TeXhax Digest Thursday, February 11, 1988 Volume 88 : Issue 14 [SCORE.STANFORD.EDU]TEXHAX14.88 Editor: Malcolm Brown Today's Topics: RE: bug in bibtex alphabetic styles? AmS-TeX and \mathcode PICTEX request Interesting note of the month An interesting puzzle. Something similar to the dangerous bend Re: file extensions Re: some questions about fonts hanging sectionhead macro of david f rogers, TeXhax 88(10) Re: day field in bibliography? TeX for CTOS... LaTeX Notes (Re: TeXhax Digest V88 #12) PXtoPK problems on VMS TeX pascal vs. TeX in C dvi2la75 no indent after section header including GEMdrawings: report Unix_DVI to VMS_DVI conversion BibTeX 0.99b \dotfill in \halign ---------------------------------------------------------------------- Mail-From: PATASHNIK created at 6-Feb-88 08:01:42 Date: Sat 6 Feb 88 08:01:42-PST From: Oren Patashnik Subject: RE: bug in bibtex alphabetic styles? > From: Marc Shapiro > Subject: bug in bibtex alphabetic styles? > It looks like I've found a bug in the alphabetic styles of bibtex, but > it's so big I can't imagine noone found it before. [He goes on to > describe the alphabetizing for the `alpha' standard style. The crux: > author = "M. Stella Atkins and Gamik Bobloian", > precedes > author = "Noga Alon and Amnon Barak and Udi Manber", > in the reference list, which is different from what you might expect.] This is definitely intentional, because the labels that `alpha' generates for these two entries, assuming they both appeared in 1987, are [AB87] and [ABM87]; that is, `alpha' alphabetizes first by label, then by author. This label-first alphabetizing scheme is best (for `alpha') for two reasons. First, it's the label that a reader uses to find the reference in the reference list, so the entries should be alphabetized first by label. (Otherwise, for example, a reader might have to search all the `A' entries to find [AB87]; this wouldn't be so hard for a 7-entry reference list, but it would be quite annoying for a 50-page bibliography that has three pages of `A' listings.) And second, if you have entries for [AB87a] and [AB87b], you'd like them to appear together; but by the author-first scheme, they might be separated (and perhaps by several pages) in the 50-page bibliography. The problem you bring up is one of the reasons I personally prefer the `plain' style to the `alpha' style (but I've done enough flaming about bibliography styles during the past week to omit the other, more important reasons). --Oren Patashnik ------------------------------ Date: Sun, 7 Feb 1988 02:21 EST From: Jim Walker Subject: AmS-TeX and \mathcode Is there a more recent version of AmS-TeX than 1.1? In particular I am wondering about the fact that amstex.tex contains \mathcode declarations for the digits 0-9 that prevents them from changing fonts in math mode. For instance that prevents AmS-TeX's own \oldnos macro from working. I am tempted to take them out, or reset the \mathcodes in amsppt.sty, but surely there must have been a reason?? %%% Barbara Beeton responds: From: Barbara Beeton Subject: Re: AmS-TeX and \mathcode the suppression of \oldnos in amstex is certainly a bug, and mike spivak knows about it. he is working on a new version of amstex, which will contain a few bug fixes (including this) and, principally, extensive documentation of the entire macro set. i have tried, but failed, to determine when he expects to be finished. if there is sufficient demand, it may be possible to construct an interim fix; this has not been a high priority item at the math society, since oldstyle numerals are not used in our journals. i can't make any promises, but will look into it. i know there was a reason for the \mathcode declarations (and the \oldnos glitch was unforeseen), but don't remember what it was. -- barbara beeton ------------------------------ Date: Sun, 07 Feb 88 21:14:10 SET To: texhax@score.stanford.EDU Subject: PICTEX request Some times ago I read about PICTEX. Who can help me recieving this package/macros?Has anyone a description? -UTE- ------------------------------ Date: Sun, 7 Feb 88 15:32 PST From: Don Hosek Subject: Interesting note of the month On page 387 of the TeXbook (that's appendix D for those of you keeping score at home), Donald Knuth writes, Some implentations of TeX display the output as you are running, so that you can choose column widths interactively until a suitable balance is obtained. To which I reply, "WHAT?" As far as I know, the only product currently existing that does something like this is Arbortext's The Publisher (I haven't seen it myself, so I *could* be wrong). But of course, those of us who get the press releases from Arbor know that the Publisher just came into being at the beginning of last year, well after I had stopped at the University bookstore to buy my TeXbook. It almost sounds like Don Knuth has something at Stanford he's not letting us know about. So, to bring this epistle (neat word, eh?) to a close, I offer this question to the rest of you out there: Have you seen such a beast? I don't think even Bart Childs (the TUG president) has, since in a recent interview (in TeXline) he spoke of TeX implementations with a built in previewer in the future tense. Hmmm. Ingenuously yours, -dh ------------------------------ Date: Sun, 7 Feb 88 16:18 PST From: Don Hosek Subject: An interesting puzzle. In _Tom Jones_, (Penguin Classics, Middlesex England. 1987.), each chapter has a long title (up to six lines) that is typeset with all but the last line being set flush right and flush left, and the last line centered. Viz, A Dialogue between the Landlady, and Susan the Chambermaid, proper to be read by all Innkeepers, and their Servants; with the Arrival, and affable Behaviour of a beautiful young Lady; which may teach Persons of Condition how they may acquire the Love of the whole World. And now for the puzzle: Can these chapter headings be typeset using TeX's line breaking algorithm, and if so how? If anybody can come up with an answer, I would appreciate hearing it. -dh (gosh I'm a busy little boy today) ------------------------------ Date: Sun, 7 Feb 88 20:15 CDT From: Subject: Something similar to the dangerous bend Could anyone on the TeX list please give me some pointers on how I could get the first character on some particular paragraphs come out in big type size as you see in some magazines, such as the New Yorker? I copied the following example from the TeX book (page 419) for producing the dangerous bend signs and modified it a little for my use. However, the first character does not align properly in the hbox and comes out more like a superscript! If someone could give me a hand with this, I sure will apreciate it very much. Here are the code and example I tried: \font\cmbxfive=cmbx10 scaled 2074 \def\f@rstc#1{\medbreak\begingroup\clubpenalty=10000 \hangindent=2pc \def\par{\endgraf\endgroup\medbreak} \noindent\hang\hangafter=-2 \hbox to 0pt{\hskip-\hangindent{\cmbxfive #1}\hfill}} \outer\def\firstc{\f@rstc} \def\endfirstc{\endgraf\endgroup} \firstc{T}his is a test. This is only a test. This is the test of the macro written for printing the first character of certain paragraphs in big type size and align it in side a box spanning the first two lines of the paragraph. This is only a test. This is the end of this test. Thank you. Murali Krishnamurthi MURALI@TAMLSR or KDM2520@TAMSIGMA \bye ------------------------------ Date: Sun 7 Feb 88 22:26:52-PST From: Eric M. Berg Subject: Re: file extensions John (Ramsdell ?) writes: >> I would like to promote the idea that LaTeX and SliTeX source files be >> given a different extension besides ".tex". [....] The dificulty with implementing this is that the part of TeX which knows about file names and such is in "TeX--The Program", not in the Plain, LaTeX, or SliTeX macro packages. In particular, that's where the default file type is defined. (see section 537 in part 29 of "TeX--The Program") Currently, LaTeX and SliTeX are implemented as macro packages on top of TeX. (That is, LATEX.EXE and SLITEX.EXE are prepared using the same IniTeX and VirTeX used to prepare