TeXhax Digest Monday, November 16, 1987 Volume 87 : Issue 93 [SCORE.STANFORD.EDU]TEXHAX93.87 Editor: Malcolm Brown Today's Topics: SLITEX FONTS Line Printers DVI2LN3 bug (RE: TexHax Digest V87 n.89) X-11 dvi previewer available Re: LaTeX Notes (TeXhax Digest #90) Re: TeXhax Digest V87 #89 TeXhax Digest V87 #89 Re: Some TeX recommendations (from TEXHAX v87 #88) Re: Bibliographies METAFONT/Icelandic Greek Font Typesetting chemical structural formulas--a solution! not for texhax but for response from a caring human Devanagari and Sinhalese Driver for AM-Varityper VT600 Postscript laser printer TeX vs. LaTeX ---------------------------------------------------------------------- Date: 6 Nov 87 09:31:00 EDT From: Subject: SLITEX FONTS I am trying to use SLITEX and although we have all our Slitex fonts loaded, tex slide &splain.tex invariably wants fonts we don't have: error opening ... amssq8.249 amssqb.249 amssq8.173 Does anyone have these fonts? or is there a simple modification of splain.tex which will make good slides without seeking these files? RSVP Thanks! J D Pearlman, Mass Gen Hosp ------------------------------ Date: Fri, 6 Nov 87 11:46:56 EST From: Gary Levin Subject: Line Printers Is there any utility for getting a rough draft of TeX output from a regular line printer / ascii terminal? One of the nice features of TROFF was that one could use NROFF (with a loss of quality) to produce readable files. `dvitty' is fine for rough pre-viewing, but not usable as machine-readable documentation. ------------------------------ Date: Sat, 7 Nov 1987 01:08 EST From: Jim Walker Subject: DVI2LN3 bug (RE: TexHax Digest V87 n.89) Gerhard Raimann asks (with a test document not reproduced here): Normally, I use PXL-files for the DVI2LN3 driver. But in this case the driver reports error messages that it does not find the fonts magnified 207% and 249%. The problem seems to be independent of the font name. If I use LN03 font files (those with extension .NFT, .109, ...) additionally, the problem disappears. So I have to offer both file types on my VAX since the DVItoVDU screen driver wants PXLs. Is there a good explanation for this behaviour of the DVI2LN3 driver ?? The reason for this problem is that DVI2LN3 does not handle roundoff error well. For instance, if you ask for a 300dpi font (cmr10, say) magnified 207%, then DVI2LN3 may look for cmr10.3111pxl or for cmr10.3110pxl, depending on whether the document was unmagnified and the font was scaled \magstep4, or whether the document was magnified \magstep1 and the font was scaled by \magstep3. I don't know what to do about this except keep 2 copies of the file around. I have informed Flavio Rose, the author of DVI2LN3, of this problem. ------------------------------ Date: Sat, 7 Nov 87 09:22:36 CST From: grunwald@m.cs.uiuc.edu (Dirk Grunwald) Subject: X-11 dvi previewer available I've got a DVI previewer available which currently runs under X-11 windows. It should be a simple matter to port it to other window systems. It is available via anon. FTP from a.cs.uiuc.edu, in files pub/iptex.tar and pub/iptex.tar.Z (compressed). Please get the compressed version if possible. The file contains a re-distribution of Chris Torek's 'iptex' program as well. The 'xdvi' program is built using the dvi-library that Chris wrote. It handles PK, GF & PXL fonts. The directory 'doc' contains a latex document describing the use & options of iptex & xdvi, and an installation guide. In brief, the previewer is structured to provide two views of your document, an overview & a detail view. You are shown one or two 'leaves' of a book. When you point the mouse at a part of the page on a leaf & click a button, that page is re-computed at an expanded size & displayed. The X-11 version attempts to size the leaves to give you the largest image possible. the previewer uses your standard fonts. For us, that means the 300dpi fonts. It shrinks the fonts to fit them on your display. Crude, but effective. If someone ports this to SunView, I'd be interested. It shouldn't take more than a couple of hours. Also, the dvi-related code is distinct from the X-11 related code. It should be easy to build previewers for any number of raster display