TeXhax@cs.washington.edu (TeXhax Digest) (06/04/90)
TeXhax Digest Sunday, June 3, 1990 Volume 90 : Issue 48
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:
Problem with BibTeX for VMS
Re: Problem with BibTeX under VMS
Macintosh LaTeX - using PostScript fonts
TeX macro
TeX and Metafont
Re: TeXhax Digest V90 #43
TeX files corruption over networks.
Printing a portion of a .dvi or .ps file through a laser printer.
Re: Printing a portion of a .dvi or .ps file through a laser printer.
-------------------------------------------------------------------------
Date: Wed, 9 May 90 14:11 MET
From: "Johannes L. Braams" <JL_Braams%pttrnl.nl@UWAVM.ACS.WASHINGTON.EDU>
Subject: Problem with BibTeX for VMS
Keywords: BibTeX, VMS
Hi all,
We are experiencing a problem with the VMS implementation of BibTeX
by Northlake software.
My diagnosis sofar is that the way filenames are handled in this
implementation causes a problem when the logical name TeX_Inputs
is defined as a search list with e.g.
$ define tex_inputs [mydir],texinputs,texsamples
In this case bibtex searches only the first member of the search
list ([mydir]) for its style file. Ofcourse it can not find the
file there. BibTeX then complains that it cann't find the file
instead of looking in TeXinputs (which is a search list as well...)
If anybody knows a fix for this, please contact me. I have located
where filenames are handled in BibTeX.ch and I suspect the routine
add_defaults, which joggles with FAB-fields like FAB$L_DNA, where
it puts defaults to be used by RMS when an incomplete filespec is
given. It might be that this causes the problem, the original
Bibtex just puts TeXinputs infront of the filename
Thanx in advance,
Johannes Braams
-------------------------------------------------------------------------------
Date: Fri, 11 May 90 15:15 PDT
From: DHOSEK@HMCVAX.CLAREMONT.EDU
Subject: Re: Problem with BibTeX under VMS
Keywords: BibTeX, VMS
[Johannes described a problem with the NLS BibTeX]
The public domain BibTeX change file for VMS (available from
ymir.claremont.edu in [anonymous.tex.sources.bibtex0_99] handles
search paths correctly. You may want to use that version instead
(it is also defined through the VMS CLI).
-dh
-------------------------------------------------------------------------------
Date: Tue, 08 May 90 09:27:21 +0100
From: J.Pearce@Cs.Ucl.AC.UK
Subject: Macintosh LaTeX - using PostScript fonts
Keywords: Macintosh, Latex, PostScript
Could anyone please tell me if it is possible to use PostScript fonts
in LaTeX on the Macintosh (TeXtures/OzTex) ?
John R. Pearce.
jpearce@uk.ac.ucl.cs
Computer Science Department
University College London.
----------------------------------------------------------------------------
Date: Mon, 07 May 90 18:02:33 EDT
From: INHB000 <INHB%MUSICB.MCGILL.CA@UWAVM.ACS.WASHINGTON.EDU>
Subject: TeX macro
Keywords: macro, TeX
Here is a solution to the question of John de Pillis, who writes:
> I would like to design a macro
> which would be called as follows:
>
> \DoWork{abc}
>
> The result within the macro is:
> 1. Counter \abc is created
> 2. A new macro, \xabc , is created.
>
> After calling macro \DoWork{abc}, I would like
> to call the induced macro \xabc
>
> Can you tell me how dto do this? I am having
> trouble with the all the backslashes among other things.
In answer to your TeXhax question, here is a solution and example:
\catcode`\@=11
\def\inewcount{\alloc@0\count\countdef\insc@unt}
\catcode`\@=12
\def\DoWork#1{\expandafter\inewcount\csname#1\endcsname
\expandafter\def \csname x#1\endcsname##1{\number\csname
#1\endcsname##1}}
\DoWork{abc} \abc=17
\xabc4
\bye
What's going on here? Forget the first three lines for the moment. The
line: \expandafter\inewcount\csname#1\endcsname means first create a
new name called \#1 and then apply the procedure \inewcount to it.
Since \inewcount is the same (almost) as \newcount, this defines that
new counter. (Actually, the backslash is NOT part of the new name as
you easily discover if you do something like \catcode`/<=0, in which
case <abc will do whatever \abc would). The \expandafter tells TeX to
expand the next word (and its parameters, if any) before expanding the
word that follows. Without that you would define \csname as a
newcounter. The next string \def \csname x#1\endcsname##1 tells TeX to
define a new macro called \x#1 with one parameter and finally its
definition \number\csname #1\endcsname##1 follows. The #1 refers to the
parameter to DoWork, while the ##1 (which expands to #1) refers to the
eventual parameter to \x#1. So that the result of \xabc4 is to produce
\number\abc4 and \number\abc expands to 17, so the result is 174.
