TeXhax Digest Friday, December 18, 1987 Volume 87 : Issue 103 [SCORE.STANFORD.EDU]TEXHAX103.87 Editor: Malcolm Brown Today's Topics: LaTeX problem idea for line printer output .TFM files for using built-in LW fonts with TeX. Single spaced bibliography TeX and Pyramid OsX4.0 TeX on the SUN4 Negative column widths in \halign -- bug or feature? Overfull boxes schizophrenia in minipage footnotes? PXL to raster font converter?? LaTeX list environment question Is it a bug? yet another qms driver Re: TeXhax Digest V87 #100 Undefined control sequences Strange \vboxes TeX sources on score New LaTeX envirinment dvi-->Epson for PC Single Spaced Bibliography ---------------------------------------------------------------------- Date: Thu, 10 Dec 87 11:04:05 MEZ From: A4422DAB%AWIUNI11.BITNET@CUNYVM.CUNY.EDU Subject: LaTeX problem Date: 10 December 1987, 10:55:39 MEZ From: Erich Neuwirth +43 222 486056-160 A4422DAB at AWIUNI11 To: TEXHAX at SCORE.STANFORD I have a problem with LaTeX: I want to print paragraphs with negative indentation. Something like: This is the first line and later it goes on like this and this and this. The problem arises with the first paragraph in a section: I have to use the following code: parindent -10pt \medskip \indent {\bf\indent Absze{\ss}:} {\bf Aconit.}, {\bf Apis}, {\bf Arn.}, {\bf Bellad.}, {\sl Calc.\-carb.}, {\bf Hepar}, {\bf Lach.}, {\bf Merc.\-sol.}, {\bf Silic.}, {\sl Sulf.} \medskip {\bf Atherom:} {\bf Bar.\-carb.}, {\rm Calc.\-carb.}, It will not work if I leave out any one of the two indent in the first paragraph. But all paragraphs from the second one on do not need any indent. Is this a bug or a feature? ------------------------------ Date: Thu, 10 Dec 87 16:38:15 EST From: gary%right.mcs@omnigate.clarkson.edu Subject: idea for line printer output The program ``dvitype'' seems to find the correct word breaks, unlike ``dvitty''. If it could be arranged to skip printing the font changing information (the main reason it was written), break at the actual line breaks, and replace ligatures by their actual combination (use fi in place of the ligature), that would go a long way toward producing simple line printer listings that could be used as readable documentation on machines. If there were a way to give substitute character sequences for non-printer characters (say * for \bullet), most everything but super/sub-scripts would be taken care of. Has anyone examined ``dvitype'' with this in mind? ------------------------------ Date: 10 DEC 87 17:07-EST From: DAK%CUTHRY.BITNET@WISCVM.WISC.EDU Subject: .TFM files for using built-in LW fonts with TeX. Does anyone have appropriate .TFM files and re-encoding vectors for the built-in fonts on the LaserWriter Plus? (The main interest is in Times Roman, Helvetica, Palatino, and Symbol fonts). David A. Kosower DAK@cuthry.BITnet ------------------------------ Date: Thu, 10 Dec 87 16:42:10 CST From: John Rager Subject: Single spaced bibliography Rick Zaccone asks about switching to single spacing for a bibliography. I did this in my dissertation, using \def\baselinestretch{1}. Note, however (Latex manual, p 155); that the value of \baselineskip is set (to default times \baselinestretch) only by a type-size-changing command. In addition, changing from normal size to normal size does not make a change. So, if you are going along in normal size, something like \def\baselinestretch{1}\small\normalsize will get you into single spacing. John Rager ------------------------------ Date: Thu, 10 Dec 87 17:14:07 MST From: hartzell@boulder.Colorado.EDU (George Hartzell) Subject: TeX and Pyramid OsX4.0 We are currently running TeX2.1 on a Pyramid 90x. (The unix tex distribution) We were able to compile it under OsX3.1 but have not been able to get it to work under 4.0. Does anyone know if there ahve been any changes to get it to compile under 4.0? If so we will order another tape (there are some other new goodies on there anyway!). g. ------------------------------ Date: Thu, 10 Dec 87 21:16:34 PST From: mackay@june.cs.washington.edu (Pierre MacKay) Subject: TeX on the SUN4 The SUN4 is a pretty impressive machine, and most of the necessary compatibility has been maintained. Here is the record of making Initex and virtex ==========Compilation of TeX Version 2.7 on a