TeXhax Digest Friday, 27 Nov 1992 Volume 92 : Issue 021 % The TeXhax Digest is brought to you as a service of the TeX Users Group % % and UK TeX Users Group in cooperation with the UK TeX Archive group % Today's Topics: Japanese & Chinese LaTeX and PostScript harvard bib styles in TeX Re: magnification in LaTeX Re: 4 second delay on startup \Bbb Keeping LaTeX's ideas of recto/verso in line with reality Re: Summary on Polish (La)TeX Lollipop v0.93 bibclean 2.05 released BIBCLEAN 2.05 on FILESERV/Niord Administrivia: Moderators: David Osborne and Peter Abbott Contributions: TeXhax@tex.ac.uk Administration, subscription and unsubscription requests: TeXhax-request@tex.ac.uk ------------------------------------------------------------ Date: Sat, 14 Nov 92 09:08:43 +0700 From: Wu Yihong Subject: Japanese & Chinese LaTeX and PostScript Does any one has the following softwares for useing on IBM PC ATs? 1. A software for viewing Postscript files on a screen; 2. Latex in Japanese; 3. Latex in Chinese; Thanks in Advance. Your help will be highly appreciated. Wu Yihong Department of Electrical Engineering National University of Singapore 10 Kent Ridge Crescent Singapore 0511 Tel 65-772-6658 Fax 65-779-1103 Email ELEWUYH@NUSVM.BITNET ------------------------------ Date: Sat, 14 Nov 92 10:28:39 +0000 From: dquah (Danny Quah) Subject: harvard bib styles in TeX I have been using Oren Patashnik and Karl Berry's btxmac.tex to produce `plain' references in my TeX'ed papers. Does anyone have a version of btxmac.tex that they use with the family of harvard .bst styles? Thanks. - --Danny (dquah@lse.ac.uk) St Clements 479: 071-955-7535 Economics Department, LSE Fax: 071-831-1840 Houghton Street, London WC2A 2AE, United Kingdom ------------------------------ Date: Sun, 15 Nov 92 09:00:37 -0500 From: Timothy Van Zandt Subject: Re: magnification in LaTeX Wei X. Zheng asks: ``The LaTeX Manual says, "The \magnification command in Plain TeX has no counterpart in LaTeX. Magnification of the output can often be done by the program that prints the dvi file." If anyone knows what this means and how to do magnification in LaTeX, please kindly let me know.'' Plain TeX's \magnification command sets both the magnification and adjusts the page parameters. Since Plain TeX and LaTeX use different page parameters, you can't use the \magnification command with LaTeX. To change the magnificaton, set the TeX primitive \mag to an integer equal to 1000 times the magnification factor, or to \magstep, where n=1,2,3,4,5 for a magnification of 1.2^n, or to \magstephalf, for a magnification of 1.2^.5. E.g., \mag 1200 \mag \magstep2 \mag \magstephalf Now you have to adjust your page parameters. The simplest option is to just reset all the page parameters using "true" dimensions, as in \textwidth 5.5truein \setlength{\textwidth}{5.5true} The \textwidth then ends up at 5.5in *after* magnification. Tim ------------------------------ Date: Tue, 24 Nov 92 22:49:28 -0700 From: jerry@math.ep.utexas.edu (Jerry) Subject: Re: 4 second delay on startup Lars Henrik Mathiesen on October 12 wrote: - -> When I had this problem, the reason turned out to be that I had set - -> TEXINPUTS_SUBDIR to be my equivalent of /usr/local/lib/tex/inputs. TeX - -> then tries to identify all subdirectories in that directory, which - -> entails a stat() system call for each file. In my case, TeX saw the - -> directory via NFS, and I think the large number of stat()s effectively - -> flushed the local attribute cache --- in any case, I got a 6 second - -> delay on start-up, even when running TeXs back-to-back. - -> - -> Moral: Don't let a SUBDIR path element be a directory with lots of - -> regular files. I undefined TEXINPUTS_SUBDIR and the problem went away. Sorry about the long delay in responding to those who asked for a followup. My TeX now compiles .tex files at the rate of 1/3 to 1/2 second per .dvi page output. Thanks a bunch! - -Jerry Jerry Graves UTEP Math Sysadmin jerry@math.ep.utexas.edu PHONE/FAX (915) 747-6757/6502 ------------------------------ Date: Wed, 25 Nov 92 11:49:19 -0100 From: Gregory Ovenden <076GMO@witsvma.wits.ac.za> Subject: \Bbb This is probably a stupid question, but I am a novice in both the