Thanks Brandon for the quick work on this. I updated the dialog-leak- test at http://linux.kytomaki.com/jqtest/ to include jquery r6323, but unfortunately the event.remove fix doesn't seem to correct this problem: IE7 and IE8 continue to leak memory in more or less the same fashion as described in http://groups.google.com/group/jquery-dev/browse_thread/thread/df7f91170d7cff51. :-(
Regards, Janne On 30 huhti, 01:22, DBJDBJ <[email protected]> wrote: > Thanks B, > > I was wondering : > > http://linux.kytomaki.com/jqtest/. > > Might this help for testing ? > > --DBJ > > On Apr 29, 10:49 pm, Brandon Aaron <[email protected]> wrote: > > > I've fixed the issue with jQuery.event.remove not unbinding namespaced > > events but had a very difficult time getting reliable results when testing > > the memory leak. Please test r6321 and let me know if this fixes the memory > > leak for you. > > -- > > Brandon Aaron > > > On Sat, Apr 25, 2009 at 3:09 PM, Mike Park <[email protected]> wrote: > > > > new thread, as requested (original message below): > > > ------- > > > > There is definitely a problem with the way jQuery (I'm using 1.3.2) > > > handles cleaning up namespaced events (I'm tracing the 'unload' event > > > codepath). The problem from what I can tell seems to be in > > > jQuery.event.remove: > > > > 2540 // Handle multiple events seperated by a space > > > 2541 // jQuery(...).unbind("mouseover mouseout", fn); > > > 2542 jQuery.each(types.split(/\s+/), function(index, type){ > > > 2543 // Namespaced event handlers > > > 2544 var namespaces = type.split("."); > > > 2545 type = namespaces.shift(); > > > 2546 var namespace = RegExp("(^|\\.)" + namespaces.slice > > > ().sort().join(".*\\.") + "(\\.|$)"); > > > > ...during the 'unload' event-handler codepath, this code snippet here > > > is expecting the passed in 'types' argument to contain the namespaced > > > event-identifier (ex. 'click.foo'), but instead it is getting the > > > common event-identifier instead (ex. 'click'). > > > > You can firebug-trace this yourselves with the simple examples I put > > > up here: > > > http://emparq.com/ex/ex-good.html > > > http://emparq.com/ex/ex-bad.html > > > > ...you'll notice the only difference between these two examples is > > > that one binds 'click.foo' and the other binds 'click'. In the bad > > > example, hitting reload constantly will see the process memory useage > > > go up (in essense, a leak) due to 'jQuery.event.remove' not properly > > > cleaning up the namespaced 'click.foo' event), while the good example > > > will not suffer the same results. > > > > I'm still new-ish to jQuery and I'm not ready to hazard a solution > > > just yet, but I'm guessing something either needs to be done with the > > > way 'jQuery.event.remove' handles the value of the 'types' argument, > > > or this code here: > > > > 2531 for ( var type in events ) > > > 2532 this.remove( elem, type + (types || "") ); > > > > ...needs to be modified to properly put those namespaced suffixes back > > > into the second argument. I'm not familiar enough with all of the use- > > > cases that this function handles, so I don't know which is the best > > > approach. > > > > Can anyone else shed some light on this subject? > > > > --Mike- Hide quoted text - > > > - Show quoted text - --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "jQuery Development" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en -~----------~----~----~----~------~----~------~--~---
