Shareware Grab Bag

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

/ Shareware Grab Bag / Shareware Grab Bag.iso / 050 / pasform.arc / PASFORM.DOC next >

Wrap

Text File | 1986-05-11 | 77.0 KB | 2,039 lines

PTOOLS - PASCAL SOFTWARE WRITING TOOLS. 03 Jan 79 PASCAL TTTTTTTT OOOOO/ OOOOO/ LL SSSSS PP PP TT OO OO OO OO LL SS SS AA AA TT OO /OO OO /OO LL SS SSSSSS SOFTWARE TT OO / OO OO / OO LL SSSSS CC TT OO/ OO OO/ OO LL SS AA TT OO OO OO OO LL SS SS LL TT /OOOOO /OOOOO LLLLLLL SSSSS THIS IS THE PTOOLS WRITEUP 79/01/01. - ANDY MICKEL - JOHN P. STRAIT - RICK L. MARCUS - UNIVERSITY COMPUTER CENTER - UNIVERSITY OF MINNESOTA THE PTOOLS WRITEUP DESCRIBES VARIOUS SOFTWARE-WRITING TOOLS WHICH AID THE PASCAL PROGRAMMER IN ALL PHASES OF PROGRAM CODING AND DEVELOPMENT--FROM ENTRY OF SOURCE CODE THROUGH DOCUMENTATION. THIS SHEET IS MERELY THE TABLE OF CONTENTS FOR ALL OF THE TOOLS AVAILABLE. NAME PAGES DESCRIPTION ---- ----- ----------- COMPARE 5 COMPARE TWO TEXTFILES AND REPORT THEIR DIFFERENCES. MEASURE 5 OBTAIN PERFORMANCE MEASUREMENTS OF A PASCAL PROGRAM USING AUGMENT AND ANALYZE. PASCODE 1 DECOMPILE A RELOCATABLE-BINARY FILE PRODUCED BY THE PASCAL COMPILER. PASCREF 2 CROSS REFERENCE IDENTIFIERS IN A PASCAL PROGRAM AND GIVE A PROCEDURE MAP. PASVERT 2 CONVERT A TEXTFILE FROM/TO CDC SCIENTIFIC CHARACTER SET TO/FROM CDC ASCII SUBSET CHARACTER SET. PRETTY 3 PRETTYPRINT (AUTOMATICALLY INDENT) A PASCAL PROGRAM. SPRUCE 10 PRETTYPRINT (AUTOMATICALLY INDENT AND FORMAT) A PASCAL PROGRAM ACCORDING TO VARIOUS OPTIONS. WRITE 6 TRANSLATE CHARACTER SETS IN WRITEUPS. AT THE UNIVERSITY OF MINNESOTA THIS DOCUMENT IS AVAILABLE AS AN INDEXED WRITEUP WITH CHAPTERS OF A FEW PAGES EACH. THIS ALLOWS THE RETRIEVAL OF SELECTED CHAPTERS OR THE ENTIRE WRITEUP. TO GET A WRITEUP OF A SPECIFIC TOOL AT THE UNIVERSITY OF MINNESOTA, USE ONE OF THE COMMANDS BELOW. AT OTHER SITES, THE METHOD TO RETRIEVE SEPARATE CHAPTERS MAY BE DIFFERENT OR IMPOSSIBLE. WRITEUP(PTOOLS=NAME) - PRINT THE WRITEUP FOR "NAME". WRITEUP(PTOOLS=NAME1+...+NAMEN) - PRINT THE NAMED TOOLS. WRITEUP(PTOOLS=*) - PRINT THE ENTIRE PTOOLS WRITEUP. - 1 - COMPARE - Compare Two Textfiles 03 Jan 79 Compare - Compare Two Textfiles and Report Their Differences James F. Miner Social Science Research Facilities Center Andy Mickel University Computer Center University of Minnesota Minneapolis, MN 55455 USA Copyright (c) 1977, 1978. What COMPARE Does ----------------- COMPARE is used to display the differences between two similar texts (referred to as "FILEA" and "FILEB"). Such textfiles could be Pascal source programs, character data, documentation, etc. COMPARE is line-oriented, meaning the smallest unit of comparison is the text line (ignoring trailing blanks). COMPARE generates a report of differences (mismatches or extra text) between the two textfiles. The criterion for determining the locality of differences is the number of consecutive lines on each file which must match after a prior mismatch, and can be selected as a parameter. By selecting other parameters, you can direct COMPARE to restrict the comparison to various linewidths, mark column-wise the differences in pairs of mismatched lines, generate text-editor directives to be used to convert FILEA into FILEB, or generate a listing which will flag lines on FILEB indicating their addition or deletion as a result of the application of the editor directives. How to Use COMPARE ------------------ COMPARE is available as an operating system control statement on CDC 6000/Cyber 70,170 computer systems. The general form of the control statement is: COMPARE(a,b,list,modfile/options) COMPARE. means COMPARE(FILEA,FILEB,MODS/C6,D,W120) "FILEA" and "FILEB" are the names of the two textfiles being compared, "OUTPUT" is the report file, and "MODS" is the file name for the generation of text-editor directives if the "M" option is selected--see below. The various options are: C, D, F, M, P, and W. - 1 - COMPARE - Compare Two Textfiles 03 Jan 79 Cn Match Criterion (1 <= n <= 100). C determines the number of consecutive lines on each file which must match in order that they be considered as terminating a prior mismatch. C therefore affects COMPARE's "sensitivity" to the "locality" of differences. Setting C to a large value tends to produce fewer (but longer) mismatches than does a small value. C6 appears to give good results on Pascal source files, but may be inappropriate for other applications. Default: C6. D Report Differences. D directs COMPARE to display mismatches and extra text between FILEA and FILEB in a clearly annotated report. Only one of D, F, or M can be explicitly selected at one time. Default: selected. F Select Flag-form output. F directs COMPARE to list FILEB annotated with lines prefixed by an "A" or "D" indicating "additions" or "deletions" respectively. Such modifications could have been generated with the M option. Only one of D, F, or M can be explicitly selected at one time. Default: not selected. M Produce MODS file. M directs COMPARE to produce a file of "INSERT" or "DELETE" directives ready for the CDC MODIFY or UPDATE text editors (an "IDENT" directive must be added). The insertions and deletions will convert FILEA into FILEB. FILEA and FILEB should be files with sequencing appearing in columns beyond the linewidth specified by the W option. This is true of MODIFY and UPDATE "COMPILE" files (W72 is recommended). Sequence numbers are of the form: {Blanks} IdentName {Blanks} UnsignedInteger. Only one of D, F, or M can be explicitly selected at one time. Default: not selected. P Mark Pairs of mismatched lines. P alters the action of the D directive by marking differing columns in pairs of lines which mismatch in sections of equal length. This is especially useful for comparing packed data files. Default: not selected. Wn Specify significant line Width (length) (10 <= n <= 150). W determines the fixed number of columns of each line which will be compared. W is ideal to use when sequence informa- tion is present at the right edge of the text file. Default: W120. - 2 - COMPARE - Compare Two Textfiles 03 Jan 79 Example ------- Suppose FILEA is: PROGRAM L2U(INPUT, OUTPUT); (* CONVERT CDC 6/12-ASCII LOWER-CASE LETTERS TO UPPER CASE. *) BEGIN WHILE NOT EOF(INPUT) DO BEGIN WHILE NOT EOLN(INPUT) DO BEGIN IF INPUT^ <> CHR(76) THEN WRITE(INPUT^); GET(INPUT) END; READLN; WRITELN END; (*ALL DONE.*) END. and FILEB is: PROGRAM U2L(INPUT, OUTPUT); (* CONVERT CDC ASCII UPPER-CASE LETTERS TO 6/12 LOWER CASE. *) BEGIN WHILE NOT EOF(INPUT) DO BEGIN WHILE NOT EOLN(INPUT) DO BEGIN IF INPUT^ IN ['A'..'Z'] THEN WRITE(CHR(76)); WRITE(INPUT^); GET(INPUT) END; READLN; WRITELN END; END. - 3 - COMPARE - Compare Two Textfiles 03 Jan 79 then a report from COMPARE looks like this: COMPARE,L2U,U2L,LIST/C1,D,P. 78/12/31. 