SUN4======================== start making initex Thu Dec 10 14:33:36 PST 1987 Thu Dec 10 14:36:37 PST 1987 finished making initex start making virtex Thu Dec 10 14:36:37 PST 1987 Thu Dec 10 14:39:13 PST 1987 finished all =========================================================================== Yes---that is *3 minutes* from start to finish on initex. Even the bad news is good news. At first reading it may seem to be bad news that the a.out headers have changed yet again, and undump for Motorola SUNs using OS 3.2 and up does not work. On the other hand, the SUN4 is so fast that you only have to wait 2 seconds to load a fmt file. Preloaded tex latex etc is large, and uses up disk space. (RISC compilations are necessarily larger than 680nn compilations, by something close to 1K bytes). Someone is bound to get around to an undump for the SUN4, but meanwhile, you might as well dump lots of super-fast preloaded fmt files for whatever large scale work you may be doing and alias them. E. g. alias tex 'virtex \&plain' alias latex 'virtex \&lplain' alias slitex 'virtex \&splain' alias mytex 'virtex \&myplain' This works even on SUN2s and SUN3s when you can't afford too many 1 Mbyte working copies of TeX. It is so effective on a SUN4 that preloaded TeX may go out of fashion. Pierre A. MacKay TUG Site Coordinator for Unix-flavored TeX ------------------------------ From: Oliver Schoett Subject: Negative column widths in \halign -- bug or feature? The algorithm for the calculation of column widths in \halign (TeXbook p. 245) may easily yield negative column widths and thus produce strange tables. Here is an example: \halign{#&#\cr text\cr % vanilla first column, no second column \span\hrulefill\cr} % entry of zero natural width spanning cols 1 and 2. You won't get an underlined 'text', but no rule at all. TeX calculates the width of the second column to be the negative of that of the first! (Try the algorithm with w11=17.5pt (from 'text'), w12=0pt (from \span\hrulefill), and w22=-\infty (no entry for the second column alone), all t_i zero.) The combined width of the two columns is zero, the \hrulefill duly fills this zero width and thus becomes invisible. You will find that the lines produced by this alignment have zero width also (try putting it in a \vbox!). If you use a \tabskip of zero natural width (e. g., \tabskip=0pt plus1fil) and use \halign to , the entire will be distributed over the tabskip glues, no matter how long 'text' is. The rule will then be as long as the center \tabskip (i. e., 1/3 if all three tabskips are the same). This very strange behaviour makes me think that the algorithm should be fixed to never produce negative column widths: it should calculate each w_j as max(0, ). This fix would also help in the case that all entries in a column end with \hidewidth. Again, this yields a negative column width and thus a strange table. Oliver Schoett schoett@lan.informatik.tu-muenchen.dbp.de / relay.cs.net schoett%lan.informatik.tu-muenchen.dbp.de@ - unido.uucp \ ddoinf6.bitnet ------------------------------ Date: Fri, 11 Dec 87 14:10 AST Subject: Overfull boxes Here's a problem that is driving me around the bend: Is there any way to detect an overfull \vbox and take corrective action? The problem comes up in the listprocessing of an address list for a form letter. We have two sizes of sticky labels we can use to print the addresses on, and ordinarily the small size of labels is sufficient. Yesterday I used a list where that was not the case, and that got me in trouble. Each address is set separately in a \vbox, and the \vboxes are lined up properly with the label boundaries. An overfull \vbox pushes the following labels down a bit, and after that the alignment is completely out of whack. I would like to use the small labels as much as I can, and I would like to be able to get TeX to automatically switch to the larger size as soon as an overfull box has been detected. Any overfull boxes occurred at the larger size I think I would simply throw out as inadmissible and type by hand. Any ideas, pointers, comments would be appreciated. gus gassmann ( GASSMANN @ dalac.bitnet ) ------------------------------ Date: Fri, 11 Dec 87 13:20:08 EST From: crl@maxwell.physics.purdue.edu (Charles R. LaBrec) Subject: schizophrenia in minipage