Re: GTK build crashes under X

[Stephen Berman]

On Mon, 11 Dec 2006 20:35:13 +0100 Jan Djärv <address@hidden> wrote:

> Hello.
>
> This is so far inside fontconfig and the theme engine that you use,
> so I suspect nothing Emacs does can help.  It seem to crash on
> /usr/X11R6/lib/X11/fonts/misc/cu12.pcf.gz, have you tried to remove
> that file?  It may be a bad font (I don't know what cu12.pcf is,
> Courier 12?).

According to the output of strings it is:
-MUTT-ClearlyU-Medium-R-Normal--17-120-100-100-P-123-ISO10646-1.
According to http://www.xfree86.org/current/fonts3.html,
/usr/X11R6/lib/X11/fonts/misc/cu12.pcf.gz is the "main ClearlyU font"
and this "family of fonts provides a set of 12 pt, 100 dpi
proportional fonts with many of the glyphs needed for Unicode text".
According to that source, the full font name has 101 instead of 123 in
the width slot; I wonder if that is relevant to my problem?  I grepped
the Emacs sources for ClearlyU (ignoring case) and got no hits.

> An idea why Emacs crashes and no other may be that Emacs uses only
> non-AA fonts, whereas Firefox and other uses AA fonts mostly.  But
> it is hard to tell.

The next time I get a segfault, I'll try to see if removing that font
helps (though I wonder if it will have bad consequences for other
applications ...). 

> If removing that file doesn't help, you could try to switch to a
> more basic theme.

It hadn't occurred to me to try switching themes.  I use the
gtk-qt-engine, which allows Gtk applications to use the selected KDE
theme.  I don't know if this is related, but this theme used to cause
a problem for Emacs (and only Emacs among Gtk applications), namely,
the color of the non-text part of the Emacs frame could not be changed
by setting the background-color face in Emacs (that is, the background
color only showed behind textual character; the rest of the window a
different color).  Other Gtk themes did not have this problem.  Then I
found on the web a report that compiling the gtk-qt-engine from source
solves similar issues and I tried that and it indeed solved the Emacs
problem.

> But all this is just shots in the dark, I'm afraid.

Thanks for your suggestions.

Steve Berman