TEX.EXE, just \inputing macro packages other than Plain.) However, in order to implement different default file types for (say) LaTeX, it would be necessary to prepare a WEB change file, run Tangle, and compile and link IniLaTeX.PAS and VirLaTeX.PAS to get IniLaTeX.EXE and VirLaTeX.EXE, before \inputing the LaTeX macro package (and similarly for SliTeX). (In addition, this would have the somewhat confusing side-effect. Currently, the results of running LATEX.EXE are identical to those of running TeX with the command TEX &lplain which loads the LaTeX .FMT file instead of the Plain .FMT. For the reason described above, a different .FMT file by itself can't change the default file type, so the two invocations would look for different file types.) My sense is that this suggestion isn't likely to meet with great enthusiasm, at least with the people who have to maintain TeX & friends at different sites. Eric M. Berg Grad. School of Business Stanford University ------------------------------ Date: Sun 7 Feb 88 22:57:12-PST From: Eric M. Berg Subject: Re: some questions about fonts The METAFONT sources for the invisible SliTeX fonts are distributed with the LaTeX sources on the TeX distribution tapes. They're available via anoymous Internet FTP from the directory on Score.Stanford.EDU, which doesn't help you much if you're on the Bitnet. Perhaps you can find someone on the Bitnet who can mail them to you. Eric M. Berg Graduate School of Business Stanford University ------------------------------ Date: 8-FEB-1988 15:32:08 GMT From: FPS%VAXA.CC.IMPERIAL.AC.UK@forsythe.stanford.edu Subject: hanging sectionhead macro of david f rogers, TeXhax 88(10) did i miss something in this? seems to me that the problem vanishes just as soon as you remove the spare spaces in the arguments. forget the macro, get the arguments right e.g. \def\sectionhead#1#2#3\par, \setbox0=\hbox,,\bf,#1-- \hskip0.5em- \setbox1=\vtop,\raggedright \hsize 3pc \noindent \rm,#2-- \setbox2=\vtop,\hsize 29 pc \advance \hsize by -\wd0 \advance \hsize by -\wd1 \noindent \rm,#3-- \noindent \halign,## & ## & ##\cr \box0 & \box1 & \box2 \cr- \bigskip - \sectionhead,I.-,First section-% Twas brillig and the slithy toves did gyre and gimble in the wabe: all mimsy were the borogoves, and the mome raths outgrab. \noindent ``Beware the Jabberwock, my son! The jaws that bite, the claws that catch! Beware the Jubjub bird, and shun the frumious Bandersnatch!'' \bye gives something which looks like i think you want. i couldn't resist changing the structure of the arguments a bit. i wasn't sure what the stuff at the end of dfr's macros was about so i left it out. (when in doubt, don't.) i still feel i've misunderstood something somewhere. malcolm clark janet: texline@uk.ac.ic.cc.vaxa ------------------------------ Mail-From: PATASHNIK created at 8-Feb-88 09:16:01 Date: Mon 8 Feb 88 09:16:01-PST From: Oren Patashnik Subject: Re: day field in bibliography? > Subject: day field in bibliography? > From: kelem@aerospace.aero.org > Why are there "month" and "year" fields in bibtex, but no "day" field? > If you say month="June 2",year=1982, you don't get a comma between the day > and the year. There's no `day' field in the standard styles because it comes up fairly infrequently and because the `month' field handles it fine as is--- you can get what you want by simply appending a comma after the `2'. However, as mentioned in the new BibTeX documentation, it's slightly better to use the new concatenation feature, and also to use a tie: month = jun # "~2," (asuming you have the new version of BibTeX). Furthermore, van Leunen recommends putting the day first, so you'd use month = "2~" # jun (with this form you don't need the comma at all). --Oren Patashnik ------------------------------ Subject: TeX for CTOS... Date: Mon Feb 8 07:50:11 1988 Howdee! I was wondering if there is a version of TeX (and friends) for the CTOS operating system (for Burroughs machines like the B26). We use some horrid (compared to TeX :-) ) WYSIWYG system and I'd love to use TeX instead. Thanks... Mark