PDA Software Library

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

/ PDA Software Library / pdasoftwarelib.iso / HP95_100 / CALC / CMCALC18 / CMCALC.DOC < prev next >

Wrap

Text File | 1995-03-21 | 30.9 KB | 796 lines

--------------------------------------------------------------- ----- COMPUTER MATH CALCULATOR ----- ----- FOR THE HP-100/200LX ----- ----- ----- ----- Version 18 ----- --------------------------------------------------------------- ----- 18 March 1995 ----- --------------------------------------------------------------- Copyright (c) 1991, 1992, 1994, 1995 by Eugene Dorr All rights reserved. --- --- --- Introduction --- --- --- CMCalc for the HP-100/200LX is a system-manager compliant application which provides a rich set of computer-science- oriented integer manipulations. It features variable word size (from 1 to 32 bits), selectable complement modes (unsigned, 1's complement, and 2's complement), four number bases (binary, octal, decimal and hexadecimal), and visible carry and overflow indicators. Only RPN entry is supported. Because CMCalc is "system-manager compliant" it can be started and used along with all other system-manager compliant applica- tions, including the built-in applications. You can switch between these applications without quitting any of them, and when you return to an application, you are returned right where you were when you left. Of course, this also means that CMCalc will operate ONLY on HP 100LX computers. (Note: Previous versions of CMCalc were written for the HP-95LX palmtop computer -- the predecessor to the HP-100LX. However, all references in this document to "CMCalc" are to the version of CMCalc written for the HP-100LX.) --- --- --- Manifest --- --- --- CMCalc is distributed in an archive called CMCALC18.ZIP. When "unzipped" you should have three files: CMCALC.DOC -- this documentation file CMC100.EXM -- the executable file CMC100.ICN -- an icon file for CMCalc --- --- --- Installing CMCalc --- --- --- Copy the following files to to your HP-100 (They must both be copied to the same subdirectory): CMC100.EXM CMC100.ICN Start the 'MORE' application (also known as 'Application Manager') by pressing the [&...] key. Press [F2] (Add), and enter the following values: Name: CMCalc Path: (drive):\(subdirectory)\CMC100.EXM Comment: In the Path field, (drive) and (subdirectory) should be replaced with the drive letter (a or c) and the subdirectory to which you copied the CMCalc files. Now tab to the Icon field, and the CMCalc icon should appear. If it does not, press the down-arrow key until you see the CMCalc icon. Press F10 (OK). Now press the key-combination you want to use as the hot-key for CMCalc (I use Alt+Calc; if you want to use something else, the dialog which is displayed at this point describes what your options are). Now you can start CMCalc by pressing the key combination you selected. --- --- --- Using CMCalc --- --- --- To start CMCalc once it has been installed, hold down the ALT key and press the blue HP CALC key (or whatever key you made the hot-key during the installation process). To quit CMCalc, press MENU, Q(uit). CMCalc uses the HP 100 calculator paradigm, so if you are familiar with using the built-in HP 100 calculator, operation of CMCalc should be straight-forward. CMCalc uses the RPN mode of data entry. This documentation assumes that you are familiar with RPN. The CMCalc calculator operates on integers only. If you need to operate on floating-point numbers, you can use the HP 100 built-in calculator and use CUT and PASTE to transfer numbers between CMCalc and the built-in calculator application. --- --- --- Controlling How CMCalc Works --- --- --- You can control how CMCalc operates in several ways. The following options are controlled in the Options/Number Format dialog: * you can specify how many bits will be used to store numbers (the word size -- from 1 to 32 bits); * you can specify what complement mode is to be used (the complement mode affects the results of arithmetic functions, and the display of numbers in decimal mode); * you can specify the number base you wish numbers to be displayed in (base 2 -- binary, base 8 -- octal, base 10 -- decimal, or base 16 -- hexadecimal); The following options are controlled using the Options/ Calculator Mode dialog: * you can specify what character set standard (i.e ASCII/ISO or EBCDIC) when using character