Bug 38056 - [Android] The touchmove delta used in basic-single-touch-events.html is too small for Android
Summary: [Android] The touchmove delta used in basic-single-touch-events.html is too s...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Tools / Tests (show other bugs)
Version: 528+ (Nightly build)
Hardware: Android Android
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-04-23 12:21 PDT by Ben Murdoch
Modified: 2010-04-23 20:37 PDT (History)
5 users (show)

See Also:


Attachments
Proposed Patch (4.15 KB, patch)
2010-04-23 12:36 PDT, Ben Murdoch
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Ben Murdoch 2010-04-23 12:21:33 PDT
Android is failing the basic-single-touch-events test due to an optimisation in the Android framework that does not forward touch events to WebCore if they are a small move. Currently the delta used in the singleTouchSequence function
in basic-single-touch-events (10, 10) -> (20, 15) falls within the scope of the optimisation and so we fail the test as the touch event is not sent to WebCore. I'd like to propose increasing this delta so that the test passes on Android. 

Proposed patch to follow.
Comment 1 Ben Murdoch 2010-04-23 12:36:56 PDT
Created attachment 54181 [details]
Proposed Patch
Comment 2 Jeremy Orlow 2010-04-23 13:51:13 PDT
Comment on attachment 54181 [details]
Proposed Patch

Might it be a good idea to change one of the values so that they're not both 50 and/or 20?
Comment 3 Ben Murdoch 2010-04-23 14:37:36 PDT
(In reply to comment #2)
> (From update of attachment 54181 [details])
> Might it be a good idea to change one of the values so that they're not both 50
> and/or 20?

I could do ... although all the analysis I've done I've always looked at the x and y as a pair, so it was only useful to make the x,y pair distinct from other pairs. Unless you feel strongly I think we can just leave as-is.

Thanks, Ben
Comment 4 Jeremy Orlow 2010-04-23 14:56:19 PDT
I r+'ed it because I was OK with it in its current state.  Just a suggestion.
Comment 5 Ben Murdoch 2010-04-23 14:59:49 PDT
Comment on attachment 54181 [details]
Proposed Patch

Great, thanks. Setting cq+.
Comment 6 WebKit Commit Bot 2010-04-23 20:36:55 PDT
Comment on attachment 54181 [details]
Proposed Patch

Clearing flags on attachment: 54181

Committed r58208: <http://trac.webkit.org/changeset/58208>
Comment 7 WebKit Commit Bot 2010-04-23 20:37:01 PDT
All reviewed patches have been landed.  Closing bug.