TeXhax Digest Sunday, February 10, 1991 Volume 91 : Issue 007 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: My mistake in making/modifying mf mode_def's, and more Asking TeX newequation.sty fonts, 600dpi Arabic or Persian (Farsi) Fonts for Tex/Latex Looking for Textset Inc.... found ArborText Inc. Dvi driver guidelines wanted RE: Naming of non-CM Fonts Raising PCTeX's memory limits Trans/Fig patches Re: TeXhax Digest V91 #005 Request: Question/Answers Style File for LaTeX \count 1 used for LaTeX section counter Font naming conventions --------------------------------------------------------------------------- Date: Thu, 31 Jan 91 22:37:53 EST From: ROCKY@IBM.COM Subject: My mistake in making/modifying mf mode_def's, and more Keywords: mode_def In TeXhax Volume 91 #001, a couple of minor corrections that should be made. First, in the mode IBMD that was given, the blacker value should have been .3 (not 0) and the fillin value should have been .4 (not .3). For the typewriter fonts, and slanted/italic fonts (cmtt's cmsl's and cmit's) a blacker of 0 does however make the M's and W's more legible. There is a concern that the weight of this font does not match the others however. The second correction is that because of preloaded formats and whatnot, it is probably not safe to assume that if fontmaking (or any other parameter) is unknown that it was set by the user as a MetaFont argument. The corrected mode_def appears at the end of this note along with another that I have found useful for most IBM 4216 printers with the Ricoh print engine. (I am told that IBM originally used a Canon print engine and then at some point switched to the Ricoh engine. I have also been told that Ricoh has on occasion change the print engines that they give IBM, for example in order to cut their costs.) Speaking of mode_def files, I want to mention another modification that I have found very useful. In the University of Washington's Unix TeX distribution, the file U_Wash.mf has some useful code written by Pierre MacKay: def font_mode_specials = p_p_i=pixels_per_inch/mag; if fontmaking > 0: begingroup; save d,s,p; save pixels_per_inch; string p; pixels_per_inch:=p_p_i; special jobname; mode_special(mag); if string mode: p:=mode; else: p:=substring(0,length(mode_name[mode])-1) of mode_name[mode]; fi special "mode:="&p&";"; mode_special(pixels_per_inch); mode_special(blacker); mode_special(fillin); mode_special(o_correction); endgroup; fi enddef; The above puts the mode_def name, blacker, fillin o_correction, and pixles_per_inch into the gf file output (and thus into any resulting pk file). This information however does not appear in the pxl format. It would be nice if repositories of gf and pk fonts ensure that such information is in pk and gf files. Font files get copied about and there is no widespread convention to put the mode information into say the path name. (Although around here, we do put the exact mode name into the font path, e.g. ...tex/fonts/RicohA/cmr10.300pk). Another place where this information in the pk file is helpful is in figuring out how to regenerate the font. For example if one decides to use a different blacker or fillin. One can do the arithmetic (assuming you know the pixels_per_inch value), but due to rounding, you can still only get approximate values for the magnification. Finally, for people who have asked (such as someone on comp.text.tex) how to decide on the blacker, fillin, and o_correction values. I pass my observations. For very high resolution devices, such as Autologic's 1016 dot-per-inch APS6 phototypesetter, the values for blacker, fillin, and o_correction don't matter all that much, so long as the values are within their normal ranges, between 0-1 for all values, with the values in the limit for and infinite-resolution device being 0, 0, and 1 respectively. For low resolution devices, the procedure I use is to set the blacker and o_correction to 0 and decide on a fillin value by looking at the diagonal of a lowercase `z' in the typewriter font. The diagonal should be the same thickness as the horizontal bars of the `z'. Then I decide on the blacker value. I don't have a good heuristic for this, I generally look at the smaller fonts to make sure that there isn't too much filling in. Finally, I then set the o_correction. R. Bernstein IBM Research % Corrected mode_def for IBM's 38xx printers % IBM3825's 27's and 35's have some sort of corner imaging. % % Thanks to Jim Hafner (hafner@ibm.com) for experimenting with blacker. % mode_def IBMD = % IBM mode for the IBM38xx's ``advanced function'' pixels_per_inch:=240; % printers if unknown fontmaking_set: % Unless told otherwise, fontmaking_set:= fontmaking:=1; % output TFM fi if unknown tracingtitles_set: % Unless told otherwise, tracingtitles_set:= tracingtitles:=0; % don't