entry mode; The following options are controlled from the Options/Display Options dialog: * you can specify which of the following windows will be displayed: - the RPN stack window - the flags window - the "views" window * within the "views" windows you can select which of the following views of the X register are displayed: - hexadecimal - decimal - octal - binary - ASCII/ISO - EBCDIC (The term "ASCII/ISO" as used in this document means ISO 646, or ANSI X3.4, which is a seven-bit character set. "EBCDIC" means IBM EBCDIC code page 37.) * you can specify whether you want leading zeros to be displayed; * you can specify whether to use "dotted decimal" notation. (Dotted decimal notation is the method used to represent IP addresses. Each byte is represented by a decimal number from 0 to 255, and separated from adjacent bytes by a period.) Note that there is not enough room on the display for all windows and all views at the same time. However, the Overflow and Carry flags will be shown on the status line if the flags window is not displayed. The settings you select are recorded in the file CMC100.ENV, which is stored in the _DAT subdirectory on the C: drive. Each time you start CMCalc, the settings which were in effect the last time CMCalc was used are restored from this file. (The contents of the stack, the storage registers, the overflow and carry flags, and the current F-key set are also restored.) The menu functions are summarized below. MENU/File /New -- not used /Open -- not used /Save -- not used /Exit -- exits CMCalc MENU/Edit /Insert -- not used /Delete -- not used /Cut -- not used /Copy -- copies the number in the X register, formatted in the prevailing base and display options, to the clipboard /Paste -- pastes a character-formatted number from the clipboard into the X register MENU/Clear /Flags -- zeros the Carry and Overflow flags /Stack -- zeros all registers in the RPN stack /Registers -- zeros all storage registers 0-9 and A-F /Calc Line -- zeros the X register MENU/Options /Number format... -- controls the prevailing base, the complement mode and the wordsize /Calulator Modes... -- controls the character code used in character entry mode /Display Options... -- controls what information is displayed, and in what format MENU/Quit -- ends CMCalc MENU/Help /Using CMCalc... -- provides reference information /Controlling CMCalc... -- provides reference information /About CMCalc... -- provides version and contact info /Registering CMCalc... -- provide registration info --- --- --- Functions --- --- --- In the following descriptions, upper case X, Y, Z and T refer to the stack registers, lower case x, y, z and t refer to the values contained in the corresponding stack registers. No stack movement occurs unless explicitly stated. The Carry and Overflow flags are unaffected unless stated otherwise. The Carry flag is usually used to indicate the value of a bit in the result which could not be accomodated in the current size of the register, for example, a carry during addition, a borrow during subtraction, or a bit shifted out of a register during a shift function. The Overflow flag is used to indicate that the result could not be expressed within the current register size and complement mode. For example, if the wordsize is 8 and the complement mode is 2's complement, then the largest positive number which can be expressed is 127, and the largest negative number which can be expressed is -128. If the CHS (change sign) function is applied to -128, then the Overflow flag will be set, since +128 cannot be expressed in 8 bits. Some functions (the ones which refer to bit positions) use the absolute (positive) value of x. This is indicated by the notation |x|. As with the HP 100 built-in calculator, many functions are assigned to F-keys, which are labelled on the bottom of the display. Four groups of functions are available, and the different groups are accessed by pressing F10 (labelled 'More'). In the following descriptions, functions which are assigned to F-keys carry the notation '(Fkey)'. -------------------- ARITHMETIC FUNCTIONS -------------------- + Add -- (+) x is added to y, result returned in X. The stack drops. Carry and Overflow are set as appropriate for the complement mode. - Subtract -- (-) x is subtracted from y, result returned in X. The stack drops. Carry and Overflow are set as appropriate for the complement