Subject: TeXhax Digest V89 #106 From: TeXhax Digest Errors-To: TeXhax-request@cs.washington.edu Maint-Path: TeXhax-request@cs.washington.edu To: TeXhax-Distribution-List:; Reply-To: TeXhax@cs.washington.edu TeXhax Digest Friday, December 1, 1989 Volume 89 : Issue 106 Moderators: Tiina Modisett and Pierre MacKay %%% The TeXhax digest is brought to you as a service of the TeX Users Group %%% %%% in cooperation with the UnixTeX distribution service at the %%% %%% University of Washington %%% Today's Topics: Printing Bibtex Database Re: Brit Scientific-TeX conversion? Passing vertical position to driver? TeX-to-ASCII "driver" Re: Bechtolsheim book, style file question, multi-line headers LaTeX cites as superscript numbers? RE: LaTeX cites as superscript numbers? LaTeX or TeX for technical documentation with version control Several questions regarding \special Problems with \uppercase{\ss\oe\aa} ------------------------------------------------------------------------------ Date: Mon, 20 Nov 89 02:45:43 -0800 FROM: Subject: Printing Bibtex Database Keywords: BibTeX, database >> From: guy@phy.duke.edu (Guy Metcalfe) >> Subject: Printing a BibTeX database I do this one very often using Unix environment. 1. Assuming all database files extensin .bib, RUN the following shell file on a Unix machine: cat *.bib | grep '@' <$*\ |sed 's/(/{/' | sed 's/\@[a-zA-Z]*/\\nocite/'|sed 's/,/\}/'>mytemp.tex 2. Prepare a small latex files containing a few lines of tex, % \documentstyle{article} \begin{document} This is a listing of .... database files with ..bst format. % \input{mytemp}, % and the following modifications \let\mybib=\bibitem \def\bibitem#1{\mybib[#1]{#1}} % together with \bibliography{ref1,ref2} % bib database files \bibliographystyle{plain} % or whatever you want. \end{document} You will get a listing of database entries in a compact form, formatted in the desired bst style, with their key values for further use. Mustafa Akgul Bilkent University Ankara Turkey akgul@trbilun.bitnet ------------------------------------------------------------------------------ Date: 20 November 1989 00:19:46 CST From: "ashish sen 996-2175 & 274-1947" Subject: Re: Brit Scientific-TeX conversion? Keywords: Brit Scientific, TeX, conversion I would like to ask if anyone knows about a wordprocessor called ~Brit Scientif ic'. Does anyone know of a program to convert documents written in it to TeX? ----------------------------------------------------------------------------- Date: Mon, 20 Nov 89 14:42 EST From: MRL@PFCVAX.PFC.MIT.EDU Subject: Passing vertical position to driver? Keywords: driver, TeX, tektronix mode I have a TeX driver which can accept a /special command to insert plot files. This command allows the user to specify where the plot is to be positioned on the page. Instead of specifying an absolute position, a user here would like to be able to pass the vertical position of the TeX page in order that the plot file will then be plotted at that point. I.e. the user is creating an empty area on the page, outputs a label, and then wants to insert the plot above the label in the empty area. However, he does not know where the label is going to appear. Presently, the only way around this is to obtain the output without inserting the plots, measuring with a ruler, and then rerunning TeX with the appropriate plot insertions. It would be much nicer if the position could be passed to the driver. Is this possible? I originally thought I could simply have the driver do that for me, i.e. the special command would be placed in the TeX file immediately after the label. Since the driver knows the position, it could then output the plot using this knowledge. However, the plots are in tektronix mode, and I have to switch in and out of tektronix mode. This causes anything which was outputted up to that point on the page to be erased, so it is not simple to have the driver do all the work. (The alternative is to either have the driver convert the tektronix code to the native printer mode, or for the driver to do a prescan to find the coordinates, but both of these operations are cpu instensive). Thanks. Mark ------------------------------------------------------------------------ Date: Sun, 19 Nov 89 20:47 EST From: Subject: TeX-to-ASCII "driver" Keywords: drivers,TeX, ASCII Does anyone have a TeX-to-ASCII driver that I can get (e-mail or FTP)? What I need to do is strip the TeX commands (incl. math defs) from some of my large TeX files so only pure words (as it were) are left? I have Turbo C and MSC 5.1 if only source is available. Also, is there any program out there that will take a TeX file and convert it into a Ventura file (ie converting the standard macros to Ventura commands)?? Thanks. --------------------------------------------------------------------------- Date: Sat, 18 Nov 89 20:45:10 EST From: INHB000 Subject: Re: Bechtolsheim book, style file question, multi-line headers Keywords: Bechtolsheim book, LaTeX, style file I have two questions and an answer. The first question is can anyone give me more details on Stephan v. Bechtolsheim's tome which according to a very nice book (if you are comfortable