Re: tetex-bin's postinst: Why the testing?

[Atsuhito Kohda <kohda@pm.tokushima-u.ac.jp>]

From: Frank Küster <frank@debian.org>
Subject: tetex-bin's postinst: Why the testing?
Date: Thu, 11 Mar 2004 15:24:02 +0100

> Can anybody give a rationale for this test? Was there an else branch
> long ago?

As far as I remembered, there was no else branch.

One more point I noticed recently when I wasinvestigating
#236246, there was a code in postinst as follows (line #181);

    if [ -d /var/spool/texmf ] && [ ! -d /var/cache/fonts ]
    then
	echo -n "Moving /var/spool/texmf to /var/cache/fonts..."
	mv -f /var/spool/texmf /var/cache/fonts
	echo "done."
    fi

This was aparently intended to move old /var/spool/texmf 
for generated fonts to FHS-mandated /var/cache/fonts but,
at normal upgrading,

- tetex-base would have installed /var/cache/fonts alreday
- so at installing stage of tetex-bin even if old 
  /var/spool/texmf existed, /var/cache/fonts also existed

therefore the following test

    if [ -d /var/spool/texmf ] && [ ! -d /var/cache/fonts ]

would be almost every time false and 

      mv -f /var/spool/texmf /var/cache/fonts

wouldn't be executed.

Am I wrong or there needs some fix?

Regards,		  2004-3-12(Fri)

-- 
 Debian Developer & Debian JP Developer - much more I18N of Debian
 Atsuhito Kohda <kohda@debian.org>
 Department of Math., Univ. of Tokushima