mode. * Multiply -- (*) x is multiplied by y, result returned to X, the stack drops. Overflow is set if the result cannot be represented in the current word size and complement mode. / Divide -- (/) The quotient of y divided by x, result returned in X. The stack drops. The Carry flag is set if there is a remainder, otherwise it is cleared. Overflow is set only in 2's complement mode when the largest possible negative number is divided by -1. In all other cases it is cleared. The function is inhibited and an error message is displayed if x is 0 or -0. ABS Absolute value -- (Fkey) If x is negative (this can only be true if the complement mode is not unsigned) then it is made positive, if possible, the result is returned to X, and Overflow is reset. If it is not possible to make x positive (i.e. the complement mode is 2's complement, and x is the largest possible negative number), then X remains unchanged and the Overflow flag is set. RMD Remainder -- (Fkey) y is divided by x, the quotient is discarded, and the remainder is returned in X. The stack drops. The function is inhibited and an error message is displayed if x is 0 or -0. CHS Change Sign -- (assigned to the 'L' key) Complements x, result returned in X. Carry and Overflow are set as appropriate for the complement mode. The function is inhibited and an error message is displayed in unsigned mode. ----------------------- BOOLEAN LOGIC FUNCTIONS ----------------------- AND Bitwise AND -- (Fkey) Bitwise logical AND of x and y, result returned in X. The stack drops. OR Bitwise OR -- (Fkey) Bitwise logical OR of x and y, result returned in X. The stack drops. XOR Bitwise Exclusive OR -- (Fkey) Bitwise logical exclusive OR of x and y, result returned in X. The stack drops. NOT Bitwise NOT -- (Fkey) Bitwise logical NOT of x, result returned in X. -------------------------- BIT-MANIPULATION FUNCTIONS -------------------------- MASKL Mask left -- (Fkey) A bit mask of the size specified by |x| is created and placed in the leftmost positions of X, replacing x. MASKR Mask right -- (Fkey) A bit mask of the size specified by |x| is created and placed in the rightmost positions of X, replacing x. SB Set bit -- (Fkey) The bit specified by |x| (x=0 specifies the rightmost bit) is set in y, the result is returned in X. The stack drops. CB Clear bit -- (Fkey) The bit specified by |x| (x=0 specifies the rightmost bit) is cleared in y, the result is returned in X. The stack drops. REV Reverse bit order -- (Fkey) The order of the bits in x is reversed, the result is returned in X. REVBYTE Reverse byte order -- (Fkey) The order of the bytes in x is reversed, the result is returned in X. REVB is only valid when the word size is an integral number of bytes (in other words, a multiple of 8). BITS Bit count -- (Fkey) The number of 1-bits in x is returned in X. -------------------------- SHIFT AND ROTATE FUNCTIONS -------------------------- SL Shift left -- (Fkey or Left-arrow key) x is shifted to the left by one position, a zero bit is introduced into the rightmost bit position, and the result is returned in X. The Carry flag is set to the value of the leftmost bit of x. +---+ +----------------+ | | <-- | <----- | <-- 0 +---+ +----------------+ Carry SR Shift right -- (Fkey or Right-arrow key) x is shifted to the right by one position, a zero bit is introduced into the leftmost bit position, and the result is returned in X. The Carry flag is set to the value of the rightmost bit of x. +----------------+ +---+ 0 --> | -----> | --> | | +----------------+ +---+ Carry SLn Shift left multiple -- (Fkey) y is shifted to the left by the number of positions specified by |x|, zero bits are introduced into the vacated rightmost bit positions, the result is returned in X. The stack drops. The Carry flag is set to the value of the last bit to be shifted from the leftmost position of y. SRn Shift right multiple -- (Fkey) y is shifted to the right by the number of positions specified by |x|, zero bits are introduced into the vacated leftmost bit positions, the result is returned in X. The stack drops. The Carry flag is set to the value of the last bit to be shifted from the rightmost position of y. ASR Arithmetic shift right -- (Fkey) x is shifted to the right by one position, the leftmost bit (which, in 1's complement and 2's complement mode, represents the sign of the value of x) retains its original value (and therefore