Usenet 1993 July

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

/ Usenet 1993 July / InfoMagic USENET CD-ROM July 1993.ISO / sources / misc / volume26 / wroff / part01 next >

Wrap

Text File | 1991-12-11 | 53.6 KB | 2,004 lines

Newsgroups: comp.sources.misc From: bkw@uecok.ECOK.EDU (Bill Walker) Subject: v26i097: wroff - text formatter for HP LaserJet with PDS 25-font cartridge, Part01/05 Message-ID: <csm-v26i097=wroff.003501@sparky.IMD.Sterling.COM> X-Md4-Signature: 14fa3c967fe26482c2cb27ec89b8edfc Date: Tue, 10 Dec 1991 06:35:29 GMT Approved: kent@sparky.imd.sterling.com Submitted-by: bkw@uecok.ECOK.EDU (Bill Walker) Posting-number: Volume 26, Issue 97 Archive-name: wroff/part01 Environment: SCO, HPUX, MS-DOS, CPM-68K Here is a text formatter in the spirit of nroff, but attuned to HP Laserjet II printers with the Pacific Data Systems 25-in-One font cartridge installed. The original was posted to comp.sources.unix several years ago, and the archive copy on uunet has been messed up ever since. This new posting is much improved, and seems to have all of its parts present. Hope someone enjoys it. bw@cs.ecok.edu ------------------------------- cut here ------------------ #! /bin/sh # This is a shell archive. Remove anything before this line, then feed it # into a shell via "sh file" or similar. To overwrite existing files, # type "sh file -c". # The tool that generated this appeared in the comp.sources.unix newsgroup; # send mail to comp-sources-unix@uunet.uu.net if you want that tool. # Contents: README Fonts Fonts/PB14L MACROS TestData bin doc # doc/wroff.1.w doc/wroff.tut.w source source/format.c # Wrapped by kent@sparky on Tue Dec 10 00:26:38 1991 PATH=/bin:/usr/bin:/usr/ucb ; export PATH echo If this archive is complete, you will see the following message: echo ' "shar: End of archive 1 (of 5)."' if test -f 'README' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'README'\" else echo shar: Extracting \"'README'\" $4198 characters$ sed "s/^X//" >'README' <<'END_OF_FILE' XThis contains the newest (August, 1991) version of wroff. X XWroff is a text formatter, similar to, but not compatible with, Xnroff and troff. It has much the same command structure, but Xit is optimized for use with the Hewlett-Packard LaserJet II Xprinter and the Pacific Data Systems 25-in-One Font Cartridge. X XWroff is written in "C" and has been "ported" to X Altos 5000 (SCO UNIX on 80486) X Altos 2000 (80386) (UNIX) X Altos 3068 (68020) (UNIX) X Altos 586/986 (Xenix on 8086) X HP-UX running who knows what on what X CPM-68K (honest!) X MSDOS with Turbo C++ X X XIt does kerning, and a few other nice things. It is rather fast, Xand has had a LOT of text put through it in the past couple of years. X X XHere is a list of directories and files: X XFonts/ -- width tables for fonts XMACROS/ -- some demonstration macros and test files XMANIFEST -- list of contents for the shar archive XMakefile -- a makefile for the whole thing XREADME -- this file XTestData/ -- some interesting test data Xdoc/ -- documentation Xsource/ -- source code for wroff X*.bat -- batch files for testing under MSDOS X X XTo create wroff under UNIX, just type "make" in this directory. This will Xcreate a non-stripped executable called "a.out", which is suitable for testing. X XIf you are using HP-UX, you will need to use the HPUX.make file instead. X XIf you are using MSDOS, you must "cd source" and use the file "make.tcc" Xinstead of the UNIX make. In this case, you must perform the testing Xusing the "*.bat" files. Wroff has been tested under Turbo C++, but Xunder no other MSDOS compiler. The binaries are available from bw@cs.ecok.edu. X XThe "a.out" version expects to find the font files in a directory Xcalled "./Fonts", so it will only run in this directory. X XYou can test this version by typing X make install X make test X make sampler XYour printer (and serial line!) must be able to handle 8-bit data. XThis "make install" will not do anything drastic to your file system, Xand it can be run without being super user. X XYou should get about three pages of printing that illustrate Xmost (but not all) of the fonts available to the Pacific Data XSystems Cartridge. X XOnce you are convinced that everything is in good order, you should Xedit the Makefile to indicate a full path name for the Fonts directory, Xand the BINDIR, and then recompile everything. You will probably have to Xsay something like: X rm source/*.o X make X Xto recompile, and then X make install Xas superuser to install the fonts and the real live wroff. XThis will create some files at various places in your filesystem, so Xif you are nervous, be sure to examine the Makefile carefully. XIt shouldn't have anything in it to bother anyone. X XAfter installation, X make test Xor X make sampler X Xis useful to increase your confidence in wroff. X XHave fun! X X X X X************************************************* XAugust, 1991: X XAt the suggestion of Marc Veeneman (marc@marcal.uucp), Xwroff is now more "mailable". All ESC characters in the font Xfiles have been replaced with a much more mailable "E" character, which Xwroff replaces with a real ESC when it reads the font files. X XWroff may not accept files with super-long lines, such as those Xwhich are produced by some commercial packages. Mr. Veeneman Xalso suggests that in these cases, you can change MAXSTR in defs.h Xto something ridiculous, and defeat the miscreants. X(You may also have to make MAXBUF in the same file even more Xridiculous.) X X X************************************************************** XSeptember 4, 1991 X XStudent Connie Nabors (cnabors@student.ecok.edu) Xmodified the original wroff to allow it to be compiled Xunder Turbo C++ on MSDOS. X XAlso, the ".tm" (transparent mode) command has been modified Xto allow the new "wtbl" program to feed wroff directly. X X************************************************************** X XAlso, note that the distribution of wroff that is archived on Xuunet does not contain all of the appropriate files. New Xfiles may be requested by sending mail to X X bw@cs.ecok.edu X X X X Surface mail, e-mail, or telephone questions are welcome: X X Bill Walker X (bw@cs.ecok.edu) X Department of Computer Science X East Central University X Ada, Oklahoma 74820 X 405 332 8000 x 594 X END_OF_FILE if test 4198 -ne `wc -c <'README'`; then echo shar: \"'README'\" unpacked with wrong size! fi # end of 'README' fi if test ! -d 'Fonts' ; then echo shar: Creating directory \"'Fonts'\" mkdir 'Fonts' fi if test -f 'Fonts/PB14L' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'Fonts/PB14L'\" else echo shar: Extracting \"'Fonts/PB14L'\" $313 characters$ sed "s/^X//" >'Fonts/PB14L' <<'END_OF_FILE' XE&l0OE(1UE(sp10h13.9vs1b11T X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 X30 END_OF_FILE if test 313 -ne `wc -c <'Fonts/PB14L'`; then echo shar: \"'Fonts/PB14L'\" unpacked with wrong size! fi # end of 'Fonts/PB14L' fi if test ! -d 'MACROS' ; then echo shar: Creating directory \"'MACROS'\" mkdir 'MACROS' fi if test ! -d 'TestData' ; then echo shar: Creating directory \"'TestData'\" mkdir 'TestData' fi if test ! -d 'bin' ; then echo shar: Creating directory \"'bin'\" mkdir 'bin' fi if test ! -d 'doc' ; then echo shar: Creating directory \"'doc'\" mkdir 'doc' fi if test -f 'doc/wroff.1.w' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'doc/wroff.1.w'\" else echo shar: Extracting \"'doc/wroff.1.w'\" $15433 characters$ sed "s/^X//" >'doc/wroff.1.w' <<'END_OF_FILE' X.fo \f0Bill Walker/page #/October 27, 1989\fP X.NAME wroff X.SYNOPSIS Xwroff [-sdpk] [-b n] [-e n] [-o outfile] [-t #] filelist X X.FUNCTION Xwroff is a general text formatting program. Options: X.sp X.nf X.in +5 XSwitch Action X.in +2 X X-p Do not effect font changes (default is p) X-b begin printing with page n (default 1) X-e end printing with page n (default 32767) X-o place output on outfile X-k do not use formfeeds to advance to new page X-s scan text for errors only - do not print X-t # expand tabstops (default 8 spaces) X (use -t 0 to retain tab characters) X-v print version number -- no text processing X X.in -7 X.fi X.PP XSwitch "-p" causes special characters such as Xthose that produce italics or boldface not to be output. XAn unrecognized command will cause wroff to terminate. X.PP XWhen writing to standard output, wroff does not emit special Xcharacters. This allows work to be previewed on the screen Xbefore it is printed. However, the screen only approximates Xthe true output. X.PP XThe wroff commands are discussed in the following pages. X.BUGS XThe "preview" mechanism should provide more realistic output. X.bp X.in 5 X.nf X.ce 2 X.bd 2 X.it 1 XWroff XCommands X.sp 2 X\fT XCommand default Effect X X.cc ! X!in +2 X.bd n 1 boldface next n lines X.bp n next advance to next page X.br none break X.cc c . change command char to c X.ce n 1 center next n lines X.dc c / change delimiter to c in titles X.de name none start macro definition X.di XX none start diversion to XX X.ed none end diversion X.en none end macro definition X.et none end transparent mode X.fi true set fill on X.fm fontname P18bl mount font on post M X.fo string "/page #/" define footer X.gx x 0 move horizontally x dots X.gy y 0 move vertically y dots X.he string none define header X.hr n 0 horizontal rule, length n X.if (expr) XX if (expr) is true, call macro XX X.in n 10 set left margin to n X.it n 1 italicize next n lines X.ju true justify right margin X.lp str none label indented paragraph X.ls n 1 set line spacing to n X.m3 n 24 margin (dots) before footer X.ne n 1 need n lines X.nf n false,n=2 set fill off, theshold to n X.nj false do not justify right margin X.pc c # set page char to c X.pl n 480 set page length to n dots X (48 per vertical inch ) X.pm n 10 place paragraph labels left X by this much (characters) X.rj n current process left jagged, right justified X right margin use .rj 0 to turn off X.rm n 70 set right margin to n X.rt none reset traps X.so name none source a file X.sp n 1 space down n lines X.ti n 0 temporary indent by n X.tf string "l c r" title format X.tl XX string none set variable XX to string X.tm none enter transparent mode X.vr n 0 vertical rule, length n X.vs n 12 vertical spacing between lines X (dots) X.ul n 1 underline next n lines X.wh n XX none set trap XX at vertical position n X.. none comment line X\\fc none embedded font change X (c = B, I, M, T, U, P, 0, 1, 2) X\\{XX} none interpolate macro XX X\\(reg) none interpolate register reg as an integer X\\(regC) none interpolate register reg as a character X\\[expr] none evaluate register expression X\\f| none provide "anchor" in nonfill mode X X!cc . X.bp X\fP X.ce X\fIVariables\fP X.nf X XAll measurements in "dots" X.sp 2 X registers and default values X X # = 1 \f| start at page 1 X A = 2 \f| the next page X B = 24 \f| bottom margin of 24 dots X C = 0 \f| center this many lines X D = 0 \f| vertical length of last diversion X E = 0 \f| italicize next n lines X F = 0 \f| current font X G = 0 \f| boldface next n lines X H = TRUE \f| justify lines X I = TRUE \f| start in FILLMODE = TRUE X J = '/' \f| a SLASH mark X L = 300 \f| left margin of 300 dots - 1 inch X M = 0 \f| temporary indent X O = '#' \f| page character X P = 480 \f| page length X Q = 0 \f| number of printable characters in outbuf X R = 2100 \f| right margin of 2100 dots - 7 inches X S = 1 \f| line spacing X T = 0 \f| top margin of 0 dots X U = 30 \f| width of a space character in dots X V = 10 \f| current vertical spacing X W = 10 \f| indented paragraph start in characters X X = 0 \f| current horizontal position X Y = REG('T') \f| current vertical position X Z = FALSE \f| underline state X ! = 8 \f| tabstops every 8 characters X _ = 24 \f| margin before footer X.bp X.fi X.ce 10 X\fIWroff Commands\fP X.ce 0 X.sp 2 X.de CMD X.in 20 X.sp X.lp $1 $2 X.en X X.CMD bd X"bd n" is used to boldface the next n lines. If n is not specified, Xthe Xdefault is to boldface the next line. XSince \fBwroff\fP supports in-line font changes, this command is Xnot really very useful. X.CMD bp X"bp n" is used to advance to the next page and number it "n". If X"n" is not specified, the default is to number pages Xsequentially. X.CMD br XThe "br" command will always cause any collected output to Xbe printed. X.CMD cc XThe "command character" is the character that preceeds an embedded command. XThe character defaults to a ".", but can be changed to any Xcharacter by issuing "cc c", where c is any character. X.CMD ce X"ce n" centers the next n lines. If n is not specified, the Xdefault is to center the next line. X.CMD dc XIn titles, such as the footers and headers, the fields are Xdelineated by using a "delimiter character". For instance, Xthe default footer is "/page #/". The "/" delimits the Xfields. This delimiter (default "/") can be changed by Xusing the "dc c" command. X.CMD de XThis command starts a macro definition. Use "de name" to Xbegin the definition of a macro named "name". There is no Xdefault name. A macro definition will continue until the X"en" command is recognized. X.CMD di XStart a diversion named "name" by issuing "di name". Text Xis copied into the diversion until the "ed" command is Xfound. X.CMD ed XEnd a diversion. X.CMD en XEnd a macro. Caution: if you define a macro, be sure to Xend it! X.CMD et XEnd transparent print. Any lines between "tm" and "et" will Xbe copied to the output without any processing (other than Xfor margins ). X.CMD fi X"fi" tells \fBwroff\fP to begin to fill lines. That means Xthat both the left and the right margins will appear Xstraight. X.CMD fm XThe "fm name" command mounts a font with the given name on Xfont post "M". To access that font, use \\fM to start, and X\\fP to return to the previous font. XThere is a list of available fonts at the end of this Xdocument. X.CMD fo X"fo string" defines the page footer to be "string". The Xstring is constructed by examining the title format string, Xwhich is defined by "tf", and replacing any page character Xby the current page number. (See "pc" to change the page Xcharacter.) XThe placement of the footer on the page can be controled by Xusing "m3". X.CMD gx X"gx x" is the "go to x" command. "x" is measured in "dots" Xfrom the left printable margin. There are 300 "dots" to an Xinch. "gx" is used to move horizontally across a page. X.CMD gy X"gy y" is the "go to y" command. "y" is measured in "dots" Xfrom the top printable margin. There are 48 "dots" to an Xinch. "gy" is used to move vertically in a page. X.CMD he X"he string" sets the page header, in the same way that "fo" Xsets the footer. X.CMD hr X"hr n" draws a horizontal rule (a line) from the current x position Xto the right. This line is of length n (default 0) in dots. X.CMD if X"if (expr) macname" will evaluate the expression and execute the Xmacro names "macname" if the expression is true. XIf the expression is an arithmetic expression, the evaluator Xregards "0" as "TRUE". X.CMD in XThe "in n" command controls the left margin. The default for n is 10. XIn contrast to other commands, n is measured in deci-inches. The default Xleft margin is therefore 1 inch from the left printable margin. X.CMD it X"it n" will italicize the next n lines. The default is to italicize Xone line. Since \fBwroff\fP supports in-line font changes, this command Xis not really useful. X.CMD ju XIssuing "ju" will cause right margin justification to take place. To Xgain a "ragged right" margin, turn justification off with "nj". X.CMD lp XUse "lp str" to label a paragraph with the string "str". This label is Xplaced to the left of the paragraph a distance of 10 deci-inches. This Xdistance can be changed with the "pm" command. X.CMD ls XUse "ls n" to set the line spacing to n. (The default is 1). For instance, Xto double space output, use "ls 2". X.CMD m3 XBefore placing a footer, wroff provides a small margin. The default is to provide Xone-half inch (24 dots), and then place the footer. This can be changed Xby using "m3 n", where n is measured in dots. (48 dots per vertical inch.) X.CMD ne X"ne n" means that \fBwroff\fP should guarantee that there Xis at least n lines left on the current page (default n=1). If this much Xspace does not remain on the current page, \fBwroff\fP will turn up a new Xpage before proceeding. X.CMD nf XThe "nf" command causes \fBwroff\fP to process in non-fill Xmode. In fill mode, as many words as possible are packed on Xa line before the line is printed. If "justify" mode (see X"ju" and "nj") is also on, the right margin is also Xjustified. The "nf" command turns fill mode off, and in Xeffect causes lines to appear as they are typed. This is Xespecially useful for tables. The "anchor" character (see Xbelow) will be helpful in making tables line up properly. X.CMD nj XEnter "non-justify" mode. XIn non-justify mode, the right margin is not justified. XNon-justify mode can coexist with fill mode to pack as many Xwords as possible on a line, without squaring up the right Xmargin. The default mode is "justify". X.CMD pc XIn headers or footers, the (default) symbol "#" is used to represent Xsomething that should be replaced with the current page Xnumber. The symbol "#" can be changed to any character c by Xusing "pc c". X.CMD pl X"pl n" is used to set the length of a page to n dots. X(default is n=480, or 10 inches) This command is little Xused, unless you use odd sized paper in your laser printer. X.CMD pm XSee the "lp" command above. The "pm n" command (default Xn=10) sets the margin for a paragraph label. The parameter Xn is measured in deci-inches. X.CMD rj XUse the "rj" command to cause text to be justified on the Xright, with a ragged left margin. This is useful for X"glitzy" work. X.CMD rm XThe "rm n" command sets the right margin. "n" is measured Xin deci-inches, with a default of 70 (7 inches). X.CMD rt XWhen "traps" (see "wh") are placed on a page, the Xtraps are "sprung" as the location at which they are placed Xis passed. A "sprung" trap will not be sprung a second time Xon the same page. If you want to "reset" a trap for some Xreason (perhaps you used "gy" to move above it), use "rt" Xto reset \fIall\fP the traps on a page. It is not possible Xto selectively reset traps. X.CMD so XSource an external file. "so filename" will attempt to Xlocate file "filename" and include it as though it were part Xof the current file. Nested "so" commands are legal. i.e. Xa file that is 'sourced' may in turn 'source' another file. X.CMD sp XSpace down n lines by using "sp n". The default for n is X1. X.CMD ti XUse "ti n" (default n=0) to indent \fIonly the next line\fP Xby n deci-inches. (Think "temporary indent") X.CMD tf XA "title format" is a string that tells the tl command X(below) how to interpret a given string. XThe current title format is used to interpolate strings Xusing the "tl" command. XThe default title format is "l c r", meaning to interpret Xthe string as having three fields, delimited by "/" (see X"dc"). The first field will be left-justified, the second Xfield will be centered, and the third Xfield Xwill be Xright justified, all within printed fields computed from the Xnumber of fields specified in the title format. XTo place the second field at the right of the center printed Xfield, use "tf l r r". X.CMD tl X"tl string" is used to interpolate the string into the output, where the Xstring is formatted according to the current 'title format'. (See "tf".) X.CMD tm XEnter transparent print mode. No