20.23.25. COMPARE VERSION 3.0 CDC (78/12/19) OUTPUT OPTION = DIFFERENCES. INPUT LINE WIDTH = 120 CHARACTERS. MATCH CRITERION = 1 LINES. FILEA: L2U FILEB: U2L *********************************** MISMATCH: L2U LINES 1 THRU 3 <NOT EQUAL TO> U2L LINES 1 THRU 3: A 1. PROGRAM L2U(INPUT, OUTPUT); B 1. PROGRAM U2L(INPUT, OUTPUT); ^ ^ A 2. (* CONVERT CDC 6/12-ASCII LOWER-CASE B 2. (* CONVERT CDC ASCII UPPER-CASE LETTERS ^^^^^^^^^^^^^^^^^^^^^^^^ A 3. LETTERS TO UPPER CASE. *) B 3. TO 6/12 LOWER CASE. *) ^^^^^^^ ^ ^^^^^^^^^^^^ ^^ *********************************** MISMATCH: L2U LINE 9 <NOT EQUAL TO> U2L LINES 9 THRU 10: A 9. IF INPUT^ <> CHR(76) THEN WRITE(INPUT^); B 9. IF INPUT^ IN ['A'..'Z'] THEN WRITE(CHR(76)); B 10. WRITE(INPUT^); *********************************** EXTRA TEXT ON L2U, BETWEEN LINES 15 AND 16 OF U2L A 15. (*ALL DONE.*) How COMPARE Works ----------------- COMPARE employs a simple backtracking-search algorithm to isolate mismatches from their surrounding matches. Each mismatch requires dynamic storage roughly proportional to the size of the largest mismatch, and time roughly proportional to the square of the size of the mismatch. Thus it may not be feasible to use COMPARE on files with very long mismatches. - 4 - COMPARE - Compare Two Textfiles 03 Jan 79 History ------- COMPARE was developed as a portable-Pascal software tool by James Miner of the Social Science Research Facilities Center at the University of Minnesota, in early 1977. It was written in standard Pascal and developed initially under CDC 6000 Pascal. Although the original version simply reported differences in a textfile, COMPARE was designed to fit naturally into a larger text-editing system. Plans for COMPARE's accommodating later enhancements to generate text-editor directives were made from the beginning. In summer of 1977, John Strait at the University of Minnesota Computer Center adapted COMPARE not only to generate such a modifications file, but also flag-form output and user-selectable options. COMPARE has been distributed to several Pascal enthusiasts in the United States who have made it operational on other Pascal implementa- tions. See Pascal News #12, May, 1978, pages 20-23. In late 1978, Willett Kempton of the Anthropology Department at the University of California Berkeley, installed COMPARE (with no changes required whatsoever) under Berkeley UNIX Pascal on a PDP 11/70 computer system. He later adapted the program to note column-wise differences in pairs of different lines and made minor changes to the format of the report. Rick Marcus and Andy Mickel at the University of Minnesota Computer Center made minor enhancements to COMPARE and fully documen- ted it it for Release 3 of Pascal 6000 in December, 1978. COMPARE is a model program in many respects. It serves to illustrate just how powerful and flexible such a comparison program can be. - 5 - MEASURE - Performance Measurement of Pascal Programs 03 Jan 79 PERFORMANCE MEASUREMENT OF PASCAL PROGRAMS USING AUGMENT AND ANALYZE - Andy Mickel 78/03/14. University Computer Center University of Minnesota Minneapolis, MN 55455 What AUGMENT and ANALYZE Do --------------------------- Suppose you want to examine the execution efficiency of your Pascal program--perhaps to make improvements to those parts which take the most computer time. AUGMENT and ANALYZE are designed to obtain rough measures of such execution times, particularly for large Pascal programs. Unlike other kinds of performance measurement, AUGMENT and ANALYZE assume the PROCEDURE and FUNCTION to be the smallest unit of a program to be monitored. This is a satisfactory assumption because well-written Pascal programs produced by stepwise refinement naturally are composed of proper-sized procedures and functions. The general principle used by these programs is that the value of the non-standard Pascal function CLOCK (which returns the elapsed processing time in milliseconds) can be sampled at procedure or function entry and exit. When the the expression (exittime - entrytime), is evaluated the time spent within the particular proce- dure or function can be ascertained. AUGMENT is the program which inserts the necessary CLOCK-sampling code into your Pascal source program for every procedure and function entry and exit. It thus causes your program to capture timing information and to write it out to a file. Next you compile and execute your program, which actually produces the file of dynamic timing measurements. ANALYZE then reads the timing file produced, and writes a report, which gives the name of each procedure or function, the number of times it was called, and the execution time it consumed for all calls and per call. AUGMENT and ANALYZE therefore provide a nearly machine-indepen- dent method for gathering performance-measurement data about a Pascal program. Most Pascal implementations have the required CLOCK function which returns the elapsed processor time in milliseconds. It is sometimes necessary to exclude the monitoring of excessive- ly called procedures and functions in large programs. A feature of AUGMENT allows you to specify any number of names to be excluded. - 1 - MEASURE - Performance Measurement of Pascal Programs 03 Jan 79 How to Use AUGMENT and ANALYZE ------------------------------ Under CDC 6000/Cyber 70,170 computer systems, AUGMENT and ANALYZE are control statements. Thus the following batch commands do the job: FETCH(AUGMENT,ANALYZE) AUGMENT(the file name of your Pascal source program) PASCAL(INTER/L-,G+) ANALYZE. The program headings for AUGMENT and ANALYZE are: AUGMENT(INPUT, EXCEPT, INTER, INTER2, OUTPUT) where: INPUT is the textfile containing the Pascal source program to be augmented. EXCEPT is the textfile containing a list of names (one to a line with no leading blanks) of procedures and functions to be excluded from measurement. EXCEPT can be an empty file in which case no procedures or functions will be excluded. INTER is the textfile on which the augmented version of the Pascal source program is written. INTER2 is the binary file on which only the names of each procedure and function in the Pascal source program is written for use by ANALYZE. OUTPUT is the textfile on which error messages are written if problems occur during augmenting. A report is written on OUTPUT verifying which procedures or functions were exclu- ded, if any. The error messages from AUGMENT are: *** TOO MANY PROCEDURES AND FUNCTIONS TO AUGMENT. (A limit of 2000 is imposed.) *** "BEGIN" EXPECTED. *** "END" EXPECTED. (There is something wrong with the statement part of the Pascal source program which is being augmented; it began with some reserved word other than "BEGIN" or there weren't enough "END"s to match "BEGIN"s.) *** "PROGRAM" EXPECTED. (AUGMENT couldn't find "PROGRAM" as the first reserved word in the Pascal source program. Possibly the INPUT file was empty.) *** UNDECLARED LABEL: xxx (AUGMENT couldn't find a label referred to by a GOTO statement.) - 2 - MEASURE - Performance Measurement of Pascal Programs 03 Jan 79 ANALYZE(OUTPUT, INTER2, TIMING) where: OUTPUT is the textfile on which the performance measurement report is written or alternatively the error message: *** TIMING FILE EMPTY. INTER2 is the binary file on which the names of each procedure and function in the Pascal source program was written by AUGMENT. TIMING is the binary file containing the dynamic timing measure- ments resulting from execution of the augmented Pascal program. Note: The identifier "TIMING" is added to