footnotes? Looking at latex.tex, it appears that \footnotemark's are made using \thefootnote, but that \footnotetext uses \thempfn, which is either \thefootnote or \thempfootnote, depending on whether you are in a minipage or not. The result is to have the marks in the text and the marks in the footnote possibly be in different formats (since minipages use \alph style for \thempfootnote by default). Is this the intended behavior? Charles LaBrec crl @ maxwell.physics.purdue.edu ------------------------------ Date: Fri, 11 Dec 87 16:47:35 PST From: Paul Kuo-hwai Chang Subject: PXL to raster font converter?? We have a Imagen printer here in Math, UCLA and we use both dtroff and TeX. TeX has many fonts that are not accessible through dtroff since it uses raster fonts. We like to know if there is such a utility that allows one to convert PXL fonts under TeX to raster fonts that can be used by dtroff and dimp. Any info will be much appreciated. Please reply directly to paul@math.ucla.edu. Thanks. paul chang Mathematics, UCLA ------------------------------ Date: Fri, 11 Dec 87 15:11:14 EST From: Steve Buyske Subject: LaTeX list environment question I would like to make a construct a list environment so that I can have labels that might require two line. Something like Professional Something silly, something else silly, some more Associations: silly things, and so on. If I put the label in a \parbox, no matter how I fiddle around, I get something like: Professional Something silly, something else silly, some more Associations: silly things, and so on. with an ugly extra line. Can anyone give me some advice? My apologies if this is an easy question, but I'm new to LaTeX and desperate to get my C.V. done. ------------------------------ Date: Sat, 12 Dec 87 17:40:51 GMT From: CET1%phoenix.cambridge.ac.uk@NSS.Cs.Ucl.AC.UK Subject: Is it a bug? The TeXbook (p.213) says ``Then the expansion of the entire \csname... \endcsname text will be a single control sequence token, defined to be like \relax if it has not previously occurred.'' Apart from the fact that ``is currently undefined'' would be more accurate than ``has not previously occurred'', either there is something missing here, or there is a bug in TeX; because the definition to be ``like \relax'' is *global*. Try, for example: \def\wombat{gerbil} \begingroup \let\wombat=\undefined \csname wombat\endcsname \endgroup \show\wombat Chris Thompson JANET: cet1@uk.ac.cam.phx ARPA: cet1%phx.cam.ac.uk@nss.cs.ucl.ac.uk ------------------------------ Date: Thu, 10 Dec 87 13:36 PST From: SHECTMAN%SETI@icdc.llnl.gov Subject: yet another qms driver In regards to Coleman's response to Simpson (texhax 87-100) I have also implemented a QMS driver (800,1200,2400). It started several years ago when no driver was available, and has evolved slowly over the years. Currently we take our's two steps further than MIT. First, we down load font data with each file, so we don't have to remember the state of the printer. Unlike MIT, we download each character glyph the first time its encountered in the dvifile, not the whole font file. This does save us some time (we have both parallel and serial interfaces, and while its wonderful on the parallel, it really pays off on the serial interfaces). We've seen that alot of characters in a font are not used, so we don't bother to waste time transmitting them. We have paid the cost for this however. Until recently, there has been a QMS bug that would hang the printer controller on occasion when sending TeX data out. How often is sensitive to the type of printer (800,1200,2400) and interface type (serial vs parallel). A parallel 800 hang occurs once or twice a year, a serial 800 can happen every day. Our latest feature concerns the pixel data needed by our drivers (QMS and LN03).