[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#18384: 24.3.93; posn-col-row and posn-actual-col-row return differen
From: |
Eli Zaretskii |
Subject: |
bug#18384: 24.3.93; posn-col-row and posn-actual-col-row return different row number for point-min when header-line-format is non-nil |
Date: |
Tue, 02 Sep 2014 18:20:13 +0300 |
> From: Dmitry <dgutov@yandex.ru>
> Date: Tue, 02 Sep 2014 01:13:03 +0400
>
> 1. Open a simple buffer; maybe an empty one.
> 2. (setq header-line-format "aaaa")
> 3. (posn-actual-col-row (posn-at-point (point-min)))
> => (0 . 1)
> 4. (posn-col-row (posn-at-point (point-min)))
> => (0 . -1)
>
> I would also understand if the latter returned (0 . 0) instead of the
> same value, but -1 seems obviously wrong.
They are both wrong. The correct result is (0 . 0), of course.
> http://debbugs.gnu.org/cgi/bugreport.cgi?bug=7390#40 ?
Yes, this is indeed a fall-out of bug #7390. As you see in
http://debbugs.gnu.org/cgi/bugreport.cgi?bug=7390#49, the Y pixel
issue was fixed on the trunk (in r102407), but then my temporary
change in posn-col-row on the release branch, which shouldn't have
been merged onto the trunk, _was_ merged (in r102428) and screwed this
up. Reverting that change fixed posn-col-row.
As for posn-actual-col-row, the problem was in posn-at-point, which
was returning incorrect row count when header line is present in the
window. I fixed that as well.
Revision 117473 on the emacs-24 branch should fix both issues.