The only other thing to explain is \inewcount. In Plain TeX \newcount
is defined as an \outer macro, meaning it cannot be used inside another
definition. The reasons are not given, but presumably involve the fact
that the number of counters that TeX has is severely limited (to 256)
and a macro might be used repeatedly. The definition of \inewcount is
exactly the same as that of \newcount, except it is not made \outer.
The same limitation is built in to all the allocation macros, such as
\newdimen, \newskip,... and also \newif.
Michael Barr
inhb@mcgillb.bitnet
inhb@musicb.mcgill.ca
----------------------------------------------------------------------------
Date: Thu, 10 May 90 13:06:52 EDT
From: ericg@sne42e.orl.mmc.com (eric grajales)
Subject: TeX and Metafont
Keyword: TeX, Metafont
I would like to subscribe to your TeX and Metafont installation and
maintanance group, and recieve the various notices of new versions/bug
fixes for Tex and MF.
I downloaded TeX from UUNET and have compiled it with no problem. I
also downloaded "SeeTeX", a set of utilites to preview *.dvi files.
However I cannot get MF to produce fonts *.85gf for the "xtex"
previewer on our Sun SparcStation I. My makefile to produce the
desired fonts looks like
***********************
.SUFFIXES: .snf .bdf .pk .gf .mf
goals: snf-fonts
echo Done
#
# NAME=cmb10
# SCALE = 1300
# EXTENSION = 159
#
cmb10.159pk:
-mf '\mode:=sun;mag=1300/1000;input cmb10'
-gftopk cmb10.159gf && rm -f cmb10.159gf
cmb10.159.snf: cmb10.159pk
-mftobdf -dpi 85 cmb10.159pk
bdftosnf < cmb10.159.bdf > cmb10.159.snf
#
...
*******************************
but the MF program produces "*.3382gf" files. My "mf" script file is
********************************
case "$0" in
*/mf|mf ) virmf $*; exit;;
*/logomf|logomf ) virmf '&logoplain' $*; exit;;
*/cmmf|cmmf ) virmf '&cmplain' $*; exit;;
*) echo "don't know how to be $0" 1>&2; exit 1;;
esac
********************************
as described in the TeX documentation. If anyone has successfully generated
the required fonts I would greatly appreciate any help in this matter.
Eric J. Grajales
Senior Engine Martin Marietta
-------------------------------------------------------------------------------
Date: Tue, 8 May 90 14:51 MET
From: "Johannes L. Braams" <JL_Braams%pttrnl.nl@UWAVM.ACS.WASHINGTON.EDU>
Subject: Re: TeXhax Digest V90 #43
Keywords: LaTeX, oldstyle
Alex Ho asks about oldstyle digits in LaTeX, here's a solution:
\def\old{\protect\pold}
\ifcase\@ptsize
\def\pold{\fam\@ne\tenmi}
\or
\def\pold{\fam\@ne\elvmi}
\or
\def\pold{\fam\@ne\twlmi}
\fi
Put the above lines in a style-file and load it after one of
xxx10.sty, xxx11.sty or xxx12.sty has been loaded.
You will than be able to say something like $\old 1234$ and get
oldstyle digits.
In his answer to the question by Dale Gerdeman, Michael Barr says he
will work out someday what needs to be done to be able to define a
new font in such a way, that commands like \large have their usual
effect. In the file sfb.sty I submitted a couple of weeks ago this
has been implemented.
Regards,
Johannes Braams
-----------------------------------------------------------------------------
Date: Tue, 08 May 90 08:45:24 IST
From: "Jacques J. Goldberg" <PHR00JG%TECHNION@UWAVM.ACS.WASHINGTON.EDU>
Subject: TeX files corruption over networks.
Keywords: TeX, corruption, files, network
I failed to note the name of the author of a remark, that TeX files using
8 moments codes might get corrupted over networks due to incomplete and/or
incompatible character conversion tables.
This indeed already is a problem with 7 moments codes and is going to
become worse.
For readers who do not understand the problem, and will thus sooner or later
waste much time finding out why a transmission failed, I append below the
EBCDIC/ASCII and ASCII/EBCDIC translation tables which happen to be plugged
at this time in VM-KERMIT at my site, one of many tools widely used for
file transfers. Of course such tables exist everywhere where an XYZ
mainframe is interfaced to anything else (XYZ is the encrypted name of a
large equipment manufacturer whose OWN small and large machines use
incompatible codes).
