Canvas element that use the alpha attribute don't display correctly.

Canvas element that use the alpha attribute don't display correctly.

Current blink is delaying the creation of the backing store until drawing something.
When alpha is false, the backing store must be initialized to opaque black instead of transparent black.

BUG=352018
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=171408

[Justin Novosad, 2014-03-14 12:17:18]

The lazy allocation is a behavior we should try to keep.
You could fix this bug by changing HTMLCanvasElement::paint to handle the
!hasImageBuffer() case properly.

[Justin Novosad, 2014-03-14 12:18:00]

Also, please add a layout test for this.

[zino, 2014-03-16 08:34:47]

On 2014/03/14 12:18:00, junov wrote:
> Also, please add a layout test for this.

Thank you for review. :)

I've uploaded new patch set. Please review again.

[Justin Novosad, 2014-03-17 17:24:23]

https://codereview.chromium.org/200003002/diff/40001/Source/core/html/HTMLCan...
File Source/core/html/HTMLCanvasElement.cpp (right):

https://codereview.chromium.org/200003002/diff/40001/Source/core/html/HTMLCan...
Source/core/html/HTMLCanvasElement.cpp:321: createImageBuffer();
I think you can solve this problem without creating an ImageBuffer.

https://codereview.chromium.org/200003002/diff/40001/Source/core/html/HTMLCan...
Source/core/html/HTMLCanvasElement.cpp:333: }
You could add an else clause here that just draws a black rectangle

[zino, 2014-03-18 02:56:27]

On 2014/03/17 17:24:23, junov wrote:
>
https://codereview.chromium.org/200003002/diff/40001/Source/core/html/HTMLCan...
> File Source/core/html/HTMLCanvasElement.cpp (right):
> 
>
https://codereview.chromium.org/200003002/diff/40001/Source/core/html/HTMLCan...
> Source/core/html/HTMLCanvasElement.cpp:321: createImageBuffer();
> I think you can solve this problem without creating an ImageBuffer.
> 
>
https://codereview.chromium.org/200003002/diff/40001/Source/core/html/HTMLCan...
> Source/core/html/HTMLCanvasElement.cpp:333: }
> You could add an else clause here that just draws a black rectangle

Done.

Thank you for your detailed explanation :)
I finally understood your words.
(Maybe..? If my patch is still wrong, please let me know again.)

[Rik, 2014-03-18 03:34:20]

On 2014/03/18 02:56:27, zino wrote:
> 
> Thank you for your detailed explanation :)
> I finally understood your words.
> (Maybe..? If my patch is still wrong, please let me know again.)

It looks like you followed Justin's suggestion.
This patch looks good to me.

[Justin Novosad, 2014-03-21 12:31:46]

On 2014/03/18 03:34:20, Rik wrote:
> On 2014/03/18 02:56:27, zino wrote:
> > 
> > Thank you for your detailed explanation :)
> > I finally understood your words.
> > (Maybe..? If my patch is still wrong, please let me know again.)
> 
> It looks like you followed Justin's suggestion.
> This patch looks good to me.

lgtm.


Another problem that we can fix in a separate change is that there are probably
some CSS compositing and blending modes that could be set on the canvas element
that would requires us to draw a transparent black rect in the case where there
*is* an alpha channel.

[zino, 2014-04-11 16:31:22]

The CQ bit was checked by jinho.bang@samsung.com

[commit-bot: I haz the power, 2014-04-11 16:31:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jinho.bang@samsung.com/200003002/100001

[commit-bot: I haz the power, 2014-04-11 17:01:35]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-04-11 17:01:36]

Try jobs failed on following builders:
  tryserver.blink on mac_blink_rel

[zino, 2014-04-13 12:56:53]

The CQ bit was checked by jinho.bang@samsung.com

[commit-bot: I haz the power, 2014-04-13 12:57:03]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jinho.bang@samsung.com/200003002/120001

[commit-bot: I haz the power, 2014-04-13 13:24:10]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-04-13 13:24:11]

Try jobs failed on following builders:
  tryserver.blink on mac_blink_rel

[zino, 2014-04-13 14:34:06]

The CQ bit was checked by jinho.bang@samsung.com

[commit-bot: I haz the power, 2014-04-13 14:34:20]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jinho.bang@samsung.com/200003002/140001

[commit-bot: I haz the power, 2014-04-13 15:39:33]

Change committed as 171408