screens. This has been run on a Sun3/260 HiRes, Sun3/50 and a PC/RT. I think that it also runs on a VS-2000 w/X-11. So, spare that tree, preview first. Does wonders for laying out new styles. Dirk Grunwald Univ. of Illinois grunwald@m.cs.uiuc.edu P.S. For you non-arpa sites, I'll be posting this to comp.sources.unix some time soon, so don't send mail asking for it -- it'll be there soon. ------------------------------ Date: Mon, 9 Nov 87 14:37 N From: Subject: Re: LaTeX Notes (TeXhax Digest #90) In TeXhax digest #88 I asked a question concerning the footnote mechanism of LaTeX. This was answered by Leslie Lamport in #90, and he correctly pointed out that I had made a mistake in the example I gave. Instead of using the 'mpfootnote' counter I should have used the 'footnote' counter. But the example did illustrate a real problem in LaTeX, which was answered by another Dutch LaTeX user, Peter Wagemans from the Groningen University. The point is that inside a minipage environment, a different counter AND a different mark generator is used, and the latter problem was not answered by Leslie Lamport. Peter Wagemans suggested something I had already thought about: creating a new macro '\mpfootnotemark', an analog of \footnotemark, to be used inside minipages. Question 1: the following macro produces the desired result, but ----------- maybe someone has a better idea? \makeatletter \def\mpfootnotemark{\@ifnextchar[{\@xmpfootnotemark}% {\stepcounter{mpfootnote}% \xdef\@thefnmark{\thempfootnote} \@footnotemark}} \def\@xmpfootnotemark[#1]{% \begingroup \c@mpfootnote #1\relax \xdef\@thefnmark{\thempfootnote}% \endgroup \@footnotemark} \makeatletter Peter Wagemans also suggested storing the value of the (mp)footnote counter and using it for repeated references. Unfortunately this has to be done as follows: \newcounter{savecounter} \setcounter{savecounter}{\value{mpfootnote}} ... \mpfootnotemark[\value{savecounter}] since in LaTeX you can keep all sorts of numbers -- pages, sections, figures, tables -- but not footnote numbers. Question 2: could someone explain why? ---------- Nico Poppelier ------------------------------ Date: Mon, 9 Nov 87 11:50 EST From: Jeffrey Mark Siskind Subject: Re: TeXhax Digest V87 #89 Jeffrey Mark Siskind writes I have noticed a bug in LaTeX. When I have run it enough times so that labels and references have converged and then I delete a page, the next time I run LaTeX the table of contents (and possibly other references) is off but I don't get the appropriate warning from LaTeX. I don't know if this is repeatable so if you can't duplicate it let me know and I will try again and send you the files. The message LaTeX Warning: Label(s) may have changed. Rerun to get cross-references right. explicitly asserts that labels may have changed--that is, the values produced by a \ref or \pageref may have changed. The presence or absence of this message implies nothing about whether or not the table of contents (or list of figures, etc.) is correct. Leslie Lamport I guess then, that an additional feature should be provided in LaTeX that prints the appropriate warning when LaTeX should be rerun to get correct numbers in the toc/lof/lot. It is just as potentially dangerous to have incorrect page numbers in the toc/lof/lot as it is in \ref or \pageref. -Jeff ------------------------------ Date: Mon, 9 Nov 87 12:49:20 EST From: Charlie Martin Subject: TeXhax Digest V87 #89 From Leslie Lamport, his reply indented.... Charlie Martin writes: If anyone has gone to the trouble of defining an algorithm environment for LaTeX I'd like to hear about it before I go reinventing the famous rounded object of locomotion. I have never bothered to do this for three reasons: * Everyone has their own preferred pretty-printing style. * Each language construct requires several different formatting styles, which implies different formatting commands. For example, one might want either IF b THEN c ELSE d or IF b THEN c ELSE d depending upon the widths of b, c, and d and the initial indentation of the IF. Even if you have a reasonable algorithm for deciding which representation to use, I doubt if there's a reasonable way to write a TeX macro to implement