Convert the key binding files (browsers, textareas, enders, and inputs) over to XBL. This needs to happen sooner rather than later. We have broken basic functionality (like keyboard scrolling) in the embedded case.

putting on beta1 radar

I have converted inputs and textareas over. Next come the browser and editor DOM windows.

Cc'ing myself so that I can add the Unix-specific bindings (if you don't do it for me) as soon as the framework is in place. The XBL bindings in the text fields/areas are SO much faster than XUL -- it'll be great to have them in the editor as well!

*** Bug 26314 has been marked as a duplicate of this bug. ***

Does this include the mail/news window? Mail compose?

This should be PDT+.

Putting on PDT- radar for beta1. Concerned about regression risk at this time.

No, this HAS to be PDT+. Other PDT+ bugs depend on this.

wiped out pdt- so they will reconsider, though you should probably list the pdt+ bugs that depend on this, since they will ask :-)

Please put an estimated completion date on this.

Well, if we don't care about the embedding case, I guess we could go to beta without finishing this conversion. Your call. If it is acceptable for viewer to remain in the busted state it's in now regarding window key bindings, then I'm comfortable shoving this to M15.

Also, composer performance is adversely impacted by leaving the bindings in XUL. If you notice, typing in text fields is much faster than typing in the composer window. If an XBL conversion gets done, then the composer window will be much faster. There would undoubtedly be some regressions that would occur as fallout from this conversion however.

Trying to be pragmatic here. I think we can get by without this being in for beta. If I start getting bugs that become PDT+ that would depend on this, we can reconsider it, but I think for now we're ok.

Marking - due to hyatts recommendation.

*** Bug 27739 has been marked as a duplicate of this bug. ***

Mass-moving all M16 non-feature bugs to M17, which we still consider to be part of beta2

spam, open xptoolkit qa contact moving over to jrgm

nominating for nsbeta2

[nsbeta2-]

Adding mlk keyword because of bug 27739 (which was marked as a duplicate of this one).

Mass-moving all nsbeta2- bugs to M20

Please clean up the leak caused by the fix for bug 27739 when fixing this bug. Thanks! (Re: bug 36639)

*** Bug 40291 has been marked as a duplicate of this bug. ***

*** Bug 41232 has been marked as a duplicate of this bug. ***

It's actually not too hard to fix the keybindings leak without fixing this bug (which is probably hard). I wrote code to do this a few weeks ago, but I couldn't test very easily because I ran into bug 42895. However, I realize now I could use the steps in bug 41232 to test my fix... I'll try doing this soon.

Let bug 17390 cover the leak issues from this point on. Removing mlk keyword from this bug but adding perf keyword since perfomance issues were discussed for the real issue in this bug.

So hyatt, can we get somebody else to help you do this? or do this for you? The current keybinding stuff has significant problems with leaks and cleaning itself up. (Although dbaron has tracked down most of those issues...)

nsbeta3-, not worth the risk for seamonkey.

Workaround for embedding is to use a XUL layer. Otherwise this is huge and risky.

Adding nsbeta2 keyword to bugs with nsbeta2 triage value in status field so the queries don't get screwed up

Reassign to dr, since he just filed bug 52359.

Hmm. Can I mark this as a dup of bug 52359, or vice-versa? The only thing that's left to do here is the browserBindings and editorBindings, right?

this is a dup of bug 52359, which will be fixed rsn as part of 48758. *** This bug has been marked as a duplicate of 52359 ***