Done -- http://trac.openlayers.org/ticket/1848
Christopher
Schmidt
<[EMAIL PROTECTED] To
rta.com> Jess Lacy/Boulder/[EMAIL PROTECTED]
cc
11/19/2008 05:21 [email protected]
PM Subject
Re: [OpenLayers-Users]
[OpenLayers-Dev] Memory Leak in
OpenLayers + IE
On Wed, Nov 19, 2008 at 03:58:45PM -0800, jesslacy wrote:
>
> I ran across this (or what may be a similar) issue today in IE 6 using OL
2.7
> for a layer with the alpha hack.
Can you open a ticket for this?
http://trac.openlayers.org/wiki/FilingTickets
> In the Image.js file, the initImgDiv method calls two observers:
>
> if (this.layerAlphaHack) {
> OpenLayers.Event.observe(this.imgDiv.childNodes[0], 'load',
> OpenLayers.Function.bind(onload,
> this));
> } else {
> OpenLayers.Event.observe(this.imgDiv, 'load',
> OpenLayers.Function.bind(onload, this));
> }
> ...
> OpenLayers.Event.observe(this.imgDiv, "error",
>
> But only one observer gets cleaned up in the destroy method.
>
> Changing:
> if (this.layerAlphaHack) {
>
> OpenLayers.Event.stopObservingElement(this.imgDiv.childNodes[0].id);
> } else {
> OpenLayers.Event.stopObservingElement(this.imgDiv.id);
> }
> to
> if (this.layerAlphaHack) {
>
> OpenLayers.Event.stopObservingElement(this.imgDiv.childNodes[0].id);
> OpenLayers.Event.stopObservingElement(this.imgDiv.id);
> } else {
> OpenLayers.Event.stopObservingElement(this.imgDiv.id);
> }
> has resolved the problem for me.
>
> Regards,
> Jess Lacy
>
>
> bradleyspencer wrote:
> >
> > List,
> >
> > I'm not sure if I have this problem or how to verify it but I am having
a
> > lot of browser freeze ups with my testing.
> >
> > I am getting this with 2.5 loaded on my server and whatever is the most
> > recent online version.
> >
> > Is there any update on this?
> >
> > Cheers,
> >
> > Brad Spencer
> >
> >
> > -----Original Message-----
> > From: Christopher Schmidt [mailto:[EMAIL PROTECTED]
> > Sent: Wednesday, 28 March 2007 1:49 AM
> > To: [EMAIL PROTECTED]
> > Subject: [OpenLayers-Dev] Memory Leak in OpenLayers + IE
> >
> > So, after some investigation this morning, I've discovered a source of
a
> > memory leak in Internet Explorer + OpenLayers:
> >
> >
> > Currently, when creating Tile.Image images, we bind a load/error
> > listener:
> >
> > OpenLayers.Event.observe(image, "load",
> > OpenLayers.Util.onImageLoad.bindAsEventListener(image));
> > OpenLayers.Event.observe(image, "error",
> >
> > OpenLayers.Util.onImageLoadError.bindAsEventListener(image));
> >
> > (Util.js, line ~180)
> >
> > This is how we prevent tiles from displaying in the wrong location.
> > Unfortunately, this closure on the image is never cleaned up, so IE
> > doesn't destroy the images, even when the page is navigated away from,
> > resulting in a memory leak.
> >
> > No fix yet, just documenting this so that if anyone has an obvious 'oh,
> > just do $foo', they can feel free to offer it up :)
> >
> > Regards,
> > --
> > Christopher Schmidt
> > MetaCarta
> > _______________________________________________
> > Dev mailing list
> > [EMAIL PROTECTED]
> > http://openlayers.org/mailman/listinfo/dev
> >
> > _______________________________________________
> > Users mailing list
> > [email protected]
> > http://openlayers.org/mailman/listinfo/users
> >
> >
>
> --
> View this message in context:
http://www.nabble.com/-OpenLayers-Dev--Memory-Leak-in-OpenLayers-%2B-IE-tp13786396p20592288.html
> Sent from the OpenLayers Users mailing list archive at Nabble.com.
>
> _______________________________________________
> Users mailing list
> [email protected]
> http://openlayers.org/mailman/listinfo/users
--
Christopher Schmidt
MetaCarta
<<inline: graycol.gif>>
<<inline: pic06777.gif>>
<<inline: ecblank.gif>>
_______________________________________________ Users mailing list [email protected] http://openlayers.org/mailman/listinfo/users