further processing is done Xon text, but the text is copied directly from input to Xoutput until command "et" is encountered. No font changes Xare executed, and \fBwroff\fP is likely to lose track of Xcurrent positions on the page. Use "tm" only \fIin Xextremis\fP. X.CMD vr XDraw a vertical line n dots (48 per inch) long starting at the current Xposition. Note that the positive direction is "down" the Xpage. X.CMD vs XAdjust the vertical spacing between lines of print. This is Xoften useful to help "squeeze" a document onto a single Xpage. "vs -1" will squeeze lines closer together by 1 Xvertical dot. The default is to space by 12 dots. X.CMD ul XUse "ul n" to underline the next n lines. (default n = 1) X.CMD wh XSet a trap. XThe "wh" command has an unusual syntax. The command "wh n XNAME" will execute macro "NAME" when the current vertical position Xon the page passes position n. The syntax "wh -24 NP" would Xfire macro NP when the current vertical position was within X24 dots of the bottom of the page. The macro must be Xdefined elsewhere. Beware of firing undefined macros. A Xmacro must be defined before the trap is fired, but need not Xbe defined before the trap is set. It is legal to change a Xmacro definition without resetting a trap. X.CMD . XThe ".." command is a comment marker. Actually, Xthe command should really be "CMDCHAR .", where the CMDCHAR Xis set by the "cc" command. The rest of a line after ".." Xis taken as a comment. X.in 10 X.bp X.ce X\fBIn-Line Actions\fP X.sp 3 X.PP X\fBwroff\fP has the ability to deal with various fonts. It can also Xinterpolate the value of a variable into the output stream, or call a Xmacro from anywhere within a line. XAll of this is accomplished with "In-Line Actions". X.PP XIn-Line Actions Xcan be placed anywhere within a line of text. X(This is opposed to \fIembedded commands\fP, which were Xdiscussed above, and which must appear on a line by Xthemselves.) The "backslash" symbol (\\) is used to signal X\fBwroff\fP that an in-line action is present. X.PP XThe possible in-line actions are listed below: X.nf X.sp 2 Xcommand \f|Action X.in +2 X\\fc \f|in-line font change X \f| where c = B, I, M, T, U, P, 0, 1, 2, other X \f| B - \f|boldface X \f| I - \f|italicize X \f| M - \f|mountable X \f| T - \f|Courier (fixed) X \f| U - \f|toggle underline X \f| P - \f|return to previous X \f| 0 - \f|default font (Times Roman) X \f| 1 - \f|Presentation 18 point X \f| 2 - \f|Helvetica 6 point X \f| other - \f|interpolate "other" char X X\\{XX} \f|interpolate macro named XX X\$reg) \f|interpolate the value of register reg X \f| as an integer X\\(regC) \f|interpolate the value of register reg X \f| as a character X\\[expr] \f|evaluate register expression - X \f| do not interpolate, see \\(reg) above X\\f| \f|place "anchor" in text X.sp 2 X.in -2 X.fi END_OF_FILE if test 15433 -ne `wc -c <'doc/wroff.1.w'`; then echo shar: \"'doc/wroff.1.w'\" unpacked with wrong size! fi # end of 'doc/wroff.1.w' fi if test -f 'doc/wroff.tut.w' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'doc/wroff.tut.w'\" else echo shar: Extracting \"'doc/wroff.tut.w'\" \(16027 characters$ sed "s/^X//" >'doc/wroff.tut.w' <<'END_OF_FILE' X.fo \f0Bill Walker/page #/October 18, 1989\fP X.ce 10 X\fBWroff Tutorial\fP X.ce 0 X.sp 3 X.in +5 X.de PP X.sp X.ti +5 X.en X.de OP X.sp X.ti -5 X.en X.OP X\fBIntroduction:\fP X.ls 2 X.PP X"Wroff" is a general purpose text formatter, Xwritten in the spirit of Unix's \f2trademark of AT&T\fP X\f1*roff\fP family. The purpose Xof this document is to make the reader able to use "wroff" Xfor preparing documents. X.PP XWroff has been specifically "tuned" to the Hewlett-Packard XLaserJet II printer. Most of the system defaults will Xprovide some reasonable output no matter \fBwhat\fP input is Xpresented. The LaserJet can do \fBboldface\fP, \fUunderlining\fU, Xor even a \fB\fUcombination\fU\fP of the two. Wroff Xsupports \fIitalics\fP as well. Wroff is aware that a page has X60 lines, and that a page should be printed with a left Xmargin of 1 inch. (These can both be changed, of course.) X.PP XWroff, when processing text, will space twice after periods, Xquestion marks, exclamation marks, and colons. X.PP XThis document was prepared using\f1 wroff\fP. X.ls 1 X.bp X.OP X\fBEmbedded Commands:\fP X.PP XWroff works by reading a text file, X(created by using "vi" under Unix) and rearranging the Xcontents of the file according to instructions which you Xgive by "embedding" certain commands in the file. X.PP XAn example of an embedded command would be this: X.cc # X X#nf X#in +5 X.ce 3 XThese three lines should Xbe centered on Xthe page. X X#cc . X X.in -5 X.fi X XThe resulting output of wroff would be: X X.ce 3 XThese three lines should Xbe centered on Xthe page. X X.PP XThe three lines are centered as a result of the ".ce 3" Xcommand being embedded in the text. ".ce 3" means "center Xthe next three lines." X.PP XThis is an example of an embedded command. There are about Xforty such commands, most of which are not needed for Xordinary work. An average document will require only two or Xthree of the commands. The rest are present just to do X"fancy stuff". X.PP XAn embedded command is usually the only command on a line, and it Xusually starts with a period ("."). For this reason, Xembedded commands are often called "dot commands". XIt is unusual to want to actually begin Xa line with a period, so this works OK. If you have some reason for Xwanting to place a period at the first of a line (we did Xabove in the example Xfor ".ce"), there is a way to do it. X X.OP X\fBModes:\fP X.PP XWe need to explain the difference in "fill mode", Xand "non-fill" mode. X.PP XIn "fill mode", wroff will pack as many words as possible on Xa line, no matter how you typed the lines in the original Xfile. If you are in "non-fill" mode, the lines will appear Xin the output just the way you typed them in. Fill mode is Xusually "on", but you can turn it off with the command X".nf". If you want it back on, just use ".fi". Here is an Xexample: X X.cc # X#nf X.nf XThese lines will appear Xin non-fill mode, just the Xway they were typed in. XThis is useful for making tables and Xother things that need to line up Xin a certain way. X XHere is the output that results: X X X These lines will appear X in non-fill mode, just the X way they were typed in. X This is useful for making tables and X other things that need to line up X in a certain way. X X#fi XHere is the way things would look in