓòÉ 28.10. Vertical Scrolling ΓòÉΓòÉΓòÉ Vertical scrolling means moving the text up or down in a window. It works by changing the value of the window's display-start location. It may also change the value of window-point to keep it on the screen. In the commands scroll-up and scroll-down, the directions ``up'' and ``down'' refer to the motion of the text in the buffer at which you are looking through the window. Imagine that the text is written on a long roll of paper and that the scrolling commands move the paper up and down. Thus, if you are looking at text in the middle of a buffer and repeatedly call scroll-down, you will eventually see the beginning of the buffer. Some people have urged that the opposite convention be used: they imagine that the window moves over text that remains in place. Then ``down'' commands would take you to the end of the buffer. This view is more consistent with the actual relationship between windows and the text in the buffer, but it is less like what the user sees. The position of a window on the terminal does not move, and short scrolling commands clearly move the text up or down on the screen. We have chosen names that fit the user's point of view. The scrolling functions (aside from scroll-other-window) will have unpredictable results if the current buffer is different from the buffer that is displayed in the selected window. See Current Buffer. -- Command: scroll-up &optional count This function scrolls the text in the selected window upward count lines. If count is negative, scrolling is actually downward. If count is nil (or omitted), then the length of the scroll is next-screen-context-lines lines less than the usable height of the window (not counting its mode line). scroll-up returns nil. -- Command: scroll-down &optional count This function scrolls the text in the selected window downward count lines. If count is negative, scrolling is actually upward. If count is omitted or nil, then the length of the scroll is next-screen-context-lines lines less than the usable height of the window. scroll-down returns nil. -- Command: scroll-other-window &optional count This function scrolls the text in another window upward count lines. Negative values of count, or nil, are handled as in scroll-up. The window that is scrolled is normally the one following the selected window in the cyclic ordering of windows---the window that next-window would return. See Cyclic Window Ordering. If the selected window is the minibuffer, the next window is normally the one at the top left corner. However, you can specify the window to scroll by binding the variable minibuffer-scroll-window. This variable has no effect when any other window is selected. See Minibuffer Misc. When the minibuffer is active, it is the next window if the selected window is the one at the bottom right corner. In this case, scroll-other-window will attempt to scroll the minibuffer. If the minibuffer contains just one line, that line will be redisplayed after the echo area momentarily displays the message ``Beginning of buffer''. -- User Option: scroll-step This variable controls how scrolling is done automatically when point moves off the screen. If the value is zero, then the text is scrolled so that point is centered vertically in the window. If the value is a positive integer n, then if it is possible to bring point back on screen by scrolling n lines in either direction, that is done; otherwise, point is centered vertically as usual. The default value is zero. -- User Option: next-screen-context-lines The value of this variable is the number of lines of continuity to retain when scrolling by full screens. For example, when scroll-up executes, this many lines that were visible at the bottom of the window move to the top of the window. The default value is 2. -- Command: recenter &optional count This function scrolls the selected window to put the text where point is located at a specified screen position. If count is a nonnegative number, it puts the line containing point count lines down from the top of the window. If count is a negative number, then it counts upward from the bottom of the window, so that -1 stands for the last usable line in the window. If count is a non-nil list, then it stands for the line in the middle of the window. If count is nil, then it puts the line containing point in the middle of the window, then clears and redisplays the entire screen. When recenter is called interactively, Emacs sets count to the raw prefix argument. Thus, typing C-u as the prefix sets the count to a non-nil list, while typing C-u 4 sets count to 4, which positions the current line four lines from the top. Typing C-u 0 C-l positions the current line at the top of the window. This action is so handy that some people bind the command to a function key. For example, (defun line-to-top-of-window () "Scroll the selected window up so current line moves to the top. Replaces three keystroke sequence C-u 0 C-l." (interactive) (recenter 0)) (global-set-key "\C-cl" 'line-to-top-of-window) ΓòÉΓòÉΓòÉ 28.11. Horizontal Scrolling ΓòÉΓòÉΓòÉ Because we read English first from top to bottom and second from left to right, horizontal scrolling is not like vertical scrolling. Vertical scrolling involves selection of a contiguous portion of text to display. Horizontal scrolling causes part of each line to go off screen. The amount of horizontal scrolling is therefore specified as a number of columns rather than as a position in the buffer. It has nothing to do with the display-start position returned by window-start. Usually, no horizontal scrolling is in effect; then the leftmost column is at the left edge of the window. In this state, scrolling to the right is meaningless, since there is no data to the left of the screen to be revealed by it, so it is not allowed. Scrolling to the left is allowed; it causes the first columns of text to go off the edge of the window and can reveal additional columns on the right that were truncated before. Once a window has a nonzero amount of leftward horizontal scrolling, you can scroll it back to the right, but only so far as to reduce the net horizontal scroll to zero. There is no limit to how far left you can scroll, but eventually all the text will disappear off the left edge. -- Command: scroll-left count This function scrolls the selected window count columns to the left (or to the right if count is negative). The return value is the total amount of leftward horizontal scrolling in effect after the change---just like the value returned by window-hscroll. -- Command: scroll-right count This function scrolls the selected window count columns to the right (or to the left if count is negative). The return value is the total amount of leftward horizontal scrolling in effect after the change---just like the value returned by window-hscroll. Once you scroll a window as far right as it can go, back to its normal position where the total leftward scrolling is zero, attempts to scroll any farther have no effect. -- Function: window-hscroll &optional window This function returns the total leftward horizontal scrolling of window---the number of columns by which the text in window is scrolled left past the left margin. The value is never negative. It is zero when no horizontal scrolling has been done in window (which is usually the case). If window is nil, the selected window is used. (window-hscroll) => 0 (scroll-left 5) => 5 (window-hscroll) => 5 -- Function: set-window-hscroll window columns This function sets the number of columns from the left margin that window is scrolled to the value of columns. The argument columns should be zero or positive; if not, it is taken as zero. The value returned is columns. (set-window-hscroll (selected-window) 10) => 10 Here is how you can determine whether a given position position is off the screen due to horizontal scrolling: (save-excursion (goto-char position) (and (>= (- (current-column) (window-hscroll window)) 0) (< (- (current-column) (window-hscroll window)) (window-width window)))) ΓòÉΓòÉΓòÉ 28.12. The Size of a Window ΓòÉΓòÉΓòÉ An Emacs window is rectangular, and its size information consists of the height (the number of lines) and the width (the number of character positions in each line). The mode line is included in the height. For a window that does not abut the right hand edge of the screen, the column of `|' characters that separates it from the window on the right is included in the width. The following three functions return size information about a window: -- Function: window-height &optional window This function returns the number of lines in window, including its mode line. If window fills the entire screen, this is one less than the value of (screen-height) (since the last line is always reserved for the minibuffer). If window is nil, the function uses the selected window. (window-height) => 23 (split-window-vertically) => #<window 4 on windows.texi> (window-height) => 11 -- Function: window-width &optional window This function returns the number of columns in window. If window fills the entire screen, this is the same as the value of (screen-width). If window is nil, the function uses the selected window. (window-width) => 80 -- Function: window-edges &optional window This function returns a list of the edge coordinates of window. If window is nil, the selected window is used. The order of the list is (left top right bottom), all elements relative to 0, 0 at the top left corner of the screen. The element right of the value is one more than the rightmost column used by window, and bottom is one more than the bottommost row used by window and its mode-line. Here is the result obtained on a typical 24-line terminal with just one window: (window-edges (selected-window)) => (0 0 80 23) If window is at the upper left corner of the screen, right and bottom are the same as the values returned by (window-width) and (window-height) respectively, and top and bottom are zero. For example, the edges of the following window are `0 0 5 8'. Assuming that the screen has more than 8 columns, the last column of the window (column 7) holds a border rather than text. The last row (row 4) holds the mode line, shown here with `xxxxxxxxx'. 0 _______ 0 | | | | | | | | xxxxxxxxx 4 7 When there are side-by-side windows, any window not at the right edge of the screen has a border in its last column. This border counts as one column in the width of the window. A window never includes a border on its left, since the border there belongs to the window to the left. In the following example, let's imagine that the screen is 7 columns wide. Then the edges of the left window are `0 0 4 3' and the edges of the right window are `4 0 7 3'. ___ ___ | | | | | | xxxxxxxxx 0 34 7 ΓòÉΓòÉΓòÉ 28.13. Changing the Size of a Window ΓòÉΓòÉΓòÉ The window size functions fall into two classes: high-level commands that change the size of windows and low-level functions that access window size. Emacs does not permit overlapping windows or gaps between windows, so resizing one window affects other windows. -- Command: enlarge-window size &optional horizontal This function makes the selected window size lines bigger, stealing lines from neighboring windows. It generally tries to steal equal numbers of lines from the other windows. If a window from which lines are stolen shrinks below window-min-height, then that window disappears. If horizontal is non-nil, then this function makes window wider by size columns, stealing columns as it does lines. If a window from which lines are stolen shrinks below window-min-width, then that window disappears. If the screen is smaller than size lines (or columns), then the function makes the window occupy the entire height (or width) of the screen. If size is negative, this function shrinks the window by -size lines. If it becomes shorter than window-min-height, it disappears. enlarge-window returns nil. -- Command: enlarge-window-horizontally columns This function makes the selected window columns wider. It could be defined as follows: (defun enlarge-window-horizontally (columns) (enlarge-window columns t)) -- Command: shrink-window size &optional horizontal This function is like enlarge-window but negates the argument size, making the selected window smaller by giving lines (or columns) to the other windows. If the window shrinks below window-min-height or window-min-width, then it disappears. If size is negative, the window is enlarged by -size lines. -- Command: shrink-window-horizontally columns This function makes the selected window columns narrower. It could be defined as follows: (defun shrink-window-horizontally (columns) (shrink-window columns t)) The following two variables constrain the window size changing functions to a minimum height and width. -- User Option: window-min-height The value of this variable determines how short a window may become before it disappears. A window disappears when it becomes smaller than window-min-height, and no window may be created that is smaller. The absolute minimum height is two (allowing one line for the mode line, and one line for the buffer display). Actions which change window sizes reset this variable to two if it is less than two. The default value is 4. -- User Option: window-min-width The value of this variable determines how narrow a window may become before it disappears. A window disappears when it becomes narrower than window-min-width, and no window may be created that is narrower. The absolute minimum width is one; any value below that is ignored. The default value is 10. ΓòÉΓòÉΓòÉ 28.14. Window Configurations ΓòÉΓòÉΓòÉ Window configurations record entire screen layouts---all windows, their sizes, which buffers they contain, what part of each buffer is displayed, and the values of point and the mark. You can bring back an entire previous screen layout by restoring a window configuration that you had previously saved. -- Function: current-window-configuration This function returns a new object representing Emacs's current window configuration, namely the number of windows, their sizes and current buffers, which window is the selected window, and for each window the displayed buffer, the display-start position, and the positions of point and the mark. An exception is made for point in the current buffer, whose value is not saved. -- Function: set-window-configuration configuration This function restores the configuration of Emacs's windows and buffers to the state specified by configuration. The argument configuration must be a value that was previously returned by current-window-configuration. Here is a way of using this function to get the same effect as save-window-excursion: (let ((config (current-window-configuration))) (unwind-protect (progn (split-window-vertically nil) ...) (set-window-configuration config))) -- Special Form: save-window-excursion forms... This special form executes forms in sequence, preserving window sizes and contents, including the value of point and the portion of the buffer which is visible. However, it does not restore the value of point in the current buffer; use save-excursion for that. The return value is the value of the final form in forms. For example: (split-window) => #<window 25 on control.texi> (setq w (selected-window)) => #<window 19 on control.texi> (save-window-excursion (delete-other-windows w) (switch-to-buffer "foo") 'do-something) => do-something ;; The screen is now split again. Primitives to look inside of window configurations would make sense, but none are implemented. It is not clear they are useful enough to be worth implementing. ΓòÉΓòÉΓòÉ 29. Positions ΓòÉΓòÉΓòÉ A position is the index of a character in the text of buffer. More precisely, a position identifies the place between two characters (or before the first character, or after the last character), so we can speak of the character before or after a given position. However, the character after a position is often said to be ``at'' that position. Positions are usually represented as integers starting from 1, but can also be represented as markers---special objects which relocate automatically when text is inserted or deleted so they stay with the surrounding characters. See Markers. ΓòÉΓòÉΓòÉ 29.1. Point ΓòÉΓòÉΓòÉ Point is a special buffer position used by many editing commands, including the self-inserting typed characters and text insertion functions. Other commands move point through the text to allow editing and insertion at different places. Like other positions, point designates a place between two characters (or before the first character, or after the last character), rather than a particular character. Many terminals display the cursor over the character that immediately follows point; on such terminals, point is actually before the character on which the cursor sits. The value of point is a number between 1 and the buffer size plus 1. If narrowing is in effect (see Narrowing), then point is constrained to fall within the accessible portion of the buffer (possibly at one end of it). Each buffer has its own value of point, which is independent of the value of point in other buffers. Each window also has a value of point, which is independent of the value of point in other windows on the same buffer. This is why point can have different values in various windows that display the same buffer. When a buffer appears in only one window, the buffer's point and the window's point normally have the same value, so the distinction is rarely important. See Window Point, for more details. -- Function: point This function returns the position of point in the current buffer, as an integer. (point) => 175 -- Function: point-min This function returns the minimum accessible value of point in the current buffer. This is 1, unless narrowing is in effect, in which case it is the position of the start of the region that you narrowed to. (See Narrowing.) -- Function: point-max This function returns the maximum accessible value of point in the current buffer. This is (1+ (buffer-size)), unless narrowing is in effect, in which case it is the position of the end of the region that you narrowed to. (See Narrowing). -- Function: buffer-end flag This function returns (point-min) if flag is less than 1, (point-max) otherwise. The argument flag must be a number. -- Function: buffer-size This function returns the total number of characters in the current buffer. In the absence of any narrowing (see Narrowing), point-max returns a value one larger than this. (buffer-size) => 35 (point-max) => 36 -- Variable: buffer-saved-size The value of this buffer-local variable is the former length of the current buffer, as of the last time it was read in, saved or auto-saved. ΓòÉΓòÉΓòÉ 29.2. Motion ΓòÉΓòÉΓòÉ Motion functions change the value of point, either relative to the current value of point, relative to the beginning or end of the buffer, or relative to the edges of the selected window. ΓòÉΓòÉΓòÉ 29.2.1. Motion by Characters ΓòÉΓòÉΓòÉ These functions move point based on a count of characters. goto-char is a fundamental primitive because it is the way to move point to a specified position. -- Command: goto-char position This function sets point in the current buffer to the value position. If position is less than 1, then point is set to the beginning of the buffer. If it is greater than the length of the buffer, then point is set to the end of the buffer. If narrowing is in effect, then the position is still measured from the beginning of the buffer, but point cannot be moved outside of the accessible portion. Therefore, if position is too small, point is set to the beginning of the accessible portion of the text; if position is too large, point is set to the end. When this function is called interactively, position is the numeric prefix argument, if provided; otherwise it is read from the minibuffer. goto-char returns position. -- Command: forward-char &optional count This function moves point forward, towards the end of the buffer, count characters (or backward, towards the beginning of the buffer, if count is negative). If the function attempts to move point past the beginning or end of the buffer (or the limits of the accessible portion, when narrowing is in effect), an error is signaled with error code beginning-of-buffer or end-of-buffer. In an interactive call, count is the numeric prefix argument. -- Command: backward-char &optional count This function moves point backward, towards the beginning of the buffer, count characters (or forward, towards the end of the buffer, if count is negative). If the function attempts to move point past the beginning or end of the buffer (or the limits of the accessible portion, when narrowing is in effect), an error is signaled with error code beginning-of-buffer or end-of-buffer. In an interactive call, count is the numeric prefix argument. ΓòÉΓòÉΓòÉ 29.2.2. Motion by Words ΓòÉΓòÉΓòÉ These functions for parsing words use the syntax table to decide whether a given character is part of a word. See Syntax Tables. -- Command: forward-word count This function moves point forward count words (or backward if count is negative). Normally it returns t. If this motion encounters the beginning or end of the buffer, or the limits of the accessible portion when narrowing is in effect, point stops there and the value is nil. In an interactive call, count is set to the numeric prefix argument. -- Command: backward-word count This function just like forward-word, except that it moves backward until encountering the front of a word, rather than forward. In an interactive call, count is set to the numeric prefix argument. This function is rarely used in programs, as it is more efficient to call forward-word with negative argument. ΓòÉΓòÉΓòÉ 29.2.3. Motion to an End of the Buffer ΓòÉΓòÉΓòÉ To move point to the beginning of the buffer, write: (goto-char (point-min)) Likewise, to move to the end of the buffer, use: (goto-char (point-max)) Here are two commands which users use to do these things. They are documented here to warn you not to use them in Lisp programs, because they set the mark and display messages in the echo area. -- Command: beginning-of-buffer &optional n This function moves point to the beginning of the buffer (or the limits of the accessible portion, when narrowing is in effect), setting the mark at the previous position. If n is non-nil, then it puts point n tenths of the way from the beginning of the buffer. In an interactive call, n is the numeric prefix argument, if provided; otherwise n defaults to nil. Don't use this function in Lisp programs! -- Command: end-of-buffer &optional n This function moves point to the end of the buffer (or the limits of the accessible portion, when narrowing is in effect), setting the mark at the previous position. If n is non-nil, then it puts point n tenths of the way from the end. In an interactive call, n is the numeric prefix argument, if provided; otherwise n defaults to nil. Don't use this function in Lisp programs! ΓòÉΓòÉΓòÉ 29.2.4. Motion by Text Lines ΓòÉΓòÉΓòÉ Text lines are portions of the buffer delimited by newline characters, which are regarded as part of the previous line. The first text line begins at the beginning of the buffer, and the last text line ends at the end of the buffer whether or not the last character is a newline. The division of the buffer into text lines is not affected by the width of the window, or by how tabs and control characters are displayed. -- Command: goto-line line This function moves point to the front of the lineth line, counting from line 1 at beginning of buffer. If line is less than 1, then point is set to the beginning of the buffer. If line is greater than the number of lines in the buffer, then point is set to the end of the last line of the buffer. If narrowing is in effect, then line still counts from the beginning of the buffer, but point cannot go outside the accessible portion. So point is set at the beginning or end of the accessible portion of the text if the line number specifies a position that is inaccessible. The return value of goto-line is the difference between line and the line number of the line to which point actually was able move (before taking account of any narrowing). Thus, the value is positive if the scan encounters the end of the buffer. In an interactive call, line is the numeric prefix argument if one has been provided. Otherwise line is read in the minibuffer. -- Command: beginning-of-line &optional count This function moves point to the beginning of the current line. With an argument count not nil or 1, it moves forward count-1 lines and then to the beginning of the line. If this function reaches the end of the buffer (or of the accessible portion, if narrowing is in effect), it positions point at the beginning of the last line. No error is signaled. -- Command: end-of-line &optional count This function moves point to the end of the current line. With an argument count not nil or 1, it moves forward count-1 lines and then to the end of the line. If this function reaches the end of the buffer (or of the accessible portion, if narrowing is in effect), it positions point at the end of the last line. No error is signaled. -- Command: forward-line &optional count This function moves point forward count lines, to the beginning of the line. If count is negative, it moves point -count lines backward, to the beginning of the line. If the beginning or end of the buffer (or of the accessible portion) is encountered before that many lines are found, then point stops at the beginning or end. No error is signaled. forward-line returns the difference between count and the number of lines actually moved. If you attempt to move down five lines from the beginning of a buffer that has only three lines, point will positioned at the end of the last line, and the value will be 2. In an interactive call, count is the numeric prefix argument. -- Function: count-lines start end This function returns the number of lines between the positions start and end in the current buffer. If start and end are equal, then it returns 0. Otherwise it returns at least 1, even if start and end are on the same line. This is because the text between them, considered in isolation, must contain at least one line unless it is empty. Here is an example of using count-lines: (defun current-line () "Return the vertical position of point in the selected window. Top line is 0. Counts each text line only once, even if it wraps." (+ (count-lines (window-start) (point)) (if (= (current-column) 0) 1 0) -1)) Also see the functions bolp and eolp in Near Point. These functions do not move point, but test whether it is already at the beginning or end of a line. ΓòÉΓòÉΓòÉ 29.2.5. Motion by Screen Lines ΓòÉΓòÉΓòÉ The line functions in the previous section count text lines, delimited only by newline characters. By contrast, these functions count screen lines, which are defined by the way the text appears on the screen. A text line is a single screen line if it is short enough to fit the width of the selected window, but otherwise it may occupy several screen lines. In some cases, text lines are truncated on the screen rather than continued onto additional screen lines. Then vertical-motion moves point just like forward-line. See Truncation. Because the width of a given string depends on the flags which control the appearance of certain characters, vertical-motion will behave differently on a given piece of text found in different buffers. It will even act differently in different windows showing the same buffer, because the width may differ and so may the truncation flag. See Control Char Display. -- Function: vertical-motion count This function moves point to the start of the screen line count screen lines down from the screen line containing point. If count is negative, it moves up instead. This function returns the number of lines moved. The value may be less in absolute value than count if the beginning or end of the buffer was reached. -- Command: move-to-window-line count This function moves point with respect to the text currently displayed in the selected window. Point is moved to the beginning of the screen line count screen lines from the top of the window. If count is negative, point moves either to the beginning of the line -count lines from the bottom or else to the last line of the buffer if the buffer ends above the specified screen position. If count is nil, then point moves to the beginning of the line in the middle of the window. If the absolute value of count is greater than the size of the window, then point moves to the place which would appear on that screen line if the window were tall enough. This will probably cause the next redisplay to scroll to bring that location onto the screen. In an interactive call, count is the numeric prefix argument. The value returned is the window line number, with the top line in the window numbered 0. ΓòÉΓòÉΓòÉ 29.2.6. The User-Level Vertical Motion Commands ΓòÉΓòÉΓòÉ A goal column is useful if you want to edit text such as a table in which you want to move point to a certain column on each line. The goal column affects the vertical text line motion commands, next-line and previous-line. See Basic Editing Commands. -- User Option: goal-column This variable holds an explicitly specified goal column for vertical line motion commands. If it is an integer, it specifies a column, and these commands try to move to that column on each line. If it is nil, then the commands set their own goal columns. Any other value is invalid. -- Variable: temporary-goal-column This variable holds the temporary goal column during a sequence of consecutive vertical line motion commands. It is overridden by goal-column if that is non-nil. It is set each time a vertical motion command is invoked, unless the previous command was also a vertical motion command. -- User Option: track-eol This variable controls how the vertical line motion commands operate when starting at the end of a line. If track-eol is non-nil, then vertical motion starting at the end of a line will keep to the ends of lines. This means moving to the end of each line moved onto. The value of track-eol has no effect if point is not at the end of a line when the first vertical motion command is given. track-eol has its effect by causing temporary-goal-column to be set to 9999 instead of to the current column. -- Command: set-goal-column unset This command sets the variable goal-column to specify a permanent goal column for the vertical line motion commands. If unset is nil, then goal-column is set to the current column of point. If unset is non-nil, then goal-column is set to nil. This function is intended for interactive use; and in an interactive call, unset is the raw prefix argument. ΓòÉΓòÉΓòÉ 29.2.7. Moving over Lists and Other Balanced Expressions ΓòÉΓòÉΓòÉ Here are several functions concerned with balanced-parenthesis expressions (also called sexps in connection with moving across them in Emacs). The syntax table controls how these functions interpret various characters; see Syntax Tables. See Parsing Expressions, for lower-level primitives for scanning sexps or parts of sexps. For user-level commands, see Lists and Sexps. -- Command: forward-list arg Move forward across arg balanced groups of parentheses. (Other syntatic entities such as words or paired string quotes are ignored.) -- Command: backward-list arg Move backward across arg balanced groups of parentheses. (Other syntatic entities such as words or paired string quotes are ignored.) -- Command: up-list arg Move forward out of arg levels of parentheses. A negative argument means move backward but still to a less deep spot. -- Command: down-list arg Move forward down arg levels of parentheses. A negative argument means move backward but still go down arg level. -- Command: forward-sexp arg Move forward across arg balanced expressions. Balanced expressions include both those delimited by parentheses and other kinds, such as words and string constants. For example, ---------- Buffer: foo ---------- (concat-!- "foo " (car x) y z) ---------- Buffer: foo ---------- (forward-sexp 3) => nil ---------- Buffer: foo ---------- (concat "foo " (car x) y-!- z) ---------- Buffer: foo ---------- -- Command: backward-sexp arg Move backward across arg balanced expressions. ΓòÉΓòÉΓòÉ 29.2.8. Skipping Characters ΓòÉΓòÉΓòÉ The following two functions move point over a specified set of characters. For example, they are often used to skip whitespace. -- Function: skip-chars-forward character-set &optional limit This function moves point in the current buffer forward, skipping over a given set of characters. Emacs first examines the character following point; if it matches character-set, then point is advanced and the next character is examined. This continues until a character is found that does not match. The function returns nil. The argument character-set is like the inside of a `[...]' in a regular expression except that `]' is never special and `\' quotes `^', `-' or `\'. Thus, "a-zA-Z" skips over all letters, stopping before the first nonletter, and "^a-zA-Z" skips nonletters stopping before the first letter. See Regular Expressions. If limit is supplied (it must be a number or a marker), it specifies the maximum position in the buffer that point can be skipped to. Point will stop at or before limit. In the following example, point is initially located directly before the `T'. After the form is evaluated, point is located at the end of that line (between the `t' of `hat' and the newline). The function skips all letters and spaces, but not newlines. ---------- Buffer: foo ---------- I read "-!-The cat in the hat comes back" twice. ---------- Buffer: foo ---------- (skip-chars-forward "a-zA-Z ") => nil ---------- Buffer: foo ---------- I read "The cat in the hat-!- comes back" twice. ---------- Buffer: foo ---------- -- Function: skip-chars-backward character-set &optional limit This function moves point backward, skipping characters that match character-set. It just like skip-chars-forward except for the direction of motion. ΓòÉΓòÉΓòÉ 29.3. Excursions ΓòÉΓòÉΓòÉ It is often useful to move point ``temporarily'' within a localized portion of the program, or to switch buffers temporarily. This is called an excursion, and it is done with the save-excursion special form. This construct saves the current buffer and its values of point and the mark so they can be restored after the completion of the excursion. The forms for saving and restoring the configuration of windows are described elsewhere (see Window Configurations). -- Special Form: save-excursion forms... The save-excursion special form saves the identity of the current buffer and the values of point and the mark in it, evaluates forms, and finally restores the buffer and its saved values of point and the mark. All three saved values are restored even in case of an abnormal exit via throw or error (see Nonlocal Exits). The save-excursion special form is the standard way to switch buffers or move point within one part of a program and avoid affecting the rest of the program. It is used more than 500 times in the Lisp sources of Emacs. The values of point and the mark for other buffers are not saved by save-excursion, so any changes made to point and the mark in the other buffers will remain in effect after save-excursion exits. Likewise, save-excursion does not restore window-buffer correspondences altered by functions such as switch-to-buffer. One way to restore these correspondences, and the selected window, is to use save-window-excursion inside save-excursion (see Window Configurations). The value returned by save-excursion is the result of the last of forms, or nil if no forms are given. (save-excursion forms) == (let ((old-buf (current-buffer)) (old-pnt (point-marker)) (old-mark (copy-marker (mark-marker)))) (unwind-protect (progn forms) (set-buffer old-buf) (goto-char old-pnt) (set-marker (mark-marker) old-mark))) ΓòÉΓòÉΓòÉ 29.4. Narrowing ΓòÉΓòÉΓòÉ Narrowing means limiting the text addressable by Emacs editing commands to a limited range of characters in a buffer. The text that remains addressable is called the accessible portion of the buffer. Narrowing is specified with two buffer positions which become the beginning and end of the accessible portion. For most editing commands these positions replace the values of the beginning and end of the buffer. While narrowing is in effect, no text outside the accessible portion is displayed, and point cannot move outside the accessible portion. Values such as positions or line numbers which usually count from the beginning of the buffer continue to do so, but the functions which use them will refuse to operate on text that is inaccessible. The commands for saving buffers are unaffected by narrowing; the entire buffer is saved regardless of the any narrowing. -- Command: narrow-to-region start end This function sets the accessible portion of the current buffer to start at start and end at end. Both arguments should be character positions. In an interactive call, start and end are set to the bounds of the current region (point and the mark, with the smallest first). -- Command: narrow-to-page move-count This function sets the accessible portion of the current buffer to include just the current page. An optional first argument move-count non-nil means to move forward or backward by move-count pages and then narrow. In an interactive call, move-count is set to the numeric prefix argument. -- Command: widen This function cancels any narrowing in the current buffer, so that the entire contents are accessible. This is called widening. It is equivalent to the following expression: (narrow-to-region 1 (1+ (buffer-size))) -- Special Form: save-restriction forms... This special form saves the current bounds of the accessible portion, evaluates forms, and finally restores the saved bounds, thus restoring the same state of narrowing (or absence thereof) formerly in effect. The state of narrowing is restored even in the event of an abnormal exit via throw or error (see Nonlocal Exits). Therefore, this construct is a clean way to narrow a buffer temporarily. The value returned by save-restriction is that returned by the last of forms, or nil if no forms were given. *Note:* it is easy to make a mistake when using save-restriction. Read the entire description here before you try it. Point and the mark are not restored by this special form; use save-excursion for that. If you use both save-restriction and save-excursion together, save-excursion should come first (on the outside). Otherwise, the old point value would be restored with temporary narrowing still in effect. If the old point value were outside the limits of the temporary narrowing, this would fail to restore it accurately. The save-restriction special form records the values of the beginning and end of the accessible portion as distances from the beginning and end of the buffer. In other words, it records the amount of inaccessible text before and after the accessible portion. This technique yields correct results if the body of the form does further narrowing. However, save-restriction can become confused if the body widens and then makes changes outside the area of the saved narrowing. When this is what you want to do, save-restriction is not the right tool for the job. Here is what you must do instead: (let ((beg (point-min-marker)) (end (point-max-marker))) (unwind-protect (progn body) (narrow-to-region beg end))) Here is a simple example of correct use of save-restriction: ---------- Buffer: foo ---------- This is the contents of foo This is the contents of foo This is the contents of foo-!