the algorithm automatically. * I've found the tabbing environment easy enought to use that such macros wouldn't save me that much time. I'm not worried about formatting the if-then-else stuff correctly, (in fact, I don't really think I asked for that, but I suppose I could have been misunderstood.) If I could import stuff from tgrind directly I'd just do that and be reasonably happy. However, there are several things that are at least not intuitive with the tabbing environment as a way to get the algorithms in: 1. What happens when a line is too long? It would be nice if there were an automatic way of handling this. 2. It would be nice if there were an environment that could simply preserve the indentation of an arbitrary text, while going ahead and setting the text itself as requested. Verbatim will, of course, preserve indentation, but leaves the user with only tt font and no access to math mode. Both of these would be helpful, at least for me. I've no dount there is probably a way to do it, probably involving changing the catcode of . It would mean a long hack time for me I think, which is why I don't really want to do it. 3. Failing that, having the definition of a tabbing environment in which the character is aliased to \> would turn the program you mention from a custom hack into entab/expand. A second filter that transforms each TAB into "\>" is certainly possible, but I'd like to avoid introducing the tbl | eqn | etc hack into TeX. 4. The tabbing environment doesn't solve the possible problem of having a series of algorithms in a paper numbered separately by algorithm rather than interleaved with tables or figures. A better approach is probably to write a program in some ordinary language that takes the algorithm as input and produces a LaTeX tabbing environment.... but like most programs of that sort, it implies either a pretty complete compiler front end or enforced restrictions on programming style. All that I really want is an environment that will let me do something like \begin{algorithm} \begin{heresthecode} {\bf main}(argv, argc) \{ fprintf(\verb|"Hello, world.\n"|); \} \end{heresthecode} \caption{Here's the code.} \end{algorithm} ------------------------------ Date: Mon, 9 Nov 87 10:54:38 PST From: KARNEY%PPC.MFENET@NMFECC.ARPA Subject: Re: Some TeX recommendations (from TEXHAX v87 #88) Siskind wants a \lastFileWriteDate, etc in TeX. I can already get my operating system (VAX/VMS) to provide the same functionality. I suspect that similar tricks are possible on other systems. Here is how I do this (for my bibliography). The TeX file is biblio.tex. This includes a line "\input biblio_time". biblio_time.tex is created by the following VMS comands $ dat = f$file_attributes("biblio.tex","rdt") $ year = f$cvtime(dat,,"year") $ month = f$cvtime(dat,,"month") $ day = f$cvtime(dat,,"day") $ time = f$cvtime(dat,,"hour")*60+f$cvtime(dat,,"minute") $ open/write temp biblio_time.tex $ write temp "\year=''year' \month=''month' \day=''day' \time=''time'" $ close temp biblio.tex can then use \today etc. and this will refer to the revision date of biblio.tex. Charles Karney Plasma Physics Laboratory Phone: +1 609 683 2607 Princeton University MFEnet: Karney@PPC.MFEnet PO Box 451 ARPAnet: Karney%PPC.MFEnet@NMFECC.ARPA Princeton, NJ 08544-0451 Bitnet: Karney%PPC.MFEnet@ANLVMS.Bitnet ------------------------------ Subject: Re: Bibliographies Date: Mon, 09 Nov 87 11:02:51 PST From: Jeffrey Goldberg There is a program called Tib on the Unix TeX tape which is very similar to Unix bib (an outgrowth of refer). Some conversion needs to be done but it is fairly minor. It has all the advanatges of the bib/refer bibliography system: It will find items based on partial information instead of by some predefined keyword. Thus it is by people who didn't invent the bibliography themselves and doesn't require the memorization of keys (or a system of keys). Tib is not fully compatible with bib and refer, and it is still quite buggy in some areas, but I highly recommend it. -jeff goldberg goldberg@russell.stanford.edu ------------------------------ Date: Mon, 09 Nov 87 14:54:18 EDT From: Dimitri Vulis Subject: METAFONT/Icelandic There have been several