reading French) called ``Le petit livre de TeX'' has been published and is available for a little less than $60. The second is how do I go about adding a (LaTeX) style file to the archives? The answer is to Rich Gonzalez who wants to know how to do multi-line headers in LaTeX. The solution is that a header can include any \hbox, but you have to do your own heading definition. The following illustrates how. It could probably be simplified somewhat but you can modify it to your own needs. Of course, you can replace ``runninghead'' by such things as \sectionmark or \chaptermark, if that is what you want. Just read and modify one of the \ps@... definitions in some .sty file. \documentstyle{article} \textwidth 6.50in \oddsidemargin0pt \headheight 20pt \begin{document} \pagestyle{twolineheadings} \thispagestyle{twolineheadings} \makeatletter \def\ps@twolineheadings{%\let\@mkboth\markboth% \def\@oddfoot{}\def\@evenfoot{}\def\@evenhead{% {\hbox {\vbox{\hbox to \textwidth{\hfill runninghead}\kern 3pt \hbox to \textwidth{\hfill\rm\thepage}}}}} \let\@oddhead\@evenhead} \makeatother Just a sample document to illustrate the idea. \end{document} Michael Barr ----------------------------------------------------------------------- Date: Sat, 18 Nov 89 14:22 EST From: "YATES, JOHN H." Subject: LaTeX cites as superscript numbers? Keywords: LaTeX, citations, superscripts I would like to modify the results of LaTeX 's Bibliographic \cite{key} (UNSRT) to produce the reference number as a superscript, without the [] surrounding it. I have been trying to define a \mycite macro that could call \nocite and find the n and $^{n}$ it there, but haven't succeeded in finding how to retrieve the counter value (or parameter) to be used. LaTeX seems to obtain it from the .aux file \citation{key} and \bibcite{key}{n} entry pairs, where it is this n that I need in the superscript. I am hoping there is a simple fix, but not convinced. I don't want to rebuild LaTeX source, I will only go as far as modifying a dynamically read file (as a .STY), and hope a macro can do it. Please respond directly to me, I don't subscribe to this list. Thanks, John John H. Yates yates@a.chem.upenn.edu ------------------------------------------------------------------------------ Date: Mon, 20 Nov 89 20:20 EST From: "YATES, JOHN H." Subject: RE: LaTeX cites as superscript numbers? Keywords: LaTeX, citations, superscripts The closest solution I have found personally to my query about making LaTeX use superscripts for references is to define mycite as below to use in the running text (not, of course, tables, figures, etc.). \def\mycite#1{{\footnotesize{$^{^{\cite{#1}}}$}}} (doubly supered for the best position in appearance). I am still hoping someone can show me a trivial solution to getting rid of the brackets around this superscript. It appears that it is built in the LaTeX executable and perhaps inaccessible other than by tweaking the LaTeX source? But perhaps it can be tricked or backspaced over or... ? I currently have it in a usable form, but would like perfection if possible. Thanks, John yates@a.chem.upenn.edu P.S. Please respond directly to me, I don't read this newslist. I'm overextended with the few I do attempt to keep up with. P.P.S. As far as getting rid of the brackets in the References, that was trivial. I just modified ARTICLE.STY (in a new document style name, of course), removing the appropriate [ and ] and replacing the right one with a period (.) . ------------------------------------------------------------------------- Date: 21-NOV-1989 20:39:07.24 From: Ralph Becker-Szendy RALPH AT UHHEPG Subject: LaTeX or TeX for technical documentation with version control Keywords: LaTeX, TeX, documentation Dear TeX Gurus, We need some kind of package for technical documentation typesetting, with some kind of "version control" and "update service". Let me first describe what the end result shall be, and then you may suggest what we should do. We have to write and continuously update a technical manual for a large research project (well, large on our scale). The end result will be a folder with a few hundred pages; essentially all something like hardware and software manuals. There won't be heavy-duty mathematics (maybe a formula here and there), and drawings will be done by other programs (mostly engineering drawings and circuit diagrams); maybe the output of these drawings will be included in manual pages (with an intelligent .DVI file filter), but probably it will just be kept on seperate pages. High output quality is not required; since the nature of the manual is to be a reference work only, a 12-point mono-spaced font would already work; although high-quality output will be appreciated. The document is highly structured, with several parts, over 20 major chapters, and each chapters with sections, sub-sections, sub-sub-sections and so on. To make updating easier the structure could be reflected in page numbering. Now comes the p[roblem: A few dozen copies of this manual have to be updated continously, at many locations, literally across the globe. A reasonable schedule may be an update (involving less than 10% of