Tex and FTP/email worlds, so please bear with me. We have recently ftp-ed some papers in tex from yale. Our version of Emtex( about 12 months old) with Ams styles) does not understand a control sequence '\Bbb' which keeps on coming up in these papers. What is this sequence and where can I pick up the appropriate macro/style file? Thanks ********************************************************************* * GREGORY MARK OVENDEN * DEPARTMENT OF COMPUTATIONAL AND * * TEL: 716-3969 (O) * APPLIED MATHEMATICS * * 648-2754 (H) * UNIVERSITY OF THE WITWATERSRAND * * EMAIL: * 2050 PO WITS * * 076GMO @ WITSVMA.WITS.AC.ZA * SOUTH AFRICA * ********************************************************************* ------------------------------ Date: Wed, 25 Nov 92 15:55:52 +0000 From: David_Rhead@vme.nott.ac.uk Subject: Keeping LaTeX's ideas of recto/verso in line with reality I've had a certain amount of success in getting LaTeX-ed documents to come out duplex on a PostScript laserprinter that is a capable of duplex. Thanks to recipes provided by Peter Ilieve and Martin Powell in UKTeX earlier this year, I've now got a .sty file that goes \def\ds@twoside{\@twosidetrue \@mparswitchtrue \special{dvitops: inline statusdict /setduplexmode known { statusdict begin true setduplexmode end } if } } This seems to be all that is needed in straightforward cases. (Does it need updating in terms of setpagedevice?) However, I think that there are circumstances in which LaTeX's idea of "whether we're on a recto or a verso" can get out of synch with the printer's ideas. For example, if someone does a \setcounter{page}{...}, LaTeX will continue to base its ideas of recto/verso (and what to do when there's a \cleardoublepage) on whether the page-counter is odd/even. Consequently, the physical result may be even-numbered rectos and odd-numbered versos. Does anyone happen to know a magic bit of PostScript that could be put into a \special within the definition of \cleardoublepage to say " ... and if, after all that (which, as far as LaTeX is concerned, has just finished off a verso), the printer has physically just finished off a recto, please put a blank physical verso in"? If this bit of magic was put into the definition of \cleardoublepage, it would at least help to limit the damage if things get out of synch. Assuming that \chapter does a \cleardoublepage, the LaTeX idea of recto/verso should be brought into line with what the printer is physically doing at the start of each chapter: oddities within a chapter would not propagate to the following chapter. I.e., is there "a PostScript equivalent of \cleardoublepage" that could be put into the definition of \cleardoublepage to ensure that, whatever happens between \cleardoublepage-s, things come right again whenever there is a \cleardoublepage? David Rhead d.rhead@uk.ac.nott.vme ------------------------------ Date: Wed, 25 Nov 92 18:39:58 +0700 From: "Janusz S. Bie/n" Subject: Re: Summary on Polish (La)TeX DC fonts, Polish language, LISTSERV@PLEARN etc. - ----------------------------------------------- I would like to comment on J%org Knappen's "Summary on Polish (La)TeX" posted to INFO-TEX and TEX-EURO on 20 November. I post this comments both to INFO-TEX and TEX-EURP and I hope they will reach both of them. My first answer to Knappen's query bounced from INFO-TEX, so just in case I send a copy to INFO-TEX owner with the kind request to forward it to INFO-TEX if necessary. >The ogonek does not yet have a standard input notation, the proposed standard >control sequence is \k{a} (from the last letter of ogonek, or from >skandinavian kvist). First, this is a general problem of many letters present in DC fonts (to be more precise, in Cork layout), and ogonek is not an exception. Secondly, I never heard about the proposal to represent ogonek by \k; in all (English language) international standards I know ogonek is called *ogonek*, so deriving its standard TeX representation from Skandinavian "kvist" would be a silly idea. My guess is that such a representation is actually used somewhere in Scandinavia, but with no intention to made it general. Anyway, I would like to stress the need for developing standard representations for all