create namespaced elements when building the jQuery.support object

[bridget.almas]

To build the jQuery.support object, the code creates a number of html elements using document.createElement and then proceeds to test the browser behavior against them. This is problematic when jQuery is used in an XUL extension, as the elements are not recognized as HTML elements when the document object used to create them is an XUL document. I think using document.createElementNS, specifying the xhtml namespace, rather than document.createElement might work around this issue.

[snover]

Not sure if we are supporting an XUL context.

[snover]

#5776 is a duplicate of this ticket.

[[email protected]…]

There is a special fork of jQuery called ​jQuery-xul at github, it is compatible with FireFox xul model.

[ekesken]

Replying to [email protected]…:

There is a special fork of jQuery called ​jQuery-xul at github, it is compatible with FireFox xul model.

i tried your jquery-xul, unfortunately it didn't work for me. i've a simple toolbar application, i don't have any problem with my main application overlay, even with plane jquery-1.4.4 but in preferences dialog xul, jquery prevents xul elements' addition to dom.

there is no problem with jquery-1.2.6.

[jitter]

#8172 is a duplicate of this ticket.

[john]

Nominating ticket for 1.7 discussion.

[Rick Waldron]

+0,

[jaubourg]

-1, Well, script tag injection will also fail for one and other places too I'm sure. AFAIK, XUL extensions are not a supported environment and that has everything to do with XML being so unfriendlily pedantic.

[ajpiano]

+0, I think we'd need to a do a bit more research to find out more about that "might"

[Timmy Willison]

+0,

[danheberden]

+0, Need to see proof/working examples before committing to include this in 1.7

[paul.irish]

-1

[dmethvin]

-1, No to blind createElementNS since it isn't supported by IE<9; is there some other way to support XUL?

[reesd]

+1 on looking at improving this, it works pretty well as it is but then occasionally fails - usually because of the jQuery.support issue.

I agree it shouldn't be blind, perhaps it can detect that the innerHTML fails (completely I might add) and then look to see if its in Gecko/Firefox? Then try createElementNS? So its isolated to only Gecko cases?

I'll add there is a a hidden issue here, the user can create XUL and HTML elements. If we run the support tests against a XUL div the we are assuming the developer is creating XUL elements. That isn't always true, so its worth documenting somewhere that fact so folks are not surprised.

[Rick Waldron]

I think there maybe some confusion here, the +1/-1/+0 is for the bugs, core and ui teams to vote on pending 1.7 issues. We apreciate your feedback and it will definitely be taken into consideration, but the voting is whitelisted.

[john]

-1, We definitely need to see more proof that this could work.

[scottgonzalez]

+0, I don't know enough about XUL to know if we want to support it

[jzaefferer]

-1

[cowboy]

+0

[Rick Waldron]

Closing per 1.7 roadmap meeting resolution