[mythtv] mythfrontend segfault

[avalanche at beyondmonkey.com]

----- Original Message ----- 
From: "Isaac Richards" <ijr at po.cwru.edu>
To: "Development of mythtv" <mythtv-dev at mythtv.org>
Sent: Saturday, January 03, 2004 8:05 AM
Subject: Re: [mythtv] mythfrontend segfault


> On Saturday 03 January 2004 01:47 am, Jim Paris wrote:
> > Just playing around in the "Fix scheduling conflicts" screen with
> > day-old CVS; this may have been triggered by making changes to the
> > record/don't-record status of different shows too quickly.
> 
> Hrmph.  Can you reproduce this at all?  I'd really like to see what's past 
> stack level #19 in that trace..  If you can reproduce this easily enough, I 
> wonder if reverting out the last set of changes to viewscheduled.cpp fixes 
> it:

Problem is that the customEvents are sneaking in from the side, the dialogs
are causing the event loop to rotate. If you set a program to 'don't record', 
it will stay visible till the next customEvent, being fast enough, you can 
call the dialog again for that program. As soon as the customEvent comes in, 
the pointer is borked.
 
I put together a container that should solve problems like this. The idea 
is to let the container maintain the pointers, current selection, bounds 
etc. If a programinfo needs to be edited, it's checked out value based 
and checked back in afterwards. Fallbacks are in place and everything is 
checked so it should take the pressure off the GUI parts.

Attached is a version for viewscheduled, planning on generalizing the 
container for other screens as well.

av
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pginfocon.diff
Type: application/octet-stream
Size: 12335 bytes
Desc: not available
Url : http://mythtv.org/pipermail/mythtv-dev/attachments/20040104/676768a9/pginfocon.obj