\ We (I) got tired of needing 300,000 or so blocks for every font at every size in both QMS Portrait and Landscape orientations, and LN03 data. Not to mentioned it being replicated on every VAX or cluster around the site. The latest version includes a hacked copy of DVITYPE that scans the .DVI file to find out what fonts are needed. It then checks to see if the font has been expanded from .GF to LN03 or QMS (portrait or landscape) format. If it hasn't, it spawns off a process to do the conversion. Thus over time, we have a slowly increasing set of expanded fonts. We seem to have stabilized at about 60,000 blocks for pixel type data since we started about 6 months ago. If it ever gets to around to taking up too much space, we can wipe out all the pixel files and let it start over. All our system managers who have installed this version love the disk space they have recovered. I am in the process of getting the paperwork to release the software ready. Most or all of this can be placed on the network for people to snarf (though beware, documentation is practically non-existent, command files are simple and short but undocumented - you get what you pay for). Robert M. Shectman Lawrence Livermore National Lab shectman@seti.llnl.gov ------------------------------ From: unido!iaoobel!bossix!woerz@uunet.UU.NET (Dieter Woerz) Subject: Re: TeXhax Digest V87 #100 Date: Sat, 12 Dec 87 21:47:01 MET DST Reply-To: iaoobel!woerz@uunet.UU.NET > Here is a context diff that can be used to patch TeX version 2.5 > into version 2.7 on Unix systems. If you have older versions > to patch, look at tex_patches.sh on ~ftp/pub at june.cs.washington.edu. Hello, Is there some possiblity to get the patches for TeX via mail. I don't have access to the ARPA-Net, so I can't ftp the patches. Thanks in advance Dieter Woerz Fraunhofer Institut fuer Arbeitswirtschaft und Organisation Abt. 453 Holzgartenstrasse 17 D-7000 Stuttgart 1 W-Germany BITNET: iaoobel.uucp!woerz@unido.bitnet UUCP: ...{uunet!unido, pyramid}!iaoobel!woerz ------------------------------ Date: Sun 13 Dec 87 16:26:45-CST From: Michael O'Leary Subject: Undefined control sequences While running LaTeX this past week (version 2.07, release 28-Oct-84; Tops-20 TeX Version 1.1) I have gotten `undefined control sequence' errors for the commands \value, \samepage and \clearpage. These are all documented in the LaTeX manual, but is there something I need to do to make their definitions accessible? Michael O'Leary ------------------------------ Date: Mon, 14 Dec 87 12:02 AST From: Gus Gassmann Subject: Strange \vboxes Can anyone account for the following curious measurements when it comes to overfull boxes? % This simple TeX program \setbox0=\vtop to 2pt{\null\hbox{TEST}} \showbox0 \box0 \bye produces this output: Overfull \vbox (10.0pt too high) detected at line 1 \vbox(2.0+0.0)x26.8056 <----- positive height ..\hbox(0.0+0.0)x0.0 ..\glue(\baselineskip) 5.16669 ..\hbox(6.83331+0.0)x26.8056 ...\tenrm T ...\tenrm E ...\tenrm S ...\tenrm T > \box0= \vbox(0.0+2.0)x26.8056 <----- height is zero ..\hbox(0.0+0.0)x0.0 ..\glue(\baselineskip) 5.16669 ..\hbox(6.83331+0.0)x26.8056 ...\tenrm T ...\tenrm E ...\tenrm S ...\tenrm T Is this a bug or am I missing something? gus gassmann (GASSMANN @ dalac.bitnet) ------------------------------ Date: Mon, 14 Dec 87 12:28:28 EDT From: Dimitri Vulis To: TeXHAX Digest From pavel1.txh (TeX v. troff comparison): >... Many, many mathematical symbols, >accent marks and other symbols are available. >Cyrilic and OCR (Optical Character Recognition) >fonts are also available for TeX. Hmm. I was not aware of an OCR font for TeX. Can someone pls tell me whether this is a mistake or if this font is non-PD and costs an arm and a leg or what is the status? Supposedly, Post Office OCRs (in New York, at least) only recognize OCR-A and OCR-B; if you print the address on an envelope in anything else, it has to be processed by hand which takes a few extra days. So, it would be nice to have. DV ------------------------------ Date: Mon, 14 Dec 87 14:58:57 CST From: William LeFebvre Subject: TeX sources on score I assume that one reason most of the stuff relating to TeX is stored on score.stanford.edu is to allow ARPANet people to FTP it away at will. I sincerely hope this is the case, because I do it all the time. For example, I just recently retrieved all the newest files for LaTeX from so that we are completely up to date (I hadn't done that for a long time). I will soon be getting the up-to-date versions of the CM fonts from . I also know that "tex.web" can be found in (makes sense). But I can't for the life of me find a copy of "plain.tex". Can anyone tell me where it is? I suppose I could get a copy from june.cs.washington.edu---I haven't looked there yet. William LeFebvre