the Pascal source program by AUGMENT and must not appear in any procedure or function which is to be monitored. When you use AUGMENT and ANALYZE, it is probably a good idea to consider the file names INTER, INTER2, and TIMING reserved. In summary, there are four steps to the performance measurement process: 1) [Pascal source program] -> AUGMENT -> INTER and INTER2 ******* 2) INTER -> PASCAL Compiler -> [Pascal binary program] *************** 3) [input data for Pascal program] -> Pascal binary program -> TIMING and [results ********************* from Pascal program] 4) TIMING and INTER2 -> ANALYZE -> [performance measurement ******* report] EXAMPLE ------- Below are a test program, its augmented version, and the performance measurement report: The source of the test program: PROGRAM TEST(OUTPUT); LABEL 5; VAR N: INTEGER; PROCEDURE A; PROCEDURE B; BEGIN N := N + 1; IF ODD(N DIV 2) THEN A ELSE B END (*B*) ; BEGIN (*A*) N := N + 1; - 3 - MEASURE - Performance Measurement of Pascal Programs 03 Jan 79 IF N > 200 THEN GOTO 5; B END (*A*) ; BEGIN N := 0; A; 5: END. The augmented version of the test program: PROGRAM TEST(OUTPUT,TIMING); LABEL 5; VAR TIMING:FILE OF PACKED RECORD I:0..2000;T:0..99999999;M:0..2 END; N: INTEGER; PROCEDURE A; PROCEDURE B; BEGIN WITH TIMING^ DO BEGIN I:= 3;T:=CLOCK;M:=0 END;PUT(TIMING); N := N + 1; IF ODD(N DIV 2) THEN A ELSE B ; WITH TIMING^ DO BEGIN I:= 3;T:=CLOCK;M:=1 END;PUT(TIMING) END (*B*) ; BEGIN WITH TIMING^ DO BEGIN I:= 1;T:=CLOCK;M:=0 END;PUT(TIMING); (*A*) N := N + 1; IF N > 200 THEN BEGIN WITH TIMING^ DO BEGIN I:= 2;T:=CLOCK;M:=2 END;PUT(TIMING); GOTO 5 END ; B ; WITH TIMING^ DO BEGIN I:= 2;T:=CLOCK;M:=1 END;PUT(TIMING) END (*A*) ; BEGIN REWRITE(TIMING); WITH TIMING^ DO BEGIN I:= 1;T:=CLOCK;M:=0 END;PUT(TIMING); N := 0; A; 5: ; WITH TIMING^ DO BEGIN I:= 1;T:=CLOCK;M:=1 END;PUT(TIMING) END . - 4 - MEASURE - Performance Measurement of Pascal Programs 03 Jan 79 The report from ANALYZE: PERFORMANCE MEASUREMENT SUMMARY FOR PASCAL PROGRAM: TEST . EXECUTION TIME CALLS (MILLISECONDS) MODULE TIMES PERCENT AVERAGE MODULE PERCENT NAME CALLED OF TOTAL PER CALL TOTAL OF TOTAL ---------- ------ -------- -------- ------ -------- A 52 25.490 0.15 8 27.586 B 151 74.020 0.13 20 68.966 TEST 1 0.490 1.00 1 3.448 ========== ====== ======== ======== ====== ======== TOTALS 204 100.000 0.14 29 100.000 From the summary provided by AUGMENT and ANALYZE, you can identify which procedures and functions to improve for greater execution efficiency. In general, it pays to concentrate on proce- dures and functions which are frequently called and take a significant amount of the total execution time of the program. Procedures and functions which have a large average execution time per call, but which are only called a few times are not worth worrying about. If one or more procedures or functions seem to dominate the results, it might be a good idea to monitor the program with these modules excluded from measurement. Use the except feature provided by AUGMENT. History ------- AUGMENT and ANALYZE were conceived originally under the names PROFILE and PRINRES in 1975-1976 by S. Matwin and M. Missala, of the Polish Academy of Sciences Computer Centre, PKiN, Warwaw, Poland. The goal of the project was to build a simple tool to measure very large programs--such as the Pascal 6000 compiler itself. A paper describing their successful work entitled: "A Simple, Machine Independent Tool for Obtaining Rough Measures of Pascal Programs," appeared in SIGPLAN Notices (11:8) August, 1976, pages 42-45. In 1976, Richard J. Cichelli of Lehigh University Mathematics Department and the American Newspaper Publishers Association Research Institute, obtained the programs and documented and distributed them to the Pascal community in the United States. In 1977, Herb Rubenstein and Andy Mickel of the University of Minnesota Computer Center, modified the programs for coding style and to increase portability, fixed bugs, and improved the performance of the programs themselves. We also removed several limitations (the built-in restrictions regarding the use of non-local GOTOs within procedures and functions as well as the monitoring of procedures named NEXTCH). See "Pascal News" 12, May, 1978, page 23. The programs are now supported with the Pascal-6000 system which is distributed to CDC installations around the world. - 5 - PASCODE - List Relocatable-Binary Code in Symbolic Form 03 Jan 79 PASCODE - List Relocatable-Binary Code in Symbolic Form J. P. Strait University Computer Center University of Minnesota Copyright (C) University of Minnesota - 1978 PASCODE lists relocatable-binary code produced by the Pascal 6000 compiler in pseudo-COMPASS form. To conserve space, one word of instructions is printed per line in columns which correspond to the four parcels in central memory words. If PASCODE is presented with a Pascal binary which was not compiled with post-mortem dump info entirely suppressed, it is able to differentiate between code, constants and post-mortem dump info. If it is presented with a non-Pascal binary, or one compiled with PMD info suppressed (Pascal compiler option P0, see the Pascal writeup), PASCODE will print every word in both octal and pseudo-COMPASS. PASCODE prints correct pseudo-COMPASS for Pascal binaries, and does fairly well for non-Pascal binaries. PASCODE does not handle all of the defined loader tables, only those that the Pascal compiler produces. It does not handle more than one external reference applied to a single address field, nor does it handle an external reference and program relocation applied to the same address field. Common blocks are not handled. Only one entry point is recognized in each relocatable record. Each procedure is assumed to be terminated by an End-Of-Record mark. In particular, the only loader tables that are processed are: 3400 - PIDL - Program identification and length. 3600 - ENTR - Entry point definitions. 4000 - TEXT - Relocatable text. 4300 - REPL - Replication of text. 4400 - LINK - External reference linkage. 7700 - PRFX - Prefix. All other tables are ignored. The program heading of PASCODE is: PROGRAM PASCODE(LGO,OUTPUT) and so to decode a binary on file LGO use the control statement: FETCH(PASCODE) PASCODE. and to decode a binary on file ABC use the control statement: FETCH(PASCODE) PASCODE(ABC) - 1 - PASCREF - CROSS REFERENCE PASCAL PROGRAMS. 03 Jan 79 PASCREF -- A Cross-Reference-Listing Generator for Pascal Programs - Andy Mickel 78/07/12. - University Computer Center - University of Minnesota What PASCREF Does ----------------- PASCREF is a software tool used to produce cross-reference listings of identifiers in a Pascal program. PASCREF lists the input program and numbers each line. Below the listing a table of all identifiers (except reserved words) is given together with a series of line numbers which identify each appearance of the identifier in the program. Generally, the first occurrence of each identifier denotes the line on which it was declared. Two exceptions to the exclusion of reserved words in the table are references to the symbols "LABEL" and "GOTO". Their inclusion is useful because they should be used carefully in a Pascal program. Any non-standard Pascal 6000 reserved words also appear in the cross-reference table. Following the cross-reference listing, a map of procedures and functions is given to aid their lookup in large programs. Facts About PASCREF ------------------- 1) The listing produced is paged to fit in top-binding folders-- perfect for nice "permanent" listings for reference use. 10) PASCREF can handle programs up to 100000 lines long and up to approximately 1500 identifiers. 