the pages) every two weeks. All the places involved have VAXes running VMS, and everybody has TeX and LaTex, and some kind of laser printer (unfortunately, not all the same). For speed reasons, the updates should be done over networks. Since not all participants can be reached by "intelligent" networks (TCP/IP, DECNet) it is required that only plain-text files are to be transmitted (for example TeX source). The main problem: Updating the document should be made easy. Obviously printing 30 copies of a nearly 1000 page document every two weeks, just because a few dozen pages have changed, is out of the question. What we need is an intelligent updating service: At the beginning, and once in a while (like every year) a complete copy is printed. In between, only pages which changed are re-printed. If possible changes should be highlighted and marked, for example changes in a page marked with a side bar, or with a different font, and the change date at the bottom of the page. Furthermore, it should be "reasonably" easy to use; the updating will be done by everything from secretaries to "senior physicists", not just by TeXperts. But one should assume that anyone who works on the text can write a simple document in LaTeX (without hard things, like tables, and without special tricks). Experts for the "tricky" stuff (tricky being formulas, tables etc.) will be available, but should not be needed for just typing or changing the text. Obviously someone will have to be the "wizard", but that person can be found. Now. how to implement it ? That's where I would like to hear suggestions. They don't have to use TeX, if you have a better idea, and they may cost money (but we would prefer if they don't). Here is how I would do it: Simply store the "last completely printed" and the "last updated" copy of the whole document, and also keep a file with all page numbers, and where the pages begin/end. Someone edits the current copy. Then, determine the differences between the three versions (for example with the VMS DIFFERENCES command, or a nice Unix-style utility). Use the differences to control the typesetting, and usually only output the pages which have been changed, marking the changed as described above. Obviously, something like this CAN be done, and I could sit down for many weeks and just do it. I would probably read the difference file with some program (in whatever my favourite language of the week is, including TeX), and generate the input to the actual TeX or LaTeX pass from the three versions and the differences, and then run that through TeX/LaTeX. I would also write a pretty LaTeX style file, and a nice set of LaTeX macros for things which are used frequently. Obviously doable, but a lot of work, it can't be implemented quickly, and I don't feel like re-inventing the wheel if it can be avoided. Has someone done something like this ? Are there commercial products out there ? Is LaTeX or TeX the right way to do it, or might (for example) a desktop-publishing system and a bunch of PCs be a better alternative (every collaborator has some kind of PC standing in some corner). Suggestions are very welcome, and I will summarize to the net if anything noteworthy comes my way. Ralph Becker-Szendy University of Hawaii / High Energy Physics Group Watanabe Hall #203, 2505 Correa Road, Honolulu, HI 96822 (808)948-7391 HEPNet,SPAN: UHHEPB=::RALPH (UHHEPB = 24730) Internet: RALPH@UHHEPG.PHYS.HAWAII.EDU BITNET: RALPH@UHHEPG ----------------------------------------------------------------------------- Date: Tue, 21 Nov 89 02:03:57 EST From: Allan Adler Subject: Several questions regarding \special Keywords: TeX, \special I finally got a tex program to run with a \special in it. I enclose both the tex program and the specially included postscript file, which is named boxpath.ps . I have succeeded only to the extent of getting it to work in this case. I still don't know exactly what I did that was right. On the other hand, I have another pair of files which don't work. They are named bun.tex and topwedge.ps and are also enclosed. I would appreciate answers to the following questions: (1) Why does the first pair work and not the second ? (2) How could I look at the PostScript file produced by dvialw and realize what I have to keep in mind when I write specials ? I have been trying to read and debug that output and while I am making significant progress, I am still pretty far from understanding. I should probably also state explicitly that I am using an Apple Laserwriter and the dvi file is converted to PostScript by dvialw. Allan Adler ara@lom1.math.yale.edu P.S. It is unnecessary for those readers who know about such things to point out to me that second pair, if they were to work, would print out an incorrect statement about cuneiform. %====================================================================== %This is the texfile that worked. It is named "box.tex" This is a box:\bigskip \special{include boxpath.ps} \vskip 1 in But {\it why} is it a box ? \end %======================================================================== % This is the PostScript file boxpath.ps . %%BoundingBox: 0 0 612 