new (i.e. those which are either absent in CM or created with accents missing in CM) characters in Cork layout. For ogonek accent I propose \og. I invite comments on this ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ suggestion and proposals for other letterns and accents (thorn etc.). > Styles > plfonts.sty by L. Holenderski > polish.sty v1.0b by E. Schl"uck and M Janich, both using "-Konventions > polish.sty (anonymous) using \, > plhb.sty (which I have not yet) > polish.sty v1.0b seems to be the most complete one (including quotes > etc.). Uses box constructions for the ogoneked letters. ... > wzorce.tex and plhb.sty used to be available from listserv@plearn.bitnet, > but I could not locate them there recently. On LISTSERV@PLEARN.BITNET I maintain, among others, INF FILELIST with some Polish TeX stuff. To get the full information, send GET INF FILELIST to LISTSERV@PLEARN.BITNET. To get PLHB STY, send GET PLHB STY INF to the same address. Please note that to fit VM/CMS syntax, PLHB.STY must be referred to as PLHB STY (space instead of point). PLHB.STY is one of the several variants of the style developed at our faculty with crucial contribution from Leszek Holenderski aimed at constructing Polish letters from CM fonts and using prefixing with " for input representation of Polish letters. LISTSERV@PLEARN.BITNET should be considered the home for PLHB.STY (there is also an FTP site, but with such a slow connection that we want to use it only locally). As far as I know, the original version of Holenderski, differing slightly in input convention, was distributed by him only in private e-mail - I guess plfonts.sty is just a version received directly from him. E. Schl"uck and M Janich derived their polish.sty from PLHB.STY and german.sty. The version I got from them (I don't remember its number) had a feature which I considered a serious error - they insisted on using a German trick for allowing hyphenation after the accented letter. However, Polish words - contrary to German ones - very often end with accented letters, and in such situation the trick blocks the completely legal line break at the *interword space*! I don't know whether version 1.0b has this feature or not. > Hyphenation > polish.hyphen, available from Stuttgart > wzorce.tex by Hana Kolodzieska extended version of the above, handling > prefixes correctly. Hanna Ko{\l}odziejska's patterns for Polish are her original work, so they are not an extension of anything from Stuttgart. They happened to be distributed under different names, so my guess is that polish.hyphen in Stuttgart is just a copy of wzorce.tex. Anyway, the home of the original (dated 1987) Ko{\l}odziejska's pattern is now LISTSERV@PLEARN, they are available by sending the command GET HYPHEN POL INF. > There is a construction of the ogoneked letters using the standard cm-fonts, > available from tex.ac.uk under > [tex-archive.digests.tex88.polish]polset.tex Somewhere in Aston archive there should be an early version of PLHB.STY submitted by Malcolm Clark on my behalf (cf. TeXline 8 p.2) but I never learned it exact location. { I've just searched the UK TeX Archive and can't find PHLB.STY --Ed.} As for polset.tex, my guess is that it was also put by Malcolm Clark and that it contains the macros needed to typeset Polish letters in the conference proceedings edited by Clark - cf. Preface to "TeX Applications, Uses, Methods". It's the first time I learned the location of that macros and I will have a look at them in some future. > The pl-fonts (extended cm-fonts) are sold by Makrosoft, the adress is: >The MacroSoft address is: >ul. (street) Chroscickiego 49 >02-414 Warsaw >ph.: 48-0-22-237681 >fax 239861 >tlx 825334 >I afraid they have no e-mail. The PL fonts (plr10 etc.) has been placed by MacroSoft in public domain, but they are bundled with other stuff into so called MeX, distributed by GUST, the Polish TeX users group, which happens to be supported by MacroSoft. Let me repeat the fragments of my previous posting, which did not make its way to INFO-TEX: - ------------------------ Date: Tue, 10 Nov 92 09:39:13 CET ... Subject: Re: polish (La)TeX X-To: TEX-EURO@DHDURZ1.BITNET, info-tex@shsu.BITNET X-cc: gust@alfa.camk.edu.pl ... >Date: Wed, 4 Nov 92 18:30:55 CET >From: "J%org Knappen" ... >Therefore I'm looking for ... >--polish hyphenation patterns (I found some from 1987 in Stuttgart. Are > these the ,,official ones''?) There is a more recent version bundled with MeX (see below); I think that hyphenation patterns should be available independently, so I don't consider them "official". >--dcfonts 1.1 (Stuttgart only has dcfonts 1.0) and/or pl-fonts I was not aware of dcfonts 1.1, I would like to have them too; in 1.0 Polish ogonek was rather ugly. As for Polish fonts, some of them are bundled with MeX (see below); as they conform only partially to Cork standard, I don't consider them "official". >--information about polish TeX (in english, german, french or esperanto; >I'm not able to read polish) ... MeX distribution contains some informations in English; both MeX and its predecessors (LeX and LaLeX) were presented by their authors on European TeX conferences. - --------------------------- Let us come back to the recent Knappen's posting: > Other: A font, containing polish letters as ligatures. (no further > information available) The LeX/LaLeX fonts used /a, /c etc. ligatures for Polish letters, they are superceded by MeX fonts mentioned above (which no longer use such ligatures). >There seems to be not a standard short reference, but the following > schemes are used: > > (1) (2) (3) (4) > \k a @a "a \,a (1) extended standard TeX notation > \'c @c "c \,c (2) polish ligatures > \k e @e "e \,e (3) plfonts.sty; polish.sty, based on german.sty > \l @l "l \,l (4) polish.sty > \'n @n "n \,n > \'o @o "o \'o > \'s @s "s \'s > \'z @z "z \'z > \.z @r "x \.z The (2) convention originated at our faculty in the days of TeX implementation for RIAD computers but was later abandoned; I never heard about Polish fonts with such ligatures. I am also not familiar with the (4) convention. That's the first time I learned about the (1) convention, I wouldn't call it "extended standard TeX notation" - who is the author of this extended standard? The (3) notation used by Holenderski differs from the convention of PLHB.STY in a different way of representing \.z - PLHB.STY requires it to be represented as "r. I would like to stress that the input convention is independed of the way Polish letters are actually created - you can easily modify PLHB.STY to any convention you wish, including direct 8-bit input according e.g. to the 852 IBM PC code page. > Prospects: > I probably will install polish with dcfonts, and automatically convert the > wzorce.tex into the ^^-notation for dc-fonts. Using the DC fonts is in my opinion the right decision. I hope that the authors of the PL fonts will unbundle the relevant parts of Metafont source from the rest of MeX stuff and hence make it easy to improve the shape of ogonek (and perhaps also other Polish letters) in the universal DC fonts. Janusz S. Bie\'n Institute of Informatics, Warsaw University ------------------------------ Date: Tue, 17 Nov 92 15:30:22 -0500 From: eijkhout%edu.utk.cs@uk.ac.nsfnet-relay Subject: Lollipop v0.93 Update announcement for Lollipop TeX. I've just put a .tar.Z file and a (Macintosh) .hqx file of Lollipop v0.93 on cs.utk.edu in /pub/eijkhout/tex. This will be the last official update until the end of the year. This update incorporates most of the comments of users so far. There may be occasional minor updates before the end of the year, but these will upward compatible, and therefore go unannounced. Some global information about Lollipop TeX follows below the signature. Victor Eijkhout Department of Computer Science; University of Tennessee at Knoxville 104 Ayres Hall; 1403 Circle Dr.; Knoxville TN 37996-1301 phone: +1 615 974 8298 (secretary 8295; fax 8296); home +1 615 558 3069 Support the League for Programming Freedom! lpf@uunet.uu.net What is Lollipop? The Lollipop format tries to bridge the gap between the fact that in TeX anything is programmable, and the fact that doing so is too hard for the people such as typographers, who know actually what to program. Or even for seasoned programmers. Lollipop is a macro package that gives the user powerful tools for programming macros for lists, headings, output