3) The execution time for running PASCREF across the Pascal compiler (an 8000-line program) was 17.8 seconds on a Cyber 74. 4) The memory required to run PASCREF is proportional to the total number of identifiers and identifier occurrences in the program to be cross-referenced. For the 400-line PASCREF program itself, CM20000 was sufficient. How to Use PASCREF ------------------ PASCREF is available as a batch control statement under CDC operating systems. Its program heading is: XREF(INPUT,OUTPUT). The general form of the PASCREF control statement is: PASCREF(ProgramFileName, X-RefListingFileName/ Parameters) Two optional parameters are available: "W" and "U". U controls the width of each input line scanned: U+ = 72, U- = 120; default is U-. W controls the width of the cross-reference table written after the numbered program listing: W+ = 72, W- = 136; default is W-. W- generates nesting level indicators to the right of the listing annotating "BEGIN-END", "REPEAT-UNTIL", and "CASE-END" pairs. There- fore the default PASCREF control statement is: PASCREF(INPUT,OUTPUT/U-,W-) - 1 - PASCREF - CROSS REFERENCE PASCAL PROGRAMS. 03 Jan 79 To compile and run a Pascal program with a cross-reference listing, use the following command sequence: PASCAL,,LIST. PASCREF(LIST) LGO. PASCREF may produce the following error messages: 1) "TABLE FULL" (caused by too many identifiers). 2) "TEXT TOO LONG" (program length was more than 100000 lines). 3) "NO PROGRAM FOUND TO CROSS REFERENCE" (an empty file was given as input). Also a workspace overflow may result if insufficient memory is given to PASCREF. History ------- PASCREF was first written in January, 1971 by Niklaus Wirth at the Institut fur Informatik, E.T.H. Zurich. He revised it several times over the years for inclusion with subsequent releases of the Pascal 6000 compiler. The major improvement was employing a quicksort procedure in 1975. The program was modified slightly by Andy Mickel, Dan LaLiberte, and John Strait at the University of Minnesota to process a compiler listing with line numbers and titles and to append a map of procedures and functions. For more information see the book: "Algorithms + Data Structures = Programs," chapter 4, by Niklaus Wirth, Prentice Hall, 1976. - 2 - PASVERT - Pascal Program Character Set Conversion 03 Jan 79 PASVERT - Pascal Program Character Set Conversion J. P. Strait University Computer Center University of Minnesota Copyright (C) University of Minnesota - 1978 Program PASVERT reads a Pascal program written in either the CDC scientific character set or the CDC ASCII subset, and performs the required conversions to write the program in the other character set. It automatically supplies the appropriate "A" compiler option as the first line of the newly converted Pascal source program. The conversions for scientific to ASCII are- Pascal Symbol Old Name New Name ------------- -------- -------- string delimiter 60B equivalence 70B single quote not equal 64B not equal <> open comment 65B right arrow (* close comment 71B down arrow *) or 66B logical or OR and 67B logical and AND not 76B logical not NOT pointer 70B up arrow 76B circumflex less or equal 74B less or equal <= greater or equal 75B greater or equal >= The conversions for ASCII to scientific are- Pascal Symbol Old Name New Name ------------- -------- -------- string delimiter 70B single quote 60B equivalence pointer 76B circumflex 70B up arrow Characters within comments are converted, but those within strings are left alone. PASVERT requires a syntactically-correct (compilable) Pascal program as input. It performs no error checking. The program heading for PASVERT is- PROGRAM PASVERT(INPUT+,OUTPUT+) and so the control statement call is- PASVERT(IN,OUT/+) -or- PASVERT(IN,OUT/-) where "IN" is the file you want to convert, and "OUT" is the result. "IN" and "OUT" may not be the same actual file name. The character ('+' or '-') after the slash selects the conversion by indicating the - 1 - PASVERT - Pascal Program Character Set Conversion 03 Jan 79 appropriate setting of the "A" compiler option for the output file. That is, '+' selects a conversion from scientific to ASCII, and '-' selects a conversion from ASCII to scientific. The default value of this option is '+'. The input file is rewound before processing unless its name is actually "INPUT". The output file is not rewound. Conversion is done until an end-of-file mark is reached, but all logical record marks within that file are preserved. - 2 - PRETTY - Pascal Prettyprinting Program 03 Jan 79 PRETTY - Pascal Prettyprinting Program - Jon Hueras and Henry Ledgard 76/08/01. - Rick L. Marcus 78/09/15. What PRETTY Does ---------------- PRETTY takes as input a Pascal program or program fragment and reformats it according to a standard set of prettyprinting rules. The prettyprinted program is given as output. PRETTY makes no attempt to detect or correct syntactic errors in your program. However, syntac- tic errors may result in erroneous prettyprinting. PRETTY differs from some other prettyprinting programs in that it does not reformat your program "right down to every last semicolon." Such a philosophy requires formatting everything in a rigid fashion which is bound to be displeasing, or else it must provide you with a voluminous set of "options." Furthermore, such a scheme must do full syntax analysis of the program which means that it must be completely written and compilable. If you wish to prettyprint a program still under development, then you are out of luck. PRETTY, on the other hand, implements a strategy which states that it is not necessary to impose more than a "minimum" set of restrictions, and the prettyprinter should yield to your discretion beyond this minimum. PRETTY preserves blank lines and extra spaces you insert in the program as input, besides adding blank lines and extra spaces according to the rules for prettyprinting given below. General prettyprinting rules: 1. Spaces and blank lines already in your program are preserved by PRETTY, which adds (but does not subtract) blank lines and spaces. You can improve the readability of your programs by taking advantage of this feature. 2. All statements and declarations begin on separate lines. 3. No input line may be longer than 80 characters. Any output line longer than 80 characters is continued on a separate line. 4. The keywords "BEGIN", "END", "REPEAT", and "RECORD" are forced to stand on lines by themselves (or possibly followed by supporting comments). In addition, the "UNTIL" clause of "REPEAT-UNTIL" statement is forced to start on a new line. 5. A blank line is forced before the keywords "PROGRAM", "PROCEDURE", "FUNCTION", "LABEL", "CONST", "TYPE", and "VAR". 