show titles in the log fi if unknown proofing_set: % Unless told otherwise proofing_set:= proofing:=0; % we're not making proofs fi if unknown blacker_set: % Set this value unless it has been set previously blacker_set:= blacker:=.3; % On cmtt8 a's are already starting to fill in, fi % so no more blacker. if unknown fillin_set: fillin_set:= fillin:=.4; % This value can be checked on cmtt10's k's and z's-- fi % check the diagonal/horizontal ratio on these. if unknown o_correction_set: o_correction_set:= o_correction:=.75; % I don't believe even a value of 1 is that bad. fi enddef; % IBMD % Ricoh setup for some of the IBM 4216's (and the TI Omnilaser?) % For the TI Omnilaser, see Stan Osborne's values TUGboat 8,2,134. % The mode seems to be a better one to use for with xdvi and % IBM's megapel displays than the CanonCX (or imagen) mode. mode_def RicohA = let font_setup=white_setup; % use the write-white font_setup pixels_per_inch:=300; % Ricoh write-white engine. if unknown fontmaking_set: % Unless told otherwise, fontmaking_set:= fontmaking:=1; % output TFM fi if unknown tracingtitles_set: % Unless told otherwise, tracingtitles_set:= tracingtitles:=0; % don't show titles in the log fi if unknown proofing_set: % Unless told otherwise proofing_set:= proofing:=0; % we're not making proofs fi if unknown blacker_set: % Set this value unless it has been set previously blacker_set:= blacker:=.2; % Don't overblack the small fonts such % such as cmtt8's e's fi % so no more blacker. if unknown fillin_set: fillin_set:= fillin:=0.0; % No compensation for diagonal fillin % this value can be verified by looking at % cmtt's z's and checking (without any % blacker which just obfuscates) and checking % that the diagonal is the same as the % horizontal. One might even put in some % reverse or negative compensation fi if unknown o_correction_set: o_correction_set:= o_correction:=.75; % Overshoot for O's, G's and A's % The usual 300 dpi value of .6 is probably % okay too. fi enddef; % RicohA -------------------------------------------------------------------------- Date: Thu, 31 Jan 91 12:13:02 PST From: JSLee Subject: Asking TeX Keywords: TeX, Is there a way I can ask it how many characters have been processed? Suppose input.tex is a 7000 character file, and I want to stop input after the first 5000. ---------------------------------------------------------------------------- Date: Wed, 30 Jan 91 11:23:07 EST From: clayton@thumper.bellcore.com (R. Clayton) Subject: newequation.sty Keywords: newequation.sty This option doesn't follow the latex convention of defining with one command and redefining with another; \newequation will redefine an existing equation environment. % newequation.sty % % 12 Jan 91 Created. % clayton@thumper.bellcore.com % % newequation.sty defines the \newequation command, which defines % new equation environments. The command format is % % \newequation{name}{format} % % where "name" is the name of the new equation environment and "format" % is the format of the equation number. For example, % % \newequation{define}{(D\arabic{define})} % % The new environment follows Latex's numbered sections convention: % there's a counter called "name" and a command called \the"name" with % "format" as its initial definition. See section 5.3 in the Latex % user's guide for more on numbering. \def \newequation#1#2{ \@definecounter{#1} \@namedef{the#1}{\hbox{#2}} \@namedef{#1}{$$\refstepcounter{#1}} \@namedef{end#1}{ \eqno \csname the#1\endcsname $$\global\@ignoretrue } } -------------------------------------------------------------------------- Date: Tue, 29 Jan 91 16:18:04 GMT From: Carlos Lourenco Subject: fonts, 600dpi Keywords: fonts, 600dpi I am trying to get some pages printed in rather good quality so that they could be photographed and the photos be used to generate a lot of copies. I am using a Laser printer LN03+ connected to a VAX and the fonts I am using (I believe) are 300dpi. I have access to a 600dpi Laser printer connected to an IBM compatible. I thought of two solutions (far from being a TeXnician): 1) Use \magnification=\magstep1 or \magstep2 and then reduce back to the normal size at the photographical stage. I have tryed it but it doesn't work since I don't have all the required "magnified" fonts. 2) Use the 600dpi laser printer. To do that I would need to have 600dpi fonts and a "dvito600dpi_laser_printer". Are there such things? Where? I guess its very likely that I am missing some important point. Anyway, before I give up hope, is there any body out there that can help (or teach) me? Thanks. Carlos e-mail address: CARLOS@UXLIPL.LIP.RCCN.PT ----------------------------------------------------------------------------- Date: Tue, 29 Jan 91 10:53:11 MEZ From: Hooshang Mehrjerdian Subject: Arabic or Persian (Farsi) Fonts for Tex/Latex Keywords: fonts, Arabic. Farsi Hello there. I am working on the implementation of an English-Persian (Farsi) Translator for scientific texts and looking for Persian/Arabic fonts, to be able to print my doctoral thesis with Tex. Would you please let me know if there is any Arabic/Persian fonts available for Tex or Latex? Any other information related to this subject will be more appreciated. Thanks in advance, Hooshang Mehrjerdian Roeckumstr. 