References: <5.1.0.14.0.20050324202131.0248fbc0@pop3.web.de>
User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.0.50 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID:  <x5eke4mx70.fsf@lola.goethe.zz>
Date:         Fri, 25 Mar 2005 00:38:27 +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: David Kastrup <dak@GNU.ORG>
Subject: Re: \@reinserts
To: LATEX-L@listserv.uni-heidelberg.de
In-Reply-To:  <5.1.0.14.0.20050324202131.0248fbc0@pop3.web.de> (
              =?iso-8859-1?q?Uwe_L=FCck's_message_of?= "Thu, 24 Mar 2005
              23:16:08 +0100")
Precedence: list
Status: R

Uwe Lück <uwe.lueck@web.de> writes:

> Happy Easter Holidays!
> -- Here is a proposal for an advanced search of Easter eggs:
>
> I thought it would be a matter of an hour or so to solve the
> problems with \@reinserts in LaTeX's output routine
> (see LaTeX bugs 3370, 3451, 3719).

That is not exactly an expression of great confidence in the
competence of the LaTeX team.

> And I need this somewhat urgently. At least I will (I'm afraid) be
> unable to sleep until \output guru Frank Mittelbach will return to
> his office after his Easter holidays. Can anybody else help me with
> sleeping or \output earlier?

I'd strongly recommend that you don't wait till Tuesday with sleeping.

> You find a proposal for getting rid of \@reinsert in latex/3719.
> Yet it seems to me that it doesn't work. With the proposed
> \let\@reinserts\relax, e.g., or with another emptying of the
> \footins box, the page builder quits without comment, and I get an
> overfull page. latex/3451 seems to allude to this problem, but what
> is needed to be done actually now ...?

If your problem is actually the same as in latex/3719, you could try
using bigfoot.sty from
<URL:http://cvs.sarovar.org/cgi-bin/cvsweb.cgi/src/?cvsroot=bigfoot>:
just loading it (in its current state) should fix the problem of
jumbled footnotes (the current code has other drawbacks at the moment,
like not giving always correct headers and footers, which is why it is
not yet released).

However, it would be unreasonable to expect lineno.sty to work with
it: even when the output-hookery stuff was cleaned up and implemented
"properly", I would be skeptical that it would cooperate without
involving more amount of work than can be reasonably done before you
ought to sleep.  I'll probably look into it when I get the rest of the
problems solved, but it is likely that I'd decide to rather try doing
the line numbers with PostScript hackery or similar expedients.

--
David Kastrup, Kriemhildstr. 15, 44793 Bochum