the sign), the result is returned in X. The Carry flag is set to the value of the leftmost bit of x. +------+ | | | +----------------+ +---+ +--> | -----> | --> | | +----------------+ +---+ Carry LJ Left justify -- (Fkey) If x<0, then x is shifted left until the leftmost bit contains a one bit. The result is returned to Y. The number of shifts performed is returned in X. The stack is lifted. If x=0, the stack is lifted and X and Y are set to 0. RL Rotate left -- (Fkey) x is shifted left by one position, the value of the leftmost bit is returned in the rightmost bit position, and also in the Carry flag. The result is returned in X. +--------------------------+ +---+ | +----------------+ | | | <---+--- | <----- | <--+ +---+ +----------------+ Carry RR Rotate right -- (Fkey) x is shifted right by one position, the value of the rightmost bit is returned in the leftmost bit position, and also in the Carry flag. The result is returned in X. +--------------------------+ | +----------------+ | +---+ +--> | -----> | ---+---> | | +----------------+ +---+ Carry RLn Rotate left multiple -- (Fkey) y is shifted left by |x| positions, and as each bit is shifted out of the leftmost bit position, it is introduced into the rightmost bit position, and into the Carry flag. The result is returned in X. The stack drops. RRn Rotate right multiple -- (Fkey) y is shifted right by |x| positions, and as each bit is shifted out of the rightmost bit position, it is introduced into the leftmost bit position, and into the Carry flag. The result is returned in X. The stack drops. RLC Rotate left through carry -- (Fkey) x is shifted left one position, the value of the Carry flag is introduced into the vacated rightmost bit position, and the Carry flag is set to the value of the bit shifted out of the leftmost position. +------------------------------------+ | +---+ +----------------+ | +--- | | <-- | <----- | <--+ +---+ +----------------+ Carry RRC Rotate right through carry -- (Fkey) x is shifted right one position, the value of the Carry flag is introduced into the vacated leftmost bit position, and the Carry flag is set to the value of the bit shifted out of the rightmost position. +------------------------------------+ | +---+ +----------------+ | +--> | | --> | -----> | ---+ +---+ +----------------+ Carry RLCn Rotate Left through Carry multiple -- (Fkey) y is shifted left by |x| positions, and at each shift, the rightmost bit position receives the value of the Carry bit, and the Carry bit receives the value of the bit shifted out of the leftmost position. The result is placed in X. The stack drops. RRCn Rotate Right through Carry multiple -- (Fkey) y is shifted right by |x| positions, and at each shift, the leftmost bit position receives the value of the Carry bit, and the Carry bit receives the value of the bit shifted out of the rightmost position. The result is placed in X. The stack drops. ------------------------ STACK CONTROL OPERATIONS ------------------------ Rv Roll Down -- (assigned to the 'v' key) The contents of each stack register is shifted into the next lower register, and x is shifted into T. R^ Roll Up -- (assigned to the 'V' key) The contents of each stack register is shifted into the next higher register, and t is shifted into X. x<>y Exchange X and Y -- (assigned to the 'X' key) The contents of the X and Y registers are exchanged. ---------------------------- NUMBER STORAGE AND RETRIEVAL ---------------------------- STO Store -- (assigned to the 'N' key) The calculator waits for a register specifier - a digit key (0..9) or A..F; when a register has been specified, x is stored into that register. The value may later be recalled to X by pressing RCL and the same register specifier. x remains in X. The STO operation may be cancelled before specifying a register, by pressing ESC. RCL Recall -- (assigned to the 'M' key) As with STO, RCL must be followed by a register specifier. The stack is lifted and the value last stored into the specified register is placed into X. The value remains in the specified register. The RCL operation may be cancelled before specifying a register by pressing ESC. RCL L Last X -- (Press the 'M' (RCL) key, followed by 'L') Returns to X the value which was in X prior to the last function. ------------ NUMBER ENTRY ------------ 0..9 Digits -- Used to enter numbers. The letters A through A..F F are used for digit entry