At 12:17 PM 2/2/00 -0500, Raymond Toy wrote:
>>>>> "Andy" == Andy Piper
<andy(a)xemacs.org> writes:
Andy> Could you try this patch instead of the other one and let me
know if
it
Andy> makes any difference.
Ok, I applied that patch and I got another crash. I don't remember
what the tabs were at the time, but I clicked on one in VM and xemacs
said good-bye.
C traceback below.
If it helps, here's some debug info. In DrawHighlight, child is:
(gdb) p *child
$2 = {core = {self = 0x1958e00, widget_class = 0x16c9000, parent =
0xb81af8,
xrm_name = 12065556,
being_destroyed = 0 '\000', destroy_callbacks = 0x0,
constraints =
0x0, x = 0, y = 0,
width = 0, height = 0, border_width = 0, managed = 0
'\000', sensitive
= 0 '\000',
ancestor_sensitive = 0 '\000', event_table = 0x0, tm =
{translations =
0x0, proc_table = 0x0,
current_state = 0x0, lastEventTime = 0}, accelerators = 0x0,
border_pixel = 0,
border_pixmap = 0, popup_list = 0x0, num_popups = 0, name = 0x0,
screen = 0x0, colormap = 0,
window = 0, depth = 0, background_pixel = 0, background_pixmap =
0,
visible = 0 '\000',
mapped_when_managed = 0 '\000'}}
so child->core.constraints is NULL.
It looks like the child hasn't been made visible yet. Maybe we should check
for this. Another thought try backing out my previous abortive bug-fix
attempt (patch attached) - this may be trying to unhighlight too early.
I think I will try and resync tabs again, this may change things.
andy
--------------------------------------------------------------
Dr Andy Piper
Senior Consultant Architect, BEA Systems Ltd