paintingRoot of a renderobject should be the renderer of its enclosingLayer RenderLayer

paintingRoot of a renderobject should be the renderer of its enclosingLayer RenderLayer.

DragImage for a element is painted by providing appropriate node to Frame::nodeImage(Node* node) but the node may not have its own RL. So, paintInfo.shouldPaintWithinRoot(RenderObject*) fails at InlineBox::paint() for "static inline-block" elements which are painted in its enclosing RL(paintingRoot). 


Infrastructure for retrieving the drag image for pixel testing does not yet exist. So, ManualTests/drag-inline-block.html: Added.

BUG=101204
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=158165

[commit-bot: I haz the power, 2013-09-14 14:35:38]

No comments yet.

[pals, 2013-09-15 02:38:20]

PTAL

[eseidel, 2013-09-16 17:13:44]

Is this a behavior change?  Can we test this?

[pals, 2013-09-18 13:58:33]

Infrastructure for retrieving the drag image for pixel testing does not yet
exist. So, ManualTests/drag-inline-block.html: Added. Also updated the CL
description. Please review.

[eseidel, 2013-09-18 14:40:02]

ManualTests unfortunately basically never get run, so it's much better to test
this via a LayoutTest if possible.

https://codereview.chromium.org/23537039/diff/7001/ManualTests/drag-inline-bl...
File ManualTests/drag-inline-block.html (right):

https://codereview.chromium.org/23537039/diff/7001/ManualTests/drag-inline-bl...
ManualTests/drag-inline-block.html:31: <div draggable="true">Drag Me</div>
This test should document what the pass condition is.

"If the div above is draggable, the test passes."

But I would expect we could test this automatically by listening for drop events
from this div in another div?

testRunner.eventSender can do mouse moves, drags, etc.

[pals, 2013-09-20 13:20:30]

The issue is not with the actual drag operation. Drag-n-drop is working fine.
But drag-icon is not generated while drag is in progress. The image for the drag
icon is painted by webkit and forwarded to UI for displaying it. So, the
drag-icon is displayed by browser application. Currently, infrastructure for
retrieving the drag image for pixel testing does not yet exist. I did "git
blame" to check how previous author tested this part of code, found out that
they also tested manually (https://bugs.webkit.org/show_bug.cgi?id=110512). I
would like to analyze the feasibility to have pixel testing for drag image
though.

https://codereview.chromium.org/23537039/diff/7001/ManualTests/drag-inline-bl...
File ManualTests/drag-inline-block.html (right):

https://codereview.chromium.org/23537039/diff/7001/ManualTests/drag-inline-bl...
ManualTests/drag-inline-block.html:31: <div draggable="true">Drag Me</div>
On 2013/09/18 14:40:03, eseidel wrote:
> This test should document what the pass condition is.
> 
> "If the div above is draggable, the test passes."
> 
> But I would expect we could test this automatically by listening for drop
events
> from this div in another div?
> 
> testRunner.eventSender can do mouse moves, drags, etc.

Done. Added the pass condition as a comment on the test case.

[eseidel, 2013-09-20 18:13:12]

lgtm

[commit-bot: I haz the power, 2013-09-20 18:13:28]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sanjoy.pal@samsung.com/23537039/16001

[commit-bot: I haz the power, 2013-09-20 21:52:24]

Retried try job too often on win_blink_rel for step(s) webkit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_blink_...

[commit-bot: I haz the power, 2013-09-23 05:38:42]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sanjoy.pal@samsung.com/23537039/16001

[commit-bot: I haz the power, 2013-09-23 08:12:55]

Change committed as 158165