fill Xmode. Here is the raw text: X X#nf X.fi XThese lines will appear in Xfill mode, with the margins adjusted Xproperly. This is the usual default for Xtext, and so you can just omit the ".fi" if Xyou want to. X XHere is the way they appear after processing by wroff: X X#fi X#cc . XThese lines will appear in Xfill mode, with the margins adjusted Xproperly. This is the usual default for Xtext, and so you can just omit the ".fi" if Xyou want to. X X.OP X\fBjustify and non-justify modes:\fP X.PP XSometimes we want to place Xas many words on a line as possible, but we do not want the Xright margins to be justified. For instance, you may be Xwriting a business letter, and a ragged right margin gives Xsort of a "typed-by-hand" look to a paper. Use the commands X".ju" and ".nj" to turn right margin justification on and Xoff. X.PP XHere is an example. In the text file, we enter: X X.nf X.cc # X.fi X.nj XThese lines will appear with as Xmany words on a line as possible, but Xwith the right margin "ragged." XThis makes for a more "typed-by-hand" appearance, Xand is useful for correspondence. X XHere is the output: X X#nj X#fi XThese lines will appear with as Xmany words on a line as possible, but Xwith the right margin "ragged." XThis makes for a more "typed-by-hand" appearance, Xand is useful for correspondence. X#cc . X X.ju X.fi X.OP X\fBChanging the "command character":\fP X.PP XSeveral times in the Xabove text, we had to print a period (a "dot") as the first Xcharacter on a line. Since an embedded commands usually Xstarts with a "dot", we had to be sure that the Xdot at the first of a line could be ignored if we wanted it Xto be. X.PP XThe "dot" is called the "command character", and actually Xany character can be used. If we want some other character Xto become the command character instead, we could use the X".cc" command (for \fBc\fPhange \fBc\fPommand). Here is how it Xworks. X.cc * X XAfter this, the command character is a "#". For instance, X X*nf X.cc # XI can now type a dot at the first of a line like this: X X.45 caliber automatic X X#cc . XNow the command character is changed back to a ".". X X*fi X*cc . X.PP XWe will see another method of accomplishing the same end later. X X.OP X\fBIndenting:\fP X.PP XMost text in letters requires indented Xprinting. The usual left-hand margin in wroff is set to X1 inch from the left printable margin on a page. XHowever, this Xcan be changed in three ways. X.PP XThe left hand margin can be changed to any other value by Xusing ".in". For instance, ".in 20" changes the left margin Xto 2 inches from the left printable margin. XThe "20" means that the margin is to be set to 20 deci-inches X(tenths of an inch) from the left printable margin. XIncidentally, the left printable margin, depending on your printer, may not Xbe at the left edge of the paper. X.PP XYou can also do it slightly differently. XUsing ".in +12" would change the left margin to be 1.2 inches Xto the right of where it was. In a like fashion, using ".in X-5" would move the left margin one-half inch to the left. For Xexample: X X.nf X.cc # X.in 10 XThis would be printed at 1.0 inches from the left of the paper. X X.in +5 XThis is placed at inch 1.5. X X.in 15 XThis is also placed at inch 1.5. X X.in -10 XThis would be printed at inch 0.5 (15 -10). X X#cc . X.fi X.PP XA variation on setting a margin permanently is just to Xchange the margin for \fUonly the next line\fU. Use ".ti +5" X(for \fBt\fPemporary \fBi\fPndent 5 deci-inches) to indent only the next Xline (by 0.5 inches). This is particularly useful at the beginning of Xparagraphs. X.PP XAs an example, X X.cc # X#nf X.ti +5 XThe paragraph printed by processing this text will have an Xindented first line. The remaining lines will not be Xindented, but will appear at the left margin. X Xprocesses into: X X#fi X#ti +5 XThe paragraph printed by processing this text will have an Xindented first line. The remaining lines will not be Xindented, but will appear at the left margin. X X#cc . X.PP XAn interesting variation is an "outdent", or hanging leader. XHere is an example: X X.cc # X#nf X.ti -5 XThis paragraph has an outdent, or hanging leader in its Xfirst line. This really helps when preparing a report or Xother written work with section headings. X#fi X XThe output will appear as: X X#ti -5 XThis paragraph has an outdent, or hanging leader in its Xfirst line. This really helps when preparing a report or Xother written work with section headings. X X#cc . X.OP X\fBBoldfacing\fP, centering, and \fUunderlining\fU: X.PP XCentering text is easy. Just use ".ce" as in the Xexample at the first of the paper. One "trick" is to tell Xwroff to center a huge number of lines, and then use ".ce 0" Xto turn centering off like this: X.nf X X.cc # X.ce 100 XThese and Xno telling how Xmany other Xlines X---- X--- Xshould be centered. X.ce 0 X#cc . X.fi X X.PP XThis results in all of the lines between ".ce 100" and ".ce X0" being centered, with no requirement for having to count Xlines ahead of time. X.PP XThe same stunt works on boldface or underlining: X.cc # X#nf X X.ul 100 X.bd 100 XAll of the words Xin this block will be both boldfaced and Xunderlined X-- X-- all the way Xdown to here, whereever Xthis is. X.bd 0 X.ul 0 X X#cc . X.fi X X.PP XBoldfacing and underlining can be accomplished in two ways. XOne way is to use either ".bd" or ".ul" as above. X.PP XA second way, which is useful when you want to just Xunderline or boldface a couple of words is to use an X"in-line font change". X.PP XAn illustration probably is sufficient. X.PP XIn this paragraph, we \\fBboldfaced three words\\fP, and \\fUunderlined Xtwo\\fU words, and \\fIitalicized three words\\fP, and printed a \\\\. XHere is how wroff printed it. X.PP XIn this paragraph, we \fBboldfaced three words\fP, and \fUunderlined Xtwo\fU words, and \fIitalicized three words\fP, and printed a \\. XHere is how wroff printed it. X.PP XIn-line font changes are discussed in more detail in a later section. X.OP X\fBUse of Macros:\fP X.PP X"Macros" are collections of frequently used commands that Xcan be called by a single name. You get to invent the name. X.PP XIn typing this document, we frequently had need to X"skip 3 lines and then temporarily outdent by 5 spaces" when Xwe came to a new section. Rather than always type X.cc # X X#nf X.sp 3 X.ti -5 X X#fi Xat the start of every section, we instead defined a "macro" Xnamed ".OP" like this: X X#nf X X.de OP X.sp 3 X.ti -5 X.en X X#nf XAt the start