bug-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#9771: 24.0.90; Redisplay problems with control characters


From: Eli Zaretskii
Subject: bug#9771: 24.0.90; Redisplay problems with control characters
Date: Wed, 19 Oct 2011 13:49:54 +0200

> From: Johan Bockgård <bojohan@gnu.org>
> Cc: handa@m17n.org,  9771@debbugs.gnu.org
> Date: Wed, 19 Oct 2011 03:06:09 +0200
> 
> emacs -Q
> 
> C-u 2000 C-q 0 RET
> 
> Notice that the lines visible in the window start with "@^".
> 
> Press <up> until the top of the buffer just scrolls into view.
> 
> Notice that lines before point start with "^@", and lines after
> point start with "@^".

Thanks for the test case.

> Change BUFFER_POS_REACHED_P (or something) so that dpvec_index == 0 is
> an acceptable stop position and this problem doesn't happen.

I'm not sure a change in BUFFER_POS_REACHED_P is the right fix for
this particular problem.  Even if it is, I'm reluctant to make in
move_it_in_display_line_to a change that could have effect in many
unrelated places, since this function is such a central piece of the
display engine.  I'd rather make a change that targets this specific
problem, certainly while we are in a pretest.

I think I found such a change.  The problem is that when we start a
new redisplay cycle of a window, we begin drawing the window's first
line at the last glyph of the display vector used to display the first
character position visible in the window.  I made a simple change that
should fix this (in revision 106131), please see if it resolves the
issue.

Another problem I found was that one of the redisplay optimizations
would redraw too few lines of a window, when the last redrawn line
ends in glyphs from a display vector.  I fixed that in revision
106133.

With all the 4 parts of the original report being addressed now, I'm
closing this bug report.  If there are any left-overs, please submit
separate bug reports about them.

Thanks.





reply via email to

[Prev in Thread] Current Thread [Next in Thread]