Scandariato ------------------------------ Date: Mon, 8 Feb 88 10:55:02 pst From: lamport@src.dec.com (Leslie Lamport) Subject: LaTeX Notes (Re: TeXhax Digest V88 #12) Alastair Milne writes: It appears that in LaTex, indentation of paragraphs, whether on or not in the current scope, is turned off after section heading commands. - why is this? and - how does one keep it turned on? (I tried putting a \indent after the section header: it worked by putting the spacing of an empty paragraph between the header and the first paragraph. That paragraph was indented, alright, but it was also too far below the header.) To answer his questions: - why is this? Because the document styles were designed by someone who understands the principles of good design. Anyone wishing further details should spend some time learning about book design. There are numerous schools of design that teach the subject. - how does one keep it turned on? One designs one's own document style, presumably after consulting with a professional designer. Searching for the definitions of the sectioning commands in the .doc files, and reading the accompanying comments, should indicate how. Leslie Lamport ------------------------------ Date: Mon, 8 Feb 88 10:43 PST From: Don Hosek Subject: PXtoPK problems on VMS Be aware that you may encounter 2 different problems with VMS PXtoPK (version 2.3); the first involves a minor bug in the CHange file as distributed by Jerry Leichter (and perhaps others). In line 326, ask:=status(pk_file)<>0; should be shanged to ask:= status(pxl_file)<>0; this minor bug will cause errors opening the PXL file when prompting the user to go undetected and the program will then give a file not found error on PXLFILE.DAT when it first attempts to read from the PXL file. The second problem regards certain PXL files as distributed---notably the AMS fonts (EU*, MCY*, MSY*). These files have a record format of "undefined, maximum record length 512 bytes". This should be changed (using your favorite VMS utility) to "fixed 512 byte records". The undefined record type causes the open file command for the PXL file to fail. -dh ------------------------------ Date: Mon, 8 Feb 88 15:10:13 MST From: grandi@noao.arizona.edu (Steve Grandi) Subject: TeX pascal vs. TeX in C I have recently installed 3 new versions of TeX on a VAX-11/750 running Unix 4.3BSD: standard Pascal TeX from the Unix TeX tape, Pat Monardo's Common TeX (written in C) and Tim Morgan's TeX-to-C. I ran the same 28 page document through all three programs with the following results: Pat Monardo's Common TeX (with patches from Mark Davies) built with the BIGG define: % time lactex noaolocal This is Common TeX, Version 2.1.1 (preloa ded format=lplain 88.1.19) (./noaolocal.tex LaTeX Version 2.09 <14 Jan 1988> . . . . . 222.6u 4.8s 4:02 93% 265+1118k 76+46io 113pf+0w Tim Morgan's TeX-to-C: % time latex noaolocal This is TeX, C Version 2.9 (preloaded format=lplain 88.1.18) (noaolocal.tex LaTeX Version 2.09 <14 Jan 1988> . . . . . 258.2u 3.6s 4:33 95% 262+904k 60+43io 58pf+0w Pascal TeX from the Unix TeX tape: % time laptex noaolocal This is TeX, Version 2.9 for Berkeley UNIX (preloaded format=lplain 88.1.18) (noaolocal.tex LaTeX Version 2.09 <14 Jan 1988> . . . . . 369.8u 5.1s 6:27 96% 402+840k 57+79io 57pf+0w The result is a clear win for Monardo's C version over Morgan's C version; they both leave Pascal TeX in the dust. (For those of you unfamiliar with the output of the Unix Cshell time command, the numbers for the last run tell us that the program used 369.8 seconds in user mode, 5.1 seconds in system mode, 6 minutes and 27 seconds elapsed time and 96% of the cpu while it was running. We'll ignore the rest of the gibberish!) I've built Common TeX with VERY generous memory arrays; it now easily handles jobs (halftones or epic plots) that failed with Pascal TeX. Increasing the capacity of Morgan's TeX should also be straight forward, but Monardo's Makefile has has several prebuilt size options. The sizes of the Monardo, Morgan and Pascal TeXs "undumped" with