- ---------- Buffer: foo ---------- (save-excursion (save-restriction (goto-char 1) (forward-line 2) (narrow-to-region 1 (point)) (goto-char (point-min)) (replace-string "foo" "bar"))) ---------- Buffer: foo ---------- This is the contents of bar This is the contents of bar This is the contents of foo-!- ---------- Buffer: foo ---------- ΓòÉΓòÉΓòÉ 30. Markers ΓòÉΓòÉΓòÉ A marker is a Lisp object used to specify a position in a buffer relative to the surrounding text. A marker changes its offset from the beginning of the buffer automatically whenever text is inserted or deleted, so that it stays with the two characters on either side of it. ΓòÉΓòÉΓòÉ 30.1. Overview of Markers ΓòÉΓòÉΓòÉ A marker specifies a buffer and a position in that buffer. The marker can be used to represent a position in the functions that require one, just as an integer could be used. See Positions, for a complete description of positions. A marker has two attributes: the marker position, and the marker buffer. The marker position is an integer which is equivalent (at the moment) to the marker as a position in that buffer; however, as text is inserted or deleted in the buffer, the marker is relocated, so that its integer equivalent changes. The idea is that a marker positioned between two characters in a buffer will remain between those two characters despite any changes made to the contents of the buffer; thus, a marker's offset from the beginning of a buffer may change often during the life of the marker. If the text around a marker is deleted, the marker is repositioned between the characters immediately before and after the deleted text. If text is inserted at the position of a marker, the marker remains in front of the new text unless it is inserted with insert-before-markers (see Insertion). When text is inserted or deleted somewhere before the marker position (not next to the marker), the marker moves back and forth with the two neighboring characters. When a buffer is modified, all of its markers must be checked so that they can be relocated if necessary. This slows processing in a buffer with a large number of markers. For this reason, it is a good idea to make a marker point nowhere if you are sure you don't need it any more. Unreferenced markers will eventually be garbage collected, but until then will continue to be updated if they do point somewhere. Because it is quite common to perform arithmetic operations on a marker position, most of the arithmetic operations (including + and -) accept markers as arguments. In such cases, the current position of the marker is used. Here are examples of creating markers, setting markers, and moving point to markers: ;; Make a new marker that initially does not point anywhere: (setq m1 (make-marker)) => #<marker in no buffer> ;; Set m1 to point between the 100th and 101st characters. ;; in the current buffer: (set-marker m1 100) => #<marker at 100 in markers.texi> ;; Now insert one character at the beginning of the buffer: (goto-char (point-min)) => 1 (insert "Q") => nil ;; m1 is updated appropriately. m1 => #<marker at 101 in markers.texi> ;; Two markers that point to the same position ;; are not eq, but they are equal. (setq m2 (copy-marker m1)) => #<marker at 101 in markers.texi> (eq m1 m2) => nil (equal m1 m2) => t ;; When you are finished using a marker, make it point nowhere. (set-marker m1 nil) => #<marker in no buffer> ΓòÉΓòÉΓòÉ 30.2. Predicates on Markers ΓòÉΓòÉΓòÉ You can test an object to see whether it is a marker, or whether it is either an integer or a marker. The latter test is useful when you are using the arithmetic functions that work with both markers and integers. -- Function: markerp object This function returns t if object is a marker, nil otherwise. In particular, integers are not markers, even though many functions will accept either a marker or an integer. -- Function: integer-or-marker-p object This function returns t if object is an integer or a marker, nil otherwise. ΓòÉΓòÉΓòÉ 30.3. Functions That Create Markers ΓòÉΓòÉΓòÉ When you create a new marker, you can make it point nowhere, or point to the present position of point, or to the beginning or end of the accessible portion of the buffer, or to the same place as another given marker. -- Function: make-marker This functions returns a newly allocated marker that does not point anywhere. (make-marker) => #<marker in no buffer> -- Function: point-marker This function returns a new marker that points to the present position of point in the current buffer. See Point. For an example, see copy-marker, below. -- Function: point-min-marker This function returns a new marker that points to the beginning of the accessible portion of the buffer. This will be the beginning of the buffer unless narrowing is in effect. See Narrowing. -- Function: point-max-marker This function returns a new marker that points to the end of the accessible portion of the buffer. This will be the end of the buffer unless narrowing is in effect. See Narrowing. Here are examples of this function and point-min-marker, shown in a buffer containing a version of the source file for the text of this chapter. (point-min-marker) => #<marker at 1 in markers.texi> (point-max-marker) => #<marker at 15573 in markers.texi> (narrow-to-region 100 200) => nil (point-min-marker) => #<marker at 100 in markers.texi> (point-max-marker) => #<marker at 200 in markers.texi> -- Function: copy-marker marker-or-integer If passed a marker as its argument, copy-marker returns a new marker that points to the same place and the same buffer as does marker-or-integer. If passed an integer as its argument, copy-marker returns a new marker that points to position marker-or-integer in the current buffer. If passed an argument that is an integer whose value is less than 1, copy-marker returns a new marker that points to the beginning of the current buffer. If passed an argument that is an integer whose value is greater than the length of the buffer, then copy-marker returns a new marker that points to the end of the buffer. An error is signaled if marker is neither a marker nor an integer. (setq p (point-marker)) => #<marker at 2139 in markers.texi> (setq q (copy-marker p)) => #<marker at 2139 in markers.texi> (eq p q) => nil (equal p q) => t (copy-marker 0) => #<marker at 1 in markers.texi> (copy-marker 20000) => #<marker at 7572 in markers.texi> ΓòÉΓòÉΓòÉ 30.4. Information from Markers ΓòÉΓòÉΓòÉ This section describes the functions for accessing the components of a marker object. -- Function: marker-position marker This function returns the position that marker points to, or nil if it points nowhere. -- Function: marker-buffer marker This function returns the buffer that marker points into, or nil if it points nowhere. (setq m (make-marker)) => #<marker in no buffer> (marker-position m) => nil (marker-buffer m) => nil (set-marker m 3770 (current-buffer)) => #<marker at 3770 in markers.texi> (marker-buffer m) => #<buffer markers.texi> (marker-position m) => 3770 Two distinct markers will be found equal (even though not eq) to each other if they have the same position and buffer, or if they both point nowhere. ΓòÉΓòÉΓòÉ 30.5. Changing Markers ΓòÉΓòÉΓòÉ This section describes how to change the position of an existing marker. When you do this, be sure you know whether the marker is used outside of your program, and, if so, what effects will result from moving it---otherwise, confusing things may happen in other parts of Emacs. -- Function: set-marker marker position &optional buffer This function moves marker to position in buffer. If buffer is not provided, it defaults to the current buffer. If position is less than 1, set-marker moves marker to the beginning of the buffer. If the value of position is greater than the size of the buffer, set-marker moves marker to the end of the buffer. If position is nil or a marker that points nowhere, then marker is set to point nowhere. The value returned is marker. (setq m (point-marker)) => #<marker at 4714 in markers.texi> (set-marker m 55) => #<marker at 55 in markers.texi> (setq b (get-buffer "foo")) => #<buffer foo> (set-marker m 0 b) => #<marker at 1 in foo> -- Function: move-marker marker position &optional buffer This is another name for set-marker. ΓòÉΓòÉΓòÉ 30.6. The Mark ΓòÉΓòÉΓòÉ A special marker in each buffer is designated the mark. It records a position for the user for the sake of commands such as C-w and C-x TAB. Lisp programs should set the mark only to values that have a potential use to the user, and never for their own internal purposes. For example, the replace-regexp command sets the mark to the value of point before doing any replacements, because this enables the user to move back there conveniently after the replace is finished. Many commands are designed so that when called interactively they operate on the text between point and the mark. If you are writing such a command, don't examine the mark directly; instead, use interactive with the `r' specification. This will provide the values of point and the mark as arguments to the command in an interactive call, but will permit other Lisp programs to specify arguments explicitly. See Interactive Codes. Each buffer has its own value of the mark that is independent of the value of the mark in other buffers. When a buffer is created, the mark exists but does not point anywhere. We consider this state as ``the absence of a mark in that buffer''. In addition to the mark, each buffer has a mark ring which is a list of markers that are the previous values of the mark. When editing commands change the mark, they should normally save the old value of the mark on the mark ring. The mark ring may contain no more than the maximum number of entries specified by the variable mark-ring-max; excess entries are discarded on a first-in-first-out basis. -- Function: mark This function returns the position of the current buffer's mark as an integer. nil is returned if the mark is not yet set for this buffer. -- Function: mark-marker This function returns the current buffer's mark. This the very marker which records the mark location inside Emacs, not a copy. Therefore, changing this marker's position will directly affect the position of the mark. Don't do it unless that is the effect you want. (setq m (mark-marker)) => #<marker at 3420 in markers.texi> (set-marker m 100) => #<marker at 100 in markers.texi> (mark-marker) => #<marker at 100 in markers.texi> Like any marker, this marker can be set to point at any buffer you like. We don't recommend that you make it point at any buffer other than the one of which it is the mark. If you do, it will yield perfectly consistent, if rather odd, results. -- Command: set-mark-command jump If jump is nil, this command sets the mark to the value of point and pushes the previous value of the mark on the mark ring. The message `Mark set' is also displayed in the echo area. If jump is not nil, this command sets point to the value of the mark, and sets the mark to the previous saved mark value, which is popped off the mark ring. This function is only intended for interactive use. -- Function: set-mark position This function sets the mark to position. The old value of the mark is not pushed onto the mark ring. *Note:* use this function only if you want the user to see that the mark has moved, and you want the previous mark position to be lost. Normally, when a new mark is set, the old one should go on the mark-ring, which is why most applications should use push-mark and pop-mark, not set-mark. Novice Emacs Lisp programmers often try to use the mark for the wrong purposes. The mark saves a location for the user's convenience. An editing command should not alter the mark unless altering the mark is part of the user-level functionality of the command. (And, in that case, this effect should be documented.) To remember a location for internal use in the Lisp program, store it in a Lisp variable. For example: (let ((beg (point))) (forward-line 1) (delete-region beg (point))). -- Variable: mark-ring The value of this buffer-local variable is the list of saved former marks of the current buffer, most recent first. mark-ring => (#<marker at 11050 in markers.texi> #<marker at 10832 in markers.texi> ...) -- User Option: mark-ring-max The value of this variable is the maximum size of mark-ring. If more marks than this are pushed onto the mark-ring, it discards marks on a first-in, first-out basis. -- Function: push-mark &optional position nomsg This function sets the current buffer's mark to position, and pushes a copy of the previous mark onto mark-ring. If position is nil, then the value of point is used. push-mark returns nil. A `Mark set' message is displayed unless nomsg is non-nil. -- Function: pop-mark This function pops off the top element of mark-ring and makes that mark become the buffer's actual mark. This does not change the buffer's point and does nothing if mark-ring is empty. The return value is not useful. ΓòÉΓòÉΓòÉ 30.7. The Region ΓòÉΓòÉΓòÉ The text between point and the mark is known as the region. Various functions operate on text delimited by point and the mark, but only those functions specifically related to the region itself are described here. -- Function: region-beginning This function returns the position of the beginning of the region (as an integer). This is the position of either point or the mark, whichever is smaller. If the mark does not point anywhere, an error is signaled. -- Function: region-end This function returns the position of the end of the region (as an integer). This is the position of either point or the mark, whichever is larger. If the mark does not point anywhere, an error is signaled. Few programs need to use the region-beginning and region-end functions. A command designed to operate on a region should instead use interactive with the `r' specification, so that the same function can be called with explicit bounds arguments from programs. (See Interactive Codes.) ΓòÉΓòÉΓòÉ 31. Text ΓòÉΓòÉΓòÉ This chapter describes the functions that deal with the text in a buffer. Most examine, insert or delete text in the current buffer, often in the vicinity of point. Many are interactive. All the functions that change the text provide for undoing the changes (see Undo). Many text-related functions operate on a region of text defined by two buffer positions passed in arguments named start and end. These arguments should be either markers (see Markers) or or numeric character positions (see Positions). The order of these arguments does not matter; it is all right for start to be the end of the region and end the beginning. For example, (delete-region 1 10) and (delete-region 10 1) perform identically. An args-out-of-range error is signaled if either start or end is outside the accessible portion of the buffer. In an interactive call, point and the mark are used for these arguments. Throughout this chapter, ``text'' refers to the characters in the buffer. ΓòÉΓòÉΓòÉ 31.1. Examining Text Near Point ΓòÉΓòÉΓòÉ Many functions are provided to look at the characters around point. Several simple functions are described here. See also looking-at in Searching and Matching. -- Function: char-after position This function returns the character in the current buffer at (i.e., immediately after) position position. If position is out of range for this purpose, either before the beginning of the buffer, or at or beyond the end, than the value is nil. Remember that point is always between characters, and the terminal cursor normally appears over the character following point. Therefore, the character returned by char-after is the character the cursor is over. In the following example, assume that the first character in the buffer is `@': (char-to-string (char-after 1)) => "@" -- Function: following-char This function returns the character following point in the current buffer. This is similar to (char-after (point)). However, point is at the end of the buffer, then the result of following-char is 0. In this example, point is between the `a' and the `c'. ---------- Buffer: foo ---------- Gentlemen may cry ``Pea-!-ce! Peace!,'' but there is no peace. ---------- Buffer: foo ---------- (char-to-string (preceding-char)) => "a" (char-to-string (following-char)) => "c" -- Function: preceding-char This function returns the character preceding point in the current buffer. See above, under following-char, for an example. If point is at the beginning of the buffer, then the result of preceding-char is 0. -- Function: bobp This function returns t if point is at the beginning of the buffer. If narrowing is in effect, this means the beginning of the accessible portion of the text. See also point-min in Point. -- Function: eobp This function returns t if point is at the end of the buffer. If narrowing is in effect, this means the end of accessible portion of the text. See also point-max in See Point. -- Function: bolp This function returns t if point is at the beginning of a line. See Text Lines. -- Function: eolp This function returns t if point is at the end of a line. The end of the buffer is always considered the end of a line. ΓòÉΓòÉΓòÉ 31.2. Examining Buffer Contents ΓòÉΓòÉΓòÉ This section describes two functions that allow a Lisp program to convert any portion of the text in the buffer into a string. -- Function: buffer-substring start end This function returns a string containing a copy of the text of the region defined by positions start and end in the current buffer. If the arguments are not positions in the accessible portion of the buffer, Emacs signals an args-out-of-range error. It is not necessary for start to be less than end; the arguments can be given in either order. But most often the smaller argument is written first. ---------- Buffer: foo ---------- This is the contents of buffer foo ---------- Buffer: foo ---------- (buffer-substring 1 10) => "This is t" (buffer-substring (point-max) 10) => "he contents of buffer foo " -- Function: buffer-string This function returns the contents of the accessible portion of the current buffer as a string. This is the portion between (point-min) and (point-max) (see Narrowing). ---------- Buffer: foo ---------- This is the contents of buffer foo ---------- Buffer: foo ---------- (buffer-string) => "This is the contents of buffer foo " ΓòÉΓòÉΓòÉ 31.3. Insertion ΓòÉΓòÉΓòÉ Insertion takes place at point. Markers pointing at positions after the insertion point are relocated with the surrounding text (see Markers). When a marker points at the place of insertion, it is normally not relocated, so that it points to the beginning of the inserted text; however, when insert-before-markers is used, all such markers are relocated to point after the inserted text. Point may end up either before or after inserted text, depending on the function used. If point is left after the inserted text, we speak of insertion before point. Each of these functions signals an error if the current buffer is read-only. -- Function: insert &rest args This function inserts the strings and/or characters args into the current buffer, at point, moving point forward. An error is signaled unless all args are either strings or characters. The value is nil. -- Function: insert-before-markers &rest args This function inserts the strings and/or characters args into the current buffer, at point, moving point forward. An error is signaled unless all args are either strings or characters. The value is nil. This function is unlike the other insertion functions in that a marker whose position initially equals point is relocated to come after the newly inserted text. -- Function: insert-char character count This function inserts count instances of character into the current buffer before point. count must be a number, and character must be a character. The value is nil. -- Function: insert-buffer-substring from-buffer-or-name start end This function inserts a substring of the contents of buffer from-buffer-or-name (which must already exist) into the current buffer before point. The text inserted consists of the characters in the region defined by start and end. The value is nil. In this example, the form is executed with buffer `bar' as the current buffer. We assume that buffer `bar' is initially empty. ---------- Buffer: foo ---------- We hold these truths to be self-evident, that all ---------- Buffer: foo ---------- (insert-buffer-substring "foo" 1 20) => nil ---------- Buffer: bar ---------- We hold these truth ---------- Buffer: bar ---------- ΓòÉΓòÉΓòÉ 31.4. User-Level Insertion Commands ΓòÉΓòÉΓòÉ This section describes higher-level commands for inserting text, commands intended primarily for the user but useful also in Lisp programs. -- Command: insert-buffer from-buffer-or-name This function inserts the entire contents of from-buffer-or-name (which must exist) into the current buffer after point. It leaves the mark after the inserted text. The value is unpredictable. -- Command: quoted-insert count This function reads the next input character verbatim and inserts it. It is primarily useful for inserting control characters. You may also type up to 3 octal digits, to insert a character with that code. The argument count is the number of these characters to insert. An error is signaled if count is not a number. This function is primarily for interactive use; there is no reason to use it in a program except for installing it on a keymap. It returns nil. -- Command: self-insert-command count This function inserts the last character typed count times and returns nil. This is the function that most printing characters are bound to. In routine use, self-insert-command is the most frequently called function in Emacs, but programs rarely use it except to install it on a keymap. In an interactive call, count is the numeric prefix argument. This function calls auto-fill-hook if the current column number is greater than the value of fill-column and the character inserted is a space (see Auto Filling). This function performs abbrev expansion if Abbrev mode (see Abbrevs) is enabled and the inserted character does not have word-constituent syntax (see Syntax Class Table). This function is also responsible for calling the blink-paren-hook when the inserted character has close parenthesis syntax (see Blinking). -- Command: newline &optional number-of-newlines This function inserts newlines into the current buffer before point. If number-of-newlines is supplied, that many newline characters are inserted. In Auto Fill mode, newline can break the preceding line if number-of-newlines is not supplied. When this happens, it actually inserts two newlines at different places: one at point, and another earlier in the line. newline does not auto-fill if number-of-newlines is non-nil. The value returned is nil. In an interactive call, count is the numeric prefix argument. -- Command: split-line This function splits the current line, moving the portion of the line after point down vertically, so that it is on the next line directly below where it was before. Whitespace is inserted as needed at the beginning of the lower line, using the indent-to function. split-line returns the position of point. Programs hardly ever use this function. -- Command: open-line count This function inserts count newlines into the current buffer after point, leaving point where it was. In an interactive call, count is the numeric prefix argument. Programs hardly ever use this function. The value is unpredictable. -- Command: overwrite-mode argument This function turns Overwrite mode on or off. If argument is nil then the mode is toggled. Otherwise, if argument is a positive number (greater than zero), then the mode is turned on; any other argument turns it off. When Overwrite mode is on, self-inserting graphic characters replace existing text character for character, and do not push the existing text to the right. This function affects primarily self-insert-command. In an interactive call, argument is set to the raw prefix argument. The return value of overwrite-mode is unpredictable. -- Variable: overwrite-mode Overwrite mode is in effect when this variable is non-nil. It is automatically made buffer-local when set in any fashion. ΓòÉΓòÉΓòÉ 31.5. Deletion of Text ΓòÉΓòÉΓòÉ All of the deletion functions operate on the current buffer, and all return a value of nil. In addition to these functions, you can also delete text using the ``kill'' functions that save it in the kill ring for the user; some of these functions save text in the kill ring in some cases but not in the usual case. See The Kill Ring. -- Function: erase-buffer This function deletes the entire text of the current buffer, leaving it empty. If the buffer is read-only, it signals a buffer-read-only error. Otherwise the text is deleted with no confirmation required. The value is always nil. As a safety measure, this function is not interactively callable. -- Command: delete-region start end This function deletes the text in the current buffer in the region defined by start and end. The value is nil. -- Command: delete-char count &optional killp This function deletes count characters directly after point, or before point if count is negative. If killp is non-nil, then it saves the deleted characters in the kill ring. In an interactive call, count is the numeric prefix argument, and killp is the unprocessed prefix argument. Therefore, if a prefix argument is supplied, the text is saved in the kill ring. If no prefix argument is supplied, then one character is deleted, but not saved in the kill ring. The value returned is always nil. -- Command: delete-backward-char count &optional killp This function deletes count characters directly before point, or after point if count is negative. If killp is non-nil, then it saves the deleted characters in the kill ring. In an interactive call, count is the numeric prefix argument, and killp is the unprocessed prefix argument. Therefore, if a prefix argument is supplied, the text is saved in the kill ring. If no prefix argument is supplied, then one character is deleted, but not saved in the kill ring. The value returned is always nil. -- Command: backward-delete-char-untabify count &optional killp This function deletes count characters backward, changing tabs into spaces. When the next character to be deleted is a tab, it is first replaced with the proper number of spaces to preserve alignment and then one of those spaces is deleted instead of the tab. If killp is non-nil, then the command saves the deleted characters in the kill ring. If count is negative, then tabs are not changed to spaces, and the characters are deleted by calling delete-backward-char with count. In an interactive call, count is the numeric prefix argument, and killp is the unprocessed prefix argument. Therefore, if a prefix argument is supplied, the text is saved in the kill ring. If no prefix argument is supplied, then one character is deleted, but not saved in the kill ring. The value returned is always nil. ΓòÉΓòÉΓòÉ 31.6. User-Level Deletion Commands ΓòÉΓòÉΓòÉ This section describes higher-level commands for deleting text, commands intended primarily for the user but useful also in Lisp programs. -- Command: delete-horizontal-space This function deletes all spaces and tabs around point. It returns nil. In the following examples, assume that delete-horizontal-space is called four times, once on each line, with point between the second and third characters on the line. ---------- Buffer: foo ---------- I -!-thought I -!- thought We-!- thought Yo-!-u thought ---------- Buffer: foo ---------- (delete-horizontal-space) ; Four times. => nil ---------- Buffer: foo ---------- Ithought Ithought Wethought You thought ---------- Buffer: foo ---------- -- Command: delete-indentation &optional join-following-p This function joins the line point is on to the previous line, deleting any whitespace at the join and in some cases replacing it with one space. If join-following-p is non-nil, delete-indentation joins this line to following line instead. The value is nil. In the example below, point is located on the line starting `events', and it makes no difference if there are trailing spaces in the preceding line. ---------- Buffer: foo ---------- When in the course of human -!- events, it becomes necessary ---------- Buffer: foo ---------- (delete-indentation) => nil ---------- Buffer: foo ---------- When in the course of human-!- events, it becomes necessary ---------- Buffer: foo ---------- After the lines are joined, the function fixup-whitespace is responsible for deciding whether to leave a space at the junction. -- Function: fixup-whitespace This function replaces white space between the objects on either side of point with either one space or no space as appropriate. It returns nil. The appropriate amount of space is none at the beginning or end of the line. Otherwise, it is one space except when point is before a character with close parenthesis syntax or after a character with open parenthesis or expression-prefix syntax. See Syntax Class Table. In the example below, point is at the beginning of the second line when fixup-whitespace is called the first time. It is located directly after the `(' for the second invocation. ---------- Buffer: foo ---------- This has too many spaces -!- at the front of this line This has too many spaces at the start of (-!