of every new section, we just typed: X X.OP XThe start of the section ... X.PP X Xand went on about our business. X#cc . X.fi X.PP XMacros are really powerful. A well thought out collection Xof macros can make writing a letter or preparing a memo a Xbreeze. Macros can be defined and placed in a file for Xfuture use. The file can be included in any document by Xusing the ".so" command to read the macro file before you Xstart processing. X.PP XDon't overlook the power of macros. Here is a set we use a Xlot. They were placed in a file called "macros" by using X"vi". X X.in +5 X.OP Xfile macros contains: X.nf X.cc # X X.de DATE X.in +30 X$1 $2 $3 X.br X.en X.de TO X.in 10 X.sp 3 X.nf X.en X.de TEXT X.sp 3 X.fi X.ti +5 X.en X.de SIGN X.sp 3 X.in +30 X.nf X.en X#cc . X.in -5 X X.fi X.OP XNow when we want to write a letter, we write X.nf X.in +5 X.cc # X X.DATE October 4, 1988 X.TO XMr. Joe Smith XSomePlace XSomewhere, USA X XDear Joe: X.TEXT XHere is the letter and so forth ... X.SIGN XYours Truly, X XFred X#cc . X.fi X.in -5 X XWe print the letter by writing "wroff macros filename". X.bp X.OP X\fBScanning for errors:\fP X.PP Xwroff is somewhat sensitive to Xerrors in input. It is sometimes useful to scan a document \fIbefore\fP Xyou print it to see if all of the wroff commands have been entered Xcorrectly. Use the command line option "-s" to scan for errors Xif you are concerned. X.PP XIf you enter "wroff -s filename", then wroff will not print anything Xat all, either on the screen, or on the printer. If there is some kind Xof problem with any of the wroff commands in your paper, wroff will report Xthat problem, and the approximate location of it, so that you can fix it Xbefore you go to the trouble and expense of printing the document. X.OP X\fBFonts:\fP X.PP XOne of the biggest pleasures in using wroff is that you can combine various Xtypefaces in a document to give a pleasing appearance. X.PP XThere are several fonts available. Here is a current list: X.nf X\fT Xsymbol font X X.in +5 XB Times Roman 12 point boldface XI Times Roman 12 point italics XM mountable - default is Presentation 18 bold XT Courier 12 point XP return to previous font XU toggle underlining X0 Times Roman 12 point Normal (default) X1 Presentation 18 point - synonymous with M X2 Helvetica Normal 6 point X\fP X.fi X.in -5 X.PP XIn order to make effective use of fonts, you must Xunderstand what is meant by "proportional spacing." X.PP XYou may have noticed that in this document, not all letters in Xa word are the same "width". For instance, the letter "i" is much more Xnarrow than the letter "M". These letters were printed with a proportionally Xspaced font. X.PP XProportional spacing is pleasant to read. It gives a "typeset" quality Xto printed work. Most (but not all) of Xthis document was printed using Times Roman 12 point XNormal type. X.PP XWroff, in the default mode, allows for proportional printing, and adjusts Xthe spacing in lines accordingly. X.PP XIn the table above, the only proportional fonts are the "Times Roman" fonts. XThe other fonts are "fixed spacing fonts", which means that each letter Xtakes the same amount of space as any other letter. Fixed spacing fonts Xare especially useful for preparing portions of a document that have Xlines which need to "line-up" with previous lines. One such example is Xwhen preparing a table. X.PP XYou can switch fonts, even in the middle of a line, by using Xan "embedded font change." To do this, use a "\\f" followed by the symbol Xfrom the table above that indicates which font you want. X.PP XAs an example, the line: X.nf X.in +5 X X\\fBboldface this\\fP and \\fIitalicize this\\fP X.in -5 X.fi X Xwould print as: X.nf X X\fBboldface this\fP and \fIitalicize this\fP X X.fi X.OP X\fBBuilding tables and using '.nf':\fP X.PP XThe only fixed spacing fonts available above are the Courier 12 point, Xand the Presentation 18 point. (Point size indicates the height of letters, Xnot their width.) X.PP XIncidentally, you may occasionally want to print the character "\\". XIf you want to enter such a character, you may enter '\\\\' in the document. XThis works for other characters, too. Entering '\\x' will print the Xcharacter 'x'. This seems useless until you realize that it is an alternative Xway of protecting a line that begins with a period. X.PP XHere is an example table: X.nf X.in +5 X X\fT X X\.nf X\.in +5 X\\fT XName Address XJoe Smith Somewhere XJames Jones Nowhere XSally Sue Anywhere X\\.50 caliber Guns X\\fP X\.in -5 X X.in -5 X\fP X.PP XThis was printed by wroff as: X X X.nf X.in +5 X\fT XName Address XJoe Smith Somewhere XJames Jones Nowhere XSally Sue Anywhere X\.50 caliber Guns X\fP X.fi X.in -5 X X.PP XJust switching from a proportional font to a fixed font in the same Xline, when in fill mode gives no problems at all. Wroff may become Xconfused when you build tables in non-fill mode and mix several fonts Xtogether. A good word of advice, then, is to build tables in non-fill Xmode, using a fixed font, just as we did above. X.PP XIf for some reason Xyou absolutely \fIinsist\fP on using a proportional font to build a table, Xyou can use an "anchor". The "anchor" characters cause a specific Xplace in the text to appear in a certain column. The anchor consists of the Xstring "\\f|". XHere is an example: X.nf X.sp X.in +5 X\fT Xage \\f|14 Xdate \\f|12 Xtoday \\f|2 X\fP X.in -5 X.fi X.PP XThis produces the following output: X.sp X.in +5 X.nf Xage \f|14 Xdate \f|12 Xtoday \f|2 X.in -5 X.fi X.PP XWhen processing a line in non-fill mode under a proportional font, Xwroff will attempt to duplicate the appearance of the original line as Xbest it can. Wroff takes the original line and determines by counting Xcharacters just where on a page that it should begin printing a word. XIt assumes that each character is 0.1 inches in width. X.PP XWroff then attempts to begin each word in the line at whatever place on Xthe page is indicated by the character position of the first letter in Xthe word. This would make for some "funny" looking lines, unless you Xplaced the anchors in the line. X.OP X\fBConclusions:\fP X.PP XThere is a lot to wroff, and a lot more to Xlearn. Most of the commands are reasonable to use, and Xactually most people only need a few of the commands. XStudying the User's Guide will repay a careful reader. X.PP XAppendix A attached to this paper is a technical description Xof all of wroff's commands. After you gain skill with Xwroff, this appendix will probably