Implement WebViewClient::createGraphicsContext3D (and support for DRT)

Implement WebViewClient::createGraphicsContext3D (and support for DRT)

This is chrome side of https://bugs.webkit.org/show_bug.cgi?id=76593
This also reverts r118249

BUG=99516
TEST=manually ran Chrome w/accelerated content, DRT, test_shell_tests.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=118543

[piman, 2012-01-19 20:06:09]

kbr: please review
tony: OWNERS
darin: in case you have comments about the new approach.

[Ken Russell (switch to Gerrit), 2012-01-19 20:31:20]

I don't see how this can land without breaking things. I think that
WebKitPlatformSupportImpl::createGraphicsContext3D() needs to be left in place
until WebKit has rolled its Chromium deps forward past this change. Then, after
WebKit rolls, we can remove it.

https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/glue/webkitpl...
File webkit/glue/webkitplatformsupport_impl.h (right):

https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/glue/webkitpl...
webkit/glue/webkitplatformsupport_impl.h:1: // Copyright (c) 2011 The Chromium
Authors. All rights reserved.
Accident?

https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/support/webki...
File webkit/support/webkit_support.h (right):

https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/support/webki...
webkit/support/webkit_support.h:102: WebKit::WebGraphicsContext3D*
CreateGraphicsContext3D(
Who calls this? Is the intent that TestWebViewDelegate or others call it?

[piman, 2012-01-19 20:48:26]

> I don't see how this can land without breaking things. I think that
> WebKitPlatformSupportImpl::createGraphicsContext3D() needs to be left in place
> until WebKit has rolled its Chromium deps forward past this change. Then,
after
> WebKit rolls, we can remove it.

We don't need that. createGraphicsContext3D() is still overridden by each
implementation. It works for the same reason it worked before my previous patch
(because essentially I'm reverting it here). The webkit side hasn't landed, so
the createGraphicsContext3D([3 params]) doesn't get called. The difference
between after this and r118248 (before the previous patch landed) is just a
bunch of functions to WebViewClient implementations, that no-one calls.

https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/glue/webkitpl...
File webkit/glue/webkitplatformsupport_impl.h (right):

https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/glue/webkitpl...
webkit/glue/webkitplatformsupport_impl.h:1: // Copyright (c) 2011 The Chromium
Authors. All rights reserved.
On 2012/01/19 20:31:21, kbr wrote:
> Accident?

Fixed.
I reverted my previous patch which bumped the year. Interestingly, the CL itself
didn't do the bump, I guess the CQ or the submit logic does that
automatically...

https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/support/webki...
File webkit/support/webkit_support.h (right):

https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/support/webki...
webkit/support/webkit_support.h:102: WebKit::WebGraphicsContext3D*
CreateGraphicsContext3D(
On 2012/01/19 20:31:21, kbr wrote:
> Who calls this? Is the intent that TestWebViewDelegate or others call it?

See webkit patch, this is for DRT, its WebViewHost calls this.

[Ken Russell (switch to Gerrit), 2012-01-19 21:04:39]

On 2012/01/19 20:48:26, piman wrote:
> > I don't see how this can land without breaking things. I think that
> > WebKitPlatformSupportImpl::createGraphicsContext3D() needs to be left in
place
> > until WebKit has rolled its Chromium deps forward past this change. Then,
> after
> > WebKit rolls, we can remove it.
> 
> We don't need that. createGraphicsContext3D() is still overridden by each
> implementation. It works for the same reason it worked before my previous
patch
> (because essentially I'm reverting it here). The webkit side hasn't landed, so
> the createGraphicsContext3D([3 params]) doesn't get called. The difference
> between after this and r118248 (before the previous patch landed) is just a
> bunch of functions to WebViewClient implementations, that no-one calls.

OK, I see. Thanks.

>
https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/glue/webkitpl...
> File webkit/glue/webkitplatformsupport_impl.h (right):
> 
>
https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/glue/webkitpl...
> webkit/glue/webkitplatformsupport_impl.h:1: // Copyright (c) 2011 The Chromium
> Authors. All rights reserved.
> On 2012/01/19 20:31:21, kbr wrote:
> > Accident?
> 
> Fixed.
> I reverted my previous patch which bumped the year. Interestingly, the CL
itself
> didn't do the bump, I guess the CQ or the submit logic does that
> automatically...
> 
>
https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/support/webki...
> File webkit/support/webkit_support.h (right):
> 
>
https://chromiumcodereview.appspot.com/9254035/diff/4001/webkit/support/webki...
> webkit/support/webkit_support.h:102: WebKit::WebGraphicsContext3D*
> CreateGraphicsContext3D(
> On 2012/01/19 20:31:21, kbr wrote:
> > Who calls this? Is the intent that TestWebViewDelegate or others call it?
> 
> See webkit patch, this is for DRT, its WebViewHost calls this.

Thanks.

LGTM then.

[tony, 2012-01-19 21:09:52]

rs LGTM.

https://chromiumcodereview.appspot.com/9254035/diff/2003/webkit/support/webki...
File webkit/support/webkit_support.cc (right):

https://chromiumcodereview.appspot.com/9254035/diff/2003/webkit/support/webki...
webkit/support/webkit_support.cc:385: default:
Nit: Remove the default.  If the enum changes, you'll get a compiler warning if
there's no default.

[piman, 2012-01-19 21:25:20]

Fixed nit and added missing include.

https://chromiumcodereview.appspot.com/9254035/diff/2003/webkit/support/webki...
File webkit/support/webkit_support.cc (right):

https://chromiumcodereview.appspot.com/9254035/diff/2003/webkit/support/webki...
webkit/support/webkit_support.cc:385: default:
On 2012/01/19 21:09:53, tony wrote:
> Nit: Remove the default.  If the enum changes, you'll get a compiler warning
if
> there's no default.

Done.

[commit-bot: I haz the power, 2012-01-19 21:25:44]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/piman@chromium.org/9254035/11001

[piman, 2012-01-20 20:38:52]

PTAL, updated patch to match the WebKit side (removed web_view parameter to
WebViewClient::createGraphicsContext3D).

[Ken Russell (switch to Gerrit), 2012-01-20 20:48:45]

lgtm

[commit-bot: I haz the power, 2012-01-20 21:30:07]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/piman@chromium.org/9254035/15001

[commit-bot: I haz the power, 2012-01-20 23:39:52]

Change committed as 118543