- this list) ---------- Buffer: foo ---------- (fixup-whitespace) => nil (fixup-whitespace) => nil ---------- Buffer: foo ---------- This has too many spaces at the front of this line This has too many spaces at the start of (this list) ---------- Buffer: foo ---------- -- Command: just-one-space This command replaces any spaces and tabs around point with a single space. It returns nil. -- Command: delete-blank-lines This function deletes blank lines surrounding point. If point is on a blank line with one or more blank lines before or after it, then all but one of them are deleted. If point is on an isolated blank line, then it is deleted. If point is on a nonblank line, the command deletes all blank lines following it. A blank line is defined as a line containing only tabs and spaces. delete-blank-lines returns nil. ΓòÉΓòÉΓòÉ 31.7. The Kill Ring ΓòÉΓòÉΓòÉ Kill functions delete text like the deletion functions, but save it so that the user can reinsert it by yanking. Most of these functions have `kill-' in their name. By contrast, the functions whose names start with `delete-' normally do not save text for yanking (though they can still be undone); these are ``deletion'' functions. Most of the kill commands are primarily for interactive use, and are not described here. What we do describe are the functions provided for use in writing such commands. When deleting text for internal purposes within a Lisp function, you should normally use deletion functions, so as not to disturb the kill ring contents. See Deletion. Emacs saves the last several batches of killed text in a list. We call it the kill ring because, in yanking, the elements are considered to be in a cyclic order. The list is kept in the variable kill-ring, and can be operated on with the usual functions for lists; there are also specialized functions, described in this section, which treat it as a ring. Some people think use of the word ``kill'' in Emacs is unfortunate, since it refers to processes which specifically do not destroy the entities ``killed''. This is in sharp contrast to ordinary life, in which death is permanent and ``killed'' entities do not come back to life. Therefore, other metaphors have been proposed. For example, the term ``cut ring'' makes sense to people who, in pre-computer days, used scissors and paste to cut up and rearrange manuscripts. However, it would be difficult to change now. ΓòÉΓòÉΓòÉ 31.7.1. Data Structures in the Kill Ring ΓòÉΓòÉΓòÉ Killed text is kept as strings in a list. A short kill ring, for example, might look like this: ("some text" "a different piece of text" "yet more text") Functions that push more text into the list make the text in question into a string (using buffer-substring), add the string to the front of the list, and then look at the length of the list. If the length is longer than the value of kill-ring-max, the last entry in the list is dropped off when the new entry is put on. The kill-ring-yank-pointer global variable points to the kill ring entry that a yank function will copy. Several functions move this pointer from one entry to another, and a user can thereby specify which entry to copy. Here is a diagram that shows the variable kill-ring-yank-pointer pointing to the second entry in the kill ring ("some text" "a different piece of text" "yet more text"). kill-ring kill-ring-yank-pointer | | | ___ ___ ---> ___ ___ ___ ___ --> |___|___|------> |___|___|--> |___|___|--> nil | | | | | | | | -->"yet more text" | | | --> "a different piece of text" | --> "some text" (This circumstance occurs after C-y (yank) is immediately followed by M-y (yank-pop).) Both kill-ring and kill-ring-yank-pointer are Lisp variables whose values are normally lists. The word ``pointer'' in the name of the kill-ring-yank-pointer indicates that the variable's purpose is to identify one element of the list that will be used by default by the next yank command. The value of kill-ring-yank-pointer is always eq to one of the links in the kill ring list. The element it identifies is the car of that link. Moving kill-ring-yank-pointer to a different link is called rotating the kill ring. The functions that do this treat the kill ring (which is a list) as ring; that is to say, a change that would otherwise move the pointer past the end of the list (which would be useless) instead moves the pointer to the first link on the list. Likewise, moving back from the first link goes to the last one. kill-region is the primitive method of killing text. Any command that calls this function is a ``kill command'' (and should probably have the word ``kill'' in its name). kill-region puts the newly killed text in a new element at the beginning of the kill-ring list, and then sets kill-ring-yank-pointer to point to the first link of the list, which contains the first element. Consequently, the next yank command will yank the text just killed. In this situation, kill-ring and kill-ring-yank-pointer are eq to each other. When kill commands are interwoven with other commands, the killed portions of text are put into separate entries in the kill ring. But when two or more kill commands are executed in sequence, the text killed by the second (or third, etc.) kill command is appended to the text killed by the first command so as to make one entry in the kill ring. The kill-region function uses the last-command variable to keep track of whether the previous was a kill command, and in such cases appends the killed text to the most recent entry. ΓòÉΓòÉΓòÉ 31.7.2. Functions for Killing ΓòÉΓòÉΓòÉ -- Command: kill-region start end This function kills the text in the region defined by start and end. The text is deleted but saved in the kill ring. The value is always nil. In an interactive call, start and end are point and the mark. -- Command: kill-line &optional count This function kills the rest of the line following point, not including the newline. If point is directly before a newline, or if there is only whitespace between point and the newline, then it kills the whitespace and newline. If count is supplied, then the command kills that many lines (including the newline). (This makes executing (kill-line 2) different from executing (kill-line) twice.) If count is negative, then kill-line kills lines backwards. In an interactive call, count is the raw prefix argument (which then gets converted to a number if non-nil). The value is always nil. -- Command: zap-to-char count character In Emacs version 18, this function kills the text from point up to but not including the specified character. Thus, if the cursor is at the beginning of this sentence and the character is `s', `Thu' is deleted. If the argument is 2, `Thus, if the cur' is deleted, up to but not including the `s' in `cursor'. In Emacs version 19, this function will kill all text in the region from point up to and including the next count occurrences of character. Thus, in the example shown in the previous paragraph, the terminating `s' will be removed. The version 18 implementation kills text to the end of the buffer if the specified character is not found, but the version 19 implementation will simply signal an error. The function scans backward from point if count is negative. The value is always nil. -- Command: copy-region-as-kill start end This function saves the region defined by start and end on the kill ring, but does not delete the text from the buffer. It returns nil. In an interactive call, start and end are point and the mark. ΓòÉΓòÉΓòÉ 31.7.3. Functions for Yanking ΓòÉΓòÉΓòÉ -- Command: yank &optional arg This function inserts the text in the first entry in the kill ring directly before point. After the yank, the mark is positioned at the beginning and point is positioned after the end of the inserted text. If arg is a list (which occurs interactively when the user types C-u with no digits), then yank inserts the text as described above, but puts point before the yanked text and puts the mark after it. If arg is a number, then yank inserts the argth most recently killed text. yank does not alter the contents of the kill ring or rotate it. It returns nil. -- Command: yank-pop arg This function replaces the just-yanked text with another batch of killed text---another element of the kill ring. This command is allowed only immediately after a yank or a yank-pop. At such a time, the region contains text that was just inserted by the previous yank. yank-pop deletes that text and inserts in its place a different stretch of killed text. The text that is deleted is not inserted into the kill ring, since it is already in the kill ring somewhere. If arg is nil, then the existing region contents are replaced with the previous element of the kill ring. If arg is numeric, then the arg'th previous kill is the replacement. If arg is negative, a more recent kill is the replacement. The sequence of kills in the kill ring wraps around, so that after the oldest one comes the newest one, and before the newest one goes the oldest. The value is always nil. ΓòÉΓòÉΓòÉ 31.7.4. Internals of the Kill Ring ΓòÉΓòÉΓòÉ This section describes the lower levels of the kill ring. -- Variable: kill-ring List of killed text sequences, most recently killed first. -- Variable: kill-ring-yank-pointer This variable's value indicates which element of the kill ring is the ``front'' of the ring. More precisely, the value is a sublist of the value of kill-ring, and its car is the kill string at the front of the ring. Rotating the ring works by changing kill-ring-yank-pointer, and does not actually change the value of kill-ring. Commands which do change the kill ring also copy the new kill ring value into this variable. The effect is to rotate the ring so that the newly killed text is at front. -- Command: rotate-yank-pointer count This function rotates the kill ring count positions, which means setting kill-ring-yank-pointer to some other link in the kill ring list. It returns the new value of kill-ring-yank-pointer. -- User Option: kill-ring-max The value of this variable is the maximum length to which the kill ring can grow, before elements are thrown away on a first-in, first-out basis. The default value for kill-ring-max is 30. ΓòÉΓòÉΓòÉ 31.8. Undo ΓòÉΓòÉΓòÉ Most buffers have an undo stack which records all changes made to the buffer's text so that they can be undone. (In general, all buffers have undo stacks except special-purpose buffers for which Emacs assumes that undoing is not useful.) The size of an undo stack is limited, so large changes or a large number of changes cannot be undone. Undoing an old change is itself a change, and is added to the undo stack. However, you are not limited to undoing just the single most recent change; you can keep undoing older and older changes, even as the undo's themselves are being added to the stack. -- Command: undo &optional arg This is a user-level command to undo some previous changes. It uses undo-more and undo-start. By repeating this command you can undo earlier and earlier changes, until the information in the undo stack is used up. A numeric argument serves as a repeat count. The value is unpredictable. -- Function: undo-boundary This function places a boundary between units of undo. The undo command stops at such a boundary, and successive undo commands will undo to earlier and earlier boundaries. The return value is nil. The editor command loop automatically creates an undo boundary between keystroke commands. Thus, each undo normally undoes the effects of one command. Calling this function explicitly is useful for splitting the effects of a command into more than one unit. For example, query-replace calls this function after each replacement so that the user can undo individual replacements one by one. -- Function: undo-more count This function is used to undo count additional units of undo. It is not safe if the buffer has been changed in any fashion other than undo since the last call to undo-start. Multiple calls to undo-more have a cumulative effect, undoing farther back in time. The return value is nil. -- Function: undo-start This function prepares to undo one or more units of undo describing the most recent changes to the current buffer. It does not actually undo anything (or change the buffer at all); only undo-more does that. It returns nil. One use of this function is to break a sequence of undo's, so a subsequent call to undo-more will undo the recent run of undoing, rather than extend it into the past. The command undo calls undo-start whenever the previous command was not an undo. -- Command: buffer-enable-undo &optional buffer-or-name This function assigns an undo stack for buffer buffer-or-name, so that subsequent changes can be undone. If no argument is supplied, then the current buffer is used. If the buffer already has an undo stack, nothing is changed. This function returns nil. In an interactive call, buffer-or-name is the current buffer. You cannot specify any other buffer. -- Function: buffer-flush-undo buffer This function deassigns the undo stack of the buffer buffer, so that it will not take up space. As a result, it is no longer possible to undo either previous changes or any subsequent changes. If the buffer already has no undo stack, then this function has no effect. This function returns nil. It cannot be called interactively. ΓòÉΓòÉΓòÉ 31.9. Filling ΓòÉΓòÉΓòÉ Filling means adjusting the lengths of lines (by moving words between them) so that they are nearly (but no greater than) a specified maximum width. Additionally, lines can be justified, which means that spaces are inserted between words to make the line exactly the specified width. The width is controlled by the variable fill-column. For ease of reading, lines should be no longer than 70 or so columns. You can use Auto Fill mode (see Auto Filling) to fill text automatically as you insert it, but changes to existing text may leave it improperly filled. Then you must fill the text explicitly. Most of the functions in this section return values that are not meaningful. -- Command: fill-paragraph justify-flag This function fills the paragraph at or after point. If justify-flag is non-nil, each line is justified as well. -- Command: fill-region start end &optional justify-flag This function fills each of the paragraphs in the region from start to end. It justifies as well if justify-flag is non-nil. (In an interactive call, this is true if there is a prefix argument.) The variable paragraph-separate controls how to distinguish paragraphs. -- Command: fill-individual-paragraphs start end &optional justify-flag mail-flag This function fills each paragraph in the region according to its individual fill prefix. Thus, if the lines of a paragraph are indented with spaces, the filled paragraph will continue to be indented in the same fashion. The first two arguments, start and end, are the beginning and end of the region that will be filled. The third and fourth arguments, justify-flag and mail-flag, are optional. If justify-flag is non-nil, the paragraphs are justified as well as filled. If mail-flag is non-nil, the function is told that it is operating on a mail message and therefore should not fill the header lines. -- Command: fill-region-as-paragraph start end &optional justify-flag This function considers a region of text as a paragraph and fills it. If the region was made up of many paragraphs, the blank lines between paragraphs are removed. This function justifies as well as filling when justify-flag is non-nil. In an interactive call, any prefix argument requests justification. -- Command: justify-current-line This function inserts spaces between the words of the current line so that the line ends exactly at fill-column. It returns nil. -- User Option: fill-column This buffer-local variable specifies the maximum width of filled lines. Its value should be an integer, which is a number of columns. All the filling, justification and centering commands are affected by this variable, including Auto Fill mode (see Auto Filling). As a practical matter, if you are writing text for other people to read, you should set fill-column to no more than 70. Otherwise the line will be too long for people to read comfortably, and this can make the text seem clumsy. -- Variable: default-fill-column The value of this variable is the default value for fill-column in buffers that do not override it. This is the same as (default-value 'fill-column). The default value for default-fill-column is 70. ΓòÉΓòÉΓòÉ 31.10. Auto Filling ΓòÉΓòÉΓòÉ Filling breaks text into lines that are no more than a specified number of columns wide. Filled lines end between words, and therefore may have to be shorter than the maximum width. Auto Fill mode is a minor mode in which Emacs fills lines automatically as text as inserted. This section describes the hook and the two variables used by Auto Fill mode. For a description of functions that you can call manually to fill and justify text, see Filling. -- Variable: auto-fill-hook The value of this variable should be a function (of no arguments) to be called after self-inserting a space at a column beyond fill-column. It may be nil, in which case nothing special is done. The default value for auto-fill-hook is do-auto-fill, a function whose sole purpose is to implement the usual strategy for breaking a line. Since auto-fill-hook is not called by the run-hooks function, it will be renamed auto-fill-function in Version 19. ΓòÉΓòÉΓòÉ 31.11. Sorting Text ΓòÉΓòÉΓòÉ The sorting commands described in this section all rearrange text in a buffer. This is in contrast to the function sort, which rearranges the order of the elements of a list (see Rearrangement). The values returned by these commands are not meaningful. -- Command: sort-regexp-fields reverse record-regexp key-regexp start end This command sorts the region between start and end alphabetically as specified by record-regexp and key-regexp. If reverse is a negative integer, then sorting is in reverse order. Alphabetical sorting means that two sort keys are compared by comparing the first characters of each, the second characters of each, and so on. If a mismatch is found, it means that the sort keys are unequal; the sort key whose character is less at the point of first mismatch is the lesser sort key. The individual characters are compared according to their numerical values. Since Emacs uses the ASCII character set, the ordering in that set determines alphabetical order. The value of the record-regexp argument specifies the textual units or records that should be sorted. At the end of each record, a search is done for this regular expression, and the text that matches it is the next record. For example, the regular expression `^.+$', which matches lines with at least one character besides a newline, would make each such line into a sort record. See Regular Expressions, for a description of the syntax and meaning of regular expressions. The value of the key-regexp argument specifies what part of each record is to be compared against the other records. The key-regexp could match the whole record, or only a part. In the latter case, the rest of the record has no effect on the sorted order of records, but it is carried along when the record moves to its new position. The key-regexp argument can refer to the text matched by a subexpression of record-regexp, or it can be a regular expression on its own. If key-regexp is: `\digit' then the text matched by the digitth `$...$' parenthesis grouping in record-regexp is used for sorting. `\&' then the whole record is used for sorting. a regular expression then the function searches for a match for the regular expression within the record. If such a match is found, it is used for sorting. If a match for key-regexp is not found within a record then that record is ignored, which means its position in the buffer is not changed. (The other records may move around it.) For example, if you plan to sort all the lines in the region by the first word on each line starting with the letter `f', you should set record-regexp to `^.*$' and set key-regexp to `\<f\w*\>'. The resulting expression looks like this: (sort-regexp-fields nil "^.*$" "\\<f\\w*\\>" (region-beginning) (region-end)) If you call sort-regexp-fields interactively, you are prompted for record-regexp and key-regexp in the minibuffer. -- Command: sort-subr reverse nextrecfun endrecfun &optional startkeyfun endkeyfun This command is the general text sorting routine that divides a buffer into records and sorts them. The functions sort-lines, sort-paragraphs, sort-pages, sort-fields, sort-regexp-fields and sort-numeric-fields all use sort-subr. To understand how sort-subr works, consider the whole accessible portion of the buffer as being divided into disjoint pieces called sort records. A portion of each sort record (perhaps all of it) is designated as the sort key. The records are rearranged in the buffer in order by their sort keys. The records may or may not be contiguous. Usually, the records are rearranged in order of ascending sort key. If the first argument to the sort-subr function, reverse, is non-nil, the sort records are rearranged in order of descending sort key. The next four arguments to sort-subr are functions that are called to move point across a sort record. They are called many times from within sort-subr. 1. nextrecfun is called with point at the end of a record. This function moves point to the start of the next record. The first record is assumed to start at the position of point when sort-subr is called. (Therefore, you should usually move point to the beginning of the buffer before calling sort-subr.) 2. endrecfun is called with point within a record. It moves point to the end of the record. 3. startkeyfun is called to move point from the start of a record to the start of the sort key. This argument is optional. If supplied, the function should either return a non-nil value to be used as the sort key, or return nil to indicate that the sort key is in the buffer starting at point. In the latter case, and endkeyfun will be called to find the end of the sort key. 4. endkeyfun is called to move point from the start of the sort key to the end of the sort key. This argument is optional. If startkeyfun returns nil and this argument is omitted (or nil), then the sort key extends to the end of the record. There is no need for endkeyfun if startkeyfun returns a non-nil value. As an example of sort-subr, here is the complete function definition for sort-lines: (defun sort-lines (reverse beg end) "Sort lines in region alphabetically; arg means reverse order. Called from a program, there are three arguments: REVERSE (non-nil means reverse order), and BEG and END (the region to sort)." (interactive "P\nr") (save-restriction (narrow-to-region beg end) (goto-char (point-min)) (sort-subr reverse 'forward-line 'end-of-line))) Here forward-line moves point to the start of the next record, and end-of-line moves point to the end of record. We do not pass the arguments startkeyfun and endkeyfun, because the entire record is used as the sort key. The sort-paragraphs function is very much the same, except that its sort-subr call looks like this: (sort-subr reverse (function (lambda () (skip-chars-forward "\n \t\f"))) 'forward-paragraph))) -- Command: sort-lines reverse start end This command sorts lines in the region between start and end alphabetically. If reverse is non-nil, the sort is in reverse order. -- Command: sort-paragraphs reverse start end This command sorts paragraphs in the region between start and end alphabetically. If reverse is non-nil, the sort is in reverse order. -- Command: sort-pages reverse start end This command sorts pages in the region between start and end alphabetically. If reverse is non-nil, the sort is in reverse order. -- Command: sort-fields field start end This command sorts lines in the region between start and end, comparing them alphabetically by the fieldth field of each line. Fields are separated by whitespace and numbered starting from 1. If field is negative, sorting is by the -fieldth field from the end of the line. This command is useful for sorting tables. -- Command: sort-numeric-fields field start end This command sorts lines in the region between start and end, comparing them numerically by the fieldth field of each line. Fields are separated by whitespace and numbered starting from 1. The specified field must contain a number in each line of the region. If field is negative, sorting is by the -fieldth field from the end of the line. This command is useful for sorting tables. -- Command: sort-columns reverse &optional beg end This command sorts the lines in the region between beg and end, comparing them alphabetically by a certain range of columns. For the purpose of this command, the region includes the entire line that point is in and the entire line containing end. The column positions of beg and end bound the range of columns to sort on. If reverse is non-nil, the sort is in reverse order. One unusual thing about this command is that the entire line containing point, and the entire line containing the mark, are included in the region sorted. Note that sort-columns uses the sort utility program, and so cannot work properly on text containing tab characters. Use M-x untabify to convert tabs to spaces before sorting. The sort-columns function doesn't work in VMS, because the subprocess facilities are lacking. ΓòÉΓòÉΓòÉ 31.12. Indentation ΓòÉΓòÉΓòÉ The indentation functions are used to examine, move to, and change whitespace that is at the beginning of a line. Some of the functions can also change whitespace elsewhere on a line. Indentation always counts from zero at the left margin. ΓòÉΓòÉΓòÉ 31.12.1. Indentation Primitives ΓòÉΓòÉΓòÉ This section describes the primitive functions used to count and insert indentation. The functions in the following sections use these primitives. -- Function: current-indentation This function returns the indentation of the current line, which is the horizontal position of the first nonblank character. If the contents are entirely blank, then this is the horizontal position of the end of the line. -- Command: indent-to column &optional minimum This function indents from point with tabs and spaces until column is reached. If minimum is specified and non-nil, then at least that many spaces are inserted even if this requires going beyond column. The value is the column at which the inserted indentation ends. -- User Option: indent-tabs-mode If this variable is non-nil, indentation functions can insert tabs as well as spaces. Otherwise, they insert only spaces. Setting this variable automatically makes it local to the current buffer. ΓòÉΓòÉΓòÉ 31.12.2. Indentation Controlled by Major Mode ΓòÉΓòÉΓòÉ An important function of each major mode is to customize the TAB key to indent properly for the language being edited. This section describes the mechanism of the TAB key and how to control it. The functions in this section return unpredictable values. -- Variable: indent-line-function This variable's value is the function to be used by TAB (and various commands) to indent the current line. The command indent-according-to-mode does no more than call this function. In Lisp mode, the value is the symbol lisp-indent-line; in C mode, c-indent-line; in Fortran mode, fortran-indent-line. In Fundamental mode, Text mode, and many other modes with no standard for indentation, the value is indent-to-left-margin (which is the default value). -- Command: indent-according-to-mode This command calls the function in indent-line-function to indent the current line in a way appropriate for the current major mode. -- Command: indent-for-tab-command This command calls the function in indent-line-function to indent the current line, except that if that function is indent-to-left-margin, insert-tab is called instead. -- Variable: left-margin This variable is the column to which the default indent-line-function will indent. (That function is indent-to-left-margin.) In Fundamental mode, LFD indents to this column. This variable automatically becomes buffer-local when set in any fashion. -- Function: indent-to-left-margin This is the default indent-line-function, used in Fundamental mode, Text mode, etc. Its effect is to adjust the indentation at the beginning of the current line to the value specified by the variable left-margin. This may involve either inserting or deleting whitespace. -- Command: newline-and-indent This function inserts a newline, then indents the new line (the one following the newline just inserted) according to the major mode. Indentation is done using the current indent-line-function. In programming language modes, this is the same thing TAB does, but in some text modes, where TAB inserts a tab, newline-and-indent indents to the column specified by left-margin. -- Command: reindent-then-newline-and-indent This command reindents the current line, inserts a newline at point, and then reindents the new line (the one following the newline just inserted). Indentation of both lines is done according to the current major mode; this means that the current value of indent-line-function is called. In programming language modes, this is the same thing TAB does, but in some text modes, where TAB inserts a tab, reindent-then-newline-and-indent indents to the column specified by left-margin. ΓòÉΓòÉΓòÉ 31.12.3. Indenting an Entire Region ΓòÉΓòÉΓòÉ This section describes commands which indent all the lines in the region. They return unpredictable values. -- Command: indent-region start end to-column This command indents each nonblank line starting between start (inclusive) and end (exclusive). If to-column is nil, indent-region indents each nonblank line by calling the current mode's indentation function, the value of indent-line-function. If to-column is non-nil, it should be an integer specifying the number of columns of indentation; then this function gives each line exactly that much indentation, by either adding or deleting whitespace. -- Variable: indent-region-function The value of this variable is a function that can be used by indent-region as a short cut. You should design the function so that it will produce the same results as indenting the lines of the region one by one (but presumably faster). If the value is nil, there is no short cut, and indent-region actually works line by line. A short cut function is useful in modes such as C mode and Lisp mode, where the indent-line-function must scan from the beginning of the function: applying it to each line would be quadratic in time. The short cut can update the scan information as it moves through the lines indenting them; this takes linear time. If indenting a line individually is fast, there is no need for a short cut. indent-region with a non-nil argument has a different definition and does not use this variable. -- Command: indent-rigidly start end count This command indents all lines starting between start (inclusive) and end (exclusive) sideways by count columns. This ``preserves the shape'' of the affected region, moving it as a rigid unit. Consequently, this command is useful not only for indenting regions of unindented text, but also for indenting regions of formatted code. For example, if count is 3, this command adds 3 columns of indentation to each of the lines beginning in the region specified. In Mail mode, C-c C-y (mail-yank-original) uses indent-rigidly to indent the text copied from the message being replied to. ΓòÉΓòÉΓòÉ 31.12.4. Indentation Relative to Previous Lines ΓòÉΓòÉΓòÉ This section describes two commands which indent the current line based on the contents of previous lines. -- Command: indent-relative &optional unindented-ok This function inserts whitespace at point, extending to the same column as the next indent point of the previous nonblank line. An indent point is a non-whitespace character following whitespace. The next indent point is the first one at a column greater than the current column of point. For example, if point is underneath and to the left of the first non-blank character of a line of text, it moves to that column by inserting whitespace. If the previous nonblank line has no next indent point (i.e., none at a great enough column position), this function either does nothing (if unindented-ok is non-nil) or calls tab-to-tab-stop. Thus, if point is underneath and to the right of the last column of a short line of text, this function moves point to the next tab stop by inserting whitespace. This command returns an unpredictable value. In the following example, point is at the beginning of the second line: This line is indented twelve spaces. -!-The quick brown fox jumped over the lazy dog. Evaluation of the expression (indent-relative nil) produces the following: This line is indented twelve spaces. -!-The quick brown fox jumped over the lazy dog. In this example, point is between the `m' and `p' of `jumped': This line is indented twelve spaces. The quick brown fox jum-!-ped over the lazy dog. Evaluation of the expression (indent-relative nil) produces the following: This line is indented twelve spaces. The quick brown fox jum -!-ped over the lazy dog. -- Command: indent-relative-maybe This command indents the current line like the previous nonblank line. The function consists of a call to indent-relative with a non-nil value passed to the unindented-ok optional argument. The value is unpredictable. If the previous line has no indentation, the current line is given no indentation (any existing indentation is deleted); if the previous nonblank line has no indent points beyond the column at which point starts, nothing is changed. ΓòÉΓòÉΓòÉ 31.12.5. Adjustable ``Tab Stops'' ΓòÉΓòÉΓòÉ This section explains the mechanism for user-specified ``tab stops'' and the mechanisms which use and set them. The name ``tab stops'' is used because the feature is similar to that of the tab stops on a typewriter. The feature works by inserting an appropriate number of spaces and tab characters to reach the designated position, like the other indentation functions; it does not affect the display of tab characters in the buffer (see Control Char Display). Note that the TAB character as input uses this tab stop feature only in a few major modes, such as Text mode. -- Function: tab-to-tab-stop This function inserts spaces or tabs up to the next tab stop column defined by tab-stop-list. It searches the list for an element greater than the current column number, and uses that element as the column to indent to. If no such element is found, then nothing is done. -- User Option: tab-stop-list This variable is the list of tab stop columns used by tab-to-tab-stops. The elements should be integers in increasing order. The tab stop columns need not be evenly spaced. Use M-x edit-tab-stops to edit the location of tab stops interactively. ΓòÉΓòÉΓòÉ 31.12.6. Indentation-Based Motion Commands ΓòÉΓòÉΓòÉ These commands, primarily for interactive use, act based on the indentation in the text. -- Command: back-to-indentation This command moves point to the first non-whitespace character in the current line (which is the line in which point is located). It returns nil. -- Command: backward-to-indentation arg This command moves point backward arg lines and then to the first nonblank character on that line. It returns nil. -- Command: forward-to-indentation arg This command moves point forward arg lines and then to the first nonblank character on that line. It returns nil. ΓòÉΓòÉΓòÉ 31.13. Counting Columns ΓòÉΓòÉΓòÉ The column functions convert between a character position (counting characters from the beginning of the buffer) and a column position (counting screen characters from the beginning of a line). Column number computations ignore the width of the window and the amount of horizontal scrolling. Consequently, a column value can be arbitrarily high. The first (or leftmost) column is numbered 0. A character counts according to the number of columns it occupies on the screen. This means control characters count as occupying 2 or 4 columns, depending upon the value of ctl-arrow, and tabs count as occupying a number of columns that depends on the value of tab-width and on the column where the tab begins. See Control Char Display. -- Function: current-column This function returns the horizontal position of point, measured in columns, counting from 0 at the left margin. The column count is calculated by adding together the widths of all the displayed representations of the characters between the start of the current line and point. For a more complicated example of the use of current-column, see the description of count-lines in Text Lines. -- Function: move-to-column column This function moves point to column in the current line. The calculation of column takes into account the widths of all the displayed representations of the characters between the start of the line and point. If the argument column is greater than the column position of the end of the line, point moves to the end of the line. If column is negative, point moves to the beginning of the line. An error is signaled if column is not an integer. The return value is the column number actually moved to. ΓòÉΓòÉΓòÉ 31.14. Case Changes ΓòÉΓòÉΓòÉ The case change commands described here work on text in the current buffer. See Character Case, for case conversion commands that work on strings and characters. -- Command: capitalize-region start end This function capitalizes all words in the region defined by start and end. To capitalize means to convert each word's first character to upper case and convert the rest of each word to lower case. The function returns nil. If one end of the region is in the middle of a word, the part of the word within the region is treated as an entire word. When capitalize-region is called interactively, start and end are point and the mark, with the smallest first. ---------- Buffer: foo ---------- This is the contents of the 5th foo. ---------- Buffer: foo ---------- (capitalize-region 1 44) => nil ---------- Buffer: foo ---------- This Is The Contents Of The 5th Foo. ---------- Buffer: foo ---------- -- Command: downcase-region start end This function converts all of the letters in the region defined by start and end to lower case. The function returns nil. When downcase-region is called interactively, start and end are point and the mark, with the smallest first. -- Command: upcase-region start end This function converts all of the letters in the region defined by start and end to upper case. The function returns nil. When upcase-region is called interactively, start and end are point and the mark, with the smallest first. -- Command: capitalize-word count This function capitalizes count words after point, moving point over as it does. To capitalize means to convert each word's first character to upper case and convert the rest of each word to lower case. If count is negative, the function capitalizes the -count previous words but does not move point. The value is nil. If point is in the middle of a word, the part of word the before point (if moving forward) or after point (if operating backward) is ignored. The rest is treated as an entire word. When capitalize-word is called interactively, count is set to the numeric prefix argument. -- Command: downcase-word count This function converts the count words after point to all lower case, moving point over as it does. If count is negative, it converts the -count previous words but does not move point. The value is nil. When downcase-word is called interactively, count is set to the numeric prefix argument. -- Command: upcase-word count This function converts the count words after point to all upper case, moving point over as it does. If count is negative, it converts the -count previous words but does not move point. The value is nil. When upcase-word is called interactively, count is set to the numeric prefix argument. ΓòÉΓòÉΓòÉ 31.15. Substituting for a Character Code ΓòÉΓòÉΓòÉ The following function replaces characters within a specified region based on their character code. -- Function: subst-char-in-region start end old-char new-char &optional noundo This function replaces all occurrences of the character old-char with the character new-char in the region of the current buffer defined by start and end. If noundo is non-nil, then subst-char-in-region does not record the change for undo and does not mark the buffer as modified. This feature is useful for changes which are not considered significant, such as when Outline mode changes visible lines to invisible lines and vice versa. subst-char-in-region does not move point and returns nil. ---------- Buffer: foo ---------- This is the contents of the buffer before. ---------- Buffer: foo ---------- (subst-char-in-region 1 20 ?i ?X) => nil ---------- Buffer: foo ---------- ThXs Xs the contents of the buffer before. ---------- Buffer: foo ---------- ΓòÉΓòÉΓòÉ 31.16. Underlining ΓòÉΓòÉΓòÉ The underlining commands are somewhat obsolete. The underline-region function actually inserts `_^H' before each appropriate character in the region. This command provides a minimal text formatting feature that might work on your printer; however, we recommend instead that you use more powerful text formatting facilities, such as Texinfo. -- Command: underline-region start end This function underlines all nonblank characters in the region defined by start and end. That is, an underscore character and a backspace character are inserted just before each non-whitespace character in the region. The backspace characters are intended to cause overstriking, but in Emacs they display as either `\010' or `^H', depending on the setting of ctl-arrow. There is no way to see the effect of the overstriking within Emacs. The value is nil. -- Command: ununderline-region start end This function removes all underlining (overstruck underscores) in the region defined by start and end. The value is nil. ΓòÉΓòÉΓòÉ 31.17. Registers ΓòÉΓòÉΓòÉ A register is a sort of variable used in Emacs editing that can hold a marker, a string, or a rectangle. Each register is named by a single character. All characters, including control and meta characters (but with the exception of C-g), can be used to name registers. Thus, there are 255 possible registers. A register is designated in Emacs Lisp by a character which is its name. The functions in this section return unpredictable values unless otherwise stated. -- Variable: register-alist This variable is an alist of elements of the form (name . contents). Normally, there is one element for each Emacs register that has been used. The object name is a character (an integer) identifying the register. The object contents is a string, marker, or list representing the register contents. A string represents text stored in the register. A marker represents a position. A list represents a rectangle; its elements are strings, one per line of the rectangle. -- Command: view-register reg This command displays what is contained in register reg. -- Function: get-register reg This function returns the contents of the register reg, or nil if it has no contents. -- Function: set-register reg value This function sets the contents of register reg to value. A register can be set to any value, but the other register functions expect only strings, markers, and lists. -- Command: point-to-register reg This command stores both the current location of point and the current buffer in register reg as a marker. -- Command: register-to-point reg This command moves point to the position stored in register reg. Since both the buffer and the location within the buffer are stored by the point-to-register function, this command can switch you to another buffer. If the register does not contain a saved position (a marker), then an error is signaled. -- Command: insert-register reg &optional beforep This command inserts contents of register reg into the current buffer. Normally, this command puts point before the inserted text, and the mark after it. However, if the optional second argument beforep is non-nil, it puts the mark before and point after. You can pass a non-nil second argument beforep to this function interactively by supplying any prefix argument. If the register contains a rectangle, then the rectangle is inserted with its upper left corner at point. This means that text is inserted in the current line and underneath it on successive lines. If the register contains something other than saved text (a string) or a rectangle (a list), currently useless things happen. This may be changed in the future. -- Command: copy-to-register reg start end &optional delete-flag This command copies the region from start to end into register reg. If delete-flag is non-nil, it deletes the region from the buffer after copying it into the register. -- Command: prepend-to-register reg start end &optional delete-flag This command prepends the region from start to end into register reg. If delete-flag is non-nil, it deletes the region from the buffer after copying it to the register. -- Command: append-to-register reg start end &optional delete-flag This command appends the region from start to end to the text already in register reg. If delete-flag is non-nil, it deletes the region from the buffer after copying it to the register. -- Command: copy-rectangle-to-register reg start end &optional delete-flag This command copies a rectangular region from start to end into register reg. If delete-flag is non-nil, it deletes the region from the buffer after copying it to the register. ΓòÉΓòÉΓòÉ 32. Searching and Matching ΓòÉΓòÉΓòÉ GNU Emacs provides two ways to search through a buffer for specified text: exact string searches and regular expression searches. After a regular expression search, you can identify the text matched by parts of the regular expression by examining the match data. ΓòÉΓòÉΓòÉ 32.1. Searching for Strings ΓòÉΓòÉΓòÉ These are the primitive functions for searching through the text in a buffer. They are meant for use in programs, but you may call them interactively. If you do so, they prompt for the search string; limit and noerror are set to nil, and repeat is set to 1. -- Command: search-forward string &optional limit noerror repeat This function searches forward from point for an exact match for string. It sets point to the end of the occurrence found, and returns t. If no match is found, the value and side effects depend on noerror. In the following example, point is positioned at the beginning of the line. Then (search-forward "fox") is evaluated in the minibuffer and point is left after the last letter of `fox': ---------- Buffer: foo ---------- -!