Fortunately, as always, there is a very simple solution. The TeX book says,
on page 43, that you may write \char98 u\char98\char98 le instead of
the word bubble . This trick is useful if the b key is broken on your
keyboard of course, but it would just the same avoid corruption if that symbol
had any special meaning for networking purposes.
Ridiculous, you say ? Nope, just try to pass a text file over mailers with
a line containing only a . ( period) in column one (and it's often enough
that the line begins with a . whatever comes behind). The leading period
simply identifies a command line in BSMTP network control.
Writing a pair of utility programs (in C please, so that they can be ported)
that would FILTER outbound and inbound TEX files, so that such characters
that might be error prone, be transmitted as \char??? sequences, is a trivial
job. HOWEVER setting up the list of characters to be filtered/restored is
not. This Forum might be the place to set this list NOW, before it is too
late, so that EVERYBODY uses the same one version of matching programs,
one to filter before mailing, the other to restore after receiving.
As a first step towards iteration I would suggest to send as \char sequences
any ASCII code below 32 and above 127 plus the . (period) {} ( curly )
[] ( square ) ^ ( caret ) and ~ ( tilde ) characters which I know not to be
universally mapped or to create network problems.
Here are the two demo tables: the input character with value n is translated
into the value sitting in location n in the table.
tdump atoe
00010203 372D2E2F 1605250B 0C0D0E0F
10111213 3C3D3226 18193F27 1C1D1E1F
405A7F7B 5B6C507D 4D5D5C4E 6B604B61
F0F1F2F3 F4F5F6F7 F8F97A5E 4C7E6E6F
7CC1C2C3 C4C5C6C7 C8C9D1D2 D3D4D5D6
D7D8D9E2 E3E4E5E6 E7E8E9AD E0BD5F6D
79818283 84858687 88899192 93949596
979899A2 A3A4A5A6 A7A8A9C0 4FD0A107
00010203 372D2E2F 1605250B 0C0D0E0F
10111213 3C3D3226 18193F27 1C1D1E1F
405A7F7B 5B6C507D 4D5D5C4E 6B604B61
F0F1F2F3 F4F5F6F7 F8F97A5E 4C7E6E6F
7CC1C2C3 C4C5C6C7 C8C9D1D2 D3D4D5D6
D7D8D9E2 E3E4E5E6 E7E8E9AD E0BD5F6D
79818283 84858687 88899192 93949596
979899A2 A3A4A5A6 A7A8A9C0 4FD0A107
tdump etoa
00010203 0009007F 0000000B 0C0D0E0F
10111213 00000800 18190000 1C1D1E1F
00000000 000A171B 00000000 00050607
00001600 00000004 00000000 1415001A
20000000 00000000 00005C2E 3C282B7C
26000000 00000000 00002124 2A293B5E
2D2F0000 00000000 00007C2C 255F3E3F
00000000 00000000 00603A23 40273D22
00616263 64656667 6869007B 00000000
006A6B6C 6D6E6F70 7172007D 00000000
007E7374 75767778 797A0000 005B0000
00000000 00000000 00000000 005D0000
7B414243 44454647 48490000 00000000
7D4A4B4C 4D4E4F50 51520000 00000000
5C005354 55565758 595A0000 00000000
30313233 34353637 38397C00 00000000
OPAL, an international collaboration of about 300 physicists in 24 laboratories
working at the new LEP 27 km particle smasher which the press wrote about last
Fall, successfully uses TeX for document circulation. Adjusting the networking
loose ends took a few months. The problem is not technical, it's managerial.
From this experience I do believe that ALL we need is to agree on one common
list of characters that everybody will mail as \char sequences, the rest is
completely trivial.
Jacques Goldberg
PS: I tried, but I cannot resist adding this. The XYZ manufacturer quoted
above sells a device to hook up ASCII terminals to their mainframes, a 7171
they name it. Would you believe that its default translation table
converts a BACKSLASH ( yes, the \ ) into something which TeX does not
understand? Fortunately "XYZperts" know a feature to correct that even without
altering this table ( SET INPUT 4A E0 in this case).
----------------------------------------------------------------------------
Date: Tue, 8 May 90 09:08 EST
From: MICHELLE@atc.bendix.com
Subject: Printing a portion of a .dvi or .ps file through a laser printer.
Keywords: .dvi, .ps, printing, laserjet
>>How can I print only a portion of .dvi or .ps file through a
>> laser printer on a main frame?
>> Sang-il (lee@hydro.stanford.edu)
There are two techniques of which I am currently aware. (I'll forward
your message to others for completeness.)
