On Mon, 17 Sep 2001, Tom Cato Amundsen wrote:
> Can someone that has installed gtk+ 1.3.7 and pygtk-1.99 test
> this little script and see if it segfaults? I'm still
> unsure if this is a local problem or a bug.
>
> import sys
> # edit this!!!
> sys.path.insert(0, "/home/tom/gtk2/lib/python2.0/site-packages")
> import gtk
>
> class VClass:
> def __init__(self):
> pass
>
> class MWin(gtk.GtkWindow, VClass):
> def __init__(self):
> gtk.GtkWindow.__init__(self, gtk.WINDOW_TOPLEVEL)
> VClass.__dict__['__init__'](self)
> print "Prepare to segfault"
> a = self.b
> print "You'll never get here!"
>
> w = MWin()
>
I have worked out what is causing this crash. It is not handling VClass
correctly in the getattr routine in gobjectmodule isn't checking for non
ExtensionClass instances.
Try applying this patch:
Index: gobjectmodule.c
===================================================================
RCS file: /cvs/gnome/gnome-python/pygtk/gobjectmodule.c,v
retrieving revision 1.45
diff -u -p -r1.45 gobjectmodule.c
--- gobjectmodule.c 2001/09/15 02:59:54 1.45
+++ gobjectmodule.c 2001/09/18 02:10:23
@@ -1085,7 +1085,7 @@ check_bases(PyGObject *self, PyExtension
else
PyErr_Clear();
}
- if (class->bases) {
+ if (PyExtensionClass_Check(class) && class->bases) {
guint i, len = PyTuple_Size(class->bases);
for (i = 0; i < len; i++) {
Applying that patch should fix the problem.
James.
--
Email: [EMAIL PROTECTED]
WWW: http://www.daa.com.au/~james/
_______________________________________________
pygtk mailing list [EMAIL PROTECTED]
http://www.daa.com.au/mailman/listinfo/pygtk
