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

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

bug#12403: 24.2; scroll-margin breaks emacs24


From: Eli Zaretskii
Subject: bug#12403: 24.2; scroll-margin breaks emacs24
Date: Wed, 12 Sep 2012 20:15:03 +0300

> Date: Mon, 10 Sep 2012 19:39:41 +0300
> From: Eli Zaretskii <eliz@gnu.org>
> Cc: 12403@debbugs.gnu.org
> 
> > Try with the following (under a graphical session):
> > 
> > emacs -q
> > 
> > open the tutorial, scroll with the arrow keys until you reach the bottom.
> > Everything looks ok.
> > 
> > kill emacs. Start again, this time:
> > 
> > C-x set-variable scroll-margin 5
> > 
> > open the tutorial, scroll down with the arrow keys until you reach the
> > bottom.  Text on the page looks *badly* garbled (this is not just
> > character left-overs from the previous screenfull, a lot of lines look
> > repeated at some point in the same line).
> 
> Confirmed :-(.  However, this bug was present even in the first
> pretest of Emacs 24, 24.0.90.
> 
> I will take a look.

Fixed in revision 108138 on the emacs-24 branch, with the following
patch:

=== modified file 'src/ChangeLog'
--- src/ChangeLog       2012-09-12 14:59:56 +0000
+++ src/ChangeLog       2012-09-12 17:10:02 +0000
@@ -1,3 +1,11 @@
+2012-09-12  Eli Zaretskii  <eliz@gnu.org>
+
+       * xdisp.c (try_window_reusing_current_matrix): Under bidi
+       reordering, locate the cursor by calling set_cursor_from_row; if
+       that fails, clear the desired glyph matrix before returning a
+       failure indication to the caller.  Fixes leaving garbled display
+       when fast scrolling with a down-key.  (Bug#12403)
+
 2012-09-12  Jan Djärv  <jan.h.d@swipnet.se>
 
        * gtkutil.c (x_wm_set_size_hint): Use 1 col for base_width so it

=== modified file 'src/xdisp.c'
--- src/xdisp.c 2012-05-30 18:09:17 +0000
+++ src/xdisp.c 2012-09-12 16:58:45 +0000
@@ -16593,28 +16593,33 @@ try_window_reusing_current_matrix (struc
            }
          if (row < bottom_row)
            {
-             struct glyph *glyph = row->glyphs[TEXT_AREA] + w->cursor.hpos;
-             struct glyph *end = row->glyphs[TEXT_AREA] + row->used[TEXT_AREA];
-
-             /* Can't use this optimization with bidi-reordered glyph
-                rows, unless cursor is already at point. */
+             /* Can't simply scan the row for point with
+                bidi-reordered glyph rows.  Let set_cursor_from_row
+                figure out where to put the cursor, and if it fails,
+                give up.  */
              if (!NILP (BVAR (XBUFFER (w->buffer), bidi_display_reordering)))
                {
-                 if (!(w->cursor.hpos >= 0
-                       && w->cursor.hpos < row->used[TEXT_AREA]
-                       && BUFFERP (glyph->object)
-                       && glyph->charpos == PT))
-                   return 0;
+                 if (!set_cursor_from_row (w, row, w->current_matrix,
+                                           0, 0, 0, 0))
+                   {
+                     clear_glyph_matrix (w->desired_matrix);
+                     return 0;
+                   }
                }
              else
-               for (; glyph < end
-                      && (!BUFFERP (glyph->object)
-                          || glyph->charpos < PT);
-                    glyph++)
-                 {
-                   w->cursor.hpos++;
-                   w->cursor.x += glyph->pixel_width;
-                 }
+               {
+                 struct glyph *glyph = row->glyphs[TEXT_AREA] + w->cursor.hpos;
+                 struct glyph *end = row->glyphs[TEXT_AREA] + 
row->used[TEXT_AREA];
+
+                 for (; glyph < end
+                        && (!BUFFERP (glyph->object)
+                            || glyph->charpos < PT);
+                      glyph++)
+                   {
+                     w->cursor.hpos++;
+                     w->cursor.x += glyph->pixel_width;
+                   }
+               }
            }
        }
 







reply via email to

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