become very useful to Xyou. X.PP XAppendix B is a listing of the raw unprocessed file that Xwas used to create this tutorial. It should provide a treasure Xtrove of examples. X.PP XEnjoy ! END_OF_FILE if test 16027 -ne `wc -c <'doc/wroff.tut.w'`; then echo shar: \"'doc/wroff.tut.w'\" unpacked with wrong size! fi # end of 'doc/wroff.tut.w' fi if test ! -d 'source' ; then echo shar: Creating directory \"'source'\" mkdir 'source' fi if test -f 'source/format.c' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'source/format.c'\" else echo shar: Extracting \"'source/format.c'\" $12233 characters$ sed "s/^X//" >'source/format.c' <<'END_OF_FILE' X/* this file contains the line formatting stuff */ X X#include <stdio.h> X#include <ctype.h> X#include "defs.h" X#include "externs.c" X Xformat(s,ifile,f) Xchar s[]; XFILE *ifile,*f; X{ X /* The idea is to collect output into a buffer (outbuf) X and when the buffer is complete, write it. X The variable s contains the text of the last line X read from the input file. The output buffer is X in variable "outbuf". */ X X int i; X X char word[MAXSTR],tstring[MAXSTR]; X int wordlength; X int diff; X int chcnt; X X /* first "doctor" the line if boldcount or italicscount > 0 */ X if (BOLDCNT > 0) { X addstr(s,"\\fB","\\fP"); X BOLDCNT--; X } X if (ITALCNT > 0) { X addstr(s,"\\fI","\\fP"); X ITALCNT--; X } X if (ULCOUNT > 0) { X addstr(s,"\\fU","\\fU"); X ULCOUNT--; X } X X /* now if the line should be centered, do it, otherwise just X format it */ X if (!FILLMODE || CNTRCNT > 0) { X /* just parse the line to get the length */ X i = 0; X while (s[i] != '\0') { X getword(word,s,&i); X parseword(tstring,word,&wordlength,&chcnt,ifile,f); X append(tstring,chcnt); X CURRX = CURRX + wordlength; X } X endln(); X X TEMPINDENT = 0; X if (CNTRCNT > 0) { X TEMPINDENT = (RMARGIN - LMARGIN - CURRX)/2; X CNTRCNT--; X } X X output (ifile,f,0); X } X else { /* don't center it */ X X /* get a word from s */ X i = 0; X while ( s[i] != '\0') { X getword(word,s,&i); X /* now add the word to the buffer */ X /* we have to parse the word to find any X embedded font changes, and to determine its printable X length */ X parseword(tstring,word,&wordlength,&chcnt,ifile,f); X X diff = RMARGIN - LMARGIN - TEMPINDENT; X if (( length(' ') + CURRX + wordlength) >= diff) { X endln(); X output(ifile,f,diff); X } X append(tstring,chcnt); X CURRX = CURRX + wordlength ; X /* note CURRX is current printable position on line */ X X } X } /* of don't center it */ X} X X X/* getword looks inside string s, starts at position i, and returns X the next word in t */ XVOID getword(t,s,i) Xchar t[]; Xchar s[]; Xint *i; X{ X int j; X /* first skip any white space */ X while ( s[*i] != '\0' && isspace(s[*i]) ) (*i)++; X X /* now copy into string t */ X j = 0; X while ( s[*i] != '\0' && !isspace(s[*i]) ) { X t[j] = s[*i]; X (*i)++; X j++; X } X t[j] = '\0'; X /* note that s[*i] is either '\0' or it is white space */ X X} X X X/* append adds the string to the existing output buffer. Note that X the buffer does NOT have a NULL at its end. */ XVOID append(s,chcnt) Xchar s[]; Xint chcnt; X{ X int j; X if (shadd && obufpos != 0 && s[0] != '\0') { X outbuf[obufpos] = ' '; X obufpos++; X outbuf[obufpos] = SHIFTOUT; X obufpos++; X CHARCNT = CHARCNT + 1; X CURRX = CURRX + length(' '); X } X else X shadd = TRUE; X X j = 0; X while (s[j] != '\0') { X outbuf[obufpos] = s[j]; X j++; X obufpos++; X } X if (LASTPERIOD) { X outbuf[obufpos] = ' '; X obufpos++; X outbuf[obufpos] = SHIFTOUT; X obufpos++; X CHARCNT = CHARCNT + 1; X CURRX = CURRX + length(' '); X LASTPERIOD = FALSE; X } X /* WARNING --- outbuf does NOT have a NULL at its end */ X CHARCNT = CHARCNT + chcnt ; X X X} X X X/* we have to parse the word to find any X embedded font changes, and to determine its printable X length. The word is copied into tstring with any control characters X expanded and embedded. This needs to be a FSM. */ X XVOID parseword(tstring,word,wordlength,chcnt,ifile,f) Xchar tstring[],word[]; Xint *wordlength; Xint *chcnt; XFILE *ifile, *f; X{ X int nextstate; X int i; /* for counter of subscript in word */ X int j; /* subscript in tstring */ X int k; /* subscript for copying */ X int m; /* general counter */ X X char tstr[MAXSTR], tempstr[MAXSTR]; /* a temporary string */ X char ch,regname; X X *chcnt = 0; X *wordlength = 0; X i = 0; X j = 0; X X nextstate = 0; X if (InExpression) X nextstate = 4; X X if (InMacro) X nextstate = 17; X X while (nextstate != 999) { X ch = word[i]; X switch(nextstate) { X case 0: /* start state */ X if (ch == '\\') X nextstate = 1; X else if (ch == '\0' || ch == EOF) X nextstate = 999; X else /* copy state */ X nextstate = 1000; X break; X X case 1 : /* it's a '\' */ X i++; X ch = word[i]; X if (ch == '\\') X nextstate = 2; X else if (ch == '(') X nextstate = 3; X else if (ch == '{') X nextstate = 17; X else if (ch == '[') X nextstate = 4; X else if (ch == 'f') X nextstate = 8; X else X /* not a state -- must be an escaped X character, so just copy it */ X nextstate = 1000; X break; X X case 2: /* it is an escaped '\' */ X tstring[j] = ch; X j++; X tstring[j] = SHIFTOUT; X j++; X *wordlength = *wordlength + length('\\'); X *chcnt = *chcnt + 1; X i++; X nextstate = 0; /* start it over */ X break; X X case 3: /* it is an in-line register */ X /* first get the name of the register -- it should X be a single character */ X i++; X regname = word[i]; X i++; /* advance to after the register name */ X if (word[i] == 'C') { X /* interpolate as a character */ X sprintf(tempstr,"%c",REG(regname)); X i++; X } X else /* interpolate as a number */ X sprintf(tempstr,"%d",REG(regname)); X if (word[i]!= ')') { X fprintf(stderr,"error in register name -- one letter only!