-The quick brown fox jumped over the lazy dog. ---------- Buffer: foo ---------- (search-forward "fox") => t ---------- Buffer: foo ---------- The quick brown fox-!- jumped over the lazy dog. ---------- Buffer: foo ---------- If limit is non-nil, then it is the upper bound to the search. (It must be a position in the current buffer.) No match extending after that position is accepted. What happens when the search fails depends on the value of noerror. If noerror is nil, a search-failed error is signaled. If noerror is t, search-forward returns nil and doesn't signal an error. If noerror is neither nil nor t, then search-forward moves point to limit and returns nil. If repeat is non-nil, then the search is repeated that many times. Point is positioned at the end of the last match. -- Command: search-backward string &optional limit noerror repeat This function searches backward from point for string. It is just like search-forward except that it searches backwards and leaves point at the beginning of the match. -- Command: word-search-forward string &optional limit noerror repeat This function searches forward from point for a ``word'' match for string. It sets point to the end of the occurrence found, and returns t. A word search differs from a simple string search in that a word search *requires* that the words it searches for are present as entire words (searching for the word `ball' will not match the word `balls'), and punctuation and spacing are ignored (searching for `ball boy' will match `ball. Boy!'). In this example, point is first placed at the beginning of the buffer; the search leaves it between the y and the !. ---------- Buffer: foo ---------- -!-He said "Please! Find the ball boy!" ---------- Buffer: foo ---------- (word-search-forward "Please find the ball, boy.") => t ---------- Buffer: foo ---------- He said "Please! Find the ball boy-!-!" ---------- Buffer: foo ---------- If limit is non-nil (it must be a position in the current buffer), then it is the upper bound to the search. The match found must not extend after that position. If noerror is t, then word-search-forward returns nil when a search fails, instead of signaling an error. If noerror is neither nil nor t, then word-search-forward moves point to limit and returns nil. If repeat is non-nil, then the search is repeated that many times. Point is positioned at the end of the last match. -- Command: word-search-backward string &optional limit noerror repeat This function searches backward from point for a word match to string. This function is just like word-search-forward except that it searches backward and normally leaves point at the beginning of the match. ΓòÉΓòÉΓòÉ 32.2. Regular Expressions ΓòÉΓòÉΓòÉ A regular expression (regexp, for short) is a pattern that denotes a (possibly infinite) set of strings. Searching for matches for a regexp is a very powerful operation. This section explains how to write regexps; the following section says how to search for them. ΓòÉΓòÉΓòÉ 32.2.1. Syntax of Regular Expressions ΓòÉΓòÉΓòÉ Regular expressions have a syntax in which a few characters are special constructs and the rest are ordinary. An ordinary character is a simple regular expression which matches that character and nothing else. The special characters are `$', `^', `.', `*', `+', `?', `[', `]' and `\'; no new special characters will be defined in the future. Any other character appearing in a regular expression is ordinary, unless a `\' precedes it. For example, `f' is not a special character, so it is ordinary, and therefore `f' is a regular expression that matches the string `f' and no other string. (It does not match the string `ff'.) Likewise, `o' is a regular expression that matches only `o'. Any two regular expressions a and b can be concatenated. The result is a regular expression which matches a string if a matches some amount of the beginning of that string and b matches the rest of the string. As a simple example, we can concatenate the regular expressions `f' and `o' to get the regular expression `fo', which matches only the string `fo'. Still trivial. To do something more powerful, you need to use one of the special characters. Here is a list of them: . (Period) is a special character that matches any single character except a newline. Using concatenation, we can make regular expressions like `a.b' which matches any three-character string which begins with `a' and ends with `b'. * is not a construct by itself; it is a suffix that means the preceding regular expression is to be repeated as many times as possible. In `fo*', the `*' applies to the `o', so `fo*' matches one `f' followed by any number of `o's. The case of zero `o's is allowed: `fo*' does match `f'. `*' always applies to the smallest possible preceding expression. Thus, `fo*' has a repeating `o', not a repeating `fo'. The matcher processes a `*' construct by matching, immediately, as many repetitions as can be found. Then it continues with the rest of the pattern. If that fails, backtracking occurs, discarding some of the matches of the `*'-modified construct in case that makes it possible to match the rest of the pattern. For example, matching `ca*ar' against the string `caaar', the `a*' first tries to match all three `a's; but the rest of the pattern is `ar' and there is only `r' left to match, so this try fails. The next alternative is for `a*' to match only two `a's. With this choice, the rest of the regexp matches successfully. + is a suffix character similar to `*' except that it must match the preceding expression at least once. So, for example, `ca+r' will match the strings `car' and `caaaar' but not the string `cr', whereas `ca*r' would match all three strings. ? is a suffix character similar to `*' except that it can match the preceding expression either once or not at all. For example, `ca?r' will match `car' or `cr'; nothing else. [ ... ] `[' begins a character set, which is terminated by a `]'. In the simplest case, the characters between the two form the set. Thus, `[ad]' matches either one `a' or one `d', and `[ad]*' matches any string composed of just `a's and `d's (including the empty string), from which it follows that `c[ad]*r' matches `cr', `car', `cdr', `caddaar', etc. Character ranges can also be included in a character set, by writing two characters with a `-' between them. Thus, `[a-z]' matches any lower case letter. Ranges may be intermixed freely with individual characters, as in `[a-z$%.]', which matches any lower case letter or `$', `%' or a period. Note that the usual special characters are not special any more inside a character set. A completely different set of special characters exists inside character sets: `]', `-' and `^'. To include a `]' in a character set, make it the first character. For example, `[]a]' matches `]' or `a'. To include a `-', write `---', which is a range containing only `-', or write `-' as the first character in the range. To include `^', make it other than the first character in the set. [^ ... ] `[^' begins a complement character set, which matches any character except the ones specified. Thus, `[^a-z0-9A-Z]' matches all characters except letters and digits. `^' is not special in a character set unless it is the first character. The character following the `^' is treated as if it were first (thus, `-' and `]' are not special there). Note that a complement character set can match a newline, unless newline is mentioned as one of the characters not to match. ^ is a special character that matches the empty string, but only at the beginning of a line in the text being matched. Otherwise it fails to match anything. Thus, `^foo' matches a `foo' which occurs at the beginning of a line. When matching a string, `^' matches at the beginning of the string or after a newline character `\n'. $ is similar to `^' but matches only at the end of a line. Thus, `x+$' matches a string of one `x' or more at the end of a line. When matching a string, `$' matches at the end of the string or before a newline character `\n'. \ has two functions: it quotes the special characters (including `\'), and it introduces additional special constructs. Because `\' quotes special characters, `\$' is a regular expression which matches only `$', and `\[' is a regular expression which matches only `[', and so on. Note that `\' also has special meaning in the read syntax of Lisp strings (see String Type), and must be quoted with `\'. For example, the regular expression that matches the `\' character is `\\'. To write a Lisp string that contains `\\', Lisp syntax requires you to quote each `\' with another `\'. Therefore, the read syntax for this string is "\\\\". *Note:* for historical compatibility, special characters are treated as ordinary ones if they are in contexts where their special meanings make no sense. For example, `*foo' treats `*' as ordinary since there is no preceding expression on which the `*' can act. It is poor practice to depend on this behavior; better to quote the special character anyway, regardless of where it appears. For the most part, `\' followed by any character matches only that character. However, there are several exceptions: characters which, when preceded by `\', are special constructs. Such characters are always ordinary when encountered on their own. Here is a table of `\' constructs: \| specifies an alternative. Two regular expressions a and b with `\|' in between form an expression that matches anything that either a or b matches. Thus, `foo\|bar' matches either `foo' or `bar' but no other string. `\|' applies to the largest possible surrounding expressions. Only a surrounding `$ ... $' grouping can limit the grouping power of `\|'. Full backtracking capability exists to handle multiple uses of `\|'. $ ... $ is a grouping construct that serves three purposes: 1. To enclose a set of `\|' alternatives for other operations. Thus, `$foo\|bar$x' matches either `foox' or `barx'. 2. To enclose a complicated expression for a suffix character such as `*' to operate on. Thus, `ba$na$*' matches `bananana', etc., with any (zero or more) number of `na' strings. 3. To record a matched substring for future reference. This last application is not a consequence of the idea of a parenthetical grouping; it is a separate feature which happens to be assigned as a second meaning to the same `$ ... $' construct because there is no conflict in practice between the two meanings. Here is an explanation of this feature: \digit after the end of a `$ ... $' construct, the matcher remembers the beginning and end of the text matched by that construct. Then, later on in the regular expression, you can use `\' followed by digit to mean ``match the same text matched the digitth time by the `$ ... $' construct.'' The strings matching the first nine `$ ... $' constructs appearing in a regular expression are assigned numbers 1 through 9 in the order that the open parentheses appear in the regular expression. `\1' through `\9' can be used to refer to the text matched by the corresponding `$ ... $' construct. For example, `$.*$\1' matches any newline-free string that is composed of two identical halves. The `$.*$' matches the first half, which may be anything, but the `\1' that follows must match the same exact text. \` matches the empty string, provided it is at the beginning of the buffer. \' matches the empty string, provided it is at the end of the buffer. \b matches the empty string, provided it is at the beginning or end of a word. Thus, `\bfoo\b' matches any occurrence of `foo' as a separate word. `\bballs?\b' matches `ball' or `balls' as a separate word. \B matches the empty string, provided it is not at the beginning or end of a word. \< matches the empty string, provided it is at the beginning of a word. \> matches the empty string, provided it is at the end of a word. \w matches any word-constituent character. The editor syntax table determines which characters these are. See Syntax Tables. \W matches any character that is not a word-constituent. \scode matches any character whose syntax is code. Here code is a character which represents a syntax code: thus, `w' for word constituent, `-' for whitespace, `(' for open parenthesis, etc. See Syntax Tables, for a list of the codes. \Scode matches any character whose syntax is not code. Not every string is a valid regular expression. For example, any string with unbalanced square brackets is invalid, and so is a string that ends with a single `\'. If an invalid regular expression is passed to any of the search functions, an invalid-regexp error is signaled. -- Function: regexp-quote string This function returns a regular expression string which matches exactly string and nothing else. This allows you to request an exact string match when calling a function that wants a regular expression. (regexp-quote "^The cat$") => "\\^The cat\\$" One use of regexp-quote is to combine an exact string match with context described as a regular expression. For example, this searches for the string which is the value of string, surrounded by whitespace: (re-search-forward (concat "\\s " (regexp-quote string) "\\s ")) ΓòÉΓòÉΓòÉ 32.2.2. Complex Regexp Example ΓòÉΓòÉΓòÉ Here is a complicated regexp, used by Emacs to recognize the end of a sentence together with any whitespace that follows. It is the value of the variable sentence-end. First, we show the regexp as a string in Lisp syntax to enable you to distinguish the spaces from the tab characters. The string constant begins and ends with a double-quote. `\"' stands for a double-quote as part of the string, `\\' for a backslash as part of the string, `\t' for a tab and `\n' for a newline. "[.?!][]\"')}]*\$$\\|\t\\| \$[ \t\n]*" In contrast, if you evaluate the variable sentence-end, you will see the following: sentence-end => "[.?!][]\"')}]*\$$\\| \\| \$[ ]*" In this case, the tab and carriage return are the actual characters. This regular expression contains four parts in succession and can be deciphered as follows: [.?!] The first part of the pattern consists of three characters, a period, a question mark and an exclamation mark, within square brackets. The match must begin with one of these three characters. []\"')}]* The second part of the pattern matches any closing braces and quotation marks, zero or more of them, that may follow the period, question mark or exclamation mark. The \" is Lisp syntax for a double-quote in a string. The `*' at the end indicates that the immediately preceding regular expression (a character set, in this case) may be repeated zero or more times. \$$\\|\t\\|@ @ \$ The third part of the pattern matches the whitespace that follows the end of a sentence: the end of a line, or a tab, or two spaces. The double backslashes are needed to prevent Emacs from reading the parentheses and vertical bars as part of the search pattern; the parentheses are used to mark the group and the vertical bars are used to indicated that the patterns to either side of them are alternatives. The dollar sign is used to match the end of a line. The tab character is written using `\t' and the two spaces are written as themselves. [ \t\n]* Finally, the last part of the pattern indicates that the end of the line or the whitespace following the period, question mark or exclamation mark may, but need not, be followed by additional whitespace. ΓòÉΓòÉΓòÉ 32.3. Regular Expression Searching ΓòÉΓòÉΓòÉ In GNU Emacs, you can search for the next match for a regexp either incrementally or not. Incremental search commands are described in the The GNU Emacs Manual. See Regular Expression Search. Here we describe only the search functions useful in programs. The principal is re-search-forward. -- Command: re-search-forward regexp &optional limit noerror repeat This function searches forward in the current buffer for a string of text that is matched by the regular expression regexp. The function skips over any amount of text that is not matched by regexp, and leaves point at the end of the first string found that does match. If the search is successful (i.e., if text matching regexp is found), then point is left at the end of that text, and the function returns t. What happens when the search fails depends on the value of noerror. If noerror is nil, a search-failed error is signaled. If noerror is t, re-search-forward returns nil and doesn't signal an error. If noerror is neither nil nor t, then search-forward moves point to limit and returns nil. If limit is non-nil (it must be a position in the current buffer), then it is the upper bound to the search. No match extending after that position is accepted. If repeat is supplied (it must be a positive number), then the search is repeated that many times (each time starting at the end of the previous time's match). The call succeeds if all these searches succeeded, and point is left at the end of the match found by the last search. Otherwise the search fails. In the following example, point is initially located directly before the `T'. After evaluating the form, point is located at the end of that line (between the `t' of `hat' and before the newline). ---------- Buffer: foo ---------- I read "-!-The cat in the hat comes back" twice. ---------- Buffer: foo ---------- (re-search-forward "[a-z]+" nil t 5) => t ---------- Buffer: foo ---------- I read "The cat in the hat-!- comes back" twice. ---------- Buffer: foo ---------- -- Command: re-search-backward regexp &optional limit noerror repeat This function searches backward in the current buffer for a string of text that is matched by the regular expression regexp, leaving point at the beginning of the first text found. This function is analogous to re-search-forward, but they are not simple mirror images. re-search-forward finds the match whose beginning is as close as possible. If re-search-backward were a perfect mirror image, it would find the match whose end is as close as possible. However, in fact it finds the match whose beginning is as close as possible. The reason is that matching a regular expression at a given spot always works from beginning to end, and is done at a specified beginning position. Thus, true mirror-image behavior would require a special feature for matching regexps from end to beginning. -- Function: string-match regexp string &optional start This function returns the index of the start of the first match for the regular expression regexp in string, or nil if there is no match. If start is non-nil, the search starts at that index in string. For example, (string-match "quick" "The quick brown fox jumped quickly.") => 4 (string-match "quick" "The quick brown fox jumped quickly." 8) => 27 The index of the first character of the string is 0, the index of the second character is 1, and so on. After this function returns, the index of the first character beyond the match is available as (match-end 0). See Match Data. (string-match "quick" "The quick brown fox jumped quickly." 8) => 27 (match-end 0) => 32 The match-end function is described along with match-beginning; see Match Data. -- Function: looking-at regexp This function determines whether the text in the current buffer directly following point matches the regular expression regexp. ``Directly following'' means precisely that: the search is ``anchored'' and it must succeed starting with the first character following point. The result is t if so, nil otherwise. Point is not moved, but the match data is updated and can be used with match-beginning or match-end. See Match Data. In this example, point is located directly before the `T'. If it were anywhere else, the result would be nil. ---------- Buffer: foo ---------- I read "-!-The cat in the hat comes back" twice. ---------- Buffer: foo ---------- (looking-at "The cat in the hat$") => t ΓòÉΓòÉΓòÉ 32.4. The Match Data ΓòÉΓòÉΓòÉ Emacs keeps track of the positions of the start and end of segments of text found during a regular expression search. This means, for example, that you can search for a complex pattern, such as a date in an Rmail message, and extract parts of it. -- Function: match-beginning count This function returns the position of the start of text matched by the last regular expression searched for. count, a number, specifies a subexpression whose start position is the value. If count is zero, then the value is the position of the text matched by the whole regexp. If count is greater than zero, then the value is the position of the beginning of the text matched by the countth subexpression, regardless of whether it was used in the final match. Subexpressions of a regular expression are those expressions grouped inside of parentheses, `$...$'. The countth subexpression is found by counting occurrences of `$' from the beginning of the whole regular expression. The first subexpression is numbered 1, the second 2, and so on. The match-end function is similar to the match-beginning function except that it returns the position of the end of the matched text. Here is an example, with a comment showing the numbers of the positions in the text: (string-match "\\(qu\$\$ick\$" "The quick fox jumped quickly.") => 4 ;^^^^^^^^^^ ;0123456789 (match-beginning 1) ; The beginning of the match => 4 ; with `qu' is at index 4. (match-beginning 2) ; The beginning of the match => 6 ; with `ick' is at index 6. (match-end 1) ; The end of the match => 6 ; with `qu' is at index 6. (match-end 2) ; The end of the match => 9 ; with `ick' is at index 9. Here is another example. Before the form is evaluated, point is located at the beginning of the line. After evaluating the search form, point is located on the line between the space and the word `in'. The beginning of the entire match is at the 9th character of the buffer (`T'), and the beginning of the match for the first subexpression is at the 13th character (`c'). (list (re-search-forward "The \$cat \$") (match-beginning 0) (match-beginning 1)) => (t 9 13) ---------- Buffer: foo ---------- I read "The cat -!-in the hat comes back" twice. ^ ^ 9 13 ---------- Buffer: foo ---------- (Note that in this case, the index returned is a buffer position; the first character of the buffer counts as 1.) It is essential that match-beginning be called after the search desired, but before any other searches are performed. match-beginning may not give the desired results if any other Lisp programs are executed between the search and it, since they may do other searches. This example shows misuse of match-beginning. (re-search-forward "The \$cat \$") => t (foo) ; Perhaps foo does more regexp searching. (match-beginning 0) => 61 ; Unexpected result! See the discussion of store-match-data for an example of how to save and restore the match data around a search. -- Function: match-end count This function returns the position of the end of text matched by the last regular expression searched for. This function is otherwise similar to match-beginning. -- Function: replace-match replacement &optional fixedcase literal This function replaces the text matched by the last search with replacement. If fixedcase is non-nil, then the case of the replacement text is not changed; otherwise, the replacement text is converted to a different case depending upon the capitalization of the text to be replaced. If the original text is all upper case, the replacement text is converted to upper case, except when all of the words in the original text are only one character long. In that event, the replacement text is capitalized. If all of the words in the original text are capitalized, then all of the words in the replacement text are capitalized. If literal is non-nil, then replacement is inserted exactly as it is, the only alterations being case changes as needed. If it is nil (the default), then the character `\' is treated specially. If a `\' appears in replacement, then it must be part of one of the following sequences: `\&' `\&' stands for the entire text being replaced. `\n' `\n' stands for the nth subexpression in the original regexp. Subexpressions are those expressions grouped inside of `$...$'. n is a digit. `\\' `\\' stands for a single `\' in the replacement text. replace-match leaves point at the end of the replacement text, and returns t. ΓòÉΓòÉΓòÉ 32.5. Saving and Restoring the Match Data ΓòÉΓòÉΓòÉ -- Function: match-data This function returns a new list containing all the information on what text the last search matched. Element zero is the position of the beginning of the match for the whole expression; element one is the position of the end of the match for the expression. The next two elements are the positions of the beginning and end of the match for the first subexpression. In general, element number 2n corresponds to (match-beginning n); and element number 2n + 1 corresponds to (match-end n). All the elements are markers, or the integer 0 for a match at the beginning of a string (with string-match), or nil if there was no match for that subexpression. As with other functions that get information about a search, there must be no possibility of intervening searches between the call to a search function and the call to match-data that is intended to save the match-data for that search. (match-data) => (#<marker at 9 in foo> #<marker at 17 in foo> #<marker at 13 in foo> #<marker at 17 in foo>) In version 19, all elements will be markers or nil if matching was done on a buffer, and all will be integers or nil if matching was done on a string with string-match. -- Function: store-match-data match-list This function sets the match data within Emacs Lisp from the elements of match-list, which should be a list created by a previous call to match-data. store-match-data may be used together with match-data to perform a search without changing the match-data. This is useful when such searches occur in subroutines whose callers may not expect searches to go on. Here is how: (let ((data (match-data))) (unwind-protect ... ; May change the original match data. (store-match-data data))) All asynchronous process functions (filters and sentinels) and some modes that use recursive-edit should save and restore the match data if they do a search or if they let the user type arbitrary commands. Here is a function which will restore the match data if the buffer associated with it still exists. (defun restore-match-data (data) "Restore the match data DATA unless the buffer is missing." (catch 'foo (let ((d data)) (while d (and (car d) (null (marker-buffer (car d))) ;; match-data buffer is deleted. (throw 'foo nil)) (setq d (cdr d))) (store-match-data data)))) ΓòÉΓòÉΓòÉ 32.6. Standard Regular Expressions Used in Editing ΓòÉΓòÉΓòÉ -- Variable: page-delimiter This is the regexp describing line-beginnings that separate pages. The default value is "^\014" (i.e., "^^L" or "^\C-l"). -- Variable: paragraph-separate This is the regular expression for recognizing the beginning of a line that separates paragraphs. (If you change this, you may have to change paragraph-start also.) The default value is "^[ \t\f]*$", which is a line that consists entirely of spaces, tabs, and form feeds. -- Variable: paragraph-start This is the regular expression for recognizing the beginning of a line that starts or separates paragraphs. The default value is "^[ \t\n\f]", which matches a line starting with a space, tab, newline, or form feed. -- Variable: sentence-end This is the regular expression describing the end of a sentence. (All paragraph boundaries also end sentences, regardless.) The default value is: "[.?!][]\"')}]*\$$\\|\t\\| \$[ \t\n]*" This means a period, question mark or exclamation mark, followed by a closing brace, followed by tabs, spaces or new lines. For a detailed explanation of this regular expression, see Regexp Example. ΓòÉΓòÉΓòÉ 32.7. Searching and Case ΓòÉΓòÉΓòÉ By default, searches in Emacs ignore the case of the text they are searching through; if you specify searching for `FOO', then `Foo' or `foo' is also considered a match. Regexps, and in particular character sets, are included: thus, `[aB]' would match `a' or `A' or `b' or `B'. If you do not want this feature, set the variable case-fold-search to nil. Then all letters must match exactly, including case. This is a per-buffer-local variable; altering the variable affects only the current buffer. ( See Intro to Buffer-Local.) Alternatively, you may change the value of default-case-fold-search, which is the default value of case-fold-search for buffers that do not override it. -- User Option: case-replace This variable determines whether query-replace should preserve case in replacements. If the variable is nil, then case need not be preserved. -- User Option: case-fold-search This buffer-local variable determines whether searches should ignore case. If the variable is nil they do not ignore case; otherwise they do ignore case. -- Variable: default-case-fold-search The value of this variable is the default value for case-fold-search in buffers that do not override it. This is the same as (default-value 'case-fold-search). ΓòÉΓòÉΓòÉ 33. Syntax Tables ΓòÉΓòÉΓòÉ A syntax table provides Emacs with the information that determines the syntactic use of each character in a buffer. This information is used by the parsing commands, the complex movement commands, and others to determine where words, symbols, and other syntactic constructs begin and end. A syntax table is a vector of 256 elements; it contains one entry for each of the 256 ASCII characters of an 8-bit byte. Each element is an integer that encodes the syntax of the character in question. Syntax tables are used only for moving across text, not for the GNU Emacs Lisp reader. GNU Emacs Lisp uses built-in syntactic rules when reading Lisp expressions, and these rules cannot be changed. Each buffer has its own major mode, and each major mode has its own idea of the syntactic class of various characters. For example, in Lisp mode, the character `;' begins a comment, but in C mode, it terminates a statement. To support these variations, Emacs makes the choice of syntax table local to the each buffer. Typically, each major mode has its own syntax table and installs that table in each buffer which uses that mode. Changing this table alters the syntax in all those buffers as well as in any buffers subsequently put in that mode. Occasionally several similar modes share one syntax table. See Example Major Modes, for an example of how to set up a syntax table. -- Function: syntax-table-p object This function returns t if object is a vector of length 256 elements. This means that the vector may be a syntax table. However, according to this test, any vector of length 256 is considered to be a syntax table, no matter what its contents. ΓòÉΓòÉΓòÉ 33.1. Syntax Descriptors ΓòÉΓòÉΓòÉ This section describes the syntax classes and flags that denote the syntax of a character, and how they are represented as a syntax descriptor, which is a Lisp string that you pass to modify-syntax-entry to specify the desired syntax. Emacs defines twelve syntax classes. Each syntax table contains a mapping that puts each character into one class. There is no necessary relationship between the class of a character in one syntax table and its class in any other table. Each class is designated by a mnemonic character which serves as the name of the class when you need to specify a class. Usually the designator character is one which is frequently put in that class; however, its meaning as a designator is unvarying and independent of how it is actually classified. A syntax descriptor is a Lisp string which specifies a syntax class, a matching character (unused except for parenthesis classes) and flags. The first character is the designator for a syntax class. The second character is the character to match; if it is unused, put a space there. Then come the characters for any desired flags. If no matching character or flags are needed, one character is sufficient. Thus, the descriptor for the character `*' in C mode is `. 23' (i.e., punctuation, matching character slot unused, second character of a comment-starter, first character of an comment-ender), and the entry for `/' is `. 