On Oct 22, 2011, at 8:53 AM, Mr. Gecko wrote:
> static NSAutoreleasePool *pool = nil;
>
> void runloop(CFRunLoopObserverRef observer, CFRunLoopActivity activity, void
> *info) {
> if (activity & kCFRunLoopEntry) {
> if (pool!=nil) [pool drain];
> pool = [NSAutoreleasePool new];
> } else if (activity & kCFRunLoopExit) {
> [pool drain];
> pool = nil;
> }
> }
I very much doubt this is kosher. An autorelease pool exists in a stack. If
the outer autorelease pool is popped, then yours is deallocated whether or not
you've released or drained it. After that happens, you have a dangling
pointer. It really makes no sense to have an autorelease pool with a lifetime
greater than the scope of its usage.
Have you tried running a timer to see if that provokes draining of the
autorelease pool managed by the framework? Have you tried posting a custom
event?
Even if those fail, the best approach is to just use local autorelease pools in
whatever methods you introduce which are creating temporary objects that would
otherwise live longer than you'd like.
Regards,
Ken
_______________________________________________
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
