Prepare ui/compositor for WebLayer type change

Prepare ui/compositor for WebLayer type change

This is chromium-side preparation for https://bugs.webkit.org/show_bug.cgi?id=94229.
After that lands and rolls, I'll delete the #else branches.

BUG=none


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=152228

[jamesr, 2012-08-16 23:28:34]

Prepwork for https://bugs.webkit.org/show_bug.cgi?id=94174

[piman, 2012-08-16 23:55:30]

http://codereview.chromium.org/10832355/diff/1/ui/compositor/compositor.cc
File ui/compositor/compositor.cc (right):

http://codereview.chromium.org/10832355/diff/1/ui/compositor/compositor.cc#ne...
ui/compositor/compositor.cc:172: root_layer_->SetCompositor(NULL);
I'm pretty sure we still want to do this

http://codereview.chromium.org/10832355/diff/1/ui/compositor/layer.cc
File ui/compositor/layer.cc (right):

http://codereview.chromium.org/10832355/diff/1/ui/compositor/layer.cc#newcode307
ui/compositor/layer.cc:307: layer_mask ? layer_mask->web_layer() : 0);
nit: we use NULL rather than 0 in Chrome-land

http://codereview.chromium.org/10832355/diff/1/ui/compositor/layer.cc#newcode412
ui/compositor/layer.cc:412: WebKit::WebLayer* new_layer = 0;
nit: NULL

http://codereview.chromium.org/10832355/diff/1/ui/compositor/layer.cc#newcode420
ui/compositor/layer.cc:420: solid_color_layer_.reset();
Maybe we could reset all the concrete layers outside of the if in the common
path - that'll be less error-prone the day we need to add another layer type.

http://codereview.chromium.org/10832355/diff/1/ui/compositor/layer.cc#newcode803
ui/compositor/layer.cc:803: texture_layer_.reset();
Same as above, maybe reset all layer types outside of the test?
Not that this is only called at initialization time, so it's probably not even
necessary.

[jamesr, 2012-08-18 00:15:36]

PTAL.  The WebKit side looks ready to land, modulo this.

[piman, 2012-08-18 00:34:18]

lgtm

http://codereview.chromium.org/10832355/diff/1006/ui/compositor/layer.cc
File ui/compositor/layer.cc (right):

http://codereview.chromium.org/10832355/diff/1006/ui/compositor/layer.cc#newc...
ui/compositor/layer.cc:806: texture_layer_.reset();
nit: you can get rid of those 2 as well.

[jamesr, 2012-08-18 00:35:41]

http://codereview.chromium.org/10832355/diff/1006/ui/compositor/layer.cc
File ui/compositor/layer.cc (right):

http://codereview.chromium.org/10832355/diff/1006/ui/compositor/layer.cc#newc...
ui/compositor/layer.cc:806: texture_layer_.reset();
On 2012/08/18 00:34:18, piman wrote:
> nit: you can get rid of those 2 as well.

I see, I missed that this was private and just called once.  Will kill!

[commit-bot: I haz the power, 2012-08-18 00:49:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamesr@chromium.org/10832355/11001

[commit-bot: I haz the power, 2012-08-18 03:06:07]

Change committed as 152228

[piman, 2012-08-21 04:38:52]

http://codereview.chromium.org/10832355/diff/11001/ui/compositor/layer.cc
File ui/compositor/layer.cc (right):

http://codereview.chromium.org/10832355/diff/11001/ui/compositor/layer.cc#new...
ui/compositor/layer.cc:443: parent_->web_layer_->replaceChild(web_layer_,
new_layer);
The problem is that once we get here, the old concrete layer has been deleted
above, and web_layer_ points to crap.

[piman, 2012-08-21 04:46:22]

On 2012/08/21 04:38:52, piman wrote:
> http://codereview.chromium.org/10832355/diff/11001/ui/compositor/layer.cc
> File ui/compositor/layer.cc (right):
> 
>
http://codereview.chromium.org/10832355/diff/11001/ui/compositor/layer.cc#new...
> ui/compositor/layer.cc:443: parent_->web_layer_->replaceChild(web_layer_,
> new_layer);
> The problem is that once we get here, the old concrete layer has been deleted
> above, and web_layer_ points to crap.

https://chromiumcodereview.appspot.com/10827435/