There exists a program called "DVISELECT" which runs on Unix which
permits you to specify the pages to be printed. The output of this program
is then run through the appropriate laser-printer specific program
to produce laser-able code. (Unfortunately, I don't know where to
get it or even if your machine could support it.)
More primitive is the technique I use. We have Talaris laser printers with
QUIC and EXCL code supported. So, after I DVI-TO-QUIC or DVI-TO-EXCL
I merely edit the resulting file to extract the pages I want. It's not
elegant, but it works.
Sorry I couldn't be of more help; hopefully, some one else will be able
to pick up the ball where I left off.
Michelle McElvany
Allied-Signal Aerospace Company
Columbia, MD
michelle@atc.bendix.com
------------------------------------------------------------------------------
Date: Tue, 8 May 1990 12:24:40.01 CST
From: <bed_gdg%shsu.bitnet@forsythe.stanford.edu> (George D. Greenwade)
Subject: Re: Printing a portion of a .dvi or .ps file through a laser printer.
Keywords: .dvi, .ps, printing, laserjet
We have a strictly VMS shop, so my answer is representative only for only our
one shop under that specific platform. We have an interpreter called LN03DVI
which is my personal favorite for LN03 printing. It allows for the use of
various modifiers. To get what you want, there are a few possibilities under
LN03DVI.
-n
First, you can specify LN03DVI-n# filename, where the -n is the number (#) of
pages to print from the beginning of the document. Thus to print the first 3
pages of JUNK.DVI (which we will assume to be a 900 page document!), use
LN03DVI-n3 JUNK which creates JUNK.LN3 for you to submit as a print job.
-s
Alternately, you can specify LN03DVI-s# filename, where -s identifies the page
number (the \thepage numbers associated with the DVI file) to begin printing
on. Thus, to print the last 20 pages of JUNK.DVI, use LN03DVI-s981 JUNK which
creates another JUNK.LN3 for your submission.
Mixing -s and -n
Finally, you can specify page to begin (-s) and number of pages to print (-n)
in a mixed combination to get a specific range of pages. Thus, LN03DVI-s20-n5
would begin on DVI page 20 and print the following five pages to a .LN# file
for printing.
OTHER VMS INTERPRETERS WE USE AND THEIR QUIRKS
Other interpreters we use include DVIALW (which creates a filename.DVI-ALW;
basically a PostScript output designed for the Apple LaserWriter), DVIJEP
(which creates a filename.DVI-JEP; Hewlett-Packard LaserJet Plus), DVIJET
(which creates a filename.DVI-JET; Hewlett-Packard 2686A Laser Jet), DVIL3P
(which creates a filename.DVI-L3P; Digital LN03-Plus Laser), DVIL75 (which
creates a filename.DVI-L75; Digital LA75 Dot Matrix printers), and DVIPRX
(which creates a filename.DVI-PRX; Printronix 300/600 Dot Matrix). The syntax
for each of these is DVIALW-o# filename to print a specific single page (#; -o9
would print page 9, for example), or DVIL75-o#:# filename to print a range of
pages (#:#; -o1:5 would print pages 1 to 5, for example).
There are a host of output devices available, each with its own quirks, syntax,
limitations, and features. For example, this has not even attempted to
identify how to landscape print, change magnification, select fonts, etc. Each
of these are powerful tools. Without trying to sound mean, hard, or flaming, I
would strongly advise you to contact your systems people and see what (if
anything) they can additionally tell you. My experience is that they love to
show off their magnificent capabilities to users who have specific questions
such as this. More important, due to the wide diversity of what is available,
only they know what is possible there (although you are certainly free to tell
them more in an effort to expand your local capabilities, if necessary).
Greetings from Huntsville,
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
George D. Greenwade, Director Bitnet: BED_GDG@SHSU
Center for Business and Economic Research THEnet: SHSU::BED_GDG
Sam Houston State University Voice: (409) 294-1518
Huntsville, TX 77341-2056 Internet: bed_gdg%shsu.decnet@utadnx.cc.utexas.edu
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-----------------------------------------------------------------------
%%% 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 <your name> % 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
**************************
-------sommer@BINAH.CC.BRANDEIS.EDU (06/05/90)
I hope this gets through to somebody at TeXhax. Some of my attempts to use the mail features of newsrdr have bombed. If this is successful, will you please let me know if you are interested in receiving or forwarding an endnotes style file to karen schaffer who requested info on endnotes in place of footnotes in an earlier edition of TeXhax? I tried mailing directly to schaffer%ellvax@... but my e-mail got bounced back to me. Thanks. (RSVP: sommer@brandeis.bitnet or sommer@binah.cc.brandeis.edu)