Hello,
I want to be provocative and spark a little bit of debate, so I'm going to use the closing of Ticket #1416 as an example. And please correct me if I'm wrong. This ticket has a stack trace that shows a crash during a contactMovedEvent. There is also a patch attached. The patch has not been applied. The ticket is now closed. Reason, changes were made to the contact list redrawing code which dramatically reduce the number of times this event fires, so the hope is that this no longer happens, and if it does, then new bugs can be filed. I don't know that this is a good enough reason to close the ticket. My reasoning: 1) As far as I can tell, the bug is a threading issue, is intermittent, and therefore not easily replicated. Even so, it is one of the many little intermittent bugs that cause crashes in Wengophone. 2) Though likely to happen less, no ultimate cause of the crash has been found. Is it simultaneous calls to contactMovedEvent? Some memory incorrectly freed? Etc? Without a proven cause, it is not likely that the bug is gone (that would be very lucky). 3) Without a bug, and the report gone, this crash will remain lurking in the code, and stack traces are hard enough to come by. My thinking is we should have an all out war on threading issues and look to kill them wherever we can. What are your thoughts? -Lukas Lukas Oberhuber
_______________________________________________ Wengophone-devel mailing list [email protected] http://dev.openwengo.com/mailman/listinfo/wengophone-devel