in hexadecimal mode. . When the prevailing base is decimal, and dotted decimal mode is enabled, the period can be used in digit entry to indicate byte boundaries. " or , Activates character entry mode. In character entry mode, the character value (interpreted as either ASCII/ISO or EBCDIC, depending on the current character code setting) of each keypress is appended to x. Note that in character entry mode, bits are allowed to shift out of the left side of X: in numeric entry mode, this loss of bits is disallowed. To terminate character entry mode, press ENTER or ESC. Functions which are normally invoked with a character key (such as +, *, STO, RCL, etc.) are not accessible until character entry mode is terminated. Selection of ASCII/ISO or EBCDIC character codes is done in the Options/ Calculator Mode dialog. ENTER Enter -- (ENTER) Copies x into Y, and disables stack lift. If numeric or character entry mode is active, it is terminated. The space bar and the equal key ('=') may also be used for this function (except when in character entry mode). <-- Backspace -- (assigned to the backarrow key) When numeric (or character) entry mode is active (the cursor is displayed on the entry line) the last digit (or character) entered is deleted. If no digits remain, entry mode is terminated. When entry mode is not active, the X register is set to zero, and stack lift is disabled. DEL Clear X -- (DEL) The value of X is set to zero. Stack lift is disabled. --- --- --- Error Messages --- --- --- Can't divide by 0 A divide (/) or RMD function was attempted when the value in X was 0 or (in 1's complement mode) -0. Division by 0 is undefined. Can't divide 0 by 0 A divide (/) or RMD function was attempted when both X and Y were 0 or (in 1's complement mode) -0. Division by 0 is undefined. Exceeds current word size A mask (MASKL or MASKR), bit-addressing (SB or CB), shift multiple (SLn or SRn), or rotate multiple (RLn or RRn) function was attempted and |x| was greater than the currently selected word size; or a rotate through carry multiple (RLCn or RRCn) function was attempted and |x| was greater than the currently selected word size + 1. Current mode is unsigned You attempted to change the sign (using CHS) of a number, but the currently selected complement mode is unsigned. A negative number cannot be expressed in unsigned mode. You can display the currently selected complement mode by pressing MENU O, N. You can change the complement mode by pressing MENU O, N, and then pressing the first character of the desired complement mode. Word size not byte integral REVB (reverse bytes) was attempted when the word size was set to a size which is not a multiple of 8 (the size of a byte). REVB can only be performed when the word size is set to 8 (in which case REVB has no effect), 16, 24 or 32. You can display the currently selected word size by pressing MENU O, N. You can change the word size when the Number Format dialog is active by pressing W, and then pressing the up- or down-arrow keys until the desired wordsize is displayed, and then pressing F10 (OK). --- --- --- The HP 100 Calculator Paradigm --- --- --- I have tried to make CMCalc adhere to the HP 100 calculator paradigm, but there are some exceptions: * Algebraic entry mode is not supported * CMCalc uses the "large" font instead of the "extra-large" font on the calc line * Help information is presented in dialog-style windows instead of via the built-in help system * STO @ and RCL @ are not supported * STO arithmetic is not supported * Shift-V performs roll-up * Square root and inverse are omitted Since the "B" and "C" keys are used for entering hexadecimal numbers, they are unavailable for the standard LASTx and x<>y functions. Therefore... * LASTx is obtained by RCL L * x<>y is assigned to the X key --- --- --- CMCalc Compared to the HP-16C --- --- --- For those familiar with the HP-16C, CMCalc does not provide programmability, and omits functions directy related to programmability (there is no indirect register, no x?y or x?0 comparisons, no flag manipulaton functions, and no bit test (B?) function). CMCalc also omits floating-point mode, and the square root function, since these are available in the built-in HP 100 calculator. The three DBL functions have been omitted. Word size is limited to a maximum of 32 bits (versus the 64 bit maximum on the HP-16C). Whereas the HP-16C had a variable