TeXhax Digest Sunday, May 24, 1987 Volume 87 : Issue 40 [SCORE.STANFORD.EDU]TEXHAX40.87 Editor: Malcolm Brown Today's Topics: Common Tex DVI2LN3 and font arithmetic (please help) Magnifications and finding font files Suppressing underscore expansion Re: Latex comment environment Reply to Buyske and two queries tty and Justification Typefaces in TeX LaTeX problems/wishes DVI to HPGL ? \vbox \not= \vbox\unvbox\vbox ---------------------------------------------------------------------- Date: 14 May 87 15:49 +0330 From: Vincent Gogan To: texhax@SCORE.STANFORD.EDU Subject: Common Tex Notes regarding Common Tex 2.0 as available from LaTeX-Styles@cs.rochester.edu: - building a preloaded TeX is accomplished a little differently than the standard TeX (pascal) distribution. Instead of: virtex ** &plain \read 0 to\blort \blort=^\ (where ^\ is a quit character typed by the user in order to dump core) you type: virtex **&plain *^\ - I only have had common tex running for a couple of days but already I have noticed some annoying bugs. - frameboxes have sides which do not meet. This does not mean just the corners, but a whole side. (rounding error?) - some words were broken up with spaces inserted between characters (kerning error?) Anyway, I'm now basically waiting for 2.1. Vincent Gogan Memorial University of Newfoundland UUCP: vincent@garfield.uucp CDN: vincent@garfield.mun.cdn CSNET: vincent%garfield.mun.cdn@ubc.csnet ARPA: vincent%garfield.mun.cdn@ubc.csnet@relay.cs.net BITNET: vincent@mun.bitnet ------------------------------ Date: Fri, 15 May 87 08:51:31 adt From: dalcs!stars!rountree@seismo.CSS.GOV (Eric Rountree) To: texhax-request@score.stanford.edu Subject: DVI2LN3 and font arithmetic (please help) We are running TeX version 2.0 on a MicroVAX II with VMS version 4.5. We have version 7 of Flavio Rose's DVI2LN3 driver. We experience the following: - The DVI2LN3 driver does not like some font specifications: 11 point font has to specified as e.g. cmr10 at 10.95 true pt (when we used the earlier versions of TeX and the driver, we had to specify 10.9 true pt). It would be nice if we could just say 11 true pt. The commands: \magnification=\magstep1 \font\testfont=cmr10 scaled\magstep3 (for 21 point fonts) result in an error of the form 'cannot load font magnified 207%' from DVI2LN3. Please note that the command: \font\testfont=cmr10 at 20.74 true pt *works* (this is magnified 207%) The sequence: \magnification=\magstep1 \font\testfont=cmr10 scaled\magstep2 also works. Can anyone shed some light on this for me? I am new at this stuff, so bear with me if this is trivial to you. Eric Rountree System Manager Business School Site Academic Computing Services Dalhousie University Halifax, Nova Scotia Canada UUCP: seismo!dalcs!stars!rountree ------------------------------ Date: Tue 19 May 87 11:17:17-PDT From: Tomas G. Rokicki Subject: Magnifications and finding font files To: texhax-request@Score.Stanford.EDU When searching for a font file based on parameters in the .dvi file, it is possible that calculations will yield a resolution that is off by one. It has been suggested on this forum that the value be rounded to the nearest magstep, and this value used instead. Unfortunately, with MetaFont, people are creating fonts at odd sizes, and using them at odd sizes; for example, \font\myfont=myfont scaled 5000. So, you should first try and load the font at the calculated size. If that fails, you should then try to round the value according to magsteps, and load that size. In my drivers, I have a routine which looks something like (for a 300 dpi printer): int fixsize(dpi) int dpi ; { static int avail_sizes[] = { 300, 329, 360, 432, 518, 622, 746, 896, 999999999 } ; int *i ; for (i=avail_sizes; *i < dpi - 1, i++) ; if (abs(*i-dpi)==1) return(*i) ; else return(0) ; } That's from memory, so don't quote me. After failing to find the correct size, you call this routine. If it returns 0, you load the tfm and report a missing font file; otherwise, you try to find the size it returns. Note that a new size is only returned if the difference between the original and new size is 1. If a set of fonts at, say, 200 dots per inch is also available on your system, you want to have those sizes in the avail_sizes array as well. (Some people like to do things like \font\tiny=cmr10 scaled 666 to load a 200 dpi font.) ------------------------------ Date: 15-MAY-1987 10:23:59 From: THOWARD%graphics.computer-science.manchester.ac.uk@Cs.Ucl.AC.UK To: texhax <@Cs.Ucl.AC.UK:texhax@score.stanford.edu> Subject: Suppressing underscore expansion It's often useful to create index entries on the fly, such as when the thing you're indexing is first defined in a macro. But when there's an underscore involved, you get an undesired expansion: \index{p\_p} % regular indexing, which as expected writes % \indexentry{p\_p}{1} % on the idx file \newcommand{\indirect}[1]{\index{#1}} \indirect{q\_q} % ``indirect indexing'', which writes the following: % \indexentry{q\unhbox \voidb@x \kern .06em % \vbox {\hrule width.3em}q}{1} % Yuk. I can't figure out how to stop this with \protect or \string or anything ---can anyone help? Thanks, Toby Howard Computer Graphics Unit Department of Computer Science University of Manchester Oxford Road Manchester PHONE: 061-273-7121 ext 5429/5406 M13 9PL ARPA: thoward%cgu.cs.man.ac.uk@cs.ucl.ac.uk United Kingdom JANET: thoward@uk.ac.man.cs.cgu ------------------------------ Date: Fri, 15 May 87 05:53:31 PDT From: Z3000PA@AWITUW01.BITNET Subject: Re: Latex comment environment To: texhax@score.stanford.edu May I add some remarks to this debate? 