X-VM-v5-Data: ([nil nil nil nil nil nil nil t nil] ["2892" "Mon" "8" "March" "93" "13:33:33" "+0100" "rolf.lindgren@USIT.UIO.NO" "rolf.lindgren@USIT.UIO.NO" nil "59" "Re: Fonts and LaTeX 3" nil nil nil "3"]) Return-Path: Received: from sc.ZIB-Berlin.DE (mailserv) by dagobert.ZIB-Berlin.DE (4.1/SMI-4.0/1.9.92 ) id AA13157; Mon, 8 Mar 93 13:36:30 +0100 Received: from vm.urz.Uni-Heidelberg.de (vm.hd-net.uni-heidelberg.de) by sc.ZIB-Berlin.DE (4.1/SMI-4.0-sc/19.6.92) id AA03725; Mon, 8 Mar 93 13:36:11 +0100 Message-Id: <9303081236.AA03725@sc.zib-berlin.dbp.de> Received: from DHDURZ1 by vm.urz.Uni-Heidelberg.de (IBM VM SMTP V2R2) with BSMTP id 3472; Mon, 08 Mar 93 13:36:19 CET Received: from DHDURZ1 by DHDURZ1 (Mailer R2.08 R208004) with BSMTP id 5100; Mon, 08 Mar 93 13:36:07 CET Received: from DHDURZ1 by DHDURZ1 (Mailer R2.08 R208004) with BSMTP id 5098; Mon, 08 Mar 93 13:36:03 CET Reply-To: Mailing list for the LaTeX3 project In-Reply-To: <"alf.uib.no.033:08.02.93.11.29.40"@uib.no> Date: Mon, 8 Mar 93 13:33:33 +0100 From: rolf.lindgren@USIT.UIO.NO Sender: Mailing list for the LaTeX3 project To: Multiple Recipients of Subject: Re: Fonts and LaTeX 3 Status: R X-Status: X-Keywords: X-UID: 1001 spqr@MINSTER.YORK.AC.UK writes: > > A solution to some of this is to write a program, say, xfm2sty, that takes > > a font metric file (or a bunch of metric files of the same font family) > > and writes a macro that contains commands to manipulate fonts. Say, Adobe > > Garamond gives rise to a macro \AdobeGaramond that switches to Adobe > > Garamond Regular at \family{garamond}\selectfont\rm, SemiBold at \bf, etc. > i think a "program" is overkill. a Makefile, perhaps? its just a > question of tacking the appropriate tools together, isn't it? I don't think it is, because: a) Most TeX users can't run Makefiles (because they run emtex on DOS-machines). b) Most TeX users can't use PS fonts (because they use HP LaserJets or dot matrix printers). So I want a program that can operate on HP softfonts (of which I truly know nothing: I assume that there exists some metric format for them) as well as PS fonts. Something that runs DC encoding isn't necessarily good enough: Sometimes I want to use ligatures from expert fonts and have to use my own encoding because there isn't room in the DC encoding for, say, fj and ffj ligatures. On the other hand, only six of Adobes' fonts come with expert fonts, I won't die if I prepare them by hand. But if I had a program that could read the tfm files (and/or the vf files) it could create the appropriate encoding command as well because it would know where I had used nonstandard positions. On the other hand, I know that too... perhaps it's altogether fitting and proper that advanced font hacking should require advanced thinking and little automization. Now, to approximate proper leading for a PS font, you have to look in the afm file for the line where the x height is given. A Makefile can find this by calling some variant of grep (1) I suppose. But, say, a style file that sets up things for Futura, of which there are 20 variants, should know to separate Futura from Futura-Condensed. How can even a complex Makefile read all the 20 Futura afm's, look into them, and decide whether there should be one style file for Futura and one for Futura-Condensed, or a style file for both, and set up the style file with commands to allow swithcing between the 20 Futuras? (and create style files with 8-character names!) Now if there were such a thing as a standard repository of style files for _all_ the font families of the Adobe PS library (1265 fonts for ca. 400 families at last count). Then these ~400 style files would constitute a de facto standard for font handling, and standards is a good thing to have. Good standards is even better to have :-) But I envision that creating these style files might reqire more care and thought than what a Makefile can provide. > > sebastian > > > [1] tools available from > ftp.tex.ac.uk:pub/archive/fonts/postscript/dcps, but don't use the > font names there >