LaTeX follow: % ls -l /local/bin/{lactex,latex,laptex} -rwxr-xr-x 1 root 1496064 Jan 19 05:20 /local/bin/lactex -rwxr-xr-x 1 root 768000 Jan 18 09:08 /local/bin/latex -rwxr-xr-x 1 root 843776 Jan 18 07:45 /local/bin/laptex Of course, a TeX can't be called TeX unless it passes the Trip test. Morgan gives details of his code's experiences while Monardo asserts that his does. Mark Davies, who fixed a few bugs in Monardo's code and incorporated the 2.1 fixes, says the following: >Common TeX with these changes almost passes the version of the Trip Test >distributed with the UNIX TeX 2.1 release (the dvi file produced is fine, >but there are a couple of discrepancies in the log file that I am not sure >of yet). Since Pat Monardo says he will soon be back working on Common TeX (after finishing a C version of Metafont) and will incorporate the 2.1 to 2.9 fixes, and since the Morgan release seems solid as well (and provides the complete path from tex.web to C code) I think the status of public domain TeX in C is pretty solid. I hope Pat fixes the extra new-line in the banner message that bothers me everytime I fire it up. One final performance point: On a "turbo" PC/XT clone, PC TeX version 2.1 takes 6:55 clock time to process the same file. Not bad for a $1200 computer! Steve Grandi, National Optical Astronomy Observatories, Tucson AZ, 602-325-9228 UUCP: {arizona,decvax,hao,ihnp4}!noao!grandi or uunet!noao.arizona.edu!grandi Internet: grandi@noao.arizona.edu SPAN/HEPNET: 5356::GRANDI or DRACO::GRANDI ------------------------------ Date: 8 Feb 88 14:20 -0600 From: Jim Walker Subject: dvi2la75 I think I have asked this question before and thrown away the answers; my apologies. Does anybody have (i.e. send me, because I am not on internet) a dvi2la75 program. I'd prefer, if possible, one that uses .pxl files, because I already have dvi2ln3. Thanks, Jim Walker (Great White North Version) ------------------------------ Date: Mon, 8 Feb 88 21:47:43 EST From: Charlie Martin Subject: no indent after section header Alastair -- If you read the *comments* in the appropriate .doc files, there is a comment that explains it. Seems a sort of dorky way to document something, but then *I'm* just a software engineer, not a logician. ------------------------------ Subject: including GEMdrawings: report Date: Tue, 09 Feb 88 02:38:09 -0800 From: Jerry Sweet Tom Pierpoint of Sypro reports success in including GEMdraw pictures in LaTeX documents. The arrangement used to do this requires the following hardware/software configuration on an IBM PC/AT clone: 1. PC TeX with LaTeX 2. ArborText DVILASER/HP driver (interprets \special{hp: plotfile #1}) 3. HP LaserJet Series II (presumably a Plus will work too) 4. Digital Research GEMdraw, current version (no trashcan! Apple's lawyers should be happy now :-<). 5. Public-domain lptx program (might be available on Simtel20) to capture GEMdraw printer output to a file (no intrinsic GEMdraw facility to do this)---if not available, should be simple to write a home-grown utility to do the same thing. 6. Little program to zero out the first two bytes of the captured file (would otherwise reset the printer with ESC-E sequence), easily home-grown. Courage, the worst is yet to come. Run lptx to redirect the output of GEMdraw to a file (say, plot.hp). Run GEMdraw to produce a drawing. Print the drawing in portrait mode (as opposed to landscape mode). "GEMdraw thinks it's got the world's fastest printer", says Tom---the output is captured in disk file plot.hp. Use your home-grown utility to zero out the first two bytes of plot.hp. Use lptx to restore the normal printer output system hooks. Follow the instructions in the ArborText DVILASER/HP manual about how to insert a drawing in your text using a simple little macro that involves \hbox, \vbox to 5.25in, \special{hp: plotfile plot.hp}, etc. Run LaTeX. Run DVIHP. Copy result to printer. All done. Much of this can be done in a batch file. Note that the plot.hp file is just a raw HP PCL file; the DVIHP back end just inserts the whole