Department of Computer Science Rice University ------------------------------ Date: Mon, 14 Dec 87 15:10:52 pst From: lamport@src.dec.com (Leslie Lamport) Subject: New LaTeX envirinment I've received numerous requests for a LaTeX environment that worked like `verbatim' except allowed embedded commands, such as font-changing and \input commands. I finally got tired of repeating my advice on how to roll your own and wrote one. It's contained in the following document-style-option file. Just cut out the following and put it into the file alltt.sty. The document-style option `alltt' will then define the `alltt' environment, which is described in the comments below. Please let me know if you find any bugs in it. I trust it will find its way into the document-style repository; I probably won't make it part of the standard LaTeX distribution. %%%%%%%%%%%%%%%%%%%%%% file alltt.sty %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % ALLTT DOCUMENT-STYLE OPTION - released 17 December 1987 % for LaTeX version 2.09 % Copyright (C) 1987 by Leslie Lamport % Defines the `alltt' environment, which is like the `verbatim' % environment except that `\', `\{', and `\}' have their usual meanings. % Thus, other commands and environemnts can appear within an `alltt' % environment. Here are some things you may want to do in an `alltt' % environment: % % * Change fonts--e.g., by typing `{\em empasized text\/}'. % % * Insert text from a file foo.tex by typing `input{foo}'. Beware that % each stars a new line, so if foo.tex ends with a % you can wind up with an extra blank line if you're not careful. % % * Insert a math formula. Note that `$' just produces a dollar sign, % so you'll have to type `\(...\)' or `\[...\]'. Also, `^' and `_' % just produce their characters; use `\sp' or `\sb' for super- and % subscripts, as in `\(x\sp{2}\)'. \def\docspecials{\do\ \do\$\do\&% \do\#\do\^\do\^^K\do\_\do\^^A\do\%\do\~} \def\alltt{\trivlist \item[]\if@minipage\else\vskip\parskip\fi \leftskip\@totalleftmargin\rightskip\z@ \parindent\z@\parfillskip\@flushglue\parskip\z@ \@tempswafalse \def\par{\if@tempswa\hbox{}\fi\@tempswatrue\@@par} \obeylines \tt \catcode``=13 \@noligs \let\do\@makeother \docspecials \frenchspacing\@vobeyspaces} \let\endalltt=\endtrivlist ------------------------------ Date: Mon, 14 Dec 87 22:08 EST From: <16448591%VUVAXCOM.BITNET@forsythe.stanford.edu> (Mark Schaffer) Subject: dvi-->Epson for PC I am looking for a MS/PC dos program that will convert dvi files to something that my Epson MX printer can output. Public domain programs are preferred since I cannot afford to spend much on this. Any ideas, suggestions, info would be greatly appreciated. | Mark Schaffer | BITNET: 164485913@vuvaxcom | | Villanova University | UUCP: ...{ihnp4!psuvax1,burdvax,cbmvax,pyrnj,bpa} | | (Go Wildcats!) | !vu-vlsi!excalibur!164485913 | ------------------------------ Date: Tue, 15 Dec 87 09:58 N From: Subject: Single Spaced Bibliography From: Nico Poppelier (Poppelier@Hutruu51.Bitnet) In TeXhax #100 Rick Zaccone writes : > I have a friend whi is typesetting his dissertation using LaTeX to > typeset his dissertation. > ... > That is, setting \baselinestretch back to one seems to be ignored. > What's wrong here? There is more in LaTeX about vertical spacing than just \baselineskip and \baselinestretch. A bibliography, for instance, is a form of the list environment, and the list environment has a lot of format parameters: have a look at the diagram on page 113 of the LaTeX book and the summary in subsection C.5.3. The relevant parameter in this case is \itemsep, the amount of vertical space between items in the list. On page 167 of the book Leslie Lamport explains: '\itemsep The amount of extra vertical space (in addition to \parsep) inserted between successive list items.' Setting this parameter to 0 might be the answer. Nico Poppelier Theoretical Nuclear Physics University of Utrecht The Netherlands ------------------------------ %%% %%% subscriptions, address changes to: texhax-request@score.stanford.edu %%% please send a valid arpanet address!! %%% %%% submissions to: texhax@score.stanford.edu %%% %%% BITNET redistribution: TEX-L@MARIST.BITNET (list server) %%% %%%\bye %%% ------------------------------ End of TeXhax Digest ************************** -------