\n"); X exit(1); X } X i++; /* to skip the ')' */ X /* now append tempstr to the output string */ X k = 0; X while (tempstr[k] != '\0') { X tstring[j] = tempstr[k]; X j++; X tstring[j] = SHIFTOUT; X *wordlength = *wordlength + length(tempstr[k]); X *chcnt = *chcnt + 1; X j++; X k++; X } X nextstate = 0; X X break; X X case 4: /* it is an expression */ X /* don't forget to advance i to end of expression */ X /* the end of the expression is signified by ']' */ X InExpression = TRUE; X tstr[0] = '\0'; X if (word[i] == '[') i++; /* skip the '[' */ X k = 0; X while (word[i] != '\0' && word[i] != ']') { X tstr[k] = word[i]; X i++; X k++; X } X tstr[k] = ' '; X tstr[k + 1] = '\0'; X strcat(expstr,tstr); X if (word[i] == ']') { X InExpression = FALSE; X eval(expstr); X expstr[0] = '\0'; X } X X if (word[i] == ']') i++; /* skip the whatever */ X nextstate = 0; X X break; X X case 8: /* it is a font change, I guess. find out which one */ X i++; X ch = word[i]; X switch (ch) { X case '0' : /* return to default font */ X pushf(cfont,&fstk); X pushi(clen,&ipstk); X clen = defl; X strcpy(cfont,prinit); X k = 0; X if (!FONTCHANGE) X cfont[0] = '\0'; X while (cfont[k] != '\0') { X tstring[j] = cfont[k]; X k++; X j++; X } X break; X case 'M': /* mountable font */ X case '1' : /* Presentation 18 point */ X pushf(cfont,&fstk); X pushi(clen,&ipstk); X strcpy(cfont,Pres); X clen = &P18bl[0]; X k = 0; X if (!FONTCHANGE ) cfont[0] = '\0'; X while (cfont[k] != '\0') { X tstring[j] = cfont[k]; X k++; X j++; X } X break; X case '2' : /* Helv. 6 point font */ X pushf(cfont,&fstk); X pushi(clen,&ipstk); X X sprintf(cfont,"%c&10O%c(0U%c(s1p6vsb4T",ESC,ESC,ESC); X clen = &H6nl[0]; X k = 0; X if (!FONTCHANGE ) cfont[0] = '\0'; X while (cfont[k] != '\0') { X tstring[j] = cfont[k]; X k++; X j++; X } X break; X case 'B' : /* Times Roman bold face */ X pushf(cfont,&fstk); X pushi(clen,&ipstk); X clen = &H12bl[0]; X sprintf(cfont,"%c&l0O%c(0U%c(s1p12vs3b5T",ESC,ESC,ESC); X if (!FONTCHANGE ) cfont[0] = '\0'; X k = 0; X while (cfont[k] != '\0') { X tstring[j] = cfont[k]; X k++; X j++; X } X break; X case 'I' : /* Times Roman 12 point italics */ X pushf(cfont,&fstk); X pushi(clen,&ipstk); X sprintf(cfont,"%c&l0O%c(0U%c(s1p12v1sb5T",ESC,ESC,ESC); X clen = &H12il[0]; X if (!FONTCHANGE ) cfont[0] = '\0'; X k = 0; X while (cfont[k] != '\0') { X tstring[j] = cfont[k]; X k++; X j++; X } X break; X case 'T' : /* Courier */ X pushf(cfont,&fstk); X pushi(clen,&ipstk); X sprintf(cfont,"%c&10O%c(0U%c(sp10h12vsb3T",ESC,ESC,ESC); X clen = &C12nl[0]; X if (!FONTCHANGE ) cfont[0] = '\0'; X k = 0; X while (cfont[k] != '\0') { X tstring[j] = cfont[k]; X k++; X j++; X } X break; X case 'U' : /* underline toggle */ X if (ULSTATE == FALSE) { X sprintf(tempstr,"%c&dD",ESC); X ULSTATE = TRUE; X } X else { X sprintf(tempstr,"%c&d@",ESC); X ULSTATE = FALSE; X } X if (!FONTCHANGE ) tempstr[0] = '\0'; X k = 0; X while (tempstr[k] != '\0') { X tstring[j] = tempstr[k]; X k++; X j++; X } X break; X case 'P' : /* pop font stack */ X popf(cfont,&fstk); X popi(&clen,&ipstk); X k = 0; X if (!FONTCHANGE ) cfont[0] = '\0'; X while (cfont[k] != '\0') { X tstring[j] = cfont[k]; X k++; X j++; X } X break; X X case '|' : /* it's an anchor */ X m = 0; X while (m <= j) { X if (tstring[m] == SHIFTOUT) { X tstring[m] = ANCHOR; X *chcnt = *chcnt - 1; X } X m++; X } X sprintf(tempstr,"%c*p%dX",ESC,TEMPINDENT + LMARGIN + i*length('w') ); X k = 0; X if (!FONTCHANGE ) tempstr[0] = '\0'; X while (tempstr[k] != '\0') { X tstring[j] = tempstr[k]; X k++; X j++; X } X break; X X } /* of switch */ X i++; X nextstate = 0; /* start over */ X break; X X case 17: /* expand in-line string macro */ X /* don't forget to advance i */ X InMacro = TRUE; X if (word[i] == '{') i++; /* skip the '{' */ X k = 0; X while (word[i] != '\0' && word[i] != '}') { X tstr[k] = word[i]; X i++; X k++; X } X tstr[k] = ' '; X tstr[k + 1] = '\0'; X strcat(macline,tstr); X if (word[i] == '}') { X /* skip the '}' */ X i++; X /* end the macline and process the X macro */ X InMacro = FALSE; X procmacro (macline,ifile,f); X macline[0] = '\0'; X tstr[0] = '\0'; X X } X X nextstate = 0; X X break; X X X case 999 : /* at the end state */ X break; X X case 1000 : /* copy state */ X default: X tstring[j] = ch; X j++; X tstring[j] = SHIFTOUT; X nextstate = 0; /* start it over */ X *wordlength = *wordlength + length(ch); X *chcnt = *chcnt + 1; X j++; X i++; X break; X X } /* of switch */ X } /* of while nextstate */ X tstring[j] = '\0'; X k = j - 1; X while (k > 0 && (isspace(tstring[k]) || tstring[k] == SHIFTOUT) ) k--; X if (k > 0 && ispunct(tstring[k]) ) X LASTPERIOD = TRUE; X else X LASTPERIOD = FALSE; X X} /* of parseword */ X X X X/* length returns the length of a character, measured in dots */ Xint length(c) Xchar c; X{ X /* temporarily */ X return *(clen + (c & 0xff)); X} X X/* clean out the buffers, print anything collected, and reset counters */ XVOID brek(ifile,ofile) XFILE *ifile,*ofile; X{ X /* be sure to empty the buffer again */ X outbuf[obufpos] = '\0'; X if (outbuf[0] != '\0') X output(ifile,ofile,0); X outbuf[0] = '\0'; X obufpos = 0; X CURRX = 0; X CHARCNT = 0; X} X X X/* process a line, but do not justify it, and preserve the spaces */ Xnonfill(s,ifile,f) Xchar s[]; XFILE *ifile,*f; X{ X /* The idea is to collect output into a buffer (outbuf) X and when the buffer is complete, write it. X The variable s contains the text of the last line X read from the input file. The output buffer is X in variable "outbuf". */ X X int i; X X char word[MAXSTR],tstring[MAXSTR]; X int wordlength; X int chcnt; X X /* first "doctor" the line if boldcount or italicscount > 0 */ X if (BOLDCNT > 0) { X addstr(s,"\\fB","\\fP"); X BOLDCNT--; X } X if (ITALCNT > 0) { X addstr(s,"\\fI","\\fP"); X ITALCNT--; X } X if (ULCOUNT > 0) { X addstr(s,"\\fU","\\fU"); X ULCOUNT--; X } X X /* now if the line should be centered, do it, otherwise just X format it */ X if ( CNTRCNT > 0) { X /* just parse the line to get the length */ X i = 0; X while (s[i] != '\0') { X getword(word,s,&i); X parseword(tstring,word,&wordlength,&chcnt,ifile,f); X append(tstring,chcnt); X CURRX = CURRX + wordlength; X } X endln(); X X TEMPINDENT = (RMARGIN - LMARGIN - CURRX)/2; X CNTRCNT--; X X output (ifile,f,0); X } X else { /* don't center it */ X X /* expand s to include any font changes */ X parseword(tstring,s,&wordlength,&chcnt,ifile,f); X strcpy(outbuf,tstring); X CURRX = CURRX + wordlength ; X TEMPINDENT = 0; X output(ifile,f,0); X X } /* of don't center it */ X} X END_OF_FILE if test 12233 -ne `wc -c <'source/format.c'`; then echo shar: \"'source/format.c'\" unpacked with wrong size! fi # end of 'source/format.c' fi echo shar: End of archive 1 $of 5$. cp /dev/null ark1isdone MISSING="" for I in 1 2 3 4 5 ; do if test ! -f ark${I}isdone ; then MISSING="${MISSING} ${I}" fi done if test "${MISSING}" = "" ; then echo You have unpacked all 5 archives. rm -f ark[1-9]isdone else echo You still must unpack the following archives: echo " " ${MISSING} fi exit 0 exit 0 # Just in case... -- Kent Landfield INTERNET: kent@sparky.IMD.Sterling.COM Sterling Software, IMD UUCP: uunet!sparky!kent Phone: (402) 291-8300 FAX: (402) 291-4362 Please send comp.sources.misc-related mail to kent@uunet.uu.net.