Fix background blending for some cases where it did not work at all. The fix adds the blendMode par…

Fix background blending for some cases where it did not work at all. The fix adds the blendMode parameter to the Image subclass drawing function call, instead of using the default normal blend mode. In some cases this involved adding a new BlendMode parameter to the draw* function. The fixed cases were:
 - background data URI SVG images, when a border-style was added to the container
 - tiled (repeated) gradient backgrounds
 - cross-fade image backgrounds

R=eseidel@chromium.org
BUG=268821
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=155773

[Horia Olaru, 2013-08-07 21:31:44]

Please have a look :)

https://codereview.chromium.org/22596003/diff/3001/LayoutTests/TestExpectations
File LayoutTests/TestExpectations (right):

https://codereview.chromium.org/22596003/diff/3001/LayoutTests/TestExpectatio...
LayoutTests/TestExpectations:1253: crbug.com/268821
css3/compositing/background-blend-mode-crossfade-image-gradient.html [
NeedsRebaseline ]
I have a couple of questions about these files:
- are the bots going to add pixel results when rebaselining, or should I add
some extra keyword
- do I need to submit another patch to remove these lines once they are
rebaselined, or is it done automatically

https://codereview.chromium.org/22596003/diff/3001/Source/core/platform/graph...
File Source/core/platform/graphics/ImageBuffer.h (right):

https://codereview.chromium.org/22596003/diff/3001/Source/core/platform/graph...
Source/core/platform/graphics/ImageBuffer.h:137: void
drawPattern(GraphicsContext*, const FloatRect&, const FloatSize&, const
FloatPoint&, CompositeOperator, const FloatRect&, BlendMode);
I would like to move the BlendMode parameter in a future patch, to be next to
the composite operator, because I think that is where it belongs. Would that be
acceptable?

Also, I am thinking of removing the default "BlendModeNormal" in the future.
There have been quite a few bugs because various draw* functions did not having
the BlendMode parameter passed in because of it. How does that sound? I haven't
played with this yet, so I am not sure if there are cases where someone would
actually use the default.

Where should I discuss such issues, before starting to work on them?

[eseidel, 2013-08-07 23:10:56]

lgtm

Seems reasonable.  Schenny is my go-to guy for graphics changes like this.

https://codereview.chromium.org/22596003/diff/3001/LayoutTests/TestExpectations
File LayoutTests/TestExpectations (right):

https://codereview.chromium.org/22596003/diff/3001/LayoutTests/TestExpectatio...
LayoutTests/TestExpectations:1253: crbug.com/268821
css3/compositing/background-blend-mode-crossfade-image-gradient.html [
NeedsRebaseline ]
Ojan's bot takes care of this automagically.  You're correct to add these lines.

On 2013/08/07 21:31:44, Horia Olaru wrote:
> I have a couple of questions about these files:
> - are the bots going to add pixel results when rebaselining, or should I add
> some extra keyword
> - do I need to submit another patch to remove these lines once they are
> rebaselined, or is it done automatically

[commit-bot: I haz the power, 2013-08-07 23:11:10]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/olaru@adobe.com/22596003/3001

[commit-bot: I haz the power, 2013-08-08 01:36:39]

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, 2013-08-08 09:26:10]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/olaru@adobe.com/22596003/3001

[commit-bot: I haz the power, 2013-08-08 14:19:41]

Change committed as 155773