Bug 101549 – [EFL][WK2][AC] Fixed layers are shaking when scrolling

• Component: WebKit EFL
• Status: RESOLVED FIXED

[Hurnjoo Lee]

At the time of adjusting positions for fixed layers, the locations of
fixed layers are calculated incorrectly because the contentsSize is empty.

[Hurnjoo Lee]

Created attachment 172936 [details]
proposed patch

[WebKit Review Bot]

Comment on attachment 172936 [details]
proposed patch

Rejecting attachment 172936 [details] from commit-queue.

New failing tests:
fast/repaint/selection-after-delete.html
Full output: http://queues.webkit.org/results/14771173

[Yael]

Comment on attachment 172936 [details]
proposed patch

View in context: https://bugs.webkit.org/attachment.cgi?id=172936&action=review

> Source/WebKit2/UIProcess/efl/PageViewportControllerClientEfl.cpp:83
> +    drawingArea()->layerTreeCoordinatorProxy()->setContentsSize(contentsSize);

Isn't PageViewportController already setting the contentsSize?
Per https://bugs.webkit.org/show_bug.cgi?id=99850#c7 the viewport client should not be calling this explicitly.

[Yael]

(In reply to comment #0)
> At the time of adjusting positions for fixed layers, the locations of
> fixed layers are calculated incorrectly because the contentsSize is empty.

Do you have a test case for this? I'd like to take a look at this.
Thanks.

[Hurnjoo Lee]

(In reply to comment #3)
> (From update of attachment 172936 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=172936&action=review
> 
> > Source/WebKit2/UIProcess/efl/PageViewportControllerClientEfl.cpp:83
> > +    drawingArea()->layerTreeCoordinatorProxy()->setContentsSize(contentsSize);
> 
> Isn't PageViewportController already setting the contentsSize?
> Per https://bugs.webkit.org/show_bug.cgi?id=99850#c7 the viewport client should not be calling this explicitly.

Yes, PageViewportController is setting its own the contentSize already but LayerTreeRenderer's contentsSize is never set.
In Qt port, set LayerTreeRenderer's contentsSize at PageViewportControllerClientQt::didChangeContentsSize.

[Hurnjoo Lee]

Created attachment 173190 [details]
test case

[Hurnjoo Lee]

(In reply to comment #4)
> (In reply to comment #0)
> > At the time of adjusting positions for fixed layers, the locations of
> > fixed layers are calculated incorrectly because the contentsSize is empty.
> 
> Do you have a test case for this? I'd like to take a look at this.
> Thanks.

I've attached test case.
Also you can test on sites.
http://jquerymobile.com/demos/1.1.0/docs/toolbars/bars-fixed.html
http://www.noobcube.com/wp-content/uploads/demos/062709-fixed-header-footer/demo/

[Yael]

Comment on attachment 172936 [details]
proposed patch

Sorry it took me so long to get to this.
You are right, in the Qt port the same call is being made indirectly.
I'll try to cq the patch because the ews error did not seem related to your change.

[WebKit Review Bot]

Comment on attachment 172936 [details]
proposed patch

Clearing flags on attachment: 172936

Committed r134098: <http://trac.webkit.org/changeset/134098>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.