On Jun 4, 2011, at 10:20 AM, John Fabiani wrote:
> Even with Ed recent updates I still get a sge fault.
The updates didn't have anything to do with that issue.
> johnf@linux-asn9:~/dabo/ide> python ClassDesigner.py
> XXX lineno: 568, opcode: 0
> Segmentation fault
>
> To repeat just open ClassDesigner.py. Add 3 vert sizers, then add a grid to
> any of the sizers.
Actually, just adding a grid by itself is enough. I just grabbed 2.7
and started playing with it, and can repro the problem consistently.
I tried a bunch of stuff like commenting out the grid filling code, but
nothing seemed to make a difference. I stepped through the code, but there
isn't a single line that causes it to blow up. Instead, it seems to be in all
the *after stuff: the refreshes/updates/layout calls.
Since I couldn't step through to locate the problem, I added the
@loggit decorator to all dGrid methods, hoping to find the one that triggered
the error, as it would be the last method logged. But when I ran with these
decorators, the grid was added normally without any bus error/ seg fault.
Instead of getting more frustrated, this gives me a clue: there isn't
any specific code that's causing the error; instead, it seems to be caused by
lots of method calls happening in a very short period, since added the loggit
decorators slowed down the actual processing without changing the code being
run. I'm going to start hunting for places where we can delay calls or avoid
multiple repetitions of calls to see if I can find a solution for this issue.
-- Ed Leafe
_______________________________________________
Post Messages to: [email protected]
Subscription Maintenance: http://leafe.com/mailman/listinfo/dabo-dev
Searchable Archives: http://leafe.com/archives/search/dabo-dev
This message:
http://leafe.com/archives/byMID/[email protected]