routines and a number of other things. The Lollipop macros take the specifications for a macro, and then construct that macro. The status of Lollipop Lollipop is not finished, but it is already quite powerful. (It has been used to typeset my book 'TeX by Topic', for instance.) I suggest that anyone who is interested format the manual and have a look at the examples in it. They document the current power of Lollipop (really: all examples, including of output routines, are formatted on the fly). Since Lollipop is still under development I will be very accessible for questions, bug reports and suggestions. Email preferred: eijkhout@cs.utk.edu. (And I will be flattered if you actually use Lollipop, but beware that changes may occur in future versions.) Lollipop can be ftp'ed from the directory /pub/eijkhout/tex on cs.utk.edu (log in as 'anonymous') where it is stored as 'lollipop.tar.Z'. Uncompress and 'tar -fx' it. The tar file unpacks itself to a subdirectory ./Lollipop-xyz where 'xyz' is the current version. Like a lot of TeXware, Lollipop is free. Make and give away all the copies that you want. Just don't ask money for it, other than reasonable expenses for copying discs or the manual. Share and enjoy! ------------------------------ Date: Tue, 24 Nov 92 10:47:05 -0700 From: Nelson H. F. Beebe"Nelson H. F. Beebe" Subject: bibclean 2.05 released bibclean 2.05 has today been released into the public domain use. This new version replaces the previous versions 1.05 and 1.06 which will remain available for some time. You can use Internet anonymous ftp to ftp.math.utah.edu and fetch in binary mode any of these files in pub/tex/bib: - -rw-r--r-- 1 beebe staff 689588 Nov 24 10:24 bibclean-2.05.tar.z - -rw-r--r-- 1 beebe staff 512514 Nov 24 10:24 bibclean-2.05.zip - -rw-r--r-- 1 beebe staff 708075 Nov 24 10:24 bibclean-2.05.zoo For each of these, there are corresponding files with suffix -lst that show the contents: - -rw-r--r-- 1 beebe staff 5501 Nov 24 10:24 bibclean-2.05.tar-lst - -rw-r--r-- 1 beebe staff 5840 Nov 24 10:24 bibclean-2.05.zip-lst - -rw-r--r-- 1 beebe staff 4884 Nov 24 10:25 bibclean-2.05.zoo-lst If you lack anonymous ftp capability, then send an e-mail message with the text "send bibclean-2.05.zip from tex/bib" to tuglib@math.utah.edu (change the file name to whichever format you prefer); the file will be returned in a large number of smaller uuencoded pieces. The previous distributions of bibclean were about 10% of the size of the 2.05 version, so here is why. Bibclean 2.05 is a considerable extension of the previous 1.x versions. The code has grown from a total of 3814 lines in bibclean 1.05 to 13372 lines in 2.05. Much of this growth is in additional files that are provided for enhanced portability. bibclean.c itself grew from 1395 lines to 5632 lines, and that file incorporates an extensive revision history log. Besides prettyprinting of BibTeX bibliography files, the new version now provides conversion of Scribe bibliography files to BibTeX form, and syntax checking of citation tags and string values. These enhancements have proved valuable, in that new errors have been uncovered in almost every bibliography that has been processed with bibclean 2.05. For testing the Scribe conversion, I've used a 36K-line 1800+ entry collection from the Computer Science Department at the University of Arizona. The syntax checking is based on patterns provided in initialization files, allowing extensions and modifications to be made WITHOUT changing the source code of bibclean. System-wide, project-wide, directory-wide, and file-specific customizations are supported. Because bibclean has many new run-time options, besides UNIX manual pages (with PostScript, text, and help forms thereof for other systems), the executable program incorporates an interactive help facility that allows the help display to be scrolled up and down. The help text displayed this way is extracted automatically from the manual pages, ensuring that it remains up-to-date when the documentation is modified in the future. The new distribution incorporates extensive tests of the correctness of bibclean. Testing has been carried out on at least these systems, representing 30 C and