14' (i.e., punctuation, matching character slot unused, first character of a comment-starter, second character of a comment-ender). ΓòÉΓòÉΓòÉ 33.1.1. Table of Syntax Classes ΓòÉΓòÉΓòÉ Here is a summary of the classes, the characters that stand for them, their meanings, and examples of their use. -- Syntax class: whitespace character Whitespace characters (designated with `@ ' or `-') separate symbols and words from each other. Typically, whitespace characters have no other syntactic use, and multiple whitespace characters are syntactically equivalent to one. Space, tab, newline and formfeed are almost always considered whitespace. -- Syntax class: word constituent Word constituents (designated with `w') are parts of normal English words and are typically used in variable and command names in programs. All upper and lower case letters and the digits are typically word constituents. -- Syntax class: symbol constituent Symbol constituents (designated with `_') are the extra characters that are used in variable and command names along with word constituents. For example, the symbol constituents class is used in Lisp mode to indicate that certain characters may be part of symbol names even though they are not part of English words. These characters are `$&*+-_<>'. In standard C, the only non-word-constituent character that is valid in symbols is underscore (`_'). -- Syntax class: punctuation character Punctuation characters (`.') are those characters that are used as punctuation in English, or are used in some way in a programming language to separate symbols from one another. Most programming language modes, including Emacs Lisp mode, have no characters in this class since the few characters that are not symbol or word constituents all have other uses. -- Syntax class: open parenthesis character -- Syntax class: close parenthesis character Open and close parenthesis characters are characters used in dissimilar pairs to surround sentences or expressions. Such a grouping is begun with an open parenthesis character and terminated with a close. Each open parenthesis character matches a particular close parenthesis character, and vice versa. Normally, Emacs indicates momentarily the matching open parenthesis when you insert a close parenthesis. See Blinking. The class of open parentheses is designated with `(', and that of close parentheses with `)'. In English text, and in C code, the parenthesis pairs are `()', `[]', and `{}'. In Emacs Lisp, the delimiters for lists and vectors (`()' and `[]') are classified as parenthesis characters. -- Syntax class: string quote String quote characters (designated with `"') is used to delimit string constants in many languages, including Lisp and C. The same string quote character appears at the beginning and the end of a string. Such quoted strings do not nest. The parsing facilities of Emacs consider a string as a single token. The usual syntactic meanings of the characters in the string are suppressed. The Lisp modes have two string quote characters: double-quote (`"') and vertical bar (`|'). `|' is not used in Emacs Lisp, but it is used in Common Lisp. C also has two string quote characters: double-quote for strings, and single-quote (`'') for character constants. English text has no string quote characters because English is not a programming language. Although quotation marks are used in English, we do not want them to turn off the usual syntactic properties of other characters in the quotation. -- Syntax class: escape An escape character (designated with `\') starts an escape sequence such as is used in C string and character constants. The character `\' belongs to this class in both C and Lisp. (In C, it is used thus only inside strings, but it turns out to cause no trouble to treat it this way throughout C code.) -- Syntax class: character quote A character quote character (designated with `/') quotes the following character so that it loses its normal syntactic meaning. This differs from an escape character in that only the character immediately following is ever affected. This class is not currently used in any standard Emacs modes. -- Syntax class: paired delimiter Paired delimiter characters (designated with `$') are like string quote characters except that the syntactic properties of the characters between the delimiters are not suppressed. Only TeX mode uses a paired identical delimiter presently---the `$' that begins and ends math mode. -- Syntax class: expression prefix An expression prefix operator (designated with `'') is used for syntactic operators that are part of an expression if they appear next to one but are not part of an adjoining symbol. These characters in Lisp include the apostrophe, `'' (used for quoting), and the comma, `,' (used in macros). -- Syntax class: comment starter -- Syntax class: comment ender The comment starter and comment ender characters are used in different languages to delimit comments. These classes are designated with `<' and `>', respectively. English text has no comment characters. In Lisp, the semi-colon (`;') starts a comment and a newline or formfeed ends one. ΓòÉΓòÉΓòÉ 33.1.2. Syntax Flags ΓòÉΓòÉΓòÉ In addition to the classes, entries for characters in a syntax table can include flags. There are four possible flags, represented by the characters `1', `2', `3', and `4'. All are used to describe multi-character comment delimiters. A flag indicates that the character for which the entry is being made can also be part of a comment sequence, in addition to the syntactic properties associated with its character class. The flags are independent of the class and each other for the sake of characters such as `*' in C mode, which is a punctuation character, and the second character of a start-of-comment sequence (`/*'), and the first character of an end-of-comment sequence (`*/'). The flags for a character c are: o `1' means c is the start of a two-character comment start sequence. o `2' means c is the second character of such a sequence. o `3' means c is the start of a two-character comment end sequence. o `4' means c is the second character of such a sequence. ΓòÉΓòÉΓòÉ 33.2. Syntax Table Functions ΓòÉΓòÉΓòÉ In this section we describe functions for creating, accessing and altering syntax tables. -- Function: make-syntax-table &optional table This function constructs a copy of table and returns it. If table is not supplied (or is nil), it returns a copy of the current syntax table. Otherwise, an error is signaled if table is not a syntax table. -- Function: copy-syntax-table &optional table This function is identical to make-syntax-table. -- Command: modify-syntax-entry char syntax-descriptor &optional table This function sets the syntax entry for char according to syntax-descriptor. The syntax is changed only for table, which defaults to the current buffer's syntax table, and not in any other syntax table. The argument syntax-descriptor specifies the desired syntax; this is a string beginning with a class designator character, and optionally containing a matching character and flags as well. See Syntax Descriptors. This function always returns nil. The old syntax information in the table for this character is discarded. An error is signaled if the first character of the syntax descriptor is not one of the twelve syntax class designator characters. An error is also signaled if char is not a character. Examples: ;; Put the space character in class whitespace. (modify-syntax-entry ?\ " ") => nil ;; Make `$' an open parenthesis character, ;; with `^' as its matching close. (modify-syntax-entry ?$ "(^") => nil ;; Make `^' a close parenthesis character, ;; with `$' as its matching open. (modify-syntax-entry ?^ ")$") => nil ;; Make `/' a punctuation character, ;; the first character of a start-comment sequence, ;; and the second character of an end-comment sequence. ;; This is used in C mode. (modify-syntax-entry ?/ ".13") => nil -- Function: char-syntax character This function returns the syntax class of character, represented by its mnemonic designator character. This only returns the class, not any matching parenthesis or flags. An error is signaled if char is not a character. The first example shows that the syntax class of space is whitespace (represented by a space). The second example shows that the syntax of `/' is punctuation in C-mode. This does not show the fact that it is also a comment sequence character. The third example shows that open parenthesis is in the class of open parentheses. This does not show the fact that it has a matching character, `)'. (char-to-string (char-syntax ?\ )) => " " (char-to-string (char-syntax ?/)) => "." (char-to-string (char-syntax ?\()) => "(" -- Function: set-syntax-table table This function makes table the syntax table for the current buffer. It returns table. -- Function: syntax-table This function returns the current syntax table, which is the table for the current buffer. -- command: describe-syntax This function describes the syntax specifications of the current syntax table. It makes a listing in the `*Help*' buffer, and then pops up a window to display it. It returns nil. A portion of a description is shown here: (describe-syntax) => nil ---------- Buffer: *Help* ---------- C-q \ which means: escape C-r .. C-_ which means: whitespace ! . which means: punctuation ( () which means: open, matches ) ) )( which means: close, matches ( * .. + _ which means: symbol , . which means: punctuation - _ which means: symbol . . which means: punctuation / . 13 which means: punctuation, is the first character of a comment-start sequence, is the first character of a comment-end sequence 0 .. 9 w which means: word ---------- Buffer: *Help* ---------- ΓòÉΓòÉΓòÉ 33.3. Parsing and Moving Over Balanced Expressions ΓòÉΓòÉΓòÉ Here are several functions for parsing and scanning balanced expressions. The syntax table controls the interpretation of characters, so these functions can be used for Lisp expressions when in Lisp mode and for C expressions when in C mode. See List Motion, for convenient higher-level functions for moving over balanced expressions. -- Function: parse-partial-sexp start limit &optional target-depth stop-before state This function parses an expression in the current buffer starting at start, not scanning past limit. Parsing stops at limit or when certain criteria described below are met; point is set to the location where parsing stops. The value returned is a description of the status of the parse at the point where it stops. Normally, start is assumed to be the top level of an expression to be parsed, such as the beginning of a function definition. Alternatively, you might wish to resume parsing in the middle of an expression. To do this, you must provide a state argument that describes the initial status of parsing. If state is omitted (or nil), parsing assumes that start is the beginning of a new parse at level 0. If the third argument target-depth is non-nil, parsing stops if the depth in parentheses becomes equal to target-depth. The depth starts at 0, or at whatever is given in state. If the fourth argument stop-before is non-nil, parsing stops when it comes to any character that starts a sexp. The fifth argument state is a seven-element list of the same form as the value of this function, described below. The return value of one call may be used to initialize the state of the parse on another call to parse-partial-sexp. The result is a list of seven elements describing the final state of the parse: 1. The depth in parentheses, starting at 0. 2. The character position of the start of the innermost containing parenthetical grouping; nil if none. 3. The character position of the start of the last complete subexpression terminated; nil if none. 4. Non-nil if inside a string. (It is the character that will terminate the string.) 5. t if inside a comment. 6. t if point is just after a quote character. 7. The minimum parenthesis depth encountered during this scan. Elements 1, 4, 5, and 6 are significant in the argument state. This function is used to determine how to indent lines in programs written in languages that have nested parentheses. -- Function: scan-lists from count depth This function scans forward count balanced parenthetical groupings from character number from. It returns the character number of the position thus found. If depth is nonzero, parenthesis depth counting begins from that value. The only candidates for stopping are places where the depth in parentheses becomes zero; scan-lists counts count such places and then stops. Thus, a positive value for depth means go out levels of parenthesis. Comments are ignored if parse-sexp-ignore-comments is non-nil. If the beginning or end of the buffer (or its accessible portion) is reached and the depth is not zero, an end-of-file error is signaled. If the depth is zero but the count is not used up, nil is returned. -- Function: scan-sexps from count Scan from character number from by count balanced expressions. It returns the character number of the position thus found. Comments are ignored if parse-sexp-ignore-comments is non-nil. If the beginning or end of (the accessible part of) the buffer is reached in the middle of a parenthetical grouping, an end-of-file error is signaled. If the beginning or end is reached between groupings but before count is used up, nil is returned. -- Function: backward-prefix-chars This function moves point backward over any number of chars with expression prefix syntax. -- Variable: parse-sexp-ignore-comments If the value is non-nil, then comments are treated as whitespace by the functions in this section and by forward-sexp. This works only when the comment terminator is something like `*/', and appears only to end a comment. If comments are terminated by newlines, you must make this variable nil, since not every newline is the end of a comment. (In version 19, this limitation is removed.) ΓòÉΓòÉΓòÉ 33.4. Some Standard Syntax Tables ΓòÉΓòÉΓòÉ Each of the major modes in Emacs has its own syntax table. Here are several of them: -- Function: standard-syntax-table This function returns the standard syntax table, which is the syntax table used in Fundamental mode. -- Variable: text-mode-syntax-table The value of this variable is the syntax table used in Text mode. -- Variable: c-mode-syntax-table The value of this variable is the syntax table in use in C-mode buffers. -- Variable: emacs-lisp-mode-syntax-table The value of this variable is the syntax table used in Emacs Lisp mode by editing commands. (It has no effect on the Lisp read function.) ΓòÉΓòÉΓòÉ 33.5. Syntax Table Internals ΓòÉΓòÉΓòÉ Each element of a syntax table is an integer that translates into the full meaning of the entry: class, possible matching character, and flags. However, it is not common for a programmer to work with the entries directly in this form since the Lisp-level syntax table functions usually work with syntax descriptors (see Syntax Descriptors). The low 8 bits of each element of a syntax table indicates the syntax class. Integer Class 0 whitespace 1 punctuation 2 word 3 symbol 4 open parenthesis 5 close parenthesis 6 expression prefix 7 string quote 8 paired delimiter 9 escape 10 character quote 11 comment-start 12 comment-end The next 8 bits are the matching opposite parenthesis (if the character has parenthesis syntax); otherwise, they are not meaningful. The next 4 bits are the flags. ΓòÉΓòÉΓòÉ 34. Abbrevs And Abbrev Expansion ΓòÉΓòÉΓòÉ An abbreviation or abbrev is a string of characters that may be expanded to a longer string. The user can insert the abbrev string and find it replaced automatically with the expansion of the abbrev. This saves typing. The set of abbrevs currently in effect is recorded in an abbrev table. Each buffer has a local abbrev table, but normally all buffers in the same major mode share one abbrev table. There is also a global abbrev table. Normally both are used. An abbrev table is represented as an obarray containing a symbol for each abbreviation. The symbol's name is the abbreviation. Its value is the expansion; its function definition is the hook; its property list cell contains the use count, the number of times the abbreviation has been expanded. Because these symbols are not inturned in the usual obarray, they will never appear as the result of reading a Lisp expression; in fact, they will never be used except by the code that handles abbrevs. Therefore, it is safe to use them in an extremely nonstandard way. See Creating Symbols. For the user-level commands for abbrevs, see Abbrev Mode. ΓòÉΓòÉΓòÉ 34.1. Setting Up Abbrev Mode ΓòÉΓòÉΓòÉ Abbrev mode is a minor mode controlled by the value of the variable abbrev-mode. -- Variable: abbrev-mode A non-nil value of this variable turns on the automatic expansion of abbrevs when their abbreviations are inserted into a buffer. If the value is nil, abbrevs may be defined, but they are not expanded automatically. This variable automatically becomes local when set in any fashion. -- Variable: default-abbrev-mode This is the value abbrev-mode for buffers that do not override it. This is the same as (default-value 'abbrev-mode). ΓòÉΓòÉΓòÉ 34.2. Abbrev Tables ΓòÉΓòÉΓòÉ This section describes how to create and manipulate abbrev tables. -- Function: make-abbrev-table This function creates and returns a new, empty abbrev table---an obarray containing no symbols. It is a vector filled with nils. -- Function: clear-abbrev-table table This function undefines all the abbrevs in abbrev table table, leaving it empty. The function returns nil. -- Function: define-abbrev-table tabname definitions This function defines tabname (a symbol) as an abbrev table name, i.e., as a variable whose value is an abbrev table. It defines abbrevs in the table according to definitions, a list of elements of the form (abbrevname expansion hook usecount). The value is always nil. -- Variable: abbrev-table-name-list This is a list of symbols whose values are abbrev tables. define-abbrev-table adds the new abbrev table name to this list. -- Function: insert-abbrev-table-description name human This function inserts before point a description of the abbrev table named name. The argument name is a symbol whose value is an abbrev table. The value is always nil. If human is non-nil, a human-oriented description is inserted. Otherwise the description is a Lisp expression---a call to define-abbrev-table which would define name exactly as it is currently defined. ΓòÉΓòÉΓòÉ 34.3. Defining Abbrevs ΓòÉΓòÉΓòÉ These functions define an abbrev in a specified abbrev table. define-abbrev is the low-level basic function, while add-abbrev is used by commands that ask for information from the user. -- Function: add-abbrev table type arg This function adds an abbreviation to abbrev table table. The argument type is a string describing in English the kind of abbrev this will be (typically, "global" or "mode-specific"); this is used in prompting the user. The argument arg is the number of words in the expansion. The return value is the symbol which internally represents the new abbrev, or nil if the user declines to redefine an existing abbrev. -- Function: define-abbrev table name expansion hook This function defines an abbrev in table named name, to expand to expansion, and call hook. The return value is an uninterned symbol which represents the abbrev inside Emacs; its name is name. The argument name should be a string. The argument expansion should be a string, or nil, to undefine the abbrev. The argument hook is a function or nil. If hook is non-nil, then it is called with no arguments after the abbrev is replaced with expansion; point is located at the end of expansion. The use count of the abbrev is initialized to zero. -- User Option: only-global-abbrevs If this variable is non-nil, it means that the user plans to use global abbrevs only. This tells the commands that define mode-specific abbrevs to define global ones instead. This variable does not alter the functioning of the functions in this section; it is examined by their callers. ΓòÉΓòÉΓòÉ 34.4. Saving Abbrevs in Files ΓòÉΓòÉΓòÉ A file of saved abbrev definitions is actually a file of Lisp code. The abbrevs are saved in the form of a Lisp program to define the same abbrev tables with the same contents. Therefore, you can load the file with load (see How Programs Do Loading). However, the function quietly-read-abbrev-file is provided as a more convenient interface. User-level facilities such as save-some-buffers can save abbrevs in a file automatically, under the control of variables described here. -- User Option: abbrev-file-name This is the default file name for reading and saving abbrevs. -- Function: quietly-read-abbrev-file filename This function reads abbrev definitions from a file named filename, previously written with write-abbrev-file. If filename is nil, the file specified in abbrev-file-name is used. save-abbrevs is set to t so that changes will be saved. This function does not display any messages. It returns nil. -- User Option: save-abbrevs A non-nil value for save-abbrev means that Emacs should save abbrevs when files are saved. abbrev-file-name specifies the file to save the abbrevs in. -- Variable: abbrevs-changed This variable is set non-nil by defining or altering any abbrevs. This serves as a flag for various Emacs commands to offer to save your abbrevs. -- Command: write-abbrev-file filename Save all abbrev definitions, in all abbrev tables, in the file filename as a Lisp program which will define the same abbrevs when loaded. This function returns nil. ΓòÉΓòÉΓòÉ 34.5. Looking Up and Expanding Abbreviations ΓòÉΓòÉΓòÉ Abbrevs are usually expanded by commands for interactive use, or automatically by self-insert. This section describes the subroutines used in writing such functions, as well as the variables they use for communication. -- Function: abbrev-symbol abbrev table This function returns the symbol representing the abbrev named abbrev. The value returned is nil if that abbrev is not defined. The optional second argument table is the abbrev table to look it up in. By default, this function tries first the current buffer's local abbrev table, and second the global abbrev table. -- User Option: abbrev-all-caps When this is set non-nil, an abbrev entered entirely in upper case is expanded using all upper case. Otherwise, an abbrev entered entirely in upper case is expanded by capitalizing each word of the expansion. -- Function: abbrev-expansion abbrev &optional table This function returns the string that abbrev would expand into (as defined by the abbrev tables used for the current buffer). The optional argument table specifies the abbrev table to use; if it is specified, the abbrev is looked up in that table only. -- Variable: abbrev-start-location This is the buffer position for expand-abbrev to use as the start of the next abbrev to be expanded. nil means use the word before point as the abbrev. abbrev-start-location is set to nil each time expand-abbrev is called. This variable is set by abbrev-prefix-mark. -- Variable: abbrev-start-location-buffer The value of this variable is the buffer for which abbrev-start-location has been set. Trying to expand an abbrev in any other buffer clears abbrev-start-location. This variable is set by abbrev-prefix-mark. -- Variable: last-abbrev This is the abbrev-symbol of the last abbrev expanded. This information is left by expand-abbrev for the sake of the unexpand-abbrev command. -- Variable: last-abbrev-location This is the location of the last abbrev expanded. This contains information left by expand-abbrev for the sake of the unexpand-abbrev command. -- Variable: last-abbrev-text This is the exact expansion text of the last abbrev expanded, as results from case conversion. Its value is nil if the abbrev has already been unexpanded. This contains information left by expand-abbrev for the sake of the unexpand-abbrev command. ΓòÉΓòÉΓòÉ 34.6. Standard Abbrev Tables ΓòÉΓòÉΓòÉ Here we list the variables that hold the abbrev tables for the preloaded major modes of Emacs. -- Variable: global-abbrev-table This is the abbrev table for mode-independent abbrevs. The abbrevs defined in it apply to all buffers. Each buffer may also have a local abbrev table, whose abbrev definitions take precedence over those in the global table. -- Variable: local-abbrev-table The value of this buffer-local variable is the (mode-specific) abbreviation table of the current buffer. -- Variable: fundamental-mode-abbrev-table This is the local abbrev table used in Fundamental mode. It is the local abbrev table in all buffers in Fundamental mode. -- Variable: text-mode-abbrev-table This is the local abbrev table used in Text mode. -- Variable: c-mode-abbrev-table This is the local abbrev table used in C mode. -- Variable: lisp-mode-abbrev-table This is the local abbrev table used in Lisp mode and Emacs Lisp mode. ΓòÉΓòÉΓòÉ 35. Processes ΓòÉΓòÉΓòÉ In the terminology of operating systems, a process is a space in which a program can execute. Emacs runs in a process. Emacs Lisp programs can invoke other programs in processes of their own. These are called subprocesses or child processes of the Emacs process, which is their parent process. A subprocess of Emacs may be synchronous or asynchronous. depending on how it is created. When you create a synchronous subprocess, the Lisp program waits for the subprocess to terminate before continuing execution. When you create an asynchronous subprocess, it can run in parallel with the Lisp program. This kind of subprocess is represented within Emacs by a Lisp object which is also called a ``process''. Lisp programs can use this object to communicate with the subprocess or to control it. For example, you can send signals, obtain status information, receive output from the process, or send input to it. -- Function: processp object This function returns t if object is a process, nil otherwise. ΓòÉΓòÉΓòÉ 35.1. Functions that Create Subprocesses ΓòÉΓòÉΓòÉ There are three functions that create a new Unix subprocess in which to run a program. One of them, start-process, creates an asynchronous process and returns a process object ( see Asynchronous Processes). The other two, call-process and call-process-region, create a synchronous process and do not return a process object (see Synchronous Processes). Synchronous and asynchronous processes are explained in following sections. Since the three functions are all called in a similar fashion, their common arguments are described here. In all cases, the program to be run is specified with the function's program argument. An error is signaled if the file is not found or cannot be executed. The actual file containing the program is found by following normal Unix rules: if an absolute file name is given, then the program must be found in the specified file; if a relative file name is given, then the directories in exec-path are searched sequentially for a suitable file. The variable exec-path is initialized when Emacs is started, based on the value of the environment variable PATH. The standard Unix abbreviations, `~', `.', and `..', are interpreted as usual, but environment variable substitutions (`$HOME', etc.) are not recognized; use substitute-in-file-name to perform them ( see File Name Expansion). Each of the subprocess-creating functions has a buffer-or-name argument which specifies where the standard output from the program will go. If buffer-or-name is nil, then the output will be discarded (by directing it to `/dev/null') unless a filter function is specified to handle it. (See Filter Functions, and Streams.) Normally, you should avoid having multiple processes send output to the same buffer because the outputs will be intermixed randomly. All three of the subprocess-creating functions have a &rest argument, args. The args must all be strings, and they are supplied to program as separate command line arguments. Wildcard characters and other shell constructs are not allowed in these strings, since they are passed directly to the specified program. *Note:* the argument program contains only the name of the program; it may not contain any command-line arguments. Such arguments must be provided via args. The subprocess gets its current directory from the value of default-directory (see File Name Expansion). The subprocess inherits its environment from Emacs; but you can specify overrides for it with process-environment. -- Variable: process-environment This variable is a list of strings to append to the environment of processes as they are created. Each string assigns a value to a shell environment variable. (This applies both to asynchronous and synchronous processes.) process-environment => ("l=/usr/stanford/lib/gnuemacs/lisp" "PATH=.:/user/lewis/bin:/usr/class:/nfsusr/local/bin" "USER=lewis" "TERM=ibmapa16" "SHELL=/bin/csh" "HOME=/user/lewis") -- Variable: exec-directory The value of this variable is the name of a directory (a string) that contains programs that come with GNU Emacs, that are intended for Emacs to invoke. The program loadst is an example of such a program; it is used by the display-time command to print the current time (and certain other information) once per minute. The default value is the name of a directory whose name ends in `etc'. We call the directory `emacs/etc', since its name usually ends that way. We sometimes refer to ``the directory `emacs/etc','' when strictly speaking we ought to say, ``the directory named by the variable exec-directory.'' Most of the time, there is no difference. -- User Option: exec-path The value of this variable is a list of directories to search for programs to run in subprocesses. Each element is either the name of a directory (i.e., a string), or nil, which stands for the default directory (which is the value of default-directory). The value of exec-path is used by call-process and start-process when the program argument is not an absolute file name. ΓòÉΓòÉΓòÉ 35.2. Creating a Synchronous Process ΓòÉΓòÉΓòÉ After a synchronous process is created, Emacs waits for the process to terminate before continuing. Starting Dired is an example of this: it runs ls in a synchronous process, then modifies the output slightly. Because the process is synchronous, the entire directory listing arrives in the buffer before Emacs tries to do anything with it. While Emacs waits for the synchronous subprocess to terminate, the user can quit by typing C-g, and the process is killed by sending it a SIGKILL signal. See Quitting. The synchronous subprocess functions return nil in version 18. In version 19, they will return an indication of how the process terminated. -- Function: call-process program &optional infile buffer-or-name display &rest args This function calls program in a separate process and waits for it to finish. The standard input for the process comes from file infile if infile is not nil and from `/dev/null' otherwise. The process output gets inserted in buffer buffer-or-name before point, if that argument names a buffer. If buffer-or-name is t, output is sent to the current buffer; if buffer-or-name is nil, output is discarded. If buffer-or-name is the integer 0, the output is discarded and call-process returns immediately. In this case, the process is not truly synchronous, since it can run in parallel with Emacs; but you can think of it as synchronous in that Emacs is essentially finished with the subprocess as soon as this function returns. If display is non-nil, then call-process redisplays the buffer as output is inserted. Otherwise the function does no redisplay, and the results become visible on the screen only when Emacs redisplays that buffer in the normal course of events. The remaining arguments, args, are strings that are supplied as the command line arguments for the program. The examples below are both run with the buffer `foo' current. (call-process "pwd" nil t) => nil ---------- Buffer: foo ---------- /usr/user/lewis/manual ---------- Buffer: foo ---------- (call-process "grep" nil "bar" nil "lewis" "/etc/passwd") => nil ---------- Buffer: bar ---------- lewis:5LTsHm66CSWKg:398:21:Bil Lewis:/user/lewis:/bin/csh ---------- Buffer: bar ---------- The dired-readin function contains a good example of the use of call-process: (call-process "ls" nil buffer nil dired-listing-switches dirname) -- Function: call-process-region start end program &optional delete buffer-or-name display &rest args This function sends the text between start to end as standard input to a process running program. It deletes the text sent if delete is non-nil, which may be useful when the output is going to be inserted back in the current buffer. If buffer-or-name names a buffer, the output is inserted in that buffer at point. If buffer-or-name is t, the output is sent to the current buffer. If buffer-or-name is nil, the output is discarded. If buffer-or-name is the integer 0, the output is discarded and call-process returns immediately, as in call-process. If display is non-nil, then call-process-region redisplays the buffer as output is inserted. Otherwise the function does no redisplay, and the results become visible on the screen only when Emacs redisplays that buffer in the normal course of events. The remaining arguments, args, are strings that are supplied as the command line arguments for the program. In the following example, we use call-process-region to run the cat utility, with standard input being the first five characters in buffer `foo' (the word `input'). cat copies its standard input into its standard output. Since the argument buffer-or-name is t, this output is inserted in the current buffer. ---------- Buffer: foo ---------- input-!- ---------- Buffer: foo ---------- (call-process-region 1 6 "cat" nil t) => nil ---------- Buffer: foo ---------- inputinput-!