6. A space is forced before and after the symbols ":=" and "=". Additionally, a space is forced after the symbol ":". Note that only "="s in declarations are formatted; "="s in expressions are ignored. - 1 - PRETTY - Pascal Prettyprinting Program 03 Jan 79 Indentation rules: 1. The body of a declaration is indented from the "LABEL", "CONST", "TYPE", or "VAR" keyword which heads it. 2. The bodies of "BEGIN-END", "REPEAT-UNTIL", "WHILE", "WITH", and "CASE" statements, as well as "RECORD-END" structures and "CASE" variants (to one level) are indented from keywords which head them. 3. An "IF-THEN-ELSE" statement is indented as follows: IF <expression> THEN <statement> ELSE <statement> How to Use PRETTY ----------------- PRETTY is available as a KRONOS/NOS control statement on CDC 6000/Cyber 70,170 computer systems. Two file names are required: the local file name of program to be prettyprinted, and the local file name for the prettyprinted listing. The general form is: PRETTY(myprog,newprog) The parameters "myprog" and "newprog" are optional but must be specified in the order given. If "myprog" is omitted, the filename, INPUT, is assumed. If "newprog" is omitted, the filename, OUTPUT, is assumed. Data on INPUT is by convention from cards or an interactive terminal; data on OUTPUT goes to the line printer or to an interactive terminal. PRETTY formats only one record of the file named INPUT. ***NOTE*** You must delete line numbers from your program if you wish to use PRETTY. Use your favorite text editor. If the line numbers are not deleted the result will not be "pretty," nor will it compile. When to Use PRETTY ------------------ PRETTY is especially useful to reformat a Pascal program or program fragment that you have typed without regard to formatting (such as entering every line left-justified to save time). Because the program need not be syntactically correct, PRETTY provides the fastest means of obtaining a more readable version of your program. PRETTY is best used to prettyprint a Pascal program already on a file or punched on cards. PRETTY changes neither the cards nor the file used as input. Prettyprinting a program line-by-line as it is entered at a terminal is considerably less useful. - 2 - PRETTY - Pascal Prettyprinting Program 03 Jan 79 Examples -------- For Interactive Processing: 1. A Pascal program (or fragment) is on the file "myprog" and the desired prettyprinted listing is to be on the file "newprog": old,myprog -or- get,myprog (if myprog is not already a local x,rfl,20000 file) x,pretty(myprog,newprog) "newprog" is now the prettyprinted version of myprog. For Batch Processing: 2. A program is on cards (file INPUT) and the prettyprinted program is to be on the printer (file OUTPUT). The command sequence is: JOB statement. USER statement. PRETTY. 7-8-9 end of record Pascal program 6-7-8-9 end of information 3. In the same run as a compile and execute, a Pascal program is prettyprinted: JOB statement. USER statement. PRETTY,,TEMP. ("TEMP" here directs PRETTY to put the pretty- printed results on a file named TEMP.) PASCAL,TEMP. LGO. 7-8-9 Pascal Program 7-8-9 data, if any 6-7-8-9 History ------- PRETTY prettyprints a Pascal program or program fragment given as input. It was originally named PRETTYPRINT and was written in Pascal by Jon Hueras and Henry Ledgard, University Computing Center and Computer and Information Science Dept., University of Massachusetts at Amherst, in August, 1976. (See "An Automatic Formatting Program for Pascal", by Jon Hueras and Henry Ledgard, Pascal Newsletter 6, page 70, November, 1976.) Small corrections were made to PRETTY by Charles Fischer, Dept. of Computer Science, University of Wisconsin, Madison, in 1977. PRETTY was modified for CDC-6000 Pascal by Rick L. Marcus, University Computer Center, University of Minnesota, in September, 1978. - 3 - SPRUCE - Pascal Program Formatter 03 Jan 79 PASCAL PROGRAM FORMATTER - Michael N. Condict 1975 and 1978. - Rick L. Marcus 78/11/24. - Andy Mickel 78/12/01. What Spruce Does ---------------- Spruce is a flexible prettyprinter for Pascal programs. It takes as input a syntactically-correct Pascal program and produces as output an equivalent but reformatted Pascal program. The resulting program consists of the same sequence of Pascal symbols and comments, but they are rearranged with respect to line boundaries and columns for readability. Spruce maintains consistent spacing between symbols, breaks control and data structures onto new lines if necessary, indents lines to reflect the syntactic level of statements and declarations, and more. Miscellaneous features such as supplying line-numbers and automatic comments, or deleting all unnecessary blanks to save space, are described below. The flexibility of Spruce is accomplished by allowing you to supply various directives (options) which override the default values. Rather than being a rigid prettyprinter which decides for you how your program is to be formatted, you have the ability to control how formatting is done. Not only can prettyprinter directives be given prior to execution ,but also during execution through the use of directives embedded in your program. Experience with Spruce over the last three years has shown that most users can find a set of values for the directives which produce satisfactory results. The default values are typical. How To Use Spruce ----------------- Spruce is available as a KRONOS/NOS control statement on CDC 6000/Cyber 70,170 computer systems. Two parameters are required: "source", which is a correct (compilable) Pascal program to be prettyprinted, and "list", which is to be the reformatted version of the source program. The general form is: SPRUCE(source,list) or, if you wish to supply the initial values of Spruce directives on the control statement: SPRUCE(source,list/FormatterDirectiveList) See the section below for typical command sequences for running Spruce. Directives to Spruce may always be specified in the program - 1 - SPRUCE - Pascal Program Formatter 03 Jan 79 itself inside comments with a special syntax. Thus the first line of a program is an ideal spot for a comment containing directives. Subsequent use of embedded directives allows you to change the kind of prettyprinting for different sections of your program. The syntax of these special comments is given below (The syntax is given using "EBNF"--Extended Backus-Naur Form--see Communications ACM, November, 1977, page 822.): DirectiveComment = "(*" DirectiveList "*)" | "(*$" CompilerOptionList CommentText DirectiveList "*)". DirectiveList = "[" Directive {"," Directive} "]" CommentText. Directive = Letter Setting. Letter = "A" | "B" | "C" | "D" | "E" | "F" | "G" | "H" | "I" | "L" | "N" | "P" | "R" | "S" | "W". Setting = Switch | Value | Range. Switch = "+" | "-" . Value = "=" UnsignedInteger. Range = "=" UnsignedInteger "-" UnsignedInteger ["<" | ">"]. UnsignedInteger = Digit{Digit}. CommentText = {Any character except "]" or close-comment}. Note: As defined above, a Directive may be within a comment