UKTeX Digest Monday, 11 Feb 1991 Volume 91 : Issue 6 Today's Topics: 1270dpi output Greek typesetting RE: Problem from TeXbook p.377 RE: 1270dpi output RE: Greek typesetting dvips 5.47, PC executable, in Aston Archive SB30TeX, TeX, INITeX, LaTeX, \@outputpage problem New smaller.sty CWeb What is the PSS number of tex.ac.uk Moderator: Peter Abbott (Aston University) Editor: David Osborne (University of Nottingham) Contributions: UKTeX@uk.ac.tex Administration, subscription and unsubscription requests: UKTeX-request@uk.ac.tex UKTeX back issues: stored in the Aston archive, in the directory DISK$TEX:[TEX-ARCHIVE.DIGESTS.UKTEX.91] TeXhax back issues:stored in the Aston archive, in the directory DISK$TEX:[TEX-ARCHIVE.DIGESTS.TEXHAX.91] Latest TeXhax: #06 TeXMaG back issues: stored in the Aston archive, in the directory DISK$TEX:[TEX-ARCHIVE.DIGESTS.TEX-MAG] Latest TeXMaG: V5 N1 ------------------------------------------------------------ Date: Sat, 02 Feb 91 00:23:00 +0000 From: CBTS8001%IE.UCC.IRUCCVAX@UK.AC.EARN-RELAY Subject: 1270dpi output A long while ago (was it the Exeter TeX Conference) I heard of a service offered by (? Imperial, ? UCL) to typeset DVI files at 1270dpi, taking the DVI file thru the network and snailmailing the output back. There was an academic discount on offer to other universities I think. As we don't have anything here which can take DVI (or even PostScript) is there anyone who can offer this service. My preference is for DVIs which referece PostScript fonts, as the user wants Times Roman (!) but will be using some math symbols from CM, and we don't have a DVIxPS here yet (altho I guess I could FTP it and install it---is there a compiled version for VMS at Aston?) ///Peter ------------------------------ Date: Sat, 02 Feb 91 02:21:00 +0000 From: CBTS8001%IE.UCC.IRUCCVAX@UK.AC.EARN-RELAY Subject: Greek typesetting A user here wants to do real Greek, and I know there are fonts or macros to do this (Silvio Levy?) but where are they? There seems to be no mention in the lists at Aston or Heidelberg. ///Peter ------------------------------ Date: Sun, 03 Feb 91 20:53:26 +0000 From: CA_ROWLEY@UK.AC.OPEN.ACS.VAX Subject: RE: Problem from TeXbook p.377 Chris (T) > > In the footnote on page 377 of the TeXbook (Appendix D), Knuth shows > a set of macros designed to test for various kinds of unusual space > token, and determine whether they are `funny' (character code not 32) > and whether they are `explicit' or `implicit'. > > A user here at Cambridge (Robert Hunt ) has > suggested to me that in the line > > \ifcat\noexpand#1\noexpand~\explicitfalse % active funny space > > that occurs in the macro \ssss, the \ifcat can never give the result > `true'. I think he is right, but am not at all sure that I have > considered all the possibilities. Can anyone design an argument for > \stest that will provoke this \ifcat to deliver `true'? I agree with Chris T and Robert H: thus we are saying that this part of the code does not do what Don implies it should (assuming that I am interpreting his characteristically cryptic comment). Moreover, this implies that, as in the following example, the whole of this code does not perform to spec: > (On the > understanding that altering \catcode`\~ would be cheating, of course.) > One might think that > > \catcode`|=\active \let|= \ftoken % the space after the = is necessary! > \newtoks\t \t={|}\stest\t > > would do the trick, but it doesn't (the \ifcat compares 10 and 13). > Here, I claim that | has been made into an implicit, funny space, but this code will claim that it is an explicit, funny space. In contrast, the code will claim, and I shall agree, that \ftoken is an implicit, funny space and that \stoken is an implicit, serious space. This leads to the question: can this code be fixed? I think that this can be