On Fri, Feb 19, 2010 at 8:15 AM, Roland King <[email protected]> wrote:
> why would that leak it? Those calls are paired, a table enters editing mode,
> the variable is set and what it's set to is retained, when the table exits
> editing mode again the cached value is released and nil'ed. The variable is
> also released in dealloc (not shown here). I do not know of a case where
> setEditing is called with YES twice without an intervening setEditing:NO.
Just because you don't know of a case doesn't mean that you shouldn't
be prepared for it. If I were you, I would do something like:
-(void)setEditing:(BOOL)editing animated:(BOOL)animated
{
if(editing) {
[preEditPath release];
preEditPath = [[[self tableView] indexPathForSelectedRow] retain];
}
[super setEditing: editing animated: animated];
if(!editing && preEditPath)
{
[[self tableView] selectRowAtIndexPath: preEditPath animated:
YES scrollPosition: UITableViewScrollPositionNone];
[preEditPath release ];
preEditPath = nil;
}
}
That is, release preEditPath in the editing case before setting it, to
avoid a leak in the case you are called twice with a value of YES, as
well as avoid passing a nil preEditPath to
-selectRowAtIndexPath:animated:scrollPosition: in the case that you
are called twice with a value of NO.
--
Clark S. Cox III
[email protected]
_______________________________________________
Cocoa-dev mailing list ([email protected])
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 [email protected]