specifying a Pascal CompilerOptionList. On most implementations this is a "$" followed by a series of letters and values ("+", "-", or digits), separated by commas. See your local manual. Examples of DirectiveComments: (*[A=15, E=3, N=1,1<]*) - good for publication quality. (*[G=0, W=1-100, C+]*) - good for compact storage. (*$U+ [R=1-72, I=2]*) - an example of a DirectiveList with a CompilerOptionList. Directives to Spruce -------------------- A=n Align declarations. The A directive forces the alignment of ":" and "=" in declarations. If A is set to a value greater than 0, then n should be equal to the maximum identifier length for that section of your program. The A directive visually clarifies the declaration part of your program. See example below. Default: A=0 (no alignment). B+ or B- Bunch statements and declarations reasonably. B+ will place as many statements or declarations onto one - 2 - SPRUCE - Pascal Program Formatter 03 Jan 79 line as will fit within the specified write margins (W directive) subject to readability constraints. Bunching (B+) when the display is off (D-) has no effect. In general, B+ saves paper and prevents your program from becoming overly stretched in the vertical direction. See example below. Default: B- (one statement or statement part per line). C+ or C- Fully Compress program. C+ removes all non-essential blanks, end-of-lines, and com- ments from your program. A compilable, packed program will be written within the specified write margins (W directive). The number of spaces specified by the G directive will still be written between symbols. C+ might save some space on long-term storage media such as disk; you might store a program in compressed form and expand it later by reformat- ting with C-. Default: C-. D+ or D- Turn Display on or off. D allows you to selectively display portions of your program during prettyprinting. Therefore, D must be switched on and off with directives which are appropriately placed in your program. D is perhaps useful to obtain program fragments for publication (such as one or more procedures) without having to print the whole program. Default: D+. E=n Supply END comments. The E directive generates comments after "END" symbols if none are already there. Common Pascal coding styles fre- quently employ these comments. E=1 creates comments after the "END" symbol in compound statements which are within structured statements, as well as those constituting proce- dure and function bodies. The comments take the form: (*StatementPart*) or (*ProcedureName*). E=2 creates comments after the "BEGIN" and "END" symbols constituting procedure and function bodies only. E=0 creates no comments at all. E=3 means E=1 and E=2. See example below. Default: E=2. F+ or F- Turn Formatting (prettyprinting) on or off. F allows you to format selected portions of your program. F- causes Spruce to copy the input program directly with no changes. Therefore by switching F on and off with directives which are appropriately placed in your program, you can preserve text which is already properly formatted (such as comments). Default: F+ (of course!). G=n Specify symbol Gap. The G directive determines the number of spaces placed between Pascal symbols during prettyprinting. G=0 still places one space between two identifiers and reserved words. The symbols [ ] ( ) , and : are handled independently of G. Default: G=1. - 3 - SPRUCE - Pascal Program Formatter 03 Jan 79 I=n Specify Indent tab. I indents each nesting level of statements and declarations a given number of columns. Using I=2 or I=1 helps prevent excessively-narrow lines within the specified write margins (W directive) where there are heavily-nested constructs. Default: I=3. L=n Specify Line-wraparound indent tab. L determines the indentation of the remainder of statements or declarations which are too long to fit on one line. Default: L=3. N=x-y< or N=x-y> Generate line-numbers on the left or right. The N directive indicates the starting line-number (x) and the increment (y) for each succeeding line-number. If y > 0 then line-numbers are written outside the specified write margins for the formatted program in sequential order start- ing at x; y = 0 shuts off line-numbering. "<" writes up to 4-digit, right-justified line-numbers together with a trail- ing space to the left of each line. ">" writes 6-digit, zero-filled line-numbers to the right of each line. Use the N directive along with the W directive. Default: N=0-0> (no line-numbers). P=n Specify spacing between Procedure and function declarations. The P directive determines the number of blank lines to be placed between procedure and function declarations. n>2 makes procedures and functions visually stand out. Default: P=2. R=x-y Specify Read margins. The R directive indicates which columns are significant when Spruce reads from input file. R allows Spruce to accept files which have line-numbers in the first (x-1) columns or after the yth column. Default: R=1-999 (large enough to read to end-of-line in most cases). S=n Specify Statement separation. The S directive determines the number of spaces between statements bunched on the same line by the use of the B+ directive. Note that this directive is only in effect if B+ is used. Default: S=3. W=x-y Specify Write margins. The W directive indicates which columns are used for writing the reformatted program on the output file. Any line-numbers generated (N directive) are written outside these margins. Default: W=1-72. - 4 - SPRUCE - Pascal Program Formatter 03 Jan 79 Examples -------- The A Directive --------------- Here is a sample program fragment before using Spruce: PROGRAM SAMPLE(OUTPUT); CONST A=6; ABC='LETTERS'; THREE=3; TYPE RANGE=1..6; COLOUR=(RED,BLUE); VAR I,I2,I33,I444,I5555:RANGE; YES,NO,MAYBE:BOOLEAN; BEGIN END. Here is the output from Spruce with all defaults set: PROGRAM SAMPLE(OUTPUT); CONST A = 6; ABC = 'LETTERS'; THREE = 3; TYPE RANGE = 1 .. 6; COLOUR = (RED, BLUE); VAR I, I2, I33, I444, I5555: RANGE; YES, NO, MAYBE: BOOLEAN; BEGIN END (*SAMPLE*). Here is the output from Spruce having added a line with the A=5 directive: (*[A=5] ALIGN DECLARATIONS. *) PROGRAM SAMPLE(OUTPUT); CONST A = 6; ABC = 'LETTERS'; THREE = 3; TYPE RANGE = 1 .. 