MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
References: <85wsoeto1k.fsf@lola.goethe.zz>
            <18385.42387.268037.81145@morse.mittelbach-online.de>
            <20080307234555.GA6607@irwin.vpn.uni-freiburg.de>
Message-ID:  <18386.38365.176041.393185@morse.mittelbach-online.de>
Date:         Sat, 8 Mar 2008 14:34:21 +0100
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
Sender: Mailing list for the LaTeX3 project <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
From: Frank Mittelbach <frank.mittelbach@LATEX-PROJECT.ORG>
Subject: Re: A really, really bad bug.
To: LATEX-L@LISTSERV.UNI-HEIDELBERG.DE
In-Reply-To:  <20080307234555.GA6607@irwin.vpn.uni-freiburg.de>
Precedence: list
Status: R

Heiko,

 > On Fri, Mar 07, 2008 at 09:29:07PM +0100, Frank Mittelbach wrote:
 > 
 > > David Kastrup writes:
 > > 
 > >  > Another possibilities are to remove the grouping from the end of
 > >  > document action testing the labels. 
 > > 
 > > from briefly staring at the code I see no reason why this group is needed at
 > > all. Does anybody see one why it could make a difference?
 > 
 > You mean the following group?
 > 
 >   \def\enddocument{%
 >     ...
 >     \begingroup
 >       ...
 >       \input\jobname.aux
 >       ...
 >     \endgroup
 >     ...
 >   }
 > 
 > Then it isn't known what .aux files may contain (apart from \newlabel).
 > Many packages write definitions and other stuff in the .aux files.

well it is known to me, but in which case would that make a difference at this
point? The only thing that happens after the endgroup is

  \deadcycles\z@\@@end

and that shouldn't even output another page unless the aux files are missued
to the point that they contribute again to the page galley

so what am I missing that you think can make a problem even if such
definitions or stuff are not executed in a group?
 
frank