accomplished only by altering the logic somewhat: it is necessary to first decide, using the method of carefully comparing #1 with \string#1 as in the next part of the current version of \ssss, whether the token is a control sequence; then, if it is not, we can then test its category code to decide whether it is implicit (ie catcode 13) or explicit (otherwise). This is accomplished by \atest#1, where \atest is the following macro: \def\atest#1{\ifnum \the\catcode`#1 = 13 \explicitfalse \else \explicitrue \fi} NB not tested: try and enjoy (assuming that anyone actually wants such information)! There is also a side-issue: whether \ifcat is working to specification. Although the explanation, on p 209, of the semantics of \if and \ifcat does not explicitly cover this case (of an active character \let equivalent to another character) I would put my money on Don saying that the observed behaviour is what he intended. Chris Rowley ------------------------------ Date: Tue, 05 Feb 91 11:37:49 -0000 From: P.TAYLOR@UK.AC.RHBNC.VAX Subject: RE: 1270dpi output Peter Flynn wrote: >>> A long while ago (was it the Exeter TeX Conference) I heard of a service >>> offered by (? Imperial, ? UCL) to typeset DVI files at 1270dpi, taking >>> the DVI file thru the network and snailmailing the output back. There >>> was an academic discount on offer to other universities I think. The service is offered by the University of London Computer Centre, on a Linotronic 300. Preliminary enquiries should be addressed to Bob Colvill (071-405 8400), and technical enquiries to Martin Powell (same number). Philip Taylor Royal Holloway and Bedford New College, ``The University of London at Windsor'' ------------------------------ Date: Tue, 05 Feb 91 11:44:03 -0000 From: P.TAYLOR@UK.AC.RHBNC.VAX Subject: RE: Greek typesetting Peter Flynn wrote: >>> A user here wants to do real Greek, and I know there are fonts or macros >>> to do this (Silvio Levy?) but where are they? There seems to be no mention >>> in the lists at Aston or Heidelberg. I have, and use, the fonts regularly. If the consensus is that they are not already in the archive, I'll put them there. But I suspect they are really there all along ... ** P. ------------------------------ Date: Tue, 05 Feb 91 15:28:10 +0000 From: S.P.Q.RAHTZ@UK.AC.SOUTHAMPTON.ECS Subject: dvips 5.47, PC executable, in Aston Archive the file in the UK TeX Archive [tex-archive.drivers.dvi2ps.dvips.pc]dvips.boo is a BOO-encoded MSDOS executable of dvips. The paths compiled in assume a TeX hierarchy under d:\emtex, so users will probably need to override them. This is *just the executable* - get the manual from the main distribution. Sebastian ------------------------------ Date: Tue, 05 Feb 91 16:29:45 +0000 From: PM1MJP@UK.AC.SHEFFIELD.PRIMEA Subject: SB30TeX, TeX, INITeX, LaTeX, \@outputpage problem Does anyone have an idea what is happening here? This is from a LOG file of a run of SB30TEX using \tracingcommands=5 \tracingmacros=5. TeX clearly knows what \@@ptpage means---in fact, it has been \let to LaTeX' \@outputpage. However, just as clearly, it doesn't have a clue what \@@ptpage means, since it ignores \begingroup, the \catcode change, \let, and then inserts a \discretionary or two in the vertical list! A clue---this only goes wrong when INITeX has read these macros, and the resulting format is used. When TeX \inputs them, they work fine. Mike Piff \@@ptpage ->\begingroup \catcode `\ =10 \let \-\@dischyph \let \'\@acci \let \`\@accii \let \=\@acciii \if@specialpage \global \@specialpagefals e \@nameuse {ps@\@specialstyle }\fi \if@twoside \ifodd \count \z@ \let \ @thehead \@oddhead \let \@thefoot \@oddfoot \let \@themargin \oddsidemar gin \else \let \@thehead \@evenhead \let \@thefoot \@evenfoot \let \@the margin \evensidemargin \fi \fi \shipout \vbox {\normalsize \baselineskip \z@ \lineskip \z@ \let \par \@@par \vskip \topmargin \moveright \@thema rgin \vbox {\setbox \@tempboxa \vbox to\headheight {\vfil \hbox to\textw idth {\let \label \@gobble \let \index \@gobble \@thehead }} \dp \@tempb oxa \z@ \box \@tempboxa \vskip \headsep \box \@outputbox \baselineskip \ footskip \hbox to\textwidth {\let \label \@gobble \let \index \@gobble \ @thefoot }}}\global \@colht \textheight \endgroup \stepcounter {page}\le t \firstmark \botmark \-->\discretionary {-}{}{} \@dischyph ->\discretionary {-}{}{} \'#1->{\accent 19 #1} #1<-\@acci \@acci #1->{\accent 19 #1} ! Argument of \@acci has an extra }. \par } \@@ptpage ...hyph \let \'\@acci \let \`\@accii \let \=\@accii... \@outputpage ...empty \@@ptpage \else \ifx \@Next \@End \gdef... \@opcol ...l \else \@outputpage \global \@colht \textheight \fi ...e \@makecol \@opcol \@floatplacement \@startcolum... \newpage ...\vfil \penalty -\@M \clearpage ->\newpage \write \m@ne {}\vbox {}\penalty -\@Mi \@ndpap ...art {#1}{}\clearpage \m@rkpart {}{} ... l.33 \endpaper ? - ----------------------------------------------------------------------------- From Dr M. J. Piff, Department of Pure Mathematics, PO Box 597, Hicks Noisy Building Site, Hounsfield Road, SHEFFIELD S10 2UN, England. Tel. SHEFFIELD(0742) 768555 Extension 4431. JANET MPiff@UK.AC.SHEF.PA or MPiff@UK.AC.SHEF.IBM - ----------------------------------------------------------------------------- ------------------------------ Date: Wed, 06 Feb 91 11:08:30 +0000 From: PLA@UK.AC.EDINBURGH.COMPUTER-SCIENCE.TARDIS Subject: New smaller.sty When acronyms or upper-case abbreviations are printed, it is good typographic style to set them in a smaller type-size to avoid undue emphasis (the output from WEB files is an example). Of course, it is possible to use {\small TEXT} in body text, {\scriptsize TEXT} in footnotes, etc. This falls over in section headings (even if you can remember the correct size command) when a table of contents is required (unless you really want a \Large word in there). One way round this is to use something like {\sc text}. Unfortunately, this only works for roman text, and the small caps font is often not available over a full range of sizes. A better solution is a relative size-changing command, and one appeared in TeXHaX(?) some time ago. Usage is {\smaller TEXT} and this works no matter what the current type size. Since it invokes LaTeX size-changing commands, it has the unfortunate side-effect of re-asserting roman type-face. It is possible to use {\smaller\em TEXT} in emphasized text, but this is tedious and inelegant. There is still a problem with section headings - {\smaller\bf TEXT} works fine in the heading itself, but again falls over in the table of contents because the \bf creeps in. So, I rehashed smaller.sty (sorry, I don't remember the author), and added the necessary commands to preserve the current typeface - shamelessly stolen from Adrian Clark's dates.sty. Of course, someone may already have done this and published it - if so, I apologize for the duplication. Adrian didn't know what family 3 was and used \rm; I dug out the fact that it's something to do with the math extension font, but I doubt that \tenex is the correct command to use there. The original smaller.sty had a corresponding \larger command, but I don't see much point in having one. If you need \larger, substitute `larger' for `smaller', `l@rger' for `sm@ller' and reverse the size-changing commands as follows: \ifx\@currsize\Huge \Huge\fi % the commands *must* be in descending order \ifx\@currsize\huge \Huge\fi \ifx\@currsize\LARGE \huge\fi ... Although not extensively tested, the only problem I've had is in a BiBTeX database where it went wrong in the title of an article (it works in the title of a book and in other entries for articles). This