TeX 1995 July/TeX CD-ROM July 1995 (Disc 1)(Walnut Creek)(1995).ISO/tex-k/tex-k-archive.past/tex-k-archive.gz/tex-k-archive/[e-mail]_Wed Jun 22 01:44:59 1994.msg

Internet Message Format | 1994-10-11 | 2KB

Received: from terminus.cs.umb.edu by cs.umb.edu with SMTP id AA21576
  (5.65c/IDA-1.4.4 for <[e-mail]>); Wed, 22 Jun 1994 05:45:00 -0400
Received: by terminus.cs.umb.edu id AA15163
  (5.65c/IDA-1.4.4 for tex-k); Wed, 22 Jun 1994 05:44:59 -0400
Date: Wed, 22 Jun 1994 05:44:59 -0400
From: "K. Berry" <[e-mail]>
Message-Id: <[e-mail]>
To: [e-mail]
Subject: MakeTeXPK interface

One infinitely recurring problem, as some of you may have noticed, is
the difference between kpathsea's interface to MakeTeXPK and the
original dvips/xdvi.

I changed MTPK to echo the full filename of the generated font to
stdout, and put commentary on stderr, so that kpathsea could read it and
open it. This is useful only when the resulting file can't be found in
the search path. (Well, it does save one path lookup to echo the full
filename, but we're assuming path lookups are cheap, right? :-)

The primary application for this is (I think) when the main tex
directory is on a readonly filesystem (because of AFS or just local
preferences). Then MTPK has to write somewhere locally (/tmp/foo*pk?)
and the font has to be moved to the main filesystem by other means. (A
cron job or something that rcp's it, maybe.)

It seems like either I should change the interface back (then we'll have
another period of massive confusion but it will eventually go away), or
I should talk to Tom (Rokicki) and Paul (Vojta) about changing
xdvi/dvips so we all use the same thing.

But before I bug them about it, I wanted to ask for opinions on how
important the kpathsea feature was, and whether anyone would notice if
I went back to the old way (of doing a path lookup on the font name if
MTPK runs successfully). So, I'm asking ...