- ---------- Buffer: foo ---------- The shell-command-on-region command uses call-process-region like this: (call-process-region start end shell-file-name ; Name of program. nil ; Do not delete region. buffer ; Send output to buffer. nil ; No redisplay during output. "-c" command) ; Arguments for the shell. ΓòÉΓòÉΓòÉ 35.3. Creating an Asynchronous Process ΓòÉΓòÉΓòÉ After an asynchronous process is created, Emacs and the Lisp program can continue running immediately. The process may thereafter run in parallel with Emacs, and the two may communicate with each other using the functions described in following sections. Here we describe how to create an asynchronous process, with start-process. -- Function: start-process name buffer-or-name program &rest args This function creates a new asynchronous subprocess and starts the program program running in it. It returns a process object that stands for the new subprocess for Emacs Lisp programs. The argument name specifies the name for the process object; if a process with this name already exists, then name is modified (by adding `<1>', etc.) to be unique. The buffer buffer-or-name is the buffer to associate with the process. The remaining arguments, args, are strings that are supplied as the command line arguments for the program. In the example below, the first process is started and runs (rather, sleeps) for 100 seconds. Meanwhile, the second process is started, given the name `my-process<1>' for the sake of uniqueness. It inserts the directory listing at the end of the buffer `foo', before the first process finishes. Then it finishes, and a message to that effect is inserted in the buffer. Much later, the first process finishes, and another message is inserted in the buffer for it. (start-process "my-process" "foo" "sleep" "100") => #<process my-process> (start-process "my-process" "foo" "ls" "-l" "/user/lewis/bin") => #<process my-process<1>> ---------- Buffer: foo ---------- total 2 lrwxrwxrwx 1 lewis 14 Jul 22 10:12 gnuemacs --> /emacs -rwxrwxrwx 1 lewis 19 Jul 30 21:02 lemon Process my-process<1> finished Process my-process finished ---------- Buffer: foo ---------- -- Variable: process-connection-type This variable controls the type of device used to communicate with asynchronous subprocesses. If it is nil, then pipes are used. If it is t, then ptys are used (or pipes if ptys not supported). ptys are usually preferable for processes visible to the user, as in Shell mode, because they allow job control (C-c, C-z, etc.) to work between the process and its children whereas pipes do not. For subprocesses used for internal purposes by programs, it is often better to use a pipe, because they are more efficient. In addition, the total number of ptys is limited on many systems and it is good not to waste them. The value process-connection-type is used when start-process is called, so in order to change it for just one call of start-process, temporarily rebind it with let. (let ((process-connection-type nil)) ; Use a pipe. (start-process ...)) ΓòÉΓòÉΓòÉ 35.4. Deleting Processes ΓòÉΓòÉΓòÉ Deleting a process disconnects Emacs immediately from the subprocess, and removes it from the list of active processes. It sends a signal to the subprocess to make the subprocess terminate, but this is not guaranteed to happen immediately. (The process object itself continues to exist as long as other Lisp objects point to it.) You can delete a process explicitly at any time. Processes are deleted automatically after they terminate, but not necessarily right away. If you delete a terminated process explicitly before it is deleted automatically, no harm results. -- Variable: delete-exited-processes This variable controls automatic deletion of processes that have terminated (due to calling exit or to a signal). If it is nil, then they continue to exist until the user runs list-processes. Otherwise, they are deleted immediately after they exit. -- Function: delete-process name This function deletes the process associated with name. The argument name may be a process, the name of a process, a buffer, or the name of a buffer. The subprocess is killed with a SIGHUP signal. (delete-process "*shell*") => nil -- Function: process-kill-without-query process This function declares that Emacs need not query the user if process is still running when Emacs is exited. The process will be deleted silently. The value is t. (process-kill-without-query (get-process "shell")) => t ΓòÉΓòÉΓòÉ 35.5. Process Information ΓòÉΓòÉΓòÉ Several functions return information about processes. list-processes is provided for interactive use. -- Command: list-processes This command displays a listing of all living processes. (Any processes listed as `Exited' or `Signaled' are actually eliminated after the listing is made.) This function returns nil. -- Function: process-list This function returns a list of all processes that have not been deleted. (process-list) => (#<process display-time> #<process shell>) -- Function: get-process name This function returns the process named name, or nil if there is none. An error is signaled if name is not a string. (get-process "shell") => #<process shell> -- Function: process-command process This function returns the command that was executed to start process. This is a list of strings, the first string being the program executed and the rest of the strings being the arguments that were given to the program. (process-command (get-process "shell")) => ("/bin/csh" "-i") -- Function: process-exit-status process This function returns the exit status of process or the signal number that killed it. If process has not yet terminated, the value is 0. -- Function: process-id process This function returns the pid of process. This is an integer which distinguishes the process process from all other processes running on the same computer at the current time. The pid of a process is chosen by the operating system kernel when the process is started and remains constant as long as the process exists. -- Function: process-name process This function returns the name of process. -- Function: process-status process-name This function returns the status of process-name as a symbol. The argument process-name must be either a process or a string. If it is a string, it need not name an actual process. The possible values for an actual subprocess are: run for a process that is running. stop for a process that is stopped but continuable. exit for a process that has exited. signal for a process that has received a fatal signal. nil if process-name is not the name of an existing process. (process-status "shell") => run (process-status "never-existed") => nil x => #<process xx<1>> (process-status x) => exit For a network stream, process-status returns one of the symbols open or closed. The latter means that the other side closed the connection, or Emacs did delete-process. ΓòÉΓòÉΓòÉ 35.6. Sending Input to Processes ΓòÉΓòÉΓòÉ Asynchronous subprocesses receive input when it is sent to them by Emacs, which is done with the functions in this section. You must specify the process to send input to, and the input data to send. The data appears on the ``standard input'' of the subprocess. Some operating systems have limited space for buffered input in a pty. On these systems, the subprocess will cease to read input correctly if you send an input line longer than the system can handle. You cannot avoid the problem by breaking the input into pieces and sending them separately, for the operating system will still have to put all the pieces together in the input buffer before it lets the subprocess read the line. The only solution is to put the input in a temporary file, and send the process a brief command to read that file. -- Function: process-send-string process-name string This function sends process-name the contents of string as standard input. The argument process-name must be a process or the name of a process. The function returns nil. (process-send-string "shell<1>" "ls\n") => nil ---------- Buffer: *shell* ---------- ... introduction.texi syntax-tables.texi~ introduction.texi~ text.texi introduction.txt text.texi~ ... ---------- Buffer: *shell* ---------- -- Command: process-send-region process-name start end This function sends the text in the region defined by start and end as standard input to process-name, which is a process or a process name. An error is signaled unless both start and end are integers or markers that indicate positions in the current buffer. (It is unimportant which number is larger.) -- Function: process-send-eof &optional process-name This function makes process-name see an end-of-file in its input. The eof comes after any text already sent to it. If process-name is not supplied, or if it is nil, then this function sends the eof to the current buffer's process. An error is signaled if the current buffer has no process. The function returns process-name. (process-send-eof "shell") => "shell" ΓòÉΓòÉΓòÉ 35.7. Sending Signals to Processes ΓòÉΓòÉΓòÉ Sending a signal to a subprocess is a way of interrupting its activities. There are several different signals, each with its own meaning. For example, the signal SIGINT means that the user has typed C-c, or that some analogous thing has happened. Each signal has a standard effect on the subprocess. Most signals kill the subprocess, but some stop or resume execution instead. Most signals can optionally be handled by programs; if the program handles the signal, then we can say nothing in general about its effects. The set of signals and their names is defined by the operating system; Emacs has facilities for sending only a few of the signals that are defined. Emacs can send signals only to its own subprocesses. You can send signals explicitly by calling the function in this section. Emacs also sends signals automatically at certain times: killing a buffer sends a SIGHUP signal to all its associated processes; killing Emacs sends a SIGHUP signal to all remaining processes. (SIGHUP is a signal that usually indicates that the user hung up the phone.) Each of the signal-sending functions takes two optional arguments: process-name and current-group. The argument process-name must be either a process, the name of one, or nil. If it is nil, the process defaults to the process associated with the current buffer. An error is signaled if process-name does not identify a process. The argument current-group is a flag that makes a difference when you are running a job-control shell as an Emacs subprocess. If it is non-nil, then the signal is sent to the current process-group of the terminal which Emacs uses to communicate with the subprocess. If the process is a job-control shell, this means the shell's current subjob. If it is nil, the signal is sent to the process group of the immediate subprocess of Emacs. If the subprocess is a job-control shell, this is the shell itself. The flag current-group has no effect when a pipe is used to communicate with the subprocess, because the operating system does not support the distinction in the case of pipes. For the same reason, job-control shells won't work when a pipe is used. See process-connection-type in Asynchronous Processes. -- Function: interrupt-process &optional process-name current-group This function interrupts the process process-name by sending the Unix signal SIGINT. Outside of Emacs, typing the ``interrupt character'' (usually C-c on Berkeley Unix) sends this signal. When the argument current-group is non-nil, you can think of this function as ``typing C-c'' on the terminal by which Emacs talks to the subprocess. -- Function: kill-process &optional process-name current-group This function kills the process process-name by sending the Unix signal SIGKILL. This signal kills the subprocess immediately, and cannot be handled by the subprocess. -- Function: quit-process &optional process-name current-group This function sends the Unix signal SIGQUIT to the process process-name. This signal is the one sent by the ``quit character'' (usually C-b or C-\) when you are not inside Emacs. -- Function: stop-process &optional process-name current-group This function stops the process process-name by sending the Unix signal SIGTSTP. Use continue-process to resume its execution. On systems with job control, the ``stop character'' (usually C-z) sends this signal (outside of Emacs). When current-group is non-nil, you can think of this function as ``typing C-z'' on the terminal Emacs uses to communicate with the subprocess. -- Function: continue-process &optional process-name current-group This function resumes execution of the process process by sending it the Unix signal SIGCONT. This presumes that process-name was stopped previously. ΓòÉΓòÉΓòÉ 35.8. Receiving Output from Processes ΓòÉΓòÉΓòÉ There are two ways to receive the output that a subprocess writes to its standard output stream. The output can be inserted in a buffer, which is called the associated buffer of the process, or a function called the filter function can be called to act on the output. ΓòÉΓòÉΓòÉ 35.8.1. Process Buffers ΓòÉΓòÉΓòÉ A process can (and usually does) have an associated buffer, which is an ordinary Emacs buffer that is used for two purposes: storing the output from the process, and deciding when to kill the process. You can also use the buffer to identify a process to operate on, since in normal practice only one process is associated with any given buffer. Many applications of processes also use the buffer for editing input to be sent to the process, but this is not built into Emacs Lisp. Unless the process has a filter function (see Filter Functions), its output is inserted in the associated buffer. The position to insert the output is determined by the process-mark ( see Process Information), which is then updated to point to the end of the text just inserted. Usually, but not always, the process-mark is at the end of the buffer. If the process has no buffer and no filter function, its output is discarded. -- Function: process-buffer process This function returns the associated buffer of the process process. (process-buffer (get-process "shell")) => #<buffer *shell*> -- Function: process-mark process This function returns the marker which controls where additional output from the process will be inserted in the process buffer ( see Process Buffers). When output is inserted, the marker is updated to point at the end of the output. This causes successive batches of output to be inserted consecutively. If process does not insert its output into a buffer, then process-mark returns a marker that points nowhere. Filter functions normally should use this marker in the same fashion as is done by direct insertion of output in the buffer. A good example of a filter function that uses process-mark is found at the end of the following section. When the user is expected to enter input in the process buffer for transmission to the process, the process marker is useful for distinguishing the new input from previous output. -- Function: set-process-buffer process buffer This function sets the buffer associated with process to buffer. If buffer is nil, the process will not be associated with any buffer. -- Function: get-buffer-process buffer-or-name This function returns the process associated with buffer-or-name. If there are several processes associated with it, then one is chosen. (Presently, the one chosen is the one most recently created.) It is usually a bad idea to have more than one process associated with the same buffer. (get-buffer-process "*shell*") => #<process shell> If the process's buffer is killed, the actual child process is killed with a SIGHUP signal (see Signals to Processes). ΓòÉΓòÉΓòÉ 35.8.2. Process Filter Functions ΓòÉΓòÉΓòÉ A process filter function is a function that receives the standard output from the associated process. If a process has a filter, then all standard output from that process is passed to the filter rather than be inserted into a buffer or discarded. The process buffer is used for output from the process only when there is no filter. A filter function must accept two arguments: the associated process and a string, which is the output. The function is then free to do whatever it chooses with the output. A filter function runs only while Emacs is waiting (e.g., for terminal input, or for time to elapse, or for process output). This avoids the timing errors that could result from running filters at random places in the middle of other Lisp programs. You may explicitly cause Emacs to wait, so that filter functions will run, by calling sit-for, sleep-for or accept-process-output ( see Accepting Output). Emacs is also waiting when the command loop is reading input. Quitting is normally inhibited within a filter function---otherwise, the effect of typing C-g at command level or to quit a user command would be unpredictable. If you want to permit quitting inside a filter function, bind inhibit-quit to nil. See Quitting. Many filter functions sometimes or always insert the text in the process's buffer, mimicking the actions of Emacs when there is no filter. Such filter functions need to use set-buffer in order to be sure to insert in that buffer. To avoid setting the current buffer semipermanently, these filter functions must use unwind-protect to make sure to restore the previous current buffer. They should also update the process marker, and in some cases update the value of point. Here is how to do these things: (defun ordinary-insertion-filter (proc string) (let ((old-buffer (current-buffer))) (unwind-protect (let (moving) (set-buffer (process-buffer proc)) (setq moving (= (point) (process-mark proc))) (save-excursion ;; Insert the text, moving the process-marker. (goto-char (process-mark proc)) (insert string) (set-marker (process-mark proc) (point))) (if moving (goto-char (process-mark proc)))) (set-buffer old-buffer)))) The reason to use an explicit unwind-protect rather than letting save-excursion restore the current buffer is so as to preserve the change in point made by goto-char. To make the filter force the process buffer to be visible whenever new text arrives, insert the following line just before the unwind-protect: (display-buffer (process-buffer proc)) To force point to move to the end of the new output no matter where it was previously, eliminate the variable moving and call goto-char unconditionally. All filter functions that do regexp searching or matching should save and restore the match data. Otherwise, a filter function that runs during a call to sit-for might clobber the match data of the program that called sit-for. See Match Data. The output to the function may come in chunks of any size. A program that produces the same output twice in a row may send it as one batch of 200 characters one time, and five batches of 40 characters the next. -- Function: set-process-filter process filter This function gives process the filter function filter. If filter is nil, then the process will have no filter. -- Function: process-filter process This function returns the filter function of process, or nil if it has none. Here is an example of use of a filter function: (defun keep-output (process output) (setq kept (cons output kept))) => keep-output (setq kept nil) => nil (set-process-filter (get-process "shell") 'keep-output) => keep-output (process-send-string "shell" "ls ~/other\n") => nil kept => ("lewis@slug[8] % " "FINAL-W87-SHORT.MSS backup.otl kolstad.mss~ address.txt backup.psf kolstad.psf backup.bib~ david.mss resume-Dec-86.mss~ backup.err david.psf resume-Dec.psf backup.mss dland syllabus.mss " "#backups.mss# backup.mss~ kolstad.mss ") Here is another, more realistic example, which demonstrates how to use the process mark to do insertion in the same fashion as is done when there is no filter function: ;; Insert input in the buffer specified by my-shell-buffer ;; and make sure that buffer is shown in some window. (defun my-process-filter (proc str) (let ((cur (selected-window)) (pop-up-windows t)) (pop-to-buffer my-shell-buffer) (goto-char (point-max)) (insert str) (set-marker (process-mark proc) (point-max)) (select-window cur))) ΓòÉΓòÉΓòÉ 35.8.3. Accepting Output from Processes ΓòÉΓòÉΓòÉ Output from asynchronous subprocesses normally arrives only while Emacs is waiting for some sort of external event, such as elapsed time or terminal input. Occasionally it is useful in a Lisp program to explicitly permit output to arrive at a specific point, or even to wait until output arrives from a process. -- Function: accept-process-output &optional process This function allows Emacs to read pending output from processes. The output is inserted in the associated buffers or given to their filter functions. If process is non-nil then this function does not return until some output has been received from process. ΓòÉΓòÉΓòÉ 35.8.4. Sentinels: Detecting Process Status Changes ΓòÉΓòÉΓòÉ A process sentinel is a function that is called whenever the associated process changes status for any reason, including signals (whether sent by Emacs or caused by the process's own actions) that terminate, stop, or continue the process. The process sentinel is also called if the process exits. The sentinel receives two arguments: the process for which the event occurred, and a string describing the type of event. The string describing the event looks like one of the following: o "finished\n". o "exited abnormally with code exitcode\n". o "name-of-signal\n". o "name-of-signal (core dumped)\n". A sentinel runs only while Emacs is waiting (e.g., for terminal input, or for time to elapse, or for process output). This avoids the timing errors that could result from running them at random places in the middle of other Lisp programs. You may explicitly cause Emacs to wait, so that sentinels will run, by calling sit-for, sleep-for or accept-process-output (see Accepting Output). Emacs is also waiting when the command loop is reading input. Quitting is normally inhibited within a sentinel---otherwise, the effect of typing C-g at command level or to quit a user command would be unpredictable. If you want to permit quitting inside a sentinel, bind inhibit-quit to nil. See Quitting. All sentinels that do regexp searching or matching should save and restore the match data. Otherwise, a sentinel that runs during a call to sit-for might clobber the match data of the program that called sit-for. See Match Data. -- Function: set-process-sentinel process sentinel This function associates sentinel with process. If sentinel is nil, then the process will have no sentinel. The default behavior when there is no sentinel is to insert a message in the process's buffer when the process status changes. (defun msg-me (process event) (princ (format "Process: %s had the event `%s'" process event))) (set-process-sentinel (get-process "shell") 'msg-me) => msg-me (kill-process (get-process "shell")) -| Process: #<process shell> had the event `killed' => #<process shell> -- Function: process-sentinel process This function returns the sentinel of process, or nil if it has none. -- Function: waiting-for-user-input-p While a sentinel or filter function is running, this function returns non-nil if Emacs was waiting for keyboard input from the user at the time the sentinel or filter function was called, nil if it was not. ΓòÉΓòÉΓòÉ 35.9. Subprocess Functions for VMS ΓòÉΓòÉΓòÉ The ordinary subprocess functions do not work on VMS in version 18. Instead, these functions are available. -- Function: default-subprocess-input-handler This function is the default input handler for input from spawned subprocesses. -- Function: spawn-subprocess integer &optional filter sentinel This function spawns an asynchronous VMS subprocess for command processing. The arguments are integer, an integer to identify the subprocess in future operations; filter, a function to be called when output arrives from the subprocess; and sentinel, a function to be called when the subprocess terminates. If filter is nil, output is inserted in the current buffer. If sentinel is nil, nothing special is done when the subprocess terminates. When the filter is called, it receives two arguments; integer to identify the process, and a string containing the output. When the sentinel is called, it receives just one argument, integer. -- Function: send-command-to-subprocess integer command This function sends the string command to a VMS subprocess numbered integer. -- Function: stop-subprocess integer This function terminates the VMS subprocess numbered integer. In version 19, these functions have been eliminated, and the ordinary subprocess functions are implemented on VMS. ΓòÉΓòÉΓòÉ 35.10. TCP ΓòÉΓòÉΓòÉ Emacs Lisp programs can open TCP connections to other processes on the same machine or other machines. A network connection is handled by Lisp much like a subprocess, and is represented by a process object. However, the process you are communicating with is not a child of the Emacs process, so you can't kill it or send it signals. All you can do is send and receive data. delete-process closes the connection, but does not kill the process at the other end of it. You can distinguish process objects representing network connections from those representing subprocesses with the process-status function. -- Function: open-network-stream name buffer-or-name host service This function opens a TCP connection for a service to a host. It returns a process object to represent the connection. The name argument specifies the name for the process object. It is modified as necessary to make it unique. The buffer-or-name argument is the buffer to associate with the connection. Output from the connection is inserted in the buffer, unless you specify a filter function to handle the output. If buffer-or-name is nil, it means that the connection is not associated with any buffer. The arguments host and service specify where to connect to; host is the host name (a string), and service is the name of the service desired (a string) or an integer specifying a port number to connect to. ΓòÉΓòÉΓòÉ 36. Operating System Interface ΓòÉΓòÉΓòÉ This chapter is about starting and getting out of Emacs, access to values in the operating system environment, and terminal input, output and flow control. See Building Emacs, for related information. See also Emacs Display, for additional operating system status information which pertain to the terminal and the screen. ΓòÉΓòÉΓòÉ 36.1. Starting Up Emacs ΓòÉΓòÉΓòÉ This section describes what Emacs does when it is started, and how you can customize these actions. ΓòÉΓòÉΓòÉ 36.1.1. Summary: Sequence of Actions at Start Up ΓòÉΓòÉΓòÉ The order of operations performed (in `startup.el') by Emacs when it is started up is as follows: 1. It loads `.emacs' unless `-q' was specified on command line. (This is not done in `-batch' mode.) `.emacs' is found in the user's home directory; the `-u' option can specify the user name whose home directory should be used. 2. It loads `default.el' unless inhibit-default-init is non-nil. (This is not done in `-batch' mode or if `-q' was specified on command line.) 3. It loads the terminal-specific Lisp file, if any, except when in batch mode. 4. It runs term-setup-hook. 5. It runs window-setup-hook. 6. It displays copyleft and nonwarranty, plus basic use information, unless the value of inhibit-startup-message is non-nil. This display is also inhibited in batch mode, and if the current buffer is not `*scratch*'. 7. It processes any remaining command line arguments. -- User Option: inhibit-startup-message This variable inhibits the initial startup messages (the nonwarranty, etc.). If it is non-nil, then the messages are not printed. This variable exists so you can set it in your personal init file, once you are familiar with the contents of the startup message. Do not set this variable in the init file of a new user, or in a way that affects more than one user, because that would prevent new users from receiving the information they are supposed to see. ΓòÉΓòÉΓòÉ 36.1.2. The Init File: .emacs ΓòÉΓòÉΓòÉ When you start Emacs, it normally attempts to load the file `.emacs' from your home directory. This file, if it exists, must contain Lisp code. It is called your init file. The command line switches `-q' and `-u' can be used to control the use of the init file; `-q' says not to load an init file, and `-u' says to load a specified user's init file instead of yours. See Entering Emacs. Emacs may also have a default init file, which is the library named `default.el'. Emacs finds the `default.el' file through the standard search path for libraries ( see How Programs Do Loading). The Emacs distribution does not have any such file; you may create one at your site for local customizations. If the default init file exists, it is loaded whenever you start Emacs, except in batch mode or if `-q' is specified. But your own personal init file, if any, is loaded first; if it sets inhibit-default-init to a non-nil value, then Emacs will not load the `default.el' file. If there is a great deal of code in your `.emacs' file, you should move it into another file named `something.el', byte-compile it (see Byte Compilation), and make your `.emacs' file load the other file using load (see Loading). See Init File Examples, for examples of how to make various commonly desired customizations in your `.emacs' file. -- User Option: inhibit-default-init This variable prevents Emacs from loading the default initialization library file for your session of Emacs. If its value is non-nil, then the default library is not loaded. The default value is nil. ΓòÉΓòÉΓòÉ 36.1.3. Terminal-Specific Initialization ΓòÉΓòÉΓòÉ Each terminal type can have its own Lisp library that Emacs will load when run on that type of terminal. For a terminal type named termtype, the library is called `term/termtype'. Emacs finds the file by searching the load-path directories as it does for other files, and trying the `.elc' and `.el' suffixes. Normally, terminal-specific Lisp library is located in `emacs/lisp/term', a subdirectory of the `emacs/lisp' directory in which most Emacs Lisp libraries are kept. The library's name is constructed by concatenating the value of the variable term-file-prefix and the terminal type. Normally, term-file-prefix has the value "term/"; changing this is not recommended. The usual purpose of a terminal-specific library is to define the escape sequences used by a terminal's function keys. See the file `term/vt100.el' for an example of a terminal-specific library. Function keys are handled by a two-level procedure. The first level is dependent on the specific type of terminal and maps Emacs's input sequences to the function keys that they represent. The second level is independent of terminal type and is customized by users; function keys are mapped into meanings at this level. The terminal-specific library handles the first level of the process and the library `keypad.el' handles the second level of mapping. When the name of the terminal type contains a hyphen, only the part of the name before the first hyphen is significant in choosing the library name. Thus, terminal types `aaa-48' and `aaa-30-rv' both use the `term/aaa' library. If necessary, the library can evaluate (getenv "TERM") to find the full name of the terminal type. Your `.emacs' file can prevent the loading of the terminal-specific library by setting term-file-prefix to nil. This feature is very useful when experimenting with your own peculiar customizations. You can also arrange to override some of the actions of the terminal-specific library by setting the variable term-setup-hook. If it is not nil, Emacs calls the value of the variable term-setup-hook as a function of no arguments at the end of Emacs initialization, after Emacs has already loaded both your `.emacs' file and any terminal-specific libraries. You can use this variable to define initializations for terminals that do not have their own libraries. -- Variable: term-file-prefix If the term-file-prefix variable is non-nil, Emacs loads a terminal-specific initialization file as follows: (load (concat term-file-prefix (getenv "TERM"))) You may set the term-file-prefix variable to nil in your `.emacs' file if you do not wish to load the terminal-initialization file. To do this, put the following in your `.emacs' file: (setq term-file-prefix nil). -- Variable: term-setup-hook The value of this variable is either nil or a function to be called by Emacs after loading your `.emacs' file, the default initialization file (if any) and after loading terminal-specific Lisp code. The function is called with no arguments. You can use term-setup-hook to override the definitions made by a terminal-specific file. See also window-setup-hook in Window Systems. ΓòÉΓòÉΓòÉ 36.1.4. Command Line Arguments ΓòÉΓòÉΓòÉ You can use command line arguments to request various actions when you start Emacs. Since you do not need to start Emacs more than once per day, and will often leave your Emacs session running longer than that, command line arguments are hardly ever used. As a practical matter, it is best to avoid making the habit of using them, since this habit would encourage you to kill and restart Emacs unnecessarily often. These options exist for two reasons: to be compatible with other editors (for invocation by other programs) and to enable shell scripts to run specific Lisp programs. -- Function: command-line This function parses the command line which Emacs was called with, processes it, loads the user's `.emacs' file and displays the initial nonwarranty information, etc. -- Variable: command-line-processed The value of this variable is t once the command line has been processed. If you redump Emacs by calling dump-emacs, you must set this variable to nil first in order to cause the new dumped Emacs to process its new command line arguments. -- Variable: command-switch-alist The value of this variable is an alist of user-defined command-line options and associated handler functions. This variable exists so you can add elements to it. A command line option is an argument on the command line of the form: -option The elements of the command-switch-alist look like this: (option . handler-function) For each element, the handler-function receives the switch name as its sole argument. In some cases, the option is followed in the command line by an argument. In these cases, the handler-function can find all the remaining command-line arguments in the variable command-line-args-left. (The entire list of command-line arguments is in command-line-args.) The command line arguments are parsed by the command-line-1 function in the `startup.el' file. See also Command Line Switches and Arguments. -- Variable: command-line-args The value of this variable is the arguments passed by the shell to Emacs, as a list of strings. ΓòÉΓòÉΓòÉ 36.2. Getting out of Emacs ΓòÉΓòÉΓòÉ There are two ways to get out of Emacs: you can kill the Emacs job, which exits permanently, or you can suspend it, which permits you to reenter the Emacs process later. As a practical matter, you seldom kill Emacs---only when you are about to log out. Suspending is much more common. ΓòÉΓòÉΓòÉ 36.2.1. Killing Emacs ΓòÉΓòÉΓòÉ Killing Emacs means ending the execution of the Emacs process. It will return to its superior process. All the information in the Emacs process, aside from files that have been saved, is lost when the Emacs is killed. Because killing Emacs inadvertently can lose a lot of work, Emacs will query for confirmation before actually terminating if you have buffers that need saving or subprocesses that are running. -- Function: kill-emacs &optional no-query This function exits the Emacs process and kills it. Normally, if there are modified files or if there are running processes, kill-emacs asks the user for confirmation before exiting. However, if no-query is supplied and non-nil, then Emacs exits without confirmation. If no-query is an integer, then it is used as the exit status of the Emacs process. (This is useful primarily in batch operation; see Batch Mode.) If no-query is a string, its contents are stuffed into the terminal input buffer so that the shell (or whatever program next reads input) can read them. -- Variable: kill-emacs-hook The value of the kill-emacs-hook variable is either nil or is that of a function to be called by kill-emacs. The hook is called before anything else is done by kill-emacs. ΓòÉΓòÉΓòÉ 36.2.2. Suspending Emacs ΓòÉΓòÉΓòÉ Suspending Emacs means stopping Emacs temporarily and returning control to its superior process, which is usually the shell. This allows you to resume editing later in the same Emacs process, with the same buffers, the same kill ring, the same undo history, and so on. To resume Emacs, use the appropriate command in the parent shell---most likely fg. Some operating systems do not support suspension of jobs; on these systems, ``suspension'' actually creates a new shell temporarily as a subprocess of Emacs. Then you would exit the shell to return to Emacs. Suspension is not useful with window systems such as X Windows, because the Emacs job may not have a parent that can resume it again, and in any case you can give input to some other job such as a shell merely by moving to a different window. Therefore, suspending is not allowed under X Windows. -- Function: suspend-emacs string This function stops Emacs and returns to the superior process. It returns nil. If string is non-nil, its characters are sent to be read as terminal input by Emacs's superior shell. The characters in string will not be echoed by the superior shell; just the results will appear. Before suspending, Emacs examines the symbol suspend-hook. If it is bound, and its value is non-nil, then the value is called as a function of no arguments. If the function returns non-nil, then suspend-emacs returns immediately and suspension does not occur. After Emacs resumes, the symbol suspend-resume-hook is examined. If it is bound and non-nil, then the value is called as a function of no arguments. The next redisplay after resumption will redraw the entire screen, unless no-redraw-on-reenter is set (see Screen Attributes). In the following example, note that `pwd' is not echoed after Emacs is suspended. But it is read and executed by the shell. (suspend-emacs) => nil (setq suspend-hook (function (lambda () (not (y-or-n-p "Really suspend? "))))) => (lambda nil (not (y-or-n-p "Really suspend? "))) (setq suspend-resume-hook (function (lambda () (message "Resumed!")))) => (lambda nil (message "Resumed!")) (suspend-emacs "pwd") => nil ---------- Buffer: Minibuffer ---------- Really suspend? y ---------- Parent Shell ---------- lewis@slug[23] % /user/lewis/manual lewis@slug[24] % fg ---------- Echo Area ---------- Resumed! -- Variable: suspend-hook The value of the suspend-hook variable, if not nil, is called as a function with no arguments by suspend-emacs before Emacs is actually suspended. If the function returns non-nil, then suspension does not take place. -- Variable: suspend-resume-hook The value of the suspend-resume-hook variable, if not nil, is called as a function with no arguments after resumption of an Emacs session that was suspended with suspend-emacs. ΓòÉΓòÉΓòÉ 36.3. Operating System Environment ΓòÉΓòÉΓòÉ Emacs provides access to variables in the operating system environment through various functions. These variables include the name of the system, the user's uid, and so on. -- Variable: system-type The value of this variable is a symbol indicating the type of operating system Emacs is operating on. Here is a table of the symbols for the operating systems that Emacs can run on up to version 18.51. berkeley-unix Berkeley BSD 4.1, 4.2, or 4.3. hpux Hewlett-Packard operating system, version 5, 6, or 7. silicon-graphics-unix Silicon Graphics Iris 3.5 or 3.6. rtu RTU 3.0, UCB universe. unisoft-unix UniSoft's UniPlus 5.0 or 5.2. usg-unix-v AT&T's System V.0, System V Release 2.0, 2.2, or 3. vax-vms VMS VMS version 4 or 5. xenix SCO Xenix 386 Release 2.2. We do not wish to add new symbols to make finer distinctions unless it is absolutely