This is a regression from bug 823619. Under the new event model, when the user taps, we get these events: touchstart touchend mousemove mousedown mouseup click Apps that use gesture_detector.js (but do not use mouse_event_shim.js) will get these events from the gesture detector: tap tap dbltap That is because the gesture detector generates a tap for the touchstart/touchend pair and the mousedown/mouseup pair. This is a regression, and we should block on it. Fortunately, I've got a fix in my patch for bug 827551. The fix is to modify gesture detector so that if we ever receive a touch event, we stop listening for mouse events. I'll attach the patch soon.
Nominating blocking. Cc'ing cjones since it is a regression from 823619. (Sorry I didn't catch this one, Chris!) The only place I see a symptom of this bug is when you preview a photo in the camera app by tapping on the filmstrip. tapping on the photo zooms in even though it is supposed to be a double tap to zoom.
Created attachment 699381 [details] link to patch on github The attached pull request fixes this bug and also 827551. Looking for a reviewer...
Comment on attachment 699381 [details] link to patch on github Chris, Is this something you can review?
re 3: I did a quick review... One nit and one question. I don't understand 100% of GD but this makes sense to me. r=lightsofapollo
Comment on attachment 699381 [details] link to patch on github Switching review to James, and marking his r+. And requesting approval to land from Vivien, in case this doesn't get marked blocking. NOTE: If blocking-basecamp+ is set, just land it for now. [Approval Request Comment] Bug caused by (feature/regressing bug #): 823619 User impact if declined: taps may be generate dbltap events in some apps. Testing completed: yes, this patch fixes the photo viewer in the camera app and does not break other users of gesture detector like gallery and browser Risk to taking this patch (and alternatives if risky):
It looks like I reviewed the patch myself, but actually, I just set the flag after I got a r+ from lightsofapollo
LGTM. I added some comments on the GitHub PR, but nothing that should delaying this fix from landing.
I've updated the github PR to address nits from both Chris and James.
Device: Unagi Build: 20130114073222 Does not repro on this build and device.