was with BiBTeX 0.99c (that's what comes with Toal's PD implementation for the Archimedes) and the unsrt style. A workaround is to use \begingroup\smaller TEXT\endgroup though I don't know if the problem is in BiBTeX or my macro. BTW, I had to type this in, so I may have introduced some `typogarphicval' errors. %--------------------8<----------------------------------------------------- % smaller.sty % Paul L. Allen, Feb 1990 % % Usage: {\smaller TEXT} % - sets TEXT in a smaller size in the same typeface. % \newcount\f@m \def\sm@ller{\f@m = \the\fam \ifx\@currsize\tiny \tiny\fi \ifx\@currsize\scriptsize \tiny\fi \ifx\@currsize\footnotesize \scriptsize\fi \ifx\@currsize\small \footnotesize\fi \ifx\@currsize\normalsize \small\fi \ifx\@currsize\large \normalsize\fi \ifx\@currsize\Large \large\fi \ifx\@currsize\LARGE \Large\fi \ifx\@currsize\huge \LARGE\fi \ifx\@currsize\Huge \huge\fi \ifcase\f@m\rm\or\mit\or\cal\or\tenex % not sure about the tenex \or\it\or\sl\or\bf\or\tt\fi} \def\smaller{\protect\sm@ller} %-------------------8<------------------------------------------------------- Paul Allen ------------------------------ Date: Thu, 07 Feb 91 15:04:18 +0000 From: Dave Clark Subject: CWeb I understand that Knuth has produced a C version of his literate programming system, Web. I'm unsure where I can get hold of the CWEB system for Unix (HP-UX in particular) - have you any ideas? Best Regards, Dave Clarke Corporate Engineering - Europe Hewlett-Packard Ltd Bristol. ------------------------------ Date: Thu, 07 Feb 91 15:13:10 +0000 From: PETER@UK.CO.MEMEX Subject: What is the PSS number of tex.ac.uk This new browsing service via PSS looks good, but as I am not part of JANET I can't turn the name Uk.Ac.TeX into whatever magic number I could give to PSS. (I don't know how PSS works, I just know that I can access it.) Is this browsing service available to me, as a uk.co site, and if so, what is the magic number? Thanks, Peter Ilieve peter@memex.co.uk ------------------------------ UK TeX ARCHIVE at ASTON UNIVERSITY *** JANET Interactive and NIFTP access *** Host: uk.ac.tex username: public password: public *** Files of interest *** [tex-archive]00readme.txt [tex-archive]00directory.list [tex-archive]00directory.size [tex-archive]00directory_dates.list [tex-archive]00last30days.files *** Media distributions *** Washington Unix tape (28 March 1990) TeX 2.993(==3.0), LaTeX 2.09, Metafont 1.9 (2.0) Unix 4.2/3BSD & System V. Tar 1600bpi, blockfactor 20, 1 file. Send one 2400' tape with return labels AND return postage. VMS backup of the archive requires two 2400' tapes at 6250bpi. VMS backup of TeX 2.991 plus PSprint requires one tape. Exabyte 8mm tapes: same formats available as 1/2in tapes. The following tapes are available: SONY Video 8 cassette P5 90MP, MAXCELL Video 8 cassette P5-90, TDK Video 8 cassette P5-90MPB OzTeX (for Macintosh): Send 10 UNFORMATTED 800K disks with return postage. emTeX (for MS-DOS): Send 11 UNFORMATTED 1.44Mb or 18 720K 3.5" disks, or 12 UNFORMATTED 5.25" disks, with return postage. *** Postage rates: (cheques made payable to Aston University) *** 0.5" tapes: UK: 2.50 pounds sterling (one tape), 5.00 (two tapes). Europe: 5.00 pounds sterling (one tape), 9.00 (two tapes). Outside Europe please enquire. 8mm tapes: UK: 1.00 pound sterling. Europe: 2.00. DC600A cartridges: UK: 1.00 pound sterling. Europe: 2.00. Diskettes: UK: 1.00 pounds sterling. Europe: 2.00. *** Postal address *** Peter Abbott, Computing Service, Aston University, Aston Triangle, Birmingham B4 7ET *** UK TeX Users Group *** For details, contact Malcolm Clark or Geeti Granger IRS John Wiley & Sons Polytechnic of Central London Baffins Lane 115 New Cavendish Street Chichester London W1M 8JS W Sussex PO19 1UD email: malcolmc@uk.ac.pcl.mole End of UKTeX Digest *******************