necessary! In fact, it would be nice to eliminate a couple of possibilities in the future. -- Function: getenv var This function returns the value of the environment variable var, as a string. (getenv "USER") => "lewis" lewis@slug[10] % printenv PATH=.:/user/lewis/bin:/usr/bin:/usr/local/bin USER=lewis TERM=ibmapa16 SHELL=/bin/csh HOME=/user/lewis -- Function: user-login-name This function returns the name under which the user is logged in. This is based on the effective uid, not the real uid. (user-login-name) => "lewis" -- Function: user-real-login-name This function returns the name under which the user logged in. This is based on the real uid, not the effective uid. This differs from user-login-name only when running with the setuid bit. -- Function: user-full-name This function returns the full name of the user. (user-full-name) => "Bil Lewis" -- Function: user-real-uid This function returns the real uid of the user. (user-real-uid) => 19 -- Function: user-uid This function returns the effective uid of the user. -- Function: system-name This function returns the name of the machine you are running on. (system-name) => "prep.ai.mit.edu" -- Function: current-time-string This function returns the current time and date as a humanly-readable string. The format of the string is unvarying; the number of characters used for each part is always the same, so you can reliably use substring to extract pieces of it. However, it would be wise to count the characters from the beginning of the string rather than from the end, as additional information describing the time zone may be added in version 19. (current-time-string) => "Wed Oct 14 22:21:05 1987" -- Function: load-average This function returns the current 1 minute, 5 minute and 15 minute load averages in a list. The values are integers that are 100 times the system load averages. (The load averages indicate the number of processes trying to run.) (load-average) => (169 48 36) lewis@rocky[5] % uptime 11:55am up 1 day, 19:37, 3 users, load average: 1.69, 0.48, 0.36 -- Function: setprv privilege-name &optional setp getprv This function sets or resets a VMS privilege. (It does not exist on Unix.) The first arg is the privilege name, as a string. The second argument, setp, is t or nil, indicating whether the privilege is to be turned on or off. Its default is nil. The function returns t if success, nil if not. If the third argument, getprv, is non-nil, setprv does not change the privilege, but returns t or nil indicating whether the privilege is currently enabled. ΓòÉΓòÉΓòÉ 36.4. Terminal Input ΓòÉΓòÉΓòÉ The terminal input functions and variables keep track of or manipulate terminal input. See Emacs Display, for related functions. -- Function: recent-keys This function returns a string comprising the last 100 characters read from the terminal. These are the last 100 characters read by Emacs, no exceptions. (recent-keys) => "erminal. These are the last 100 characters read by Emacs, no exceptions. @example (recent-keys)^U^X^E" Here the string `@example' is a Texinfo command that was inserted in the source file for the manual, and `^U^X^E' are the characters that were typed to evaluate the expression (recent-keys). -- Command: open-dribble-file filename This function opens a dribble file named filename. When a dribble file is open, Emacs copies all keyboard input characters to that file. (The contents of keyboard macros are not typed on the keyboard so they are not copied to the dribble file.) You close the dribble file by calling this function with an argument of "". The function always returns nil. This function is normally used to record the input necessary to trigger an Emacs bug, for the sake of a bug report. (open-dribble-file "$j/dribble") => nil See also the open-termscript function (see Terminal Output). -- Function: set-input-mode interrupt flow quit-char This function sets the mode for reading keyboard input. If interrupt is non-null, then Emacs uses input interrupts. If it is nil, then it uses cbreak mode. If flow is non-nil, then Emacs uses xon/xoff (C-q, C-s) flow control for output to terminal. This has no effect except in cbreak mode. See Flow Control. The normal setting is system dependent. Some systems always use cbreak mode regardless of what is specified. If quit-char is non-nil, it specifies the character to use for quitting. Normally this is 7, the code for C-g. See Quitting. -- Variable: meta-flag This variable tells Emacs whether to treat the 0200 bit in keyboard input as the Meta bit. nil means no, and anything else means yes. In version 19, meta-flag will be a function instead of a variable. -- Variable: keyboard-translate-table This variable defines the translate table for keyboard input. This allows the user to redefine the keys on the keyboard without changing any command bindings. Its value must be a string or nil. If keyboard-translate-table is a string, then each character read from the keyboard is looked up in this string and the character in the string is used instead. If the string is of length n, character codes n and up are untranslated. In the example below, keyboard-translate-table is set to a string of 128 characters. Then the characters C-s and C-\ are swapped and the characters C-q and C-^ are swapped. After executing this function, typing C-\ has all the usual effects of typing C-s, and vice versa. (See Flow Control for more information on this subject.) (defun evade-flow-control () "Replace C-s with C-\ and C-q with C-^." (interactive) (let ((the-table (make-string 128 0))) (let ((i 0)) (while (< i 128) (aset the-table i i) (setq i (1+ i)))) ;; Swap C-s and C-\. (aset the-table ?\034 ?\^s) (aset the-table ?\^s ?\034) ;; Swap C-q and C-^. (aset the-table ?\036 ?\^q) (aset the-table ?\^q ?\036) (setq keyboard-translate-table the-table))) Note that this translation is the first thing that happens after a character is read from the terminal. As a result, record-keeping features such as recent-keys and open-dribble-file record the translated characters. ΓòÉΓòÉΓòÉ 36.5. Terminal Output ΓòÉΓòÉΓòÉ The terminal output functions send or keep track of output sent from the computer to the terminal. The baud-rate function tells you what Emacs thinks is the output baud rate of the terminal. -- Function: baud-rate This function returns the output baud rate of the terminal. (baud-rate) => 9600 If you are running across a network, and different parts of the network work at different baud rates, the value returned by Emacs may be different from the value used by your local terminal. Some network protocols communicate the local terminal speed to the remote machine, so that Emacs and other programs can get the proper value, but others do not. If the machine where Emacs is running has the wrong speed setting, you can specify the right speed using the stty program. However, you will have to start Emacs afresh to make this take effect. *Note:* In version 19, baud-rate is a variable so that you can change it conveniently within Emacs. -- Function: send-string-to-terminal string This function sends string to the terminal without alteration. Control characters in string will have terminal-dependent effects. One use of this function is to define function keys on terminals that have downloadable function key definitions. For example, this is how on certain terminals to define function key 4 to move forward four characters (by transmitting the characters C-u C-f to the computer): (send-string-to-terminal "\eF4\^U\^F") => nil -- Command: open-termscript filename This function is used to open a termscript file that will record all the characters sent by Emacs to the terminal. It returns nil. Termscript files are useful for investigating problems where Emacs garbles the screen, problems which are due to incorrect termcap entries or to undesirable settings of terminal options more often than actual Emacs bugs. Once you are certain which characters were actually output, you can determine reliably whether they correspond to the termcap specifications in use. See also open-dribble-file in Terminal Input. (open-termscript "../junk/termscript") => nil ΓòÉΓòÉΓòÉ 36.6. Flow Control ΓòÉΓòÉΓòÉ This section attempts to answer the question ``Why does Emacs choose to use flow-control characters in its command character set?'' For a second view on this issue, read the comments on flow control in the `emacs/INSTALL' file from the distribution; for help with termcaps and DEC terminal concentrators, see `emacs/etc/TERMS'. At one time, most terminals did not need flow control. This meant that the choice of C-s and C-q as command characters was reasonable. Emacs, for economy of keystrokes and portability, chose to use the control characters in the ASCII character set, and tried to make the assignments mnemonic (thus, C-s for search and C-q for quote). Later, some terminals were introduced which used these characters for flow control. They were not very good terminals, so Emacs maintainers did not pay attention. In later years, the practice became widespread among terminals, but by this time it was usually an option. And the majority of users, who can turn flow control off, were unwilling to switch to less mnemonic key bindings for the sake of flow control. So which usage is ``right'', Emacs's or that of some terminal and concentrator manufacturers? This is a rhetorical (or religious) question; it has no simple answer. One reason why we are reluctant to cater to the problems caused by C-s and C-q is that they are gratuitous. There are other techniques (albeit less common in practice) for flow control that preserve transparency of the character stream. Note also that their use for flow control is not an official standard. Interestingly, on the model 33 teletype with a paper tape punch (which is very old), C-s and C-q were sent by the computer to turn the punch on and off! GNU Emacs (version 18.48 and later) provides several options for coping with terminals or front-ends that insist on using flow control characters. Listed in estimated order of preference, these options are as follows: 1. Have Emacs run in cbreak mode with the kernel handling flow control. Issue (set-input-mode nil t) from `.emacs'. After doing this, it is necessary to find other keys to bind to the commands isearch-forward and quoted-insert. The usual nominees are C-^ and C-\. There are two ways to get this effect: a. Use the keyboard-translate-table to cause C-^ and C-\ to be received by Emacs as though C-s and C-q were typed. Emacs (except at its very lowest level) never knows that the characters typed were anything but C-s and C-q, so the use of these keys inside isearch-forward still works---typing C-^ while incremental searching will move the cursor to the next match, etc. For example: (setq keyboard-translate-table (make-string 128 0)) (let ((i 0)) (while (< i 128) (aset keyboard-translate-table i i) (setq i (1+ i)))) ;; Swap C-s and C-\. (aset the-table ?\034 ?\^s) (aset the-table ?\^s ?\034) ;; Swap C-q and C-^. (aset the-table ?\036 ?\^q) (aset the-table ?\^q ?\036))) b. Simply rebind the keys C-^ and C-\ to isearch-forward and quoted-insert. To use the new keys to repeat searches, it is necessary to set search-repeat-char to C-^ as well. 2. Don't use cbreak mode, but cause C-s and C-q to be bound to a null command. The problem with this solution is that the flow control characters were probably sent because whatever sent them is falling behind on the characters being sent to it. The characters that find their way to the terminal screen will not in general be those that are intended. Also, it will be be necessary to find other keys to bind to isearch-forward and quoted-insert; see the previous alternative. Here is a suitable null command: (defun noop () "Do nothing; return nil." (interactive)) 3. Don't use cbreak mode, and unset the C-s and C-q keys with the global-unset-key function. This is similar to the previous alternative, except that the flow control characters will probably cause beeps or visible bells. Note that if the terminal is the source of the flow control characters and kernel flow control handling is enabled, you probably will not have to send padding characters as specified in a termcap or terminfo entry. In this case, it may be possible to customize a termcap entry to provide better Emacs performance on the assumption that flow control is in use. This effect can also be simulated by announcing (with stty or its equivalent) that the terminal is running at a very slow speed, provided you are communicating across a network so that stty does not actually try to change the line speed. ΓòÉΓòÉΓòÉ 36.7. Batch Mode ΓòÉΓòÉΓòÉ The command line option `-batch' causes Emacs to run noninteractively. In this mode, Emacs does not read commands from the terminal, it does not alter the terminal modes, and it does not expect to be outputting to an erasable screen. The idea is that you will specify Lisp programs to run; when they are finished, Emacs should exit. The way to specify the programs to run is with `-l file', which causes the library named file to be loaded, and `-f function', which causes function to be called with no arguments. Any Lisp program output that would normally go to the echo area, either using message or using prin1, etc., with t as the stream, will actually go to Emacs's standard output descriptor when in batch mode. Thus, Emacs behaves much like a noninteractive application program. (The echo area output that Emacs itself normally generates, such as command echoing, is suppressed entirely.) -- Variable: noninteractive This variable is non-nil when Emacs is running in batch mode. ΓòÉΓòÉΓòÉ 37. Emacs Display ΓòÉΓòÉΓòÉ This chapter describes a number of features related to the display that Emacs presents to the user. ΓòÉΓòÉΓòÉ 37.1. Refreshing the Screen ΓòÉΓòÉΓòÉ -- Command: redraw-display This function clears the screen and redraws what is supposed to appear on it. ΓòÉΓòÉΓòÉ 37.2. Screen Attributes ΓòÉΓòÉΓòÉ The screen attribute functions describe and define the characteristics of the terminal. -- Function: screen-height This function returns the number of lines on the screen that are available for display. (screen-height) => 50 -- Function: screen-width This function returns the number of columns on the screen that are available for display. (screen-width) => 80 -- Function: set-screen-height lines &optional not-actual-size This function declares that the terminal can display lines lines. The sizes of existing windows will be altered proportionally to fit. If not-actual-size is non-nil, then Emacs will display lines lines of output, but will not change its value for the actual height of the screen. Knowing the correct actual size may be necessary for correct cursor positioning. If lines is different from what it was previously, then the entire screen is cleared and redisplayed using the new size. This function returns nil. -- Function: set-screen-width columns &optional not-actual-size This function declares that the terminal can display columns columns. The details are as in set-screen-height. -- Variable: no-redraw-on-reenter This variable controls whether Emacs redraws the entire screen after it has been suspended and resumed. Non-nil means yes, nil means no. On most terminals, it is necessary to redraw. Not redrawing is useful if the terminal can remember and restore the Emacs screen contents. -- Variable: inverse-video This variable controls whether Emacs uses inverse video for all text on the screen. Non-nil means yes, nil means no. The default is nil. -- User Option: mode-line-inverse-video This variable controls the use of inverse video for mode lines. If it is non-nil, then mode lines are displayed in inverse video (or another suitable display mode). Otherwise, mode lines are displayed normal, just like the rest of the screen. The default is t. ΓòÉΓòÉΓòÉ 37.3. Truncation ΓòÉΓòÉΓòÉ When a line of text extends beyond the right edge of a window, the line can either be truncated or continued on the next line. When a line is truncated, this is shown with a `$' in the rightmost column of the window. When a line is continued or ``wrapped'' onto the next line, this is shown with a `\' on the rightmost column of the window. The additional screen lines used to display a long text line are called continuation lines. (Note that wrapped lines are not filled; filling has nothing to do with truncation and continuation. See Filling.) -- User Option: truncate-lines This buffer-local variable controls how Emacs displays lines that extend beyond the right edge of the window. If it is non-nil, then Emacs does not display continuation lines; but rather each line of text will take exactly one screen line, and a dollar sign will be shown at the edge of any line that extends to or beyond the edge of the window. The default is nil. If the variable truncate-partial-width-windows is non-nil, then truncation is used for windows that are not the full width of the screen, regardless of the value of truncate-lines. -- Variable: default-truncate-lines This variable is the default value for truncate-lines in buffers that do not override it. -- User Option: truncate-partial-width-windows This variable determines how lines that are too wide to fit on the screen are displayed in side-by-side windows ( see Splitting Windows). If it is non-nil, then wide lines are truncated (with a `$' at the end of the line); otherwise they are wrapped (with a `\' at the end of the line). ΓòÉΓòÉΓòÉ 37.4. The Echo Area ΓòÉΓòÉΓòÉ The echo area is used for displaying messages made with the message primitive, and for echoing keystrokes. It is not the same as the minibuffer, despite the fact that the minibuffer appears (when active) in the same place on the screen as the echo area. The GNU Emacs Manual specifies the rules for resolving conflicts between the echo area and the minibuffer for use of that screen space (see The Minibuffer). You can write output in the echo area by using the Lisp printing funtions with t as the stream (see Output Functions), or as follows: -- Function: message string &rest arguments This function prints a one-line message in the echo area. The argument string is similar to a C language printf control string. See format in String Conversion, for the details on the conversion specifications. message returns the constructed string. (message "Minibuffer depth is %d." (minibuffer-depth)) => "Minibuffer depth is 0." ---------- Echo Area ---------- Minibuffer depth is 0. ---------- Echo Area ---------- -- Variable: cursor-in-echo-area This variable controls where the cursor is positioned when a message is displayed in the echo area. If it is non-nil, then the cursor appears at the end of the message. Otherwise, the cursor appears at point---not in the echo area at all. The value is normally nil except when bound to t for brief periods of time. ΓòÉΓòÉΓòÉ 37.5. Selective Display ΓòÉΓòÉΓòÉ Selective display is a class of minor modes in which specially marked lines do not appear on the screen, or in which highly indented lines do not appear. The first variant, explicit selective display, is designed for use in a Lisp program. The program controls which lines are hidden by altering the text. Outline mode uses this variant. In the second variant, the choice of lines to hide is made automatically based on indentation. This variant is designed as a user-level feature. The way you control explicit selective display is by replacing a newline (control-j) with a control-m. The text which was formerly a line following that newline is now invisible. Strictly speaking, it is no longer a separate line, since only newlines can separate lines; it is now part of the previous line. On its own, selective display does not affect editing commands. For example, C-f (forward-char) moves point unhesitatingly into invisible space. However, the replacement of newline characters with carriage return characters affects some editing commands. For example, next-line skips invisible lines, since it searches only for newlines. Modes that use selective display can also define commands that take account of the newlines, or which make parts of the text visible or invisible. When you write a selectively displayed buffer into a file, all the control-m's are replaced by their original newlines. This means that when you next read in the file, it looks OK, with nothing invisible. Selective display is an effect that is seen only in Emacs. -- Variable: selective-display This buffer-local variable enables selective display. This means that lines, or portions of lines, may be made invisible. o If the value of selective-display is t, then any portion of a line that follows a control-m will not be displayed. o If the value of selective-display is a positive integer, then lines that start with more than selective-display columns of indentation will not be displayed. When some portion of a buffer is invisible, the vertical movement commands operate as if that portion did not exist, allowing a single next-line command to skip any number of invisible lines. However, character movement commands (such as forward-char) will not skip the invisible portion, and it is possible (if tricky) to insert or delete parts of an invisible portion. In the examples below, what is shown is the display of the buffer foo, which changes with the value of selective-display. The contents of the buffer do not change. (setq selective-display nil) => nil ---------- Buffer: foo ---------- 1 on this column 2on this column 3n this column 3n this column 2on this column 1 on this column ---------- Buffer: foo ---------- (setq selective-display 2) => 2 ---------- Buffer: foo ---------- 1 on this column 2on this column 2on this column 1 on this column ---------- Buffer: foo ---------- -- Variable: selective-display-ellipses If this buffer-local variable is non-nil, then Emacs displays `...' at the end of a line that is followed by invisible text. This example is a continuation of the previous one. (setq selective-display-ellipses t) => t ---------- Buffer: foo ---------- 1 on this column 2on this column ... 2on this column 1 on this column ---------- Buffer: foo ---------- ΓòÉΓòÉΓòÉ 37.6. Overlay Arrow ΓòÉΓòÉΓòÉ The overlay arrow is useful for directing the user's attention to a particular line in a buffer. For example, in the modes used for interface to debuggers, the overlay arrow indicates the current line of code about to be executed. -- Variable: overlay-arrow-string This variable holds the string to display as an arrow, or nil if the arrow feature is not in use. -- Variable: overlay-arrow-position This variable holds a marker which indicates where to display the arrow. It should point at the beginning of a line. The arrow text will be displayed at the beginning of that line, overlaying any text that would otherwise appear. Since the arrow is usually short, and the line usually begins with indentation, normally nothing significant is overwritten. The overlay string is displayed only in the buffer which this marker points into. Thus, only one buffer can have an overlay arrow at any given time. ΓòÉΓòÉΓòÉ 37.7. Temporary Displays ΓòÉΓòÉΓòÉ Temporary displays are used by commands to put output into a buffer and then present it to the user for perusal rather than for editing. Many of the help commands use this feature. -- Special Form: with-output-to-temp-buffer buffer-name forms... This function executes forms while arranging to insert any output they print into the buffer named buffer-name. The buffer is then shown in some window for viewing, displayed but not selected. The buffer is named by the string buffer-name, and it need not already exist. The argument buffer-name must be a string, not a buffer. The buffer is erased initially (with no questions asked), and it is marked as unmodified after with-output-to-temp-buffer exits. with-output-to-temp-buffer first binds standard-output to the buffer, then it evaluates the forms in forms. With standard-output rebound, any output directed there will naturally be inserted into that buffer. Only Lisp output directed to the stream standard-output is affected; screen display and messages in the echo area, although output in the general sense of the word, are not affected. See Output Functions. The value of the last form in forms is returned. ---------- Buffer: foo ---------- This is the contents of foo. ---------- Buffer: foo ---------- (with-output-to-temp-buffer "foo" (print 20) (print standard-output)) => #<buffer foo> ---------- Buffer: foo ---------- 20 #<buffer foo> ---------- Buffer: foo ---------- -- Variable: temp-buffer-show-hook The value of the temp-buffer-show-hook variable is either nil or is called as a function to display a help buffer. This variable is used by with-output-to-temp-buffer. -- Function: momentary-string-display string position &optional char message This function momentarily displays string in the current buffer at position (which is a character offset from the beginning of the buffer). The display remains until the next character is typed. If the next character the user types is char, Emacs ignores it. Otherwise, that character remains buffered for subsequent use as input. Thus, typing char will simply remove the string from the display, while typing (say) C-f will remove the string from the display and later (presumably) move point forward. The argument char is a space by default. The result of momentary-string-display is not useful. If message is non-nil, it is displayed in the echo area. If it is nil, then instructions to type char are displayed there, e.g., `Type RET to continue editing'. In this example, point is initially located at the beginning of the second line: ---------- Buffer: foo ---------- This is the contents of foo. -!-This is the contents of foo. ---------- Buffer: foo ---------- (momentary-string-display "******* Important Message! *******" (point) ?\r "Type RET when done reading") => t ---------- Buffer: foo ---------- This is the contents of foo. ******* Important Message! *******This is the contents of foo. ---------- Buffer: foo ---------- ---------- Echo Area ---------- Type RET when done reading This function works by actually changing the text in the buffer. As a result, if you later undo in this buffer, you will see the message come and go. ΓòÉΓòÉΓòÉ 37.8. Waiting for Elapsed Time or Input ΓòÉΓòÉΓòÉ The waiting commands are designed to make Emacs wait for a certain amount of time to pass or until there is input. For example, you may wish to pause in the middle of a computation to allow the user time to view the display. sit-for performs a pause with an update of screen, while sleep-for performs a pause without updating the screen. -- Function: sit-for seconds This function performs redisplay (provided there is no pending input from the user), then waits seconds seconds, or until input is available. The result is t if sit-for waited the full time with no input arriving (see input-pending-p in Keyboard Input). Otherwise, nil is returned. Redisplay is always preempted if input arrives, and does not happen at all if input is available before it starts. Thus, there is no way to force screen updating if there is pending input; however, if there is no input pending, you can force an update with no delay by using (sit-for 0). The purpose of sit-for to give the user time to read text that you display. -- Function: sleep-for seconds This function simply pauses for seconds seconds without updating the display. It pays no attention to available input. It returns nil. Use sleep-for when you wish to guarantee a delay. ΓòÉΓòÉΓòÉ 37.9. Blinking ΓòÉΓòÉΓòÉ This section describes the mechanism by which Emacs shows a matching open parenthesis when the user inserts a close parenthesis. -- Variable: blink-paren-hook The value of this variable should be a function (of no arguments) to be called whenever a char with close parenthesis syntax is inserted. The value of blink-paren-hook may be nil, in which case nothing is done. *Note:* in version 18, this function is named blink-paren-hook, but since it is not called with the standard convention for hooks, it is being renamed to blink-paren-function in version 19. -- Variable: blink-matching-paren If this variable is nil, then blink-matching-open does nothing. -- Variable: blink-matching-paren-distance This variable specifies the maximum distance to scan for a matching parenthesis before giving up. -- Function: blink-matching-open This function is the default value of blink-paren-hook. It assumes that point follows a character with close parenthesis syntax and moves the cursor momentarily to the matching opening character. If that character is not already on the screen, then its context is shown by displaying it in the echo area. To avoid long delays, this function does not search farther than blink-matching-paren-distance characters. Here is an example of calling this function explicitly. (defun interactive-blink-matching-open () "Indicate momentarily the start of sexp before point." (interactive) (let ((blink-matching-paren-distance (buffer-size)) (blink-matching-paren t)) (blink-matching-open))) ΓòÉΓòÉΓòÉ 37.10. Display of Control Characters ΓòÉΓòÉΓòÉ These variables affect the way certain characters are displayed on the screen. Since they change the number of columns the characters occupy, they also affect the indentation functions. -- User Option: ctl-arrow This buffer-local variable controls how control characters are displayed. If it is non-nil, they are displayed as an uparrow followed by the character: `^A'. If it is nil, they are displayed as a backslash followed by three octal digits: `\001'. -- Variable: default-ctl-arrow The value of this variable is the default value for ctl-arrow in buffers that do not override it. This is the same as (default-value 'ctl-arrow) (see Default Value). -- User Option: tab-width The value of this variable is the spacing between tab stops used for displaying tab characters in Emacs buffers. The default is 8. Note that this feature is completely independent from the user-settable tab stops used by the command tab-to-tab-stop. See Indent Tabs. ΓòÉΓòÉΓòÉ 37.11. Beeping ΓòÉΓòÉΓòÉ You can make Emacs ring a bell (or blink the screen) to attract the user's attention. Be conservative about how often you do this; frequent bells can become irritating. Also be careful not to use beeping alone when signaling an error is appropriate. (See Errors.) -- Function: ding &optional dont-terminate This function beeps, or flashes the screen (see visible-bell below). It also terminates any keyboard macro currently executing unless dont-terminate is non-nil. -- Function: beep &optional dont-terminate This is a synonym for ding. -- Variable: visible-bell This variable determines whether Emacs will try to flash the screen to represent a bell. Non-nil means yes, nil means no. This is effective only if the termcap entry for the terminal in use has the visible bell flag (`vb') set. ΓòÉΓòÉΓòÉ 37.12. Window Systems ΓòÉΓòÉΓòÉ Emacs works with several window systems, most notably X Windows. Note that both Emacs and the X Window System use the term ``window'', but use it differently. The entire Emacs screen is a single window as far as X Windows is concerned; the individual Emacs windows are not known to X Windows at all. -- Variable: window-system This variable tells Lisp programs what window system Emacs is running under. Its value should be a symbol such as x (if Emacs is running under X Windows) or nil (if Emacs is running on an ordinary terminal). -- Variable: window-system-version This variable distinguishes between different versions of the X Window System. Its value is 10 or 11 when using X Windows; nil otherwise. -- Variable: window-setup-hook The value of the window-setup-hook variable is either nil or a function for Emacs to call after loading your `.emacs' file and the default initialization file (if any), after loading terminal-specific Lisp code, and after calling term-setup-hook. window-setup-hook is called with no arguments. This hook is used for internal purposes: setting up communication with the window system, and creating the initial window. Users should not interfere with it. ΓòÉΓòÉΓòÉ 38. Tips and Standards ΓòÉΓòÉΓòÉ This chapter describes no additional features of Emacs Lisp. Instead it gives advice on making effective use of the features described in the previous chapters. ΓòÉΓòÉΓòÉ 38.1. Writing Clean Lisp Programs ΓòÉΓòÉΓòÉ Here are some tips for avoiding common errors in writing Lisp code intended for widespread use: o Since all global variables share the same name space, and all functions share another name space, you should choose a short word to distinguish your program from other Lisp programs. Then take care to begin the names of all global variables, constants, and functions with the chosen prefix. This helps avoid name conflicts. This recommendation applies even to names for traditional Lisp primitives that are not primitives in Emacs Lisp---even to cadr. Believe it or not, there is more than one plausible way to define cadr. Play it safe; append your name prefix to produce a name like foo-cadr or mylib-cadr instead. If one prefix is insufficient, your package may use two or three alternative common prefixes, so long as they make sense. o It is often useful to put a call to provide in each separate library program, at least if there is more than one entry point to the program. o If one file foo uses a macro defined in another file bar, foo should contain (require 'bar) before the first use of the macro. (And bar should contain (provide 'bar), to make the require work.) This will cause bar to be loaded when you byte-compile foo. Otherwise, you risk compiling foo without the necessary macro loaded, and that would produce compiled code that won't work right. See Compiling Macros. o If you define a major mode, make sure to run a hook variable using run-hooks, just as the existing major modes do. See Hooks. o Please do not define C-c letter as a key. These sequences are reserved for users; they are the *only* sequences reserved for users, so we cannot do without them. Everything in Emacs that used to define such sequences has been changed, which was a lot of work. Abandoning this convention would waste that work and inconvenience the users. o It is a bad idea to define aliases for the Emacs primitives. Use the standard names instead. o Redefining an Emacs primitive is an even worse idea. It may do the right thing for a particular program, but there is no telling what other programs might break as a result. o If a file does replace any of the functions or library programs of standard Emacs, prominent comments at the beginning of the file should say which functions are replaced, and how the behavior of the replacements differs from that of the originals. o If a file requires certain standard library programs to be loaded beforehand, then the comments at the beginning of the file should say so. o Don't use next-line or previous-line in programs; nearly always, forward-line is more convenient as well as more predictable and robust. See Text Lines. o Don't use functions that set the mark in your Lisp code (unless you are writing a command to set the mark). The mark is a user-level feature, so it is incorrect to change the mark except to supply a value for the user's benefit. See The Mark. In particular, don't use these functions: - beginning-of-buffer, end-of-buffer - replace-string, replace-regexp If you just want to move point, or replace a certain string, without any of the other features intended for interactive users, you can replace these functions with one or two lines of simple Lisp code. o The recommended way to print a message in the echo area is with the message function, not princ. See The Echo Area. o When you encounter an error condition, call the function error (or signal). The function error does not return. See Signaling Errors. Do not use message, throw, sleep-for, or beep to report errors. o Avoid using recursive edits. Instead, do what the Rmail w command does: use a new local keymap that contains one command defined to switch back to the old local keymap. Or do what the edit-options command does: switch to another buffer and let the user switch back at will. See Recursive Editing. o In some other systems there is a convention of choosing variable names that begin and end with `*'. We don't use that convention in Emacs Lisp, so please don't use it in your library. The users will find Emacs more coherent if all libraries use the same conventions. o Indent each function with C-M-q (indent-sexp) using the default indentation parameters. o Don't make a habit of putting close-parentheses on lines by themselves; Lisp programmers find this disconcerting. Once in a while, when there is a sequence of many consecutive close-parentheses, it may make sense to split them in one or two significant places. o Please put a copyright notice on the file if you give copies to anyone. Use the same lines that appear at the top of the Lisp files in Emacs itself. If you have not signed papers to assign the copyright to the Foundation, then place your name in the copyright notice in place of the Foundation's name. ΓòÉΓòÉΓòÉ 38.2. Tips for Making Compiled Code Fast ΓòÉΓòÉΓòÉ Here are ways of improving the execution speed of byte-compiled lisp programs. o Use iteration rather than recursion whenever possible. Function calls are slow in Emacs Lisp even when a compiled function is calling another compiled function. o Using the primitive list-searching functions memq, assq or assoc is even faster than explicit iteration. It may be worth rearranging a data structure so that one of these primitive search functions can be used. For example, if you want to search a list of strings for a string equal to a given one, you can use an explicit loop: (let ((tail list)) (while (and tail (not (string= string (car tail)))) (setq tail (cdr tail)))) However, if you use a list of elements of the form (string), such as (("foo") ("#&") ("bar")), then you can search it with assoc: (assoc string list) The latter runs entirely in C code, so it is much faster. o Certain built-in functions are handled specially by the byte compiler avoiding the need for an ordinary function call. It is a good idea to use these functions rather than alternatives. To see whether a function is handled specially by the compiler, examine its byte-compile property. If the property is non-nil, then the function is handled specially. For example, the following input will show you that aref is compiled specially (see Array Functions) while elt is not (see Sequence Functions): (get 'aref 'byte-compile) => byte-compile-two-args (get 'elt 'byte-compile) => nil o Often macros result in faster execution than functions. For example, the following macro and the following function have the same effect when called, but code using the macro runs faster because it avoids an extra call to a user-defined function: (defmacro fast-cadr (x) (list 'car (list 'cdr x))) (defun slow-cadr (x) (car (cdr x))) ΓòÉΓòÉΓòÉ 38.3. Tips for Documentation Strings ΓòÉΓòÉΓòÉ Here are some tips for the writing of documentation strings. o Every command, function or variable intended for users to know about should have a documentation string. o An internal subroutine of a Lisp program need not have a documentation string, and you can save space by using a comment instead. o The first line of the documentation string should consist of one or two complete sentences which stand on their own as a summary. In particular, start the line with a capital letter and end with a period. The documentation string can have additional lines which expand on the details of how to use the function or variable. The additional lines should be made up of complete sentences also, but they may be filled if that looks good. o Do not start or end a documentation string with whitespace. o Format the documentation string so that it fits in an Emacs window on an 80 column screen. It is a good idea for most lines to be no wider than 60 characters. The first line can be wider if necessary to fit the information that ought to be there. However, rather than simply filling the entire documentation string, you can make it much more readable by choosing line breaks with care. Use blank lines between topics if the documentation string is long. o *Do not* indent subsequent lines of a documentation string so that the text is lined up in the source code with the text of the first line. This looks nice in the source code, but looks bizarre when users view the documentation. Remember that the indentation before the starting double-quote is not part of the string! o A variable's documentation string should start with `*' if the variable is one that users would want to set interactively often. If the value is a long list, or a function, or if the variable would only be set in init files, then don't start the documentation string with `*'. See Defining Variables. o The documentation string for a variable that is a yes-or-no flag should start with words such as ``Non-nil means...'', to make it clear both that the variable only has two meaningfully distinct values and which value means ``yes''. o When a function's documentation string mentions the value of an argument of the function, use the argument name in capital letters as if it were a name for that value. Thus, the documentation string of the function / refers to its second argument as `DIVISOR'. Also use all caps for meta-syntactic variables, such as when you show the decomposition of a list or vector into subunits, some of which may be variable. o When a documentation string refers to a Lisp symbol, write it as it would be printed (which usually means in lower case), with single-quotes around it. For example: ``lambda''. There are two exceptions: write t and nil without single-quotes. o Don't write key sequences directly in documentation strings. Instead, use the `\\[...]' construct to stand for them. For example, instead of writing `C-f', write `\\[forward-char]'. When the documentation string is printed, Emacs will substitute whatever key is currently bound to forward-char. This will usually be `C-f', but if the user has moved key bindings, it will be the correct key for that user. See Keys in Documentation. o In documentation strings for a major mode, you will want to refer to the key bindings of that mode's local map, rather than global ones. Therefore, use the construct `\\<...