From mboxrd@z Thu Jan  1 00:00:00 1970
X-Msuck: nntp://news.gmane.io/gmane.comp.tex.context/4252
Path: main.gmane.org!not-for-mail
From: "Giuseppe Bilotta" <oblomov@freemail.it>
Newsgroups: gmane.comp.tex.context
Subject: Still on footnotes
Date: Wed, 28 Feb 2001 18:17:01 +0100
Sender: owner-ntg-context@let.uu.nl
Message-ID: <000201c0a1d5$72eb40a0$a3ccfea9@nuovo>
NNTP-Posting-Host: coloc-standby.netfonds.no
Mime-Version: 1.0
Content-Type: multipart/mixed;
	boundary="----=_NextPart_000_0005_01C0A1B2.A560A980"
X-Trace: main.gmane.org 1035394933 24138 80.91.224.250 (23 Oct 2002 17:42:13 GMT)
X-Complaints-To: usenet@main.gmane.org
NNTP-Posting-Date: Wed, 23 Oct 2002 17:42:13 +0000 (UTC)
Original-To: "ConTeXt" <ntg-context@ntg.nl>
Xref: main.gmane.org gmane.comp.tex.context:4252
X-Report-Spam: http://spam.gmane.org/gmane.comp.tex.context:4252

This is a multi-part message in MIME format.

------=_NextPart_000_0005_01C0A1B2.A560A980
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

[Using the latest ConTeXt]

Hello, I noticed that there are still a couple of problems
with footnotes in ConTeXt.

The first one is the fact that sometimes body text flows in the
footnote space, thus overprinting existing footnotes. After some
debugging, I think I found that it happens only in a very particular
case: when using footnotes inside columns, without having footnotes
after the column.

Precisely, if there is a footnote in the columns, and the columns
end before the end of the page, and there is non-column text without
footnotes, the footnote from the columns is overprinted by the text
after the column. If, on the other hand, there *are* footnotes in the
post-column text, then the footnotes are all placed correctly
(both the in-column and post-column footnotes) and text doesn't
overprint footnotes. Since the presence of post-column footnotes
allows the correct placement also of in-column footnotes, I believe
that the overprint can be prevented also when there are no post-column
footnotes, if the output routines do an appropriate check.

[see attached file footbug1.tex for error details: commenting the
second footnote shows the bug]

The second problem is that sometimes the footnote text ends up
in the wrong page. This happens mostly when long footnotes happen
near the end of the page, and the size of paragraph+footnote is bigger
than would fit in the page, while the paragraph alone would: in such
a case the paragraph is not split, nor does it migrate to the next
page (to prevent underfull boxes), thus giving the awfull effect of the
footnote number at the end of a page and the relative footnote
text on the next page, which is IMHO worse than the underfull box
that might result from migrating the whole paragraph.

[see attached file footbug2.tex for two examples of such errors]

Giuseppe Bilotta

------=_NextPart_000_0005_01C0A1B2.A560A980
Content-Type: application/octet-stream;
	name="footbug1.tex"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="footbug1.tex"

\setupindenting[medium]=0A=
=0A=
\starttext=0A=
=0A=
\input knuth=0A=
=0A=
\blank=0A=
=0A=
\input zapf=0A=
=0A=
\blank=0A=
=0A=
\startcolumns=0A=
\input tufte\footnote{This was a text from Tufte.}=0A=
\stopcolumns=0A=
=0A=
\blank=0A=
=0A=
\input tufte\footnote{This was a text from Tufte.}=0A=
=0A=
\blank=0A=
=0A=
\input knuth=0A=
=0A=
\stoptext=0A=

------=_NextPart_000_0005_01C0A1B2.A560A980
Content-Type: application/octet-stream;
	name="footbug2.tex"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="footbug2.tex"

\setupindenting[medium]=0A=
=0A=
\starttext=0A=
=0A=
\input knuth=0A=
=0A=
\blank=0A=
=0A=
\input zapf=0A=
=0A=
\blank=0A=
=0A=
\startcolumns=0A=
\input tufte\footnote{This was a text from Tufte. Let's now get=0A=
this footnote a little longer, so as to trigger the second bug. We=0A=
need at least two lines, maybe three of four\dots}=0A=
\stopcolumns=0A=
=0A=
\blank=0A=
=0A=
\input tufte\footnote{\input zapf }=0A=
=0A=
\blank=0A=
=0A=
\input knuth=0A=
=0A=
\page[yes]=0A=
=0A=
\input knuth=0A=
=0A=
\blank=0A=
=0A=
\input zapf=0A=
=0A=
\blank=0A=
=0A=
\startcolumns=0A=
\input tufte=0A=
\stopcolumns=0A=
=0A=
\blank=0A=
=0A=
You don't need another footnote to trigger the bug: just=0A=
enough text to push the last line of the problematic=0A=
paragraph to the end of the page.=0A=
=0A=
\input tufte\footnote{\input zapf }=0A=
=0A=
\blank=0A=
=0A=
\input knuth=0A=
=0A=
=0A=
=0A=
=0A=
\stoptext=0A=

------=_NextPart_000_0005_01C0A1B2.A560A980--