Drop RenderObject::hasBlendMode

Drop RenderObject::hasBlendMode

Remove RenderObject::hasBlendMode in favor of RenderStyle::hasBlendMode.
Also remove the gating on the RuntimeEnabledFeatures feature in
RenderStyle, since this should be handled by the CSS parser already.
Gating also removed for the 'isolation' property.

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=185823

[fs, 2014-11-21 14:48:55]

fs@opera.com changed reviewers:
+ chrishtr@chromium.org, jchaffraix@chromium.org

[Julien - ping for review, 2014-11-21 16:29:21]

lgtm

https://codereview.chromium.org/746163002/diff/1/Source/core/rendering/Render...
File Source/core/rendering/RenderObject.h (left):

https://codereview.chromium.org/746163002/diff/1/Source/core/rendering/Render...
Source/core/rendering/RenderObject.h:597: bool hasBlendMode() const;
To be frank, I would prefer keeping hasBlendMode() but make it as simple as
hasFilter() above. It's pretty common in rendering/ to have such helper
functions. It would also remove a lot of the churn from this change.

That said, I won't fight strongly to keep it.

[fs, 2014-11-21 16:38:59]

https://codereview.chromium.org/746163002/diff/1/Source/core/rendering/Render...
File Source/core/rendering/RenderObject.h (left):

https://codereview.chromium.org/746163002/diff/1/Source/core/rendering/Render...
Source/core/rendering/RenderObject.h:597: bool hasBlendMode() const;
On 2014/11/21 16:29:20, Julien Chaffraix - PST wrote:
> To be frank, I would prefer keeping hasBlendMode() but make it as simple as
> hasFilter() above. It's pretty common in rendering/ to have such helper
> functions. It would also remove a lot of the churn from this change.
> 
> That said, I won't fight strongly to keep it.

Yeah, I can kind of agree to that - I wasn't really expecting this level of
churn...

Let's hear what Chris has to say.

[chrishtr, 2014-11-21 18:30:37]

On 2014/11/21 at 16:38:59, fs wrote:
>
https://codereview.chromium.org/746163002/diff/1/Source/core/rendering/Render...
> File Source/core/rendering/RenderObject.h (left):
> 
>
https://codereview.chromium.org/746163002/diff/1/Source/core/rendering/Render...
> Source/core/rendering/RenderObject.h:597: bool hasBlendMode() const;
> On 2014/11/21 16:29:20, Julien Chaffraix - PST wrote:
> > To be frank, I would prefer keeping hasBlendMode() but make it as simple as
> > hasFilter() above. It's pretty common in rendering/ to have such helper
> > functions. It would also remove a lot of the churn from this change.
> > 
> > That said, I won't fight strongly to keep it.
> 
> Yeah, I can kind of agree to that - I wasn't really expecting this level of
churn...
> 
> Let's hear what Chris has to say.

I prefer fewer extra redundant helper functions. I find they increase cognitive
load when reading the code. e.g. wondering whether the version on RenderObject
is different from the one on RenderStyle (and in some cases they have been).
It's also a version of code duplication.

[chrishtr, 2014-11-21 18:30:41]

lgtm

[fs, 2014-11-21 22:24:24]

On 2014/11/21 18:30:37, chrishtr wrote:
> On 2014/11/21 at 16:38:59, fs wrote:
> >
>
https://codereview.chromium.org/746163002/diff/1/Source/core/rendering/Render...
> > File Source/core/rendering/RenderObject.h (left):
> > 
> >
>
https://codereview.chromium.org/746163002/diff/1/Source/core/rendering/Render...
> > Source/core/rendering/RenderObject.h:597: bool hasBlendMode() const;
> > On 2014/11/21 16:29:20, Julien Chaffraix - PST wrote:
> > > To be frank, I would prefer keeping hasBlendMode() but make it as simple
as
> > > hasFilter() above. It's pretty common in rendering/ to have such helper
> > > functions. It would also remove a lot of the churn from this change.
> > > 
> > > That said, I won't fight strongly to keep it.
> > 
> > Yeah, I can kind of agree to that - I wasn't really expecting this level of
> churn...
> > 
> > Let's hear what Chris has to say.
> 
> I prefer fewer extra redundant helper functions. I find they increase
cognitive
> load when reading the code. e.g. wondering whether the version on RenderObject
> is different from the one on RenderStyle (and in some cases they have been).
> It's also a version of code duplication.

Thanks, I'll land this then.

[fs, 2014-11-21 22:24:32]

The CQ bit was checked by fs@opera.com

[commit-bot: I haz the power, 2014-11-21 22:25:28]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/746163002/1

[commit-bot: I haz the power, 2014-11-21 22:31:38]

Committed patchset #1 (id:1) as
https://src.chromium.org/viewvc/blink?view=rev&revision=185823