>' once in the documentation string to specify which key map to use. Do this before the first use of `\\[...]'. The text inside the `\\<...>' should be the name of the variable containing the local keymap for the major mode. ΓòÉΓòÉΓòÉ 39. GNU Emacs Internals ΓòÉΓòÉΓòÉ This chapter describes how the runnable Emacs executable is dumped with the preloaded Lisp libraries in it, how storage is allocated, and some internal aspects of GNU Emacs that may be of interest to C programmers. ΓòÉΓòÉΓòÉ 39.1. Building Emacs ΓòÉΓòÉΓòÉ The first step in building Emacs is to compile the C sources. This produces a program called temacs, also called a bare impure Emacs. It contains the Emacs Lisp interpreter and I/O routines, but not the editing commands. Then, to create a working Emacs editor, issue the command `temacs -l loadup'. This directs temacs to evaluate the Lisp files specified in the file `loadup.el'. These files set up the normal Emacs editing environment, resulting in an Emacs which is still impure but no longer bare. It takes long time to load the standard Lisp files. Luckily, you don't have to do this each time you run Emacs; temacs can dump out an executable program called xemacs which has these files preloaded. xemacs starts more quickly because it does not need to load the files. It is xemacs that is normally installed under the name emacs for users to run. To create xemacs, use the command `temacs -batch -l loadup dump'. The purpose of `-batch' here is to prevent temacs from trying to initialize any of its data on the terminal; this ensures that the tables of terminal information are empty in the dumped Emacs. When the xemacs executable is started, it will automatically load the user's `.emacs' file, or the default initialization file `default.el' if the user has none. With the `.emacs' file, you can produce a version of Emacs that suits you and is not the same as the version other people use. With `default.el', you can customize Emacs for all the users at your site who don't choose to customize it for themselves. (For further reflection: why is this different from the case of the barber who shaves every man who doesn't shave himself?) On some systems, dumping does not work. Then, you must start Emacs with the `temacs -l loadup' command each time you use it. This takes a long time, but since you need to start Emacs once a day at most---and once a week or less frequently if you never log out---the extra time is not too severe a problem. Before xemacs is dumped, the documentation strings for primitive and preloaded functions (and variables) need to be found in the file where they are stored. This is done by calling Snarf-documentation (see Accessing Documentation). These strings are omitted from temacs to save space. See Documentation Basics. -- Function: dump-emacs to-file from-file This function dumps the current state of Emacs into an executable file to-file. It takes symbols from from-file (this is normally the executable file `temacs'). If you use this function in an Emacs that was already dumped, you must set command-line-processed to nil first for good results. See Command Line Arguments. -- Command: emacs-version This function returns a string describing the version of Emacs that is running. It is useful to include this string in bug reports. (emacs-version) => "GNU Emacs 18.36.1 of Fri Feb 27 1987 on slug (berkeley-unix)" Called interactively, the function prints the same information in the echo area. -- Variable: emacs-build-time The value of this variable is the time at which Emacs was built at the local site. emacs-build-time => "Fri Feb 27 14:55:57 1987" -- Variable: emacs-version The value of this variable is the version of Emacs being run. It is a string, e.g. "18.36.1". ΓòÉΓòÉΓòÉ 39.2. Pure Storage ΓòÉΓòÉΓòÉ There are two types of storage in GNU Emacs Lisp for user-created Lisp objects: normal storage and pure storage. Normal storage is where all the new data which is created during an Emacs session is kept; see the following section for information on normal storage. Pure storage is used for certain data in the preloaded standard Lisp files: data that should never change during actual use of Emacs. Pure storage is allocated only while temacs is loading the standard preloaded Lisp libraries. In the file xemacs, it is marked as read-only (on operating systems which permit this), so that the memory space can be shared by all the Emacs jobs running on the machine at once. Pure storage is not expandable; a fixed amount is allocated when Emacs is compiled, and if that is not sufficient for the preloaded libraries, temacs crashes. If that happens, you will have to increase the compilation parameter PURESIZE in the file `config.h'. This normally won't happen unless you try to preload additional libraries or add features to the standard ones. -- Function: purecopy object This function makes a copy of object in pure storage and returns it. It copies strings by simply making a new string with the same characters in pure storage. It recursively copies the contents of vectors and cons cells. It does not make copies of symbols, or any other objects, but just returns them unchanged. It signals an error if asked to copy markers. This function is used only while Emacs is being built and dumped, and is called only in the file `emacs/lisp/loaddefs.el'. -- Variable: pure-bytes-used The value of this variable is the number of bytes of pure storage allocated so far. Typically, in a dumped Emacs, this number is very close to the total amount of pure storage available---if it were not, we would preallocate less. -- Variable: purify-flag This variable determines whether defun should make a copy of the function definition in pure storage. If it is non-nil, then the function definition is copied into pure storage. This flag is t while loading all of the basic functions for building Emacs initially (allowing those functions to be sharable and non-collectible). It is set to nil when Emacs is saved out as xemacs. The flag is set and reset in the C sources. You should not change this flag in a running Emacs. ΓòÉΓòÉΓòÉ 39.3. Garbage Collection ΓòÉΓòÉΓòÉ When a program creates a list or the user defines a new function (such as by loading a library), then that data is placed in normal storage. If normal storage runs low, then Emacs asks the operating system to allocate more memory in blocks of 1k bytes. Each block is used for one type of Lisp object, so symbols, cons cells, markers, etc. are segregated in distinct blocks in memory. (Vectors, buffers and certain other editing types, which are fairly large, are allocated in individual blocks, one per object, while strings are packed into blocks of 8k bytes.) It is quite common to use some storage for a while, then release it by, for example, killing a buffer or deleting the last pointer to an object. Emacs provides a garbage collector to reclaim this abandoned storage. (This name is traditional, but ``garbage recycler'' might be a more intuitive metaphor for this facility.) The garbage collector operates by scanning all the objects that have been allocated and marking those that are still accessible to Lisp programs. To begin with, all the symbols, their values and associated function definitions, and any data presently on the stack, are accessible. Any objects which can be reached indirectly through other accessible objects are also accessible. When this is finished, all inaccessible objects are garbage. No matter what the Lisp program or the user does, it is impossible to refer to them, since there is no longer a way to reach them. Their space might as well be reused, since no one will notice. That is what the garbage collector arranges to do. Unused cons cells are chained together onto a free list for future allocation; likewise for symbols and markers. The accessible strings are compacted so they are contiguous in memory; then the rest of the space formerly occupied by strings is made available to the string creation functions. Vectors, buffers, windows and other large objects are individually allocated and freed using malloc. Common Lisp note: unlike other Lisps, GNU Emacs Lisp does not call the garbage collector when the free list is empty. Instead, it simply requests the operating system to allocate more storage, and processing continues until gc-cons-threshold bytes have been used. This means that you can make sure that the garbage collector will not run during a certain portion of a Lisp program by calling the garbage collector explicitly just before it (provided that portion of the program does not use so much space as to force a second garbage collection). -- Command: garbage-collect This command runs a garbage collection, and returns information on the amount of space in use. (Garbage collection can also occur spontaneously if you use more than gc-cons-threshold bytes of Lisp data since the previous garbage collection.) garbage-collect returns a list containing the following information: ((used-conses . free-conses) (used-syms . free-syms) (used-markers . free-markers) used-string-chars used-vector-slots) (garbage-collect) => ((3435 . 2332) (1688 . 0) (57 . 417) 24510 3839) Here is a table explaining each element: used-conses The number of cons cells in use. free-conses The number of cons cells for which space has been obtained from the operating system, but that are not currently being used. used-syms The number of symbols in use. free-syms The number of symbols for which space has been obtained from the operating system, but that are not currently being used. used-markers The number of markers in use. free-markers The number of markers for which space has been obtained from the operating system, but that are not currently being used. used-string-chars The total size of all strings, in characters. used-vector-slots The total number of elements of existing vectors. -- User Option: gc-cons-threshold The value of this variable is the number of bytes of storage that must be allocated for Lisp objects after one garbage collection in order to request another garbage collection. A cons cell counts as eight bytes, a string as one byte per character plus a few bytes of overhead, and so on. (Space allocated to the contents of buffers does not count.) Note that the new garbage collection does not happen immediately when the threshold is exhausted, but only the next time the Lisp evaluator is called. The initial threshold value is 100,000. If you specify a larger value, garbage collection will happen less often. This reduces the amount of time spent garbage collecting, but increases total memory use. You may want to do this when running a program which creates lots of Lisp data. You can make collections more frequent by specifying a smaller value, down to 10,000. A value less than 10,000 will remain in effect only until the subsequent garbage collection, at which time garbage-collect will set the threshold back to 100,000. ΓòÉΓòÉΓòÉ 39.4. Writing Emacs Primitives ΓòÉΓòÉΓòÉ Lisp primitives are Lisp functions implemented in C. The details of interfacing the C function so that Lisp can call it are handled by a few C macros. The only way to really understand how to write new C code is to read the source, but we can explain some things here. An example of a special form is the definition of or, from `eval.c'. (An ordinary function would have the same general appearance.) DEFUN ("or", For, Sor, 0, UNEVALLED, 0, "Eval args until one of them yields non-NIL, then return that value.\n\ The remaining args are not evalled at all.\n\ If all args return NIL, return NIL.") (args) Lisp_Object args; { register Lisp_Object val; Lisp_Object args_left; struct gcpro gcpro1; if (NULL(args)) return Qnil; args_left = args; GCPRO1 (args_left); do { val = Feval (Fcar (args_left)); if (!NULL (val)) break; args_left = Fcdr (args_left); } while (!NULL(args_left)); UNGCPRO; return val; } Let's start with a precise explanation of the arguments to the DEFUN macro: 1. The first argument is the name of the Lisp symbol to define with this function; it is or. 2. The second argument is the C function name for this function. This is the name that is used in C code for calling the function. The name is, by convention, `F' prepended to the Lisp name, with all dashes (`-') in the Lisp name changed to underscores. Thus, to call this function from C code, call For. Remember that the arguments must be of type Lisp_Object; various macros and functions for creating values of type Lisp_Object are declared in the file `lisp.h'. 3. The third argument is a C variable name to use for a structure that holds the data for the subr object that represents the function in Lisp. This structure conveys the Lisp symbol name to the initialization routine that will create the symbol and store the subr object as its definition. By convention, this name is the C function name with `F' replaced with `S'. 4. The fourth argument is the minimum number of arguments that the function requires. In this case, no arguments are required. 5. The fifth argument is the maximum number of arguments that the function accepts. Alternatively, it can be UNEVALLED, indicating a special form that receives unevaluated arguments. A function with the equivalent of an &rest argument would have MANY in this position. Both UNEVALLED and MANY are macros. This argument must be one of these macros or a number at least as large as the fourth argument. 6. The sixth argument is an interactive specification, a string such as might be used as the argument of interactive in a Lisp function. In this case it is 0 (a null pointer), indicating that this function cannot be called interactively. A value of "" indicates an interactive function not taking arguments. 7. The last argument is the documentation string. It is written just like a documentation string for a function defined in Lisp, except you must write `\n\' at the end of each line. In particular, the first line should be a single sentence. After the call to the DEFUN macro, you must write the list of argument names that every C function must have, followed by ordinary C declarations for them. Normally, all the arguments must be declared as Lisp_Object. If the function has no upper limit on the number of arguments in Lisp, then in C it receives two arguments: the number of Lisp arguments, and the address of a block containing their values. These have types int and Lisp_Object *. Within the function For itself, note the use of the macros GCPRO1 and UNGCPRO. GCPRO1 is used to ``protect'' a variable from garbage collection---to inform the garbage collector that it must look in that variable and regard its contents as an accessible object. This is necessary whenever you call Feval or anything that can directly or indirectly call Feval. At such a time, any Lisp object that you intend to refer to again must be protected somehow. UNGCPRO cancels the protection of the variables that are protected in the current function. It is necessary to do this explicitly. For most data types, it suffices to know that one pointer to the object is protected; as long as the object is not recycled, all pointers to it remain valid. This is not so for strings, because the garbage collector can move them. When a string is moved, any pointers to it that the garbage collector does not know about will not be properly relocated. Therefore, all pointers to strings must be protected across any point where garbage collection may be possible. The macro GCPRO1 protects just one local variable. If you want to protect two, use GCPRO2 instead; repeating GCPRO1 will not work. There are also GCPRO3 and GCPRO4. In addition to using these macros, you must declare the local variables such as gcpro1 which they implicitly use. If you protect two variables, with GCPRO2, you must declare gcpro1 and gcpro2, as it uses them both. Alas, we can't explain all the tricky details here. Defining the C function is not enough; you must also create the Lisp symbol for the primitive and store a suitable subr object in its function cell. This is done by adding code to an initialization routine. The code looks like this: defsubr (&subr-structure-name); subr-structure-name is the name you used as the third argument to DEFUN. If you are adding a primitive to a file that already has Lisp primitives defined in it, find the function (near the end of the file) named syms_of_something, and add that function call to it. If the file doesn't have this function, or if you create a new file, add to it a syms_of_filename (e.g., syms_of_myfile). Then find the spot in `emacs.c' where all of these functions are called, and add a call to syms_of_filename there. This function syms_of_filename is also the place to define any C variables which are to be visible as Lisp variables. DEFVAR_LISP is used to make a C variable of type Lisp_Object visible in Lisp. DEFVAR_INT is used to make a C variable of type int visible in Lisp with a value that is an integer. Here is another function, with more complicated arguments. This comes from the code for the X Window System, and it demonstrates the use of macros and functions to manipulate Lisp objects. DEFUN ("coordinates-in-window-p", Fcoordinates_in_window_p, Scoordinates_in_window_p, 2, 2, "xSpecify coordinate pair: \nXExpression which evals to window: ", "Return non-nil if POSITIONS (a list, (SCREEN-X SCREEN-Y)) is in WINDOW.\n\ Returned value is list of positions expressed\n\ relative to window upper left corner.") (coordinate, window) register Lisp_Object coordinate, window; { register Lisp_Object xcoord, ycoord; if (!CONSP (coordinate)) wrong_type_argument (Qlistp, coordinate); CHECK_WINDOW (window, 2); xcoord = Fcar (coordinate); ycoord = Fcar (Fcdr (coordinate)); CHECK_NUMBER (xcoord, 0); CHECK_NUMBER (ycoord, 1); if ((XINT (xcoord) < XINT (XWINDOW (window)->left)) || (XINT (xcoord) >= (XINT (XWINDOW (window)->left) + XINT (XWINDOW (window)->width)))) { return Qnil; } XFASTINT (xcoord) -= XFASTINT (XWINDOW (window)->left); if (XINT (ycoord) == (screen_height - 1)) return Qnil; if ((XINT (ycoord) < XINT (XWINDOW (window)->top)) || (XINT (ycoord) >= (XINT (XWINDOW (window)->top) + XINT (XWINDOW (window)->height)) - 1)) { return Qnil; } XFASTINT (ycoord) -= XFASTINT (XWINDOW (window)->top); return (Fcons (xcoord, Fcons (ycoord, Qnil))); } Note that you cannot directly call functions defined in Lisp as, for example, the primitive function Fcons is called above. You must create the appropriate Lisp form, protect everything from garbage collection, and Feval the form, as was done in For above. `eval.c' is a very good file to look through for examples; `lisp.h' contains the definitions for some important macros and functions. ΓòÉΓòÉΓòÉ 39.5. Object Internals ΓòÉΓòÉΓòÉ GNU Emacs Lisp manipulates many different types of data. The actual data are stored in a heap and the only access that programs have to it is through pointers. Pointers are thirty-two bits wide in most implementations. Depending on the operating system and type of machine for which you compile Emacs, twenty-four to twenty-six bits are used to address the object, and the remaining six to eight bits are used for a tag that identifies the object's type. Because all access to data is through tagged pointers, it is always possible to determine the type of any object. This allows variables to be untyped, and the values assigned to them to be changed without regard to type. Function arguments also can be of any type; if you want a function to accept only a certain type of argument, you must check the type explicitly using a suitable predicate (see Type Predicates). ΓòÉΓòÉΓòÉ 39.5.1. Buffer Internals ΓòÉΓòÉΓòÉ Buffers contain fields not directly accessible by the Lisp programmer. We describe them here, naming them by the names used in the C code. Many are accessible indirectly in Lisp programs via Lisp primitives. name The buffer name is a string which names the buffer. It is guaranteed to be unique. See Buffer Names. save_modified This field contains the time when the buffer was last saved, as an integer. See Buffer Modification. modtime This field contains the modification time of the visited file. It is set when the file is written or read. Every time the buffer is written to the file, this field is compared to the modification time of the file. See Buffer Modification. auto_save_modified This field contains the time when the buffer was last auto-saved. last_window_start This field contains the window-start position in the buffer as of the last time the buffer was displayed in a window. undodata This field points to the buffer's undo stack. See Undo. syntax_table_v This field contains the syntax table for the buffer. See Syntax Tables. markers This field contains the chain of all markers that point into the buffer. At each deletion or motion of the buffer gap, all of these markers must be checked and perhaps updated. See Markers. backed_up This field is a flag which tells whether a backup file has been made for the visited file of this buffer. mark This field contains the mark for the buffer. The mark is a marker, hence it is also included on the list markers. See The Mark. local_var_alist This field contains the association list containing all of the variables local in this buffer, and their values. A copy of this list is returned by the function buffer-local-variables. See Buffer-Local Variables. mode_line_format This field contains a Lisp object which controls how to display the mode line for this buffer. See Mode Line Format. ΓòÉΓòÉΓòÉ 39.5.2. Window Internals ΓòÉΓòÉΓòÉ Windows have the following accessible fields: height The height of the window, measured in lines. width The width of the window, measured in columns. buffer The buffer which the window is displaying. This may change often during the life of the window. start The position in the buffer which is the first character to be displayed in the window. pointm This is the value of point in the current buffer when this window is selected; when it is not selected, it retains its previous value. left This is the left-hand edge of the window, measured in columns. (The leftmost column on the screen is column 0.) top This is the top edge of the window, measured in lines. (The top line on the screen is line 0.) next This is the window that is the next in the chain of siblings. prev This is the window that is the previous in the chain of siblings. force_start This is a flag which, if non-nil, says that the window has been scrolled explicitly by the Lisp program. At the next redisplay, if point is off the screen, instead of scrolling the window to show the text around point, point will be moved to a location that is on the screen. hscroll This is the number of columns that the display in the window is scrolled horizontally to the left. Normally, this is 0. use_time This is the last time that the window was selected. This field is used by get-lru-window. ΓòÉΓòÉΓòÉ 39.5.3. Process Internals ΓòÉΓòÉΓòÉ The fields of a process are: name A string, the name of the process. command A list containing the command arguments that were used to start this process. filter A function used to accept output from the process instead of a buffer, or nil. sentinel A function called whenever the process receives a signal, or nil. buffer The associated buffer of the process. pid An integer, the Unix process id. childp A flag, non-nil if this is really a child process. It is nil for a network connection. flags A symbol indicating the state of the process. Possible values include run, stop, closed, etc. reason An integer, the Unix signal number that the process received that caused the process to terminate or stop. If the process has exited, then this is the exit code it specified. mark A marker indicating the position of end of last output from this process inserted into the buffer. This is usually the end of the buffer. kill_without_query A flag, non-nil meaning this process should not cause confirmation to be needed if Emacs is killed. ΓòÉΓòÉΓòÉ 40. Standard Errors ΓòÉΓòÉΓòÉ Here is the complete list of the error symbols in standard Emacs, grouped by concept. The list includes each symbol's message (on the error-message property of the symbol), and a cross reference to a description of how the error can occur. Each error symbol has an error-conditions property which is a list of symbols. Normally, this list includes the error symbol itself, and the symbol error. Occasionally it includes additional symbols, which are intermediate classifications, narrower than error but broader than a single error symbol. For example, all the errors in accessing files have the condition file-error. As a special exception, the error symbol quit does not have the condition error, because quitting is not considered an error. See Errors, for an explanation of how errors are generated and handled. symbol string; reference. error "error"; see error in Errors. quit "Quit"; see Quitting. args-out-of-range "Args out of range"; see Sequences Arrays Vectors. arith-error "Arithmetic error"; see / and % in Numbers. beginning-of-buffer "Beginning of buffer"; see Motion. buffer-read-only "Buffer is read-only"; see Read Only Buffers. end-of-buffer "End of buffer"; see Motion. end-of-file "End of file during parsing"; see Input Functions. This is not a file-error. file-error See Files. This error, and its subcategories, do not have error-strings, because the error message is constructed from the data items alone when the error condition file-error is present. file-locked See File Locks. This is a file-error. file-already-exists See Writing to Files. This is a file-error. file-supersession See Buffer Modification. This is a file-error. invalid-function "Invalid function"; see Classifying Lists. invalid-read-syntax "Invalid read syntax"; see Input Functions. invalid-regexp "Invalid regexp"; see Regular Expressions. no-catch "No catch for tag"; see Catch and Throw. search-failed "Search failed"; see Searching and Matching. setting-constant "Attempt to set a constant symbol"; the values of the symbols nil and t may not be changed. void-function "Symbol's function definition is void"; see Function Cells. void-variable "Symbol's value as variable is void"; see Accessing Variables. wrong-number-of-arguments "Wrong number of arguments"; see Classifying Lists. wrong-type-argument "Wrong type argument"; see Type Predicates. ΓòÉΓòÉΓòÉ 41. Standard Buffer-Local Variables ΓòÉΓòÉΓòÉ The table below shows all of the variables that are automatically local (when set) in each buffer in Emacs Version 18 with the common packages loaded. abbrev-mode See Abbrevs. auto-fill-hook See Auto Filling. buffer-auto-save-file-name See Auto-Saving. buffer-backed-up See Backup Files. buffer-file-name See Buffer File Name. buffer-read-only See Read Only Buffers. buffer-saved-size See Point. case-fold-search See Searching and Case. ctl-arrow See Control Char Display. default-directory See System Environment. fill-column See Auto Filling. left-margin See Indentation. local-abbrev-table See Abbrevs. major-mode See Mode Help. mark-ring See The Mark. minor-modes See Minor Modes. mode-name See Mode Line Variables. overwrite-mode See Insertion. paragraph-separate See Standard Regexps. paragraph-start See Standard Regexps. require-final-newline See Insertion. selective-display See Selective Display. selective-display-ellipses See Selective Display. tab-width See Control Char Display. truncate-lines See Truncation. ΓòÉΓòÉΓòÉ 42. Standard Keymaps ΓòÉΓòÉΓòÉ The following symbols are used as the names for various keymaps. Some of these exist when Emacs is first started, others are only loaded when their respective mode is used. This is not an exhaustive list. Almost all of these maps are used as local maps. Indeed, of the modes that presently exist, only Vip mode and Terminal mode ever change the global keymap. Buffer-menu-mode-map A full keymap used by Buffer Menu mode. c-mode-map A sparse keymap used in C mode as a local map. command-history-map A full keymap used by Command History mode. ctl-x-4-map A sparse keymap for subcommands of the prefix C-x 4. ctl-x-map A full keymap for C-x commands. debugger-mode-map A full keymap used by Debugger mode. dired-mode-map A full keymap for dired-mode buffers. doctor-mode-map A sparse keymap used by Doctor mode. edit-abbrevs-map A sparse keymap used in edit-abbrevs. edit-tab-stops-map A sparse keymap used in edit-tab-stops. electric-buffer-menu-mode-map A full keymap used by Electric Buffer Menu mode. electric-history-map A full keymap used by Electric Command History mode. emacs-lisp-mode-map A sparse keymap used in Emacs Lisp mode. function-keymap The keymap for the definitions of keypad and function keys. If there are none, then it contains an empty sparse keymap. fundamental-mode-map The local keymap for Fundamental mode. It is empty and should not be changed. Helper-help-map A full keymap used by the help utility package. It has the same keymap in its value cell and in its function cell. Info-edit-map A sparse keymap used by the e command of Info. Info-mode-map A sparse keymap containing Info commands. lisp-interaction-mode-map A sparse keymap used in Lisp mode. lisp-mode-map A sparse keymap used in Lisp mode. mode-specific-map The keymap for characters following C-c. Note, this is in the global map. This map is not actually mode specific: its name was chosen to be informative for the user in C-h b (display-bindings), where it describes the main use of the C-c prefix key. mouse-map A sparse keymap for mouse commands from the X Window System. occur-mode-map A local keymap used in Occur mode. text-mode-map A sparse keymap used by Text mode. view-mode-map A full keymap used by View mode. ΓòÉΓòÉΓòÉ 43. Standard Hooks ΓòÉΓòÉΓòÉ The following is a list of hooks available with the distributed 18.52 version of GNU Emacs. Some of these hooks are called with run-hooks and can be a list of functions. Others are not called with run-hooks and may or may not allow a list of functions. For example, the suspend-hook can only reference a single function. See Hooks, for more information about using hooks. *Note:* in version 19, blink-paren-hook and auto-fill-hook are renamed to blink-paren-function and auto-fill-function respectively, since they are not called by the run-hooks function. auto-fill-hook blink-paren-hook c-mode-hook command-history-hook comment-indent-hook define-hooked-global-abbrev define-hooked-local-abbrev dired-mode-hook disabled-command-hook edit-picture-hook electric-buffer-menu-mode-hook electric-command-history-hook electric-help-mode-hook emacs-lisp-mode-hook find-file-hooks find-file-not-found-hooks fortran-comment-hook fortran-mode-hook ftp-setup-write-file-hooks ftp-write-file-hook indent-mim-hook LaTeX-mode-hook ledit-mode-hook lisp-indent-hook lisp-interaction-mode-hook lisp-mode-hook m2-mode-hook mail-mode-hook mail-setup-hook medit-mode-hook mh-compose-letter-hook mh-folder-mode-hook mh-letter-mode-hook mim-mode-hook news-mode-hook news-reply-mode-hook news-setup-hook nroff-mode-hook outline-mode-hook plain-TeX-mode-hook prolog-mode-hook protect-innocence-hook rmail-edit-mode-hook rmail-mode-hook rmail-summary-mode-hook scheme-indent-hook scheme-mode-hook scribe-mode-hook shell-mode-hook shell-set-directory-error-hook suspend-hook suspend-resume-hook temp-buffer-show-hook term-setup-hook terminal-mode-hook terminal-mode-break-hook TeX-mode-hook text-mode-hook vi-mode-hook view-hook write-file-hooks x-process-mouse-hook ΓòÉΓòÉΓòÉ 44. Index ΓòÉΓòÉΓòÉ Sorry, no fn index These words prevent "local variables" above from confusing Emacs.