queries in the past concerning the rune letters Thorn and Edh, needed for Icelandic and Old English. I would like to point out that in addition to these 4 letters, an Icelandic text needs the ligature 'fj'. The only English word containing this ligature is 'fjord' and it looks UGLY in cmr10. On the other hand, in Icelandic this letter combination is very common---you do need this ligature. By the way, the lowercase Thorn comes with MF as test.mf I think; and the uppercase Edh can be composed of uppercase D and hyphen dash. Missing are only uppercase Thorn and lowercase Edh. cmchar "The ligature fj"; beginchar(oct"000",10u#+2letter_fit#,asc_height#,desc_depth#); % written by Dimitri Vulis % mostly based on cm program for fi from romligs italcorr asc_height#*slant-serif_fit#+.5if serifs:stem# else:dot_size# fi-2u#; adjust_fit(0,serif_fit#); pickup tiny.nib; pos1(stem',0); lft x1l=hround(2.5u-.5stem'); pos11(stem',0); rt x11r=hround(w-2.5u+.5stem'); pos12(stem',0); x11=x12; bot y12=-1/3d; pickup fine.nib; numeric bulb_diam, inner_jut; if serifs: bulb_diam=hround 1/4[.8[stem,flare],dot_size]; pos2(bulb_diam,0); x2r=x12r; y2+.5bulb_diam=.8[x_height,h+oo]; top y11=x_height+min(oo,serif_drop); if rt x1r+jut+.5u+2<=lft x11l-jut: inner_jut=jut; else: rt x1r+inner_jut+.5u+2=lft x11l-inner_jut; fi else: bulb_diam=max(stem,dot_size); pos2(6/7[vair,flare],90); top y2r=h; rt x2=hround 5.3u; pickup tiny.nib; pos13(bulb_diam,0); pos14(bulb_diam,90); x13=x11-.5; top y14r=min(2x_height,h+1); top y11=x_height; if bot y14l-x_height Subject: Greek Font Hi, Does there exist somewhere a Greek font (other than the math Greek) that includes breathing marks and the various accents? I guess I could construct something using the Math Greek, but I would rather not if I didn't have to. I would appreciate the metafont code if possible, or if not, a reasonable assortment of PXL files (we have a DEC LN03 here). Also, where can one find the metafont source to fonts other than the CMR variety (public domain or for purchace). We would like to increase the variety of fonts we have here, but I haven't heard of any possibilities. Thanks. Erik Kilgore System Programmer -- VAX/VMS Homewood Computing Facilities Johns Hopkins University BITNET: ECF_EEJK@JHUVMS ARPA: ECF_EEJK%JHUVMS.BITNET@FORSYTHE.STANFORD.EDU ------------------------------ Date: Mon, 9 Nov 87 08:15:33 CST From: "Linda J. Hutchison" Subject: Typesetting chemical structural formulas--a solution! I have located the information for typesetting chemical structural formulas! (Thank you Kevin O'Kane for your help in locating Roswitha Haas.) Ms. Haas did her Master's thesis on "Typesetting Chemical Structure Formulas with the Text Formatter TeX/LaTeX." I just received a copy (printed) in the mail today. If she can locate the macros, she'll be forwarding them, as well. She did her graduate work at the University of Tennessee, Knoxville but is now with the Oakridge National Laboratories. If she is able to forward the macros via BITNET, I'll make them available to TeXhax. She is anxious for someone else to benefit from her work. If we have to do the key-entry, don't expect anything very soon. Linda Hutchison Manager of Consulting and Publications 104 Computer Science Iowa State University Ames, IA 50011 (515) 294-1578 BITNET: GA.LJH@ISUMVS ------------------------------ Date: Mon, 9 Nov 87 18:57 EST From: Subject: not for texhax but for response from a caring human I have been trying to find someone willing to respond to my messages of inquiry about how to post Liz Barnhart's multicolumn macros at the listserver or where ever it should go. Will someone answer me? bob jantzen ------------------------------ Date: Tue, 10 Nov 87 00:00:47 EST From: dow@wjh12.harvard.edu (Dominik Wujastyk) Subject: Devanagari and Sinhalese Regarding the query: > Date: Mon 2 Nov 87 09:03:16-PST > From: Emma Pease > Subject: Sinhalese and Devanagari > > Has anyone done any work with new metafont (or TeX) for Sinhalese or > Devanagari? > > Emma Pease > > ps. I am aware of the work done at Stanford on creating Tamil characters > using metafont. Yes, Frans Velthuis at the Computing Center, Rijksuniversiteit Groningen Postbox 800, Landleven 