1. I consider it usefull to "comment out" a portion of text lines, e.g. some text that I do not want to print in some interim version but that shall be printed again in a later version of the document. 2. If the lines to be commented out are valid TeX input, with braces and \if's properly paired, and if no "outer" macros are called in it, then \iffalse may be used in lieu of \begin{comment} and \fi in lieu of \end{comment}, and this works both with LaTeX and with Plain Tex! (I usually put some striking comment like %-------- after these commands, so that I can easily see the start and the end of the "commented out" portion of the input file.) ------------------------------ Date: Fri, 15 May 87 13:30:32 CDT From: MATH300%UNLCDC3.BITNET@wiscvm.wisc.edu Subject: Reply to Buyske and two queries To: texhax@score.stanford.edu 1) In reply to the question of Steve Buyske (TeXHaX v. 87 #33): I have used both AmS-TeX and LaTeX for the preparation of mathematical papers. I find that for a journal sized article, by using \include statements to organize sections the AmS-TeX package has sufficient logical structure, (\proclaim, \endproclaim, \demo, \enddemo, \roster, \endroster, etc.) I think the setting of mathematics is easier in AmS-TeX than in LaTeX. When using AmS-TeX I do miss the automatic citation numbering, and the automatic numbering of equations, and figures, and the easy cross-referencing abilities available in LaTeX. For me, for a journal sized article, this is not a serious problem. Overall, in a way I cannot quantify, AmS-TeX feels more natural and smooth to me than LaTeX. But I would happily use LaTeX for a long monograph or book, provided it was not going to be published by the AMS, which can accept documents formatted with AmS-TeX in machine readable form, e.g. on floppy disks. 2) This brings me to my first question: What journals and publishers currently accept TeX documents approved for publication in a machine readable format? (This should not include journals or series using camera-ready copy.) Here are the ones I know of: a) AMS journals: Bulletin, Notices, Transactions, Proceedings, Memoirs b) Natural Resources Modelling, published by the Rocky Mountain Math. Consortium, edited by Bob McKelvey c) The TUGBoat (of course!) 3) There has been considerable discussion and controversy in this forum concerning good and readable document design/typographic design. What are some references where one can learn the basics of document design? This would be a considerable help when writing one's own macros. Also when explaining to someone "why LaTeX looks that way" it would be nice to cite some reference, beyond saying "Leslie Lamport says so." Please be aware that I am generally satisfied with LaTeX, can do everything I want to in LaTeX, and am pleased with its appearance, but when discussing with others, I need some reinforcements! Steve Dunbar (MATH300 @ UNLCDC3.BITNET) ------------------------------ Date: Mon 18 May 87 08:30:34-PDT From: Ed Brink Subject: tty and Justification To: texhax@Score.Stanford.EDU Well, no, it isn't entirely a silly idea. Consider the plight of someone who has all this neat TeX stuff but wants to send a reasonably readable file via electricity to someone else who has no TeX. Say, an IBM installation (we still use letters that are all the same size, printing with things like SCRIPT and such). Why not a "font" that is nothing more than a stone age formatter, putting out an ASCII file (doesn't have to be EBCDIC...) which can then be read on, say, your garden variety ASCII terminal? Granted that isn't exactly tty, but I betcha that's what the original requester had in mind... ..Ed ------------------------------ Date: 18 MAY 87 01:10-N To: TEXHAX @ SCORE.STANFORD.EDU From: CBTS8001@IRUCCVAX.BITNET (Peter Flynn) To: texhax-request%su.score@wisc.wiscvm.arpa Subject:Typefaces in TeX At last some fellow feeling! I cannot agree more with Sebastain Rahtz (cmi011@ibm.soton.ac.uk) when he suggests junking TNR and the associated rubbish which comes delivered with every WYSIWYG DTP system on the market. Despite what Barbara Beeton says about Times New Roman, it is *not* always suited to long texts, as it tires the eye at sizes over about 9pt. TNR was developed for small size news work, around 7--9pt, and its thick/thin ratios and serif bracketing curves were intended to maintain legibility (note this is not the same as readability) through the hot-metal---flong---stereo letterpress printing process. I suspect TNR has been selected for use in journals and the like more on the grounds of its wide