132 D-5300 Bonn 1, Germany ------------------------------------------------------------------------- Date: Mon, 28 Jan 91 12:56:54 EST From: aed%arbortext@arbortext.com (Andrew Dobrowolski) Subject: Looking for Textset Inc.... found ArborText Inc. Keywords: Textset, ArborText Some time ago Textset Inc. changed its name to ArborText Inc. Our address and other particulars follow. Write 535 W. William St. Suite 300, Ann Arbor, MI 48103 U.S.A or phone (313) 996-3566 or fax (313) 996-3573 For product support you may also send email to help@arbortext.com on internet. Andrew Dobrowolski, ArborTeXt Inc. ------------------------------------------------------------------------------- Date: Mon, 28 Jan 91 20:40:49 MET From: Bo Thide' Subject: Dvi driver guidelines wanted Keywords: dviware, guidelines I am about to improve our generic HP LaserJet dvi driver to full PCL4 (and possibly PCL5) support. As far as I understand there exists some kind of committee who has produced guidelines on how to standardize dvi drivers. Since I am pretty sure our present driver is not compliant with this standard I would appreciate any documents, templates etc to help me not making the driver incompatible with others. Our present driver can handle PCL pictures, preloaded fonts, and very large fonts and seems to do a decent job on HP LaserJet+, LaserJet II, and LaserJet IIP printers. Bo ^ Bo Thide'-------------------------------------------------------------- |I| Swedish Institute of Space Physics, S-755 91 Uppsala, Sweden |R| Phone: (+46) 18-303671. Telex: 76036 (IRFUPP S). Fax: (+46) 18-403100 /|F|\ INTERNET: bt@irfu.se UUCP: ...!mcvax!sunic!irfu!bt ~~U~~ -----------------------------------------------------------------sm5dfw ---------------------------------------------------------------------------- Date: 6 Feb 91 17:11:00 PDT From: "JohnM" Subject: RE: Naming of non-CM Fonts Keywords: fonts, naming What Teresa Ehling suggests is very good. We have been collecting non-CM fonts, including converted HP Softfonts (via HP2TeX), and have been wrestling with some of the same issues as Teresa raises. Just to add another wrinkle, how would you handle the {\em same} font from different sources, i.e. Courier 10 point (COM10)? Without growing the fontname past 6 characters, it could be tough. I propose the use of two additional [optional] characters indicating the source of the font (so COM10 from LTI would become COM10LT, COM10 from HP would be named COM10HP, and so forth). Moderating the list of vendors would, of course, be a nightmare, but at least my Courier BoldOblique 10 point font (COBO10JM) could be distinguished from R J Drofnats' Courier BoldOblique 10 point (COBO10RJ). And, of course, those users who cannot handle 8 character fontnames would still have the problem. The format for font filenames would then become: 2-3 char 0-2 char 1-2 char 0-2 char John Mulhollen TRW Defense Systems Group johnm@afsc-bmo.af.mil ------------------------------------------------------------------------- Date: Wed, 6 Feb 91 13:22:46 EST From: Doug Ravenel Subject: Raising PCTeX's memory limits Keywords: PCTeX, multiletter control sequences How can I raise PCTeX's limit of 2500 multiletter control sequences to 3000 or more? Douglas C. Ravenel drav@troi.cc.rochester.edu Department of Mathematics University of Rochester -------------------------------------------------------------------------- Date: Wed, 6 Feb 91 09:44:41 -0500 From: beck@cs.cornell.edu (Micah Beck) Subject: Trans/Fig patches Keywords: Trans/Fig patches Patchlevel 2 of the Fig 2.1.1 Sunview graphics editor and patchlevel 1 of TransFig 2.1.1 are available for FTP from svax.cs.cornell.edu, or by mail archive server from sun.soe.clarkson.edu. They are in ~ftp/pub/transfig/fig-2.1.1-pl1 fig-2.1.1-pl2 transfig-2.1.1-pl1 Micah Beck Dept. of Computer Sci. Cornell Univ. ------------------------------------------------------------------------------ Date: Mon, 04 Feb 91 09:48:04 PLT From: Dean Guenther Subject: Re: TeXhax Digest V91 #005 Keywords: TeX, hyphenation > Is there a difference in the hyphenation rules for TeX 3.0? > > Thomas Petsche > petsche@leaning.siemens.com I had the same problem, and either Joachim Lammarsch or Peter Breitenlohner (I can't remember who gets the credit) pointed out that I needed to change plain.tex too. Add \lefthyphenmin=2 \righthyphenmin=3 % disallow x- or -xx breaks Just prior to: \input hyphen and that ought to take care of the problem. -- Dean ------------------------------------------------------------------------------ Date: Tue, 05 Feb 91 11:55:43 EST From: Tom Prusa Subject: Request: Question/Answers Style File for LaTeX Keywords: LaTeX, style file, Q & A Hello, Has anyone written a LaTeX style file for writing problems sets. For example, a file that gives a LaTeX version of Knuth's exercise macros (from TeXbook) or that plain TeX example in the latest TUGboat? A desirable file might allow automatic question numbering and allow answers to be handled like endnotes. thanks, tom -------------------------------------------------------------------------- Date: Tue, 5 Feb 91 05:24:50 PST From: Donald_Arseneau@mtsg.ubc.ca Subject: \count 1 used for LaTeX section counter Keywords: LaTeX, \count 1 Dave Steiner asked why the following didn't work: \def\section{\expandafter\count1\expandafter\c@section\oldsection} The use of \expandafter is incorrect (there would have to be one between \count and 1 to get \oldsection expanded first), but the \expandafter's are completely superfluous: they only cause \oldsection to be expanded, they do not force the counters to be incremented (assignments are not "expandable"). Therefore, I don't see how this can work at all--the number in \count1 will always lag 1 behind the true section number. The simple way to achieve the same effect is to REASSIGN \c@section to use counter 1: \countdef\c@section=1 That's it. Unfortunately, when a \section command appears near the bottom of a page, the counter will be incremented before the page break is detected, even if the section heading is carried over to the next page. Thus you will get pages like [9.3] when section 3 really starts at the top of page 10. The only way I can see to make everything right is to use "marks" (which LaTeX already inserts) to set \count1 in a modified output routine...but it is far beyond me to start tinkering with LaTeX's output routine. Donald Arseneau Arseneau@mtsg.ubc.ca asnd@triumfcl (.bitnet) ----------------------------------------------------------------------------- Date: Mon, 4 Feb 91 14:03:38 -0500 From: spooky!witr@uunet.UU.NET Subject: Font naming conventions Keywords: fonts, naming, conventions In TeXhax Digest V91 #005 "Teresa A. Ehling" submits a ``RECOMMENDATION FOR THE NAMING of NON-CM FONTS USED IN TeX''. This scheme is primarily based on the notion that >A sequence of six or fewer ordinary letters and/or digits >followed by a space should be a file name that works in >essentially the same way on all installations of TeX...." While this is certainly one way to attack the problem, I feel that it is a poorly engineered one, in that it is forever tied to the technology of n-character file names (with n equal to six or eight). I think a far better solution would be to use the font naming scheme developed by the X-Windows software (or something like it) and to correct the TeX software to work with font names of this type. This has at least four important advantages, namely: 1) The font names are un-ambiguous and complete, 2) the solution is technology-independent, as it does not rely on any particular length of file-names, and 3) It provides a path for the convergence of the screen and paper representation of fonts, and 4) it makes for easier font matching and substitution. For those who may not be familiar with the X-Windows representation of fontnames, here is an example: ``-Adobe-Times Roman-Bold-I-Normal--14-300-300-300-P-231-ISO8859-1'' Which means, Foundry=Adobe, Family=Times Roman, Weight=Bold, Slant=Italic, Width=Normal (instead of Condensed, for example), Style= (could be Serif or the like), Pointsize=14Pt, Pixelsize=300dpi (although I would prefer an SI measure), Horizontalresolution = 300dpi, Verticalresolution=300dpi, Spacing=proportional, Averagewidth=23.1 pixels, Charset encoding registry=ISO8859, charset encoding=1. This name is then converted, via a dictionary, to the appropriate filename. Fields are allowed to contain the characters ``*'' and ``?'' which are wildcards. An example would be: ``-*-Times-Bold-I-*--14-*-*-*-*-*-ISO8859-1'' Which would be asking for any 14pt Times bold italic font with the proper characterset encoding. This usage is not encouraged because, of course, you might be surpriesd by the results. Robert Withrow, R.W. Withrow Associates, Swampscott MA 01907 USA Tel: +1 617 598 4480, Fax: +1 617 598 4430, Uucp: witr@rwwa.COM ----------------------------------------------------------------------- %%% 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.nnn %%% yy = last two digits of current year %%% nnn = issue number %%% %%%\bye %%% End of TeXhax Digest ************************** -------