C++ compilers and three major operating systems, including the 6 most popular workstation architectures: ============================================================================= System O/S Compiler(s) ============================================================================= DEC VAX VMS 5.4 cc DECstation ULTRIX 4.2 cc, lcc, gcc, g++ HP 370 4.3 BSD cc, CC HP 720 HP-UX c89 HP 850 HP-UX c89 IBM 3090 AIX cc, CC IBM PC DOS 4.0 and 5.0 tcc (2.0, 3.0), tcc++ (3.0), cl (5.1, 6.0) IBM PS/2 AIX cc, CC IBM RS/6000 AIX 3.2 c89 MIPS 6280 RISCos 4.52 and 5.2 cc NeXT Mach 3.0 gcc Silicon Graphics IRIX 4.0 cc, lcc, CC Stardent 1520 OS 2.2 cc Sun 4 SunOS 4.1.1, 4.1.2, 4.1.3 acc, cc, gcc, g++, CC ============================================================================= The conversion to C++ code goes only as far as the provision of Standard C function headers which are based on the C++ syntax. Old-style (K&R) function headers are also provided; both styles are under C preprocessor control. There is no use whatever of C++ classes, I/O, or comment syntax. Testing under C++ has been so valuable in weeding out errors at compile time that I now do all my C software development using C++ compilers (yes, my DVI 3.0 project supports both C and C++, and no, it is not yet ready for release). For user convenience, the bibclean distribution contains executables in binary and uuencoded form for IBM PC DOS and VAX VMS, since C compilers are extra cost options on those systems. The top-level README file contains detailed installation instructions; here is a short summary: * For the UNIX systems above, all that is required is to select an appropriate definition of CC in the Makefile, set installation directory locations, then type "make all", "make test", and "make install". * For VAX VMS, DCL command files are provided for building and testing. * For IBM PC DOS, there are batch command files and makefiles. * For both VAX VMS and IBM PC DOC, minor customization of the command files and makefiles is needed to set appropriate directory paths the local installation. Enjoy the new bibclean, and as always, please report problems, errors, and comments back to the author. Nelson H. F. Beebe Tel: +1 801 581 5254 Center for Scientific Computing FAX: +1 801 581 4148 Department of Mathematics, 105 JWB Internet: beebe@math.utah.edu University of Utah Salt Lake City, UT 84112, USA ------------------------------ Date: Wed, 25 Nov 92 09:59:06 -0600 From: "George D. Greenwade" Subject: BIBCLEAN 2.05 on FILESERV/Niord This is a longish post regarding Nelson Beebe's new version (2.05) of his bibclean. Attached are his announcement, followed by the relevant instructions for retrieval of this package from FILESERV and Niord. Nelson, this works GREAT on VMS! Thanks!! - --George BIBCLEAN -------- The BIBCLEAN package includes the a 18 part UUENCODEd ZIP archive of Nelson Beebe's bibclean (version 2.05, 24 November 1992).... To retrieve complete package of 18 files, include: SENDME BIBCLEAN in the body of a mail message to FILESERV@SHSU.BITNET (FILESERV@SHSU.edu). The ZIP archive file is available for anonymous ftp retrieval from Niord.SHSU.edu (192.92.115.8) in [FILESERV.BIBCLEAN]BIBCLEAN-2_05.ZIP. Files in this package: (1 Block = 512 bytes) File Blocks - ----------------------------------------------------------------------------- BIBCLEAN.UUE_01OF18 thorugh BIBCLEAN.UUE_17OF18 79 (each) BIBCLEAN.UUE_18OF18 56 Approximate total blocks in full BIBCLEAN package = 1399 ------------------------------ 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. Please send contributions to: TeXhax@tex.ac.uk Administration, subscription and unsubscription requests: On Internet: send a one line mail message to TeXhax-request@tex.ac.uk SUBSCRIBE TEX-L UNSUBSCRIBE TEX-L On BITNET: send a similar one-line mail message to LISTSERV@xxx On JANET: send a similar one line mail message to TeXhax-request@uk.ac.tex Back issues of the digest are available for anonymous ftp from the UK TeX Archive, tex.ac.uk (134.151.40.18) in [tex-archive.digests.texhax.YY]texhax.NN where YY = last two digits of year, NN = issue number \bye End of TeXhax Digest [Volume 92 Issue 21] *****************************************