792 %begin(plot) % width height box_path path /box_path { 0 1 index rlineto % relative rlineto to (0, height) 1 index 0 rlineto % relative rlineto to (width, 0) 0 exch neg rlineto % relative rlineto to (0, _height), eating height neg 0 rlineto % relative rlineto to (_width, 0), eating width closepath } def .072 .072 scale 20 setlinewidth 0 700 neg rmoveto 300 500 box_path stroke %copypage %restore %end(plot) %============================================================================== Now begins the second pair. %============================================================================== %This is the file bun.tex The Sumerians also wrote the number one in the following way: \special{include topwedge.ps} \vskip 1in They pronounced it \`A\v S. This must have been very confusing because their word for six is pronounced A\v S. \end %============================================================================== %This is the file topwedge.ps %! %%BoundingBox: 0 0 2000 2000 %%Title: /usr/local/bin/dvialw halmos %%CreationDate: Sat Nov 18 00:20:39 1989 %%Creator: ara and [TeX82 DVI Translator Version 2.09b for PostScript [Apple LaserWriter laser printer]] %%Pages: (atend) %%BugHistory: Incorporates Allan Hetzel's 31-Oct-85 DARPA LASER-LOVERS PS Version 23.0 X-on/X-off bug workaround %%EndComments %%EndProlog %begin(plot) % wedge_height top_head path /top_head { gsave line_width 0 exch rlineto dup 0 exch exch rlineto line_width neg 1.2 mul 0 exch rlineto dup neg dup head_slope mul 2 div exch 2 div exch rlineto line_width 1.2 mul 0 exch rlineto dup 2 div head_slope mul line_width neg 2 mul add dup head_slope div exch rlineto % line_width 2 mul head_angle tan div neg 1 index add 0 exch exch rlineto closepath eofill grestore } def % stroke_length wedge_height top_wedge path /top_wedge { gsave % remember current position top_head % make left head 2 div 0 exch exch rmoveto 0 line_width exch rmoveto 0 line_width 2 mul neg exch rlineto neg dup 0 exch rlineto % 0 line_width 2 mul neg exch rlineto % neg 0 exch rlineto closepath eofill grestore } def .072 .072 scale /line_width 10 def /head_slope 0.5 def 3000 3000 moveto 700 200 top_wedge %0 300 exch rmoveto %700 200 top_wedge showpage %end(plot) --------------------------------------------------------------------------- Date: 21 NOV 89 16:29:07 From: Z3000PA%AWITUW01.BITNET@UWAVM.ACS.WASHINGTON.EDU Subject: Problems with \uppercase{\ss\oe\aa} Keywords: TeX, \uppercase{\ss\oe\aa} Some time ago, I posted the following problem: > The uppercase equivalent of the german eszet (\ss) is a pair of two S. > \uppercase{Stra\ss e} should produce STRASSE, but it produces STRA\ss E > (i.e. it prints the lowercase eszet amidst all the capital letters). Since I did not receive any positive answer (and one negative one), I assume that, in TeX 2.x, there is no way to persuade \uppercase to perform the \ss-->SS translation. Question 1: Can this problem (or the related problem when inputting the eszet as it's ASCII code > 128 national character) be solved with TeX 3 ? How can it be solved? In the meantime, I was told that -- although \uppercase is not clever enough to translate \ss to SS -- the small caps and caps font (\sc in LaTeX) is very good in doing this job. Therefore, I can use \def\uppercase#1{{\sc #1}} to produce something that is a (good?) replacement for the \uppercase\ss problem. If the only application of \uppercase is in the LaTeX headings pagestyle, it works pretty well. In fact, I even prefer \sc to \uppercase for the section headings in the running heads, because they are more distinct from the running text and because they preserve the information about capitalisation (which improves legibility and even understandibility, especially for german texts, where `Der Gefangene floh' means something different from `Der gefangene Floh'). Question 2: Who has tried this approach? Or who knows another (perhaps better) way to bypass the \uppercase\ss problem? Question 3: What is good / bad / wrong with this method? I assume that a similar situation holds for the french and scandinavian special characters (\oe, \OE, \aa, \AA, etc.). Question 4: How do \uppercase on the one hand and \sc on the other hand succeed in translating \oe to \OE and so on? Hubert Partl, Vienna ----------------------------------------------------------------------- %%% Further information about the TeXhax Digest, the TeX %%% Users Group, and the latest software versions is available %%% in every tenth issue of the TeXhax Digest. %%% %%% Concerning subscriptions, address changes, unsubscribing: %%% %%% BITNET: send a one-line mail message to LISTSERV@xxx %%% SUBSCRIBE TEX-L % to subscribe %%% or UNSUBSCRIBE TEX-L %%% %%% Internet: send a similar one line mail message to %%% TeXhax-request@cs.washington.edu %%% JANET users may choose to use %%% texhax-request@uk.ac.nsf %%% All submissions to: TeXhax@cs.washington.edu %%% %%% Back issues available for FTPing as: %%% machine: directory: filename: %%% JUNE.CS.WASHINGTON.EDU TeXhax/TeXhaxyy.nn %%% yy = last two digits of current year %%% nn = issue number %%% %%%\bye %%% End of TeXhax Digest ************************** -------