Summary: | [Qt][WK2] Double-tap during scroll animation leaves content in suspended state | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Andras Becsi <abecsi> | ||||
Component: | New Bugs | Assignee: | Andras Becsi <abecsi> | ||||
Status: | RESOLVED DUPLICATE | ||||||
Severity: | Normal | CC: | benjamin, cmarcelo, jturcotte, menard, webkit.review.bot | ||||
Priority: | P2 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | Unspecified | ||||||
OS: | Unspecified | ||||||
Attachments: |
|
Description
Andras Becsi
2013-02-28 08:19:30 PST
Created attachment 190733 [details]
Patch
Comment on attachment 190733 [details]
Patch
It's for those "might have been" cases that the update deferrers were good at handling.
r=me
Benjamin could you sign this off? Is this the kind of change that is considered trivial? (In reply to comment #2) > (From update of attachment 190733 [details]) > It's for those "might have been" cases that the update deferrers were good at handling. Yes, this bug was unfortunately present since we removed the deferrers. Comment on attachment 190733 [details]
Patch
Isn't panGestureCancelled() called when transitioning from a pan to another gesture?
E.g.: pan->pinch; pan->double finger scroll
If yes, this patch is wrong.
Comment on attachment 190733 [details]
Patch
Isn't panGestureCancelled() called when transitioning from a pan to another gesture?
E.g.: pan->pinch; pan->double finger scroll
If yes, this patch is wrong.
(In reply to comment #6) > (From update of attachment 190733 [details]) > Isn't panGestureCancelled() called when transitioning from a pan to another gesture? Yes. > E.g.: pan->pinch; pan->double finger scroll pan->pinch and pan->tap/double-tap, there is no double finger scroll in Qt. > > If yes, this patch is wrong. Could you elaborate? > > If yes, this patch is wrong.
>
> Could you elaborate?
You would resume the content between two interactions.
Let say you do pan->pinch->pan without ever lifting the first finger. You don't expect the content to resume at any point.
(In reply to comment #8) > > > If yes, this patch is wrong. > > > > Could you elaborate? > > You would resume the content between two interactions. > Let say you do pan->pinch->pan without ever lifting the first finger. You don't expect the content to resume at any point. Currently we only explicitly prevent the content from resuming between subsequent pan gestures and do not enforce staying suspended between pan->pinch->pan. Although it's true that the case for the pan->pinch part of the chain is currently in most cases provided as the side effect of this bug, as it was also the case when we had update deferrers. To enforce suspended state for gesture chains when the finger is not lifted, without having corner cases where the content remains suspended, I think we need to bring back the update deferrers in some form. The issue is covered by a more comprehensive patch at https://bugs.webkit.org/show_bug.cgi?id=111751 *** This bug has been marked as a duplicate of bug 111751 *** |