6; COLOUR = (RED, BLUE); VAR I, I2, - 5 - SPRUCE - Pascal Program Formatter 03 Jan 79 I33, I444, I5555: RANGE; YES, NO, MAYBE: BOOLEAN; BEGIN END (*SAMPLE*). The B Directive --------------- If the input to Spruce is: PROGRAM T(OUTPUT); CONST INCREMENT = 5; VAR I,J,N:INTEGER; BEGIN N:=0; J:=3; I:=SQR(N); N:=N+INCREMENT; IF N>73 THEN BEGIN DOTHIS; DOTHAT END ; IF N>5 THEN IF J>6 THEN DOSOMETHINGELSE; END. then the output from Spruce (using the default, B-) is: PROGRAM T(OUTPUT); CONST INCREMENT = 5; VAR I, J, N: INTEGER; BEGIN N := 0; J := 3; I := SQR(N); N := N + INCREMENT; IF N > 73 THEN BEGIN DOTHIS; DOTHAT END; IF N > 5 THEN IF J > 6 THEN DOSOMETHINGELSE; END (*T*). and the output from Spruce with B directives embedded is: (*[B+] BUNCH STATEMENTS. *) PROGRAM T(OUTPUT); CONST - 6 - SPRUCE - Pascal Program Formatter 03 Jan 79 INCREMENT = 5; VAR I, J, N: INTEGER; BEGIN N := 0; J := 3; I := SQR(N); N := N + INCREMENT; IF N > 73 THEN BEGIN DOTHIS; DOTHAT END; (*[B-] UNBUNCH. *) IF N > 5 THEN IF J > 6 THEN DOSOMETHINGELSE; END (*T*). The E Directive --------------- Suppose that a Pascal program fragment looked like: PROCEDURE SAMPLE; PROCEDURE INNER; BEGIN END; BEGIN IF X=3 THEN BEGIN X:=1; I:=I+1 END ELSE BEGIN X:=X+I; I:=0 END; WHILE (CH<>'X') AND FLAG1 DO BEGIN I:=I+3; INNER END; END; then using Spruce with E=3 produces: PROCEDURE SAMPLE; PROCEDURE INNER; BEGIN END (*INNER*); BEGIN (*SAMPLE*) IF X = 3 THEN BEGIN X := 1; I := I + 1 END (*IF*) ELSE BEGIN X := X + I; I := 0 END (*ELSE*); - 7 - SPRUCE - Pascal Program Formatter 03 Jan 79 WHILE (CH <> 'X') AND FLAG1 DO BEGIN I := I + 3; INNER END (*WHILE*); END (*SAMPLE*); How Spruce Works ---------------- Spruce parses your program by performing syntax analysis similar to the Pascal compiler: recursive descent within nested declarations and statements. It gathers characters into a buffer in which the indenting count of each character is maintained. The characters are being continually emptied from the buffer as new ones are added. Spruce has limited error-recovery facilities, and no results are guaranteed if a syntactically-incorrect program is supplied as input. The bane of most Pascal prettyprinters is the treatment of comments. Spruce considers them in the context of a declaration or statement. Therefore using comments like: CONST LS=6 (*LINESIZE*); is a good idea because Spruce will carry the comment along with the declaration. Similarly: BEGIN (* 'Z' < CH <= ' ' *) is also okay. Stand-alone comments however, receive rough treatment. Spruce always left-justifies the first line of such comments and places it on a separate line. See the F directive. Thus: CONST LS=6; (*LINESIZE*) will be reformatted as: CONST LS = 6; (*LINESIZE*) Proper treatment of comments is certainly an area of future development for Spruce. - 8 - SPRUCE - Pascal Program Formatter 03 Jan 79 Command Sequences ----------------- For Interactive Processing: 1. A Pascal program is on the file "myprog", and the spruced-up listing is to be on the file "newprog": old,myprog -or- get,myprog (if "myprog" is not already a local file) x,rfl,20000 x,spruce,myprog,newprog save,newprog "newprog" is now the spruced-up version of "myprog". 2. Rather than embedding directives into your program to change the default values, you desire to place them on the control statement: x,spruce,myprog,newprog/b+,s=2,i=2 For Batch Processing: 3. A program is on cards (file INPUT) and the spruced-up program is to go to the printer (file OUTPUT). The command sequence is: Job Statement. USER Statement. SPRUCE. 7-8-9 end-of-record Pascal program 6-7-8-9 end-of-information 4. In the same run as a compile and execute, a Pascal program is spruced-up: Job Statement. USER Statement. SPRUCE,,TEMP. ("TEMP" here directs SPRUCE to put the pretty- printed results on a file named TEMP.) PASCAL,TEMP. LGO. 7-8-9 end-of-record Pascal program 7-8-9 end-of-record data, if any 6-7-8-9 end-of-information - 9 - SPRUCE - Pascal Program Formatter 03 Jan 79 Spruce issues the following error messages: 1. " *** 'PROGRAM' EXPECTED." The Pascal program you fed to Spruce did not contain a standard Pascal program declaration. 2. " *** ERRORS FOUND IN PASCAL PROGRAM." Your program is syntactically incorrect. The output from Spruce probably does not contain all of the text from your input file. The cause could be any syntactic error, most commonly unmatched "BEGIN- END" pairs, or the lack of semicolons, string quotation marks, or the final period. 3. " *** STRING TOO LONG." Your program contains a character string (including both the quotes) which is wider than the specified write margins (W directive). 4. " *** NO PROGRAM FOUND TO FORMAT." The input file given to Spruce is empty. History ------- Spruce was originally written under the name "Format" in 1975 by Michael Condict as a class project in a Pascal programming course taught by Richard Cichelli at Lehigh University using CDC-6000 Pascal. After that, making improvements and adding directives became, tempora- rily, an obsession with the author (note limited usefulness of the D directive). Fortunately, the program eventually stabilized and is now in general use by Pascal programmers at Lehigh University and other institutions. After graduation the author transported Format in 1977 to a PDP-11 running under the Swedish Pascal compiler and RSX-11 with a total effort of 2 days. Spruce was modified for inclusion with Release 3 of CDC-6000 Pascal by Rick L. Marcus and Andy Mickel, University Computer Center, University of Minnesota, in November, 1978. - 10 - WRITE - A Writeup-Oriented Character Set Translator 03 Jan 79 WRITE - A Writeup-Oriented Character Set Translator. J. P. Strait University Computer Center University of Minnesota Copyright (C) University of Minnesota - 1978 WRITE converts the Pascal 6000 writeups to any of a variety of destination character sets and output devices. The Pascal 6000 writeups are maintained in CDC 6/12 ASCII character set and use the full ASCII character set. These writeups may be converted by using the "WRITE" program to be compatible with the character set(s) and output device(s) available at your site. The WRITE control statement has the following form: FETCH(WRITE) WRITE(input,output/options) the options are keywords which define various output characteristics. They are organized into 5 main groups. Main character set type: ------------------------ ASC - CDC ASCII subset character set (TERM,TTY in KRONOS and NOS terminology). See APPENDIX below. SCI - CDC scientific character set (TERM,TTYD in KRONOS and NOS terminology). A95 - 8-in-12 bit ASCII - characters are printed as actual ASCII values, one in each 12-bit byte. The upper bit in the byte is always set. This form corresponds to SCOPE and NOS/BE 96 character "IC=ASCII" character set. Destination printer type: ------------------------- LPT - Line printer. Uses carriage control characters '1', '0', and '+'. TELEX - Interactive terminal under KRONOS and NOS. Carriage control characters are expanded and ASCII "CR" is used for over- printing. INTRCOM - Interactive terminal under SCOPE and NOS/BE. Carriage control characters are expanded but '+' is used for over- printing. Underlining ability: -------------------- ULN - Underlining is possible on the output device. NOULN - Underlining is not possible on the output device. - 1 - WRITE - A Writeup-Oriented Character Set Translator 03 Jan 79 63/64 character set (ignored for A95 mode): ------------------------------------------- 63 - The installation character set is 63 character set. 64 - The installation character set is 64 character set. KRONOS and NOS 6/12 character set (ignored for A95 mode): --------------------------------------------------------- NORMAL - Only 6-bit display code will be used. This is required for upper-case-only line printers. EXTEND - The extended 6/12 ASCII characters are used. This option should not be used under SCOPE or NOS/BE. The default control statement is: WRITE(INPUT,OUTPUT/ASC,LPT,EXTEND,63,ULN) although the default options may be changed to be more appropriate