availability and wide range of sorts rather than for its typographical qualities. There are many excellent book faces: Bembo, Plantin, Garamond, Pilgrim and dozens of others. CMR can be spotted as an American typeface at 100 paces: the width of the round characters (CGOQ) and the weight imbalance between thick and thin are dead giveaways (compare with the namestyle on a Pampers box). No European typographer would normally select such a face unless s/he were trying to parody American usage. However, it is both legible and readable from 8pt to 12pt at least, and IT IS ALL WE'VE GOT FOR TEXT WORK (apart from a horrendous kludge on Helvetica provided in the Medley kit). Barbara is quite right to say we must press founders to consider TeX as a market. Most foundries, typesetter manufacturers and printers have *NEVER* heard of TeX, or have a vague idea that it is some crude academic attempt to make general-purpose computers (ie DEC20, PC etc etc) do typesetting. I urge all those who want more faces to call or write your local office of Lino, Mono, Intertype, etc and ask what typefaces are available for TeX. When they say "what's TeX?", explain and try to speak to someone who knows what they are doing. Even Lino, who should know better, claim here never to have heard of TeX (and that was a pro typo!) Flame off...Peter ------------------------------ Date: Mon, 18 May 87 18:52:41 edt From: salkind@nyu-acf8.arpa (Lou Salkind) To: texhax@score.stanford.edu Subject: LaTeX problems/wishes I have been setting a lot of tables in LaTeX lately, and there are several things that I can't figure out how to do conveiently: a) Having a text string implicitly inserted before each entry of a given column. This would let me conveniently set the type of one column in (say) italics. (The @{...} command inserts text and commands after each column entry, except if it appears before the first column specifier.) Ideally I would like something along the following lines: R{...} equivalent to right justified format, with ... appearing before every entry L{...} same as R{} but entry is left justified C{...} same as R{} but entry is centered Then I can say \begin{tabular}{lL{\em}} ... \end{tabular} to set two columns, the first in normal type and the second in italics. (For now I am putting \em before the second entry of every row.) b) Setting double horizontal lines, with the vertical lines being connected across the horizontal lines (normally \hline \hline leaves a gap in the vertical lines). I have tried using \hline \\[\negativelength] \hline to back up and print another line, but this does not work properly at the bottom of a table. Suggestions and comments welcome. Lou Salkind ------------------------------ Date: 18-MAY-1987 17:12:19 From: ABBOTTP%UK.AC.ASTON.MAIL%UK.AC.RL.GB@ac.uk To: TEXHAX@score.stanford.edu Aston University - Electronic Mail Date: 18-May-1987 17:08 BST From: Peter Abbott ABBOTTP@UK.AC.ASTON.MAIL Dept: Computing Service Tel No: 021 359 5492 -direct Subject: LaTex and VDU output We would like to preview LaTeX documents on terminals (VT100). We also need to be able to capture the screen output for inserting in other files (such as the VAX VMS help library). We have Andrew Trevorrow's program which will display the general layout but as he says you cannot proof read only check the format. Can anyone help please? Peter ------------------------------ Date: Tue, 19 May 87 13:28:58 BST From: Ian Moor To: texhax@score.stanford.edu Subject: DVI to HPGL ? We have an HP 7475 plotter, it has a 6 colour carousel (sp?) and a resolution of 0.025 mm (according tothe manual) the plotter will output characters, either predefined or user defined, with varying sizes as well as plot graphs. In the past I have used a very crude Pascal program to convert a text file to HPGL (HP Graphics Language) instructions to plot multicolour slides (commands in the text indicate pen and size changes). The items on SLiTeX make me wonder if anybody has considered a DVI to HPGL translator; I know very little about either form, so - Does a translator exist ? - If not, what are opinions about its feasibility ? ------------------------------ Date: Tue 19 May 87 13:40:15-EDT From: John Gourlay Subject: \vbox \not= \vbox\unvbox\vbox To: texhax@score.stanford.edu, tex%OSU-20@ohio-state.ARPA I would think that the following two ways of creating box 0 would produce the same results, but they don't. The difference is that in the second case the baselineskip glue is missing between the a and b. \setbox0=\vbox{\hbox{a}\hbox{b}\hbox{c}} \showbox0 \setbox1=\vbox{\hbox{a}} \setbox0=\vbox{\unvbox1\hbox{b}\hbox{c}} \showbox0 I solved my immediate problem by using struts, but I'm wondering if anyone can explain why TeX behaves this way and whether or not it's possible to coax TeX into producing the baselineskip in the second case. ------------------------------ %%% %%% subscriptions, address changes to: texhax-request@score.stanford.edu %%% %%% submissions to: texhax@score.stanford.edu %%% %%%\bye %%% ------------------------------ End of TeXhax Digest **************************