1, Groningen, The Netherlands VELTHUIS@HGRRUG5.BITNET has just (last week) completed version 1.0 of a Metafont description of the Devanagari script for Hindi and (with some extra characters to come) Sanskrit. This is very exciting for all those of us who work in these languages. Frans's fonts will also be useable for Marathi and other North Indian languages that use Devanagari. The package consists of the fonts at several sizes, at a desired resolution, write-white or -black, and in .pk or .pxl form, as chosen. There is also a fast preprocessor program, DEVNAG, which reads an input file and does a bit of string processing on the sections of the text marked to be in Devanagari. DEVNAG can be compiled for use on any of the following systems: VAX/VMS, SUN, Cyber, IBM/PC, Atari ST. DEVNAG takes the strain off the MF ligtable commands, and makes for a reasonable length .tfm. Finally, there is a macro file to be \input at the start of a file, to give access to the fonts. I have contributed a style file, DEV.STY, for use with LaTeX, to make size switching automatic; Frans may or may not decide to use this. In use, one types the portions to be in Devanagari using a very simple and easy transliteration. DEVNAG, the macros and fonts do the rest. Frans has had to sacrifice his salary for a while to work on these fonts, and therefore has to sell them rather than give them away, as was his original intention. I do not know his charges. Obviously the more of us that buy them, the lower the price. It would be a fine thing if an appropriate organisation such as the American Oriental Society could buy the fonts from Frans for a proper sum, and make them available in the public domain. I have been using a beta version of the fonts for some time. I can vouch for the fact that they are superb. Anyone who is in reach of Harvard is welcome to come and see some examples. I urge anyone who works on Indian languages to leap at this opportunity. Get in touch with Frans at the addresses given above. Regarding Sinhalese, I am afraid I can offer no similar information. I too would be interested in news of any further Indian fonts. I didn't know about the Stanford work on Tamil fonts: who should I contact to learn more? Dominik Wujastyk, November 10, 1987. DOW@HARVUNXW.BITNET DOW@WJH12.HARVARD.EDU ------------------------------ Date: Tue, 10 Nov 87 09:36 GMT From: Peter Flynn Subject: Driver for AM-Varityper VT600 Postscript laser printer AM have announced their 600-dpi beast running PostScript. I have seen the output and it is *very* impressive. Only thing is, somebody needs to start cooking 600-dpi drivers. Anyone care to offer, or is there a 600-dpi Post- script driver already in existence? And if so, where (please don't say I can FTP it from a DARPA site --- BITNET sites can't)? AM Varityper Ltd are at Maylands Avenue, Hemel Hempstead, Herts HP2 7ET, UK Phone +44 442 42251. Standard disclaimer --- I have no connection at all with AM and opinions expressed are not necessarily those of the firm. All charac- ters are fictitious (shome mishtake shurely --- Ed.) Peter Flynn, Computer Bureau, University College, Cork, Ireland +353 21 276871 (CBTS8001@IRUCCVAX.BITNET) (cbts8001%iruccvax.bitnet@wisc.wiscvm.edu) (CBTS8001@VAX1.UCC.HEA.IRL) (cbts8001@irl.hea.ucc.vax1 [for janet users]) Telex 75583 uncc ei Fax +353 21 277194 (pflynn@bix) (p.flynn@EuroKom) ------------------------------ Date: Tue, 10 Nov 87 12:16:06 EST From: Rick Snodgrass Subject: TeX vs. LaTeX After struggling mightily for 6 months with the intricacies of TeX macros, I decided to go with LaTeX for a book I am writing. Converting the 500 page manuscript to LaTeX was tedious, but well worth it. Unless you have very strong feelings about how the output should look, AND are willing to invest a substantial amount of time convincing TeX to generate this output, I recommend that you give LaTeX a try. The software is solid and extremely well done, and the documentation is WONDERFUL. Rick Snodgrass ------------------------------ %%% %%% 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@TAMVM1.BITNET (list server) %%% %%%\bye %%% ------------------------------ End of TeXhax Digest ************************** -------