to a particular computer system. *** *** *** *** *** The input to WRITE is assumed to be a line-printer ready file. The only carriage control characters that WRITE recognizes are '1', '0', '+', and ' '. Lines which start with any other character are ignored. If the output is intended for a line-printer (see the configuration variables), WRITE begins by printing a line consisting of the character 'Q' to de-select auto-page-eject. WRITE reads the input file line by line, and converts to ASCII codes as its internal character set. Linefeeds (LF) are used as end-of-line characters, formfeeds (FF) are used as end-of-line fol- lowed by page eject, and carriage returns (CR) are used to represent over-printing. Thus if the correct translation is not already available in this program, it is relatively easy to add a new one. Note: Some combinations may not be meaningful under a particular operating system. As an example, any of the Pascal-related writeups (and some others as well) at the University of Minnesota can be reprocessed by WRITE: WRITEUP,PASCAL=*/PT=AS,L=TEMP. FETCH(WRITE) WRITE(TEMP/TELEX) This will produce a version of the writeup for an interactive terminal. - 2 - WRITE - A Writeup-Oriented Character Set Translator 03 Jan 79 APPENDIX - ASCII Character Set Information ------------------------------------------ Andy Mickel 78/08/17. University Computer Center University of Minnesota ASCII (American Standard Code for Information Interchange) is the American variant of an officially-recognized, standard, international character set called the ISO (International Standards Organization) set. It specifies an 8-bit code for 128 characters (7 bits + 1 bit used for parity). Within the ISO code there may exist national variants for certain symbols (such as the currency symbol: $ ). The 128 characters consist of 95 which print as single graphics and 33 which are used for device control. The backspace control character is specifically used to allow overprinting of characters such as accents on some letters in foreign languages. ASCII is the only character set to receive formal approval in the United States, having been adopted unanimously (including IBM) within ANSI (American National Standards Institute) during the period 1963-65. Of course IBM then made no effort to change from EBCDIC (Extended Binary Coded Decimal Interchange Code), and so we in the computing field have 2 character sets to worry about. Nevertheless ASCII is widely accepted as the industry standard by an overwhelming number of computer manufacturers, and most modern equipment (namely interactive terminals and personal (micro) computers) use ASCII. Today's programming languages are defined in terms of ASCII. There are several official subsets of ASCII as well (which are presented below). The 63- and 64-character subsets are for 6-bit hardware representations onto which the other characters in full ASCII may be folded. Lower-case letters fold to upper case, curly braces to square brackets, etc. Even the keyboards of most interactive terminals exhibit the systematic ordering of ASCII for upper-case and control characters. The 63-character subset: 0 1 2 3 4 5 6 7 8 9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z + - * / = ( ) [ ] < > ^ ' " ! ? ; : , . @ # $ & % _ space The 64-character subset: the 63 set except [ and ] plus { } \ The 66-character subset: - 3 - WRITE - A Writeup-Oriented Character Set Translator 03 Jan 79 the 63 set plus { } \ The 89-character subset: the 63 set plus a b c d e f g h i j k l m n o p q r s t u v w x y z The 95-character subset: the 89 set plus { } \ | ` ~ The full 128-character set: the 95 set plus the 33 control characters: ACK (Acknowledge) FF (Form Feed) BEL (Bell) FS (File Separator) BS (Backspace) GS (Group Separator) CAN (Cancel) HT (Horizontal Tab) CR (Carriage Return) LF (Line Feed) DC1 (Device Control 1) NAK (Negative Acknowledge) DC2 (Device Control 2) NUL (Null) DC3 (Device Control 3) RS (Record Separator) DC4 (Device Control 4) SI (Shift In) DEL (Delete) SO (Shift Out) DLE (Data Link Escape) SOH (Start of Heading) EM (End of Medium) STX (Start of Text) ENQ (Enquiry) SUB (Substitute) EOT (End of Transmission) SYN (Synchronous Idle) ESC (Escape) US (Unit Separator) ETB (End of Transmission Block) VT (Vertical Tab) ETX (End of Text) 00 20 40 60 100 120 140 160 ---------------------------------------- 0 | NUL DLE 0 @ P ` p 1 | SOH DC1 ! 1 A Q a q 2 | STX DC2 " 2 B R b r 3 | ETX DC3 # 3 C S c s 4 | EOT DC4 $ 4 D T d t 5 | ENQ NAK % 5 E U e u 6 | ACK SYN & 6 F V f v 7 | BEL ETB ' 7 G W g w 10 | BS CAN ( 8 H X h x 11 | HT EM ) 9 I Y i y 12 | LF SUB * : J Z j z 13 | VT ESC + ; K [ k { 14 | FF FS , < L \ l | 15 | CR GS - = M ] m } 16 | SO RS . > N ^ n ~ 17 | SI US / ? O _ o DEL - 4 - WRITE - A Writeup-Oriented Character Set Translator 03 Jan 79 The 7-bit octal code for a character is the sum of the column and row numbers. For example, the code for the letter G is 7 + 100 = 107 octal. CDC KRONOS/NOS ASCII Representations (63-Character Set) ------------------------------------ --------------------- ----------------------- Internal Normal ASCII Internal Extended ASCII display ------------ display -------------- 6/12-bit 7-bit 6/12-bit 7-bit octal char octal octal char octal --------------------- ----------------------- 00 7600 01 A 101 7601 a 141 02 B 102 7602 b 142 03 C 103 7603 c 143 04 D 104 7604 d 144 05 E 105 7605 e 145 06 F 106 7606 f 146 07 G 107 7607 g 147 10 H 110 7610 h 150 11 I 111 7611 i 151 12 J 112 7612 j 152 13 K 113 7613 k 153 14 L 114 7614 l 154 15 M 115 7615 m 155 16 N 116 7616 n 156 17 O 117 7617 o 157 20 P 120 7620 p 160 21 Q 121 7621 q 161 22 R 122 7622 r 162 23 S 123 7623 s 163 24 T 124 7624 t 164 25 U 125 7625 u 165 26 V 126 7626 v 166 27 W 127 7627 w 167 30 X 130 7630 x 170 31 Y 131 7631 y 171 32 Z 132 7632 z 172 33 0 060 7633 { 173 34 1 061 7634 | 174 35 2 062 7635 } 175 36 3 063 7636 ~ 176 37 4 064 7637 DEL 177 40 5 065 7640 NUL 000 41 6 066 7641 SOH 001 42 7 067 7642 STX 002 43 8 070 7643 ETX 003 44 9 071 7644 EOT 004 45 + 053 7645 ENQ 005 46 - 055 7646 ACK 006 47 * 052 7647 BEL 007 50 / 057 7650 BS 010 - 5 - WRITE - A Writeup-Oriented Character Set Translator 03 Jan 79 51 ( 050 7651 HT 011 52 ) 051 7652 LF 012 53 $ 044 7653 VT 013 54 = 075 7654 FF 014 55 SPACE 040 7655 CR 015 56 , 054 7656 SO 016 57 . 056 7657 SI 017 60 # 043 7660 DLE 020 61 [ 133 7661 DC1 021 62 ] 135 7662 DC2 022 63 : 072 7663 DC3 023 64 " 042 7664 DC4 024 65 _ 137 7665 NAK 025 66 ! 041 7666 SYN 026 67 & 046 7667 ETB 027 70 ' 047 7670 CAN 030 71 ? 077 7671 EM 031 72 < 074 7672 SUB 032 73 > 076 7673 ESC 033 74 @ 100 7674 FS 034 75 \ 134 7675 GS 035 76 ^ 136 7676 RS 036 77 ; 073 7677 US 037 --------------------- 7400 7401 @ 100 7402 ^ 136 7403 7404 % 045 7405 7406 7407 ` 140 ----------------------- References ========== "A View of the History of the ISO Character Code" R. W. Bemer in HONEYWELL COMPUTER JOURNAL, Vol. 6, No. 4, 1972, pp 274-289. "Hints on Distributing Portable Software" William Waite in SOFTWARE PRACTICE AND EXPERIENCE, Vol. 5, 1975, pp 295-308. NOS 1.2 REFERENCE MANUAL VOLUME 1, CDC, Pub. No. 60435400, Appendix A. - 6 - /