On 28.03.2007, at 21:01, Brandon Aaron wrote: > I think the most likely candidate for the leaks is going to surround > the usage of AJAX and Events. Are you attaching events to html pulled > in via AJAX? I ask because I believe if you proceed to replace that > html without unbinding the events first, that might cause a leak ... > but I'm unsure ... just all I can think of for now.
Thanks for the tip - we'll be rechecking this. Only the reloading frames are binding events to AJAX-loaded HTML actually, but we'll recheck this. Unfortunately it's not possible to find out which frame is the leaking one - all the tools for leak checks which I've found so far seem to be almost unusable/useless. We've now also patched our copy of jQuery so that the originally IE- only unload handler for unbinding events is executing in all browsers for those parts of our page which reload - just for safety - and now we'll have to wait whether this helps. Astonishingly, we had the least problems with IE though it's not the primary browser for development for us, and this might have something to do with it. -- Markus Peter - [EMAIL PROTECTED] http://www.spin-ag.de/ SPiN AG, Bischof-von-Henle-Str. 2b, 93051 Regensburg, HRB 6295 Regensburg Aufsichtsratsvors.: Dr. Christian Kirnberger Vorstände: Fabian Rott, Paul Schmid _______________________________________________ jQuery mailing list discuss@jquery.com http://jquery.com/discuss/