number of storage registers, CMCalc has a fixed set of 16 registers: 0..9 and A..F. The HP-16C allows you to perform CHS when the complement mode is set to unsigned: CMCalc does not. --- --- --- Version History --- --- --- The first released version of CMCALC was version 16. Version 17 introduced the following changes: * The number base can be changed by pressing the UP and DOWN arrow keys. * The INV (invert bit order) function was renamed REV (reverse bit order.) * The REVB (reverse byte order) function was added. * The stack display can, as an option, be replaced with the SHOW display. * Character code was added to the SHOW display. * In support of the character code display, the option of selecting the ASCII or EBCDIC character code was added. * The selected number base is always displayed. In version 16, if the currently selected number base was decimal, no indication was displayed on the status line. * The status display (MENU Status) no longer displays the selected base, since it now always appears on the status line. The selected character code was added to the status display. * The active F-key set is remembered in the CMCALC.ENV file. * The space bar can be used to perform the same function as the ENTER key. * A display bug (leftover junk when an alarm window was displayed on top of the CMCALC display) was fixed. * The overflow flag is now set correctly during multiplication. * The most-significant-bit is now set correctly during multiplication. * Due to a bug in the HP 95 ROMs, task-switching between applications which have "fixups" can sometimes cause problems of a completely unpredictable nature. CMCALC is now free of fixups (thanks to Craig Finseth for the technique). Version 18 was a complete re-engineering of CMCALC to conform to the HP 100 calculator paradigm. Numerous changes were made, including: * A complete re-design of the display, so that both the X register in all number bases, and the stack can be shown at all times. * Allowed user configuration of the display. * Added "dotted decimal" display option. * Added CUT and PASTE operations. * Added character entry mode. --- --- --- Problems, Bugs, Comments? --- --- --- If you have problems with CMCalc, or have found a bug, or if you want to make suggestions for change or improvement you can contact me on CompuServe (75156.3132) or via the Internet (dorre@well.com). CMCalc may be freely copied and distributed ONLY in unmodified form, and you may not charge any fee for it. I neither make nor express any warranty that CMCalc operates correctly (though I believe it to do so), or that it is suitable for your purposes (though I hope it is). --- --- --- Registration --- --- --- If you continue to use version 18 of CMCalc, please send a picture postcard of your hometown (or else the place you *wish* was your hometown)! Send your postcard to: Eugene Dorr 8778 Casa Grande Drive Pittsburgh PA 15237 USA Please include the following information: - Where you obtained CMCalc from (If via anonymous ftp, from what site. If from a BBS, which one. If from a friend, the friend's name.) - Your e-mail address, if you have one (please print clearly!). Benefits of registration: You can continue to use version 18 with a clear conscience. Oh yes, if you send me your e-mail address, I will notify you of any future upgrades to CMCalc. --- --- --- Known Bugs --- --- --- Despite trying every STYLE and STATUS setting available, I cannot cause the "OK" buttons in the "Option" dialog boxes to be highlighted, thus indicating they are the default pushbuttons in the dialog. A suggestion was made that sending a SET_FOCUS message to the pushbutton might work, but I couldn't get it to. I also couldn't get STATUS_DEF_PUSHB to do anything useful, but at least was able to create the *effect* of having a default pushbutton by putting appropriate code in the window handler for the parent dialog. When pasting a value from the clipboard, I ignore the thousands separators that might be present in a number. That is, as long as you live in a country where a comma is used as the thousands separator. I cannot find any way to detect the setting of the current thousands separator. I tried m_get_settings_addr() but, true to the documentation, that function call is obsolete. If anyone out there can help me with any of these problems, I sure would appreciate it! Eugene Dorr Internet: dorre@well.com CompuServe: 75156.3132