thing as-is into the output stream. Now, wasn't that \special? -jns P.S. Not affiliated with any of the above-named companies. /j ------------------------------ Date: Tue, 9 Feb 88 03:46:08 PST From: Subject: Unix_DVI to VMS_DVI conversion Recently, someone described how to convert DVI files generated on a Unix system so they could be transferred to VAX/VMS and correctly interpreted. It involved padding the end of the file out to a 512-byte boundary with (I thought) the last character in the file. However, I tried that and one of Nelson Beebe's drivers which I run under VMS still said, "Are you sure this is a DVI file?" Could someone enlighten me, please? I remember seeing the article about it, but can't remember where (comp.text? TeXhax? UKTeX?...) It was in the last month or so, anyway. thanks in advance, Dave +---------------------------------------------------------------------------+ | Cripps Computing Centre, University of Nottingham, Nottingham NG7 2RD, UK | +---------------------------------------------------------------------------+ | JANET: cczdao@uk.ac.nott.vaxa Phone: +44 602 506101 x2064 | | BITNET: cczdao%uk.ac.nott.vaxa@ukacrl.bitnet | | ARPA: cczdao%uk.ac.nott.vaxa@nss.cs.ucl.ac.uk | | or %ukacrl.bitnet@cunyvm.cuny.edu | | UUCP: ,...!mcvax-!ukc!nott-cs!vaxa!cczdao Voice: "Dave!" | ------------------------------ Mail-From: PATASHNIK created at 9-Feb-88 05:42:38 Date: Tue 9 Feb 88 05:42:38-PST From: Oren Patashnik Subject: BibTeX 0.99b There is a minor bug in BibTeX version 0.99a, which shows up when the `alpha' standard style tries to create a label---but it shows up only for a name that contains braces other than the ones for an accented character. This bug is fixed in version 0.99b, which is now in the standard distribution area on SCORE.STANFORD.EDU. --Oren Patashnik ------------------------------ From: Julian Bradfield Date: Tue, 9 Feb 88 13:24:45 GMT Subject: \dotfill in \halign I had a similar problem when I wanted to typeset a contents page for a bulletin, in the form page no\ title ..........................................author I was unable to solve this, and so was everybody else I asked. I eventually resorted to manual kludges using the tab mechanism; this has the disadvantage that it requires me to specify the longest author's name, or in your case the longest extension. This is what I did: % nasty way to set a contents page \def\n{\hskip0.5em}\let\df=\dotfill % \n is a digit-width space \setbox0=\hbox{Agnieszka Sylwanowicz} % longest entry in the last column \newdimen\firstcolwidth % so first column (page, author, dots) \firstcolwidth=\hsize % is \hsize - \wd0 \advance\firstcolwidth by -\wd0 % now manually set up the tab control box appropriately (ugh!) \setbox\tabs=\hbox{\hbox{\hskip\wd0}\hbox{\hskip\firstcolwidth}} % and here we go (not making sufficient use of macros!) \+\n2 Foreword \& General Comments\df\ &DD\cr \+\n3 \w{The Hobbit} in Germany\df\ &Manfred Zimmerman\cr \+\n8 Some comments on the Dutch\cr \+\n\n\quad translation of \w{The Hobbit}\df\ &Ren\'ee Vink\cr \+11 \w{The Hobbit} in Norwegian\df\ &Nils-Ivar Ag\o y\cr \+17 \w{O Hobbit}\df\ &Ronald Kyrmse\cr \+19 The Japanese \w{Hobbit}\df\ &Takashi Okunishi\cr \+21 A few comments on Maria Skibniewska's\cr \+\n\n\quad translation of \w{The Hobbit}\df\ &Agnieszka Sylwanowicz\cr \+24 Finnish\df\ &Ellen Pakarinen\cr \+\n\n\ Hungarian\df\ &Andrea Fazakas\cr \input cyracc.def \+25 {\cyr Khobbit}\df\ &DD\cr \+26 \w{Bilbo le Hobbit}\df\ &DD\cr If you do find a neat way to do it, please let me know. (I'm not a very regular TeXhax reader, so please mail me.) Julian Bradfield. jcb@lfcs.ed.ac.uk ------------------------------ %%% %%% subscriptions, address changes to: texhax-request@score.stanford.edu %%% please send a valid arpanet address!! %%% %%% BITNET distribution: subscribe by sending the following %%% line to LISTSERV@TAMVM1.BITNET: %%% SUBSCRIBE TEX-L %%% %%% submissions to: texhax@score.stanford.edu %%% %%%\bye %%% ------------------------------ End of TeXhax Digest ************************** -------