[PATCH] Revised mime-type patch.
- From: "Iain" <nodatadj webtribe net>
- To: <jrb redhat com>
- Cc: "gnome mailing list" <gnome-list gnome org>, <gnome-devel-list gnome org>
- Subject: [PATCH] Revised mime-type patch.
- Date: Thu, 2 Sep 1999 23:46:39 +0100
> I would rather see the clist with one column, and the buttons state
> change depending on whether or not it can be changed. Also, the buttons
> should be moved to the right side of the clist.
Changed, see patch.
> In addition, it segfaults
> when trying to revert. I don't want to commit it as is (this is supposed
to be
> stable, after all...) but would welcome it with those bugs fixed.
This happens in the standard control centre before my additions are
made.
I've had a look at it, but not been able to work out whats doing it yet,
apart from it seems that the function context_destroy is corrupting the
memory somehow.
If you uncomment the line
/* g_hash_table_foreach (specific_types, print_mime_foreach, NULL);
*/
It happily prints all the mimetypes and then proceeds to crash when
g_hash_table_foreach (specific_types, clean_mime_foreach, NULL); calls
context_destroy.
if print_mime_foreach () is changed to
static void
print_mime_foreach (gpointer mime_info, gpointer mi, gpointer data)
{
g_print ("mime_info:%s:\n", (char *)mime_info);
g_print ("\t:%s:\n", ((MimeInfo *)mi)->mime_type);
context_destroy ((GnomeMimeContext *) mi);
}
then you'll see the memory corruption (always on the 3rd time through this
for me).
Sorry if this is completely incorrect and I'm getting completely confused.
Iain
control-center.patch
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]