Refactor cc scrollbar layers to separate solid-color vs desktop.

Refactor cc scrollbar layers architecture.

This CL separates the solid-color scrollbar functionality into its own class structure, resulting in the following hierarchy:

ScrollbarLayerInterface (pure-virtual)
 |
 --------  PaintedScrollbarLayer (for outer-viewport)
 |
 --------- SolidColorScrollbarLayer (for inner-viewport)            
                                    (Android, pinch-zoom)

ScrollbarLayerImplBase
 |
 ----- PaintedScrollbarLayerImpl (for outer-viewport)
 |
 ----- SolidColorScrollbarLayerImpl (for inner-viewport)
                                    (Android, pinch-zoom)

In this architecture, solid-color scrollbars can live side-by side with regular desktop scrollbars, and no longer require a WebScrollbar in order to be created. The scrolling LayerImpl's only know about ScrollbarLayerImplBase, and the TreeSynchronizer & LTI only deal with *Base layer types.

The ScrollbarLayerImplBase layer is not pure virtual, but rather attempts to collect any common functionality, and the result is much simpler classes for SolidColorScrollbarLayerImpl.

Work remaining:

1) Once the pinch-viewport layers are in place in CC, we can get rid of vertical adjust (I think), and so LTI::UpdateSolidColorScrollbars() can go away. When this goes away, some of the interface for ScrollbarLayerImplBase can be simplified as well.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=220637

[wjmaclean, 2013-07-04 00:35:57]

I think this is ready for a look ... I've fixed the tests, and removed the
settings solid_color_scrollbar_thickness_DIP and solid_color_scrollbar_color,
and greatly reduced the number of cases of solid_color_scrollbars (this will
take some extra work to do, and may need to wait until after the cc
pinch-viewport layers patch lands ...)

[jamesr, 2013-07-04 00:49:18]

I don't think this is quite the right way to share code here.  Solid color
scrollbars don't have any reason to inherit from ContentsScalingLayer if they
are their own type.  A lot of code is duplicated between the impl-side classes
instead.  Orientation() is now a virtual call even though it's a
construction-time invariant for all types of ScrollbarLayers.

I don't really see anything worth sharing in cc::ScrollbarLayerBase - do you
even need a base class?  What is it giving you?

https://codereview.chromium.org/18341009/diff/6001/cc/layers/solid_color_scro...
File cc/layers/solid_color_scrollbar_layer_impl.cc (right):

https://codereview.chromium.org/18341009/diff/6001/cc/layers/solid_color_scro...
cc/layers/solid_color_scrollbar_layer_impl.cc:47: gfx::Rect
SolidColorScrollbarLayerImpl::ComputeThumbQuadRect() const {
this function has a really unfortunate amount of duplication with the base class

[wjmaclean, 2013-07-04 17:14:53]

> I don't think this is quite the right way to share code here.  Solid color
> scrollbars don't have any reason to inherit from ContentsScalingLayer if they
are their own type.

OK, I've revised this so the base class does not inherit from
ContentsScalingLayer, and each sub-class separately inherits what it needs
(solid color scrollbars inherit from Layer, and regular scrollbars inherit from
ContentsScalingLayer).

> A lot of code is duplicated between the impl-side classes
instead.

I've reduced as much of this duplication as I reasonably can in the latest
patch, pushing common code into the ScrollbarLayerImplBase class.

> Orientation() is now a virtual call even though it's a construction-time
invariant for all types of >ScrollbarLayers.

I've move this call to the base class so it's not virtual.

> I don't really see anything worth sharing in cc::ScrollbarLayerBase - do you
even need a base class?  What is it giving you?

It gives us poly-morphism in TreeSynchronizer ... since the tree may contain
both types of scrollbars, and calls them via a templated function, this
simplifies things considerably.

Since the templated function works on both Layer/LayerImpl scrollbars, it is
important to keep the mechanism the same on both threads.

[tfarina, 2013-07-04 17:32:13]

Is this a bug (filed on crbug.com) you are working on that needs BUG=
line to be filled up?



--
Thiago

[tfarina, 2013-07-04 17:34:45]

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
File webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc (right):

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:20:
cc::ScrollbarOrientation convertOrientation(
CamelCase

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:22: if
(orientation == WebKit::WebScrollbar::Horizontal)
I know if/else is clear, but I'd just use the ternary operator ?: here.

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:28: const int
solid_color_scrollbar_thumb_thickness = 3;
kFoo

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:29: const
SkColor solid_color_scrollbar_color = SkColorSetARGB(128, 128, 128, 128);
kFoo

[wjmaclean, 2013-07-04 17:50:11]

Thanks!

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
File webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc (right):

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:20:
cc::ScrollbarOrientation convertOrientation(
On 2013/07/04 17:34:45, tfarina wrote:
> CamelCase

Done.

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:22: if
(orientation == WebKit::WebScrollbar::Horizontal)
On 2013/07/04 17:34:45, tfarina wrote:
> I know if/else is clear, but I'd just use the ternary operator ?: here.

Done.

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:28: const int
solid_color_scrollbar_thumb_thickness = 3;
On 2013/07/04 17:34:45, tfarina wrote:
> kFoo

Done.

https://codereview.chromium.org/18341009/diff/14002/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:29: const
SkColor solid_color_scrollbar_color = SkColorSetARGB(128, 128, 128, 128);
On 2013/07/04 17:34:45, tfarina wrote:
> kFoo

Done.

[tfarina, 2013-07-04 22:30:25]

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_impl_base.h (right):

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:1: // Copyright 2012 The Chromium Authors.
All rights reserved.
update the year in new files to 2013. applies to other files as well.

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:10: #include "cc/layers/layer.h"
do you need this include? if you do, can't you forward declare it?

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:22: float CurrentPos() const { return
current_pos_; }
unix_hacker(), applies until SetMaximum

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:35: ScrollbarOrientation Orientation()
const { return orientation_; }
unix_hacker

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:51: visible_to_total_length_ratio_(1.f),
can you do all this initialization in the source file instead? I guess clang
isn't complaining because they are all POD?

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:61: int scroll_layer_id_;
these data member variables should be private.

See:
http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml?showone=Inheri...

You have exposed other data members in other class in protected section too.

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:71: };
DISALLOW_ ....

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:74: #endif  //
CC_LAYERS_SCROLLBAR_LAYER_IMPL_BASE_H_
tinynit: add blank line above.

[wjmaclean, 2013-07-05 17:06:25]

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_impl_base.h (right):

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:1: // Copyright 2012 The Chromium Authors.
All rights reserved.
On 2013/07/04 22:30:26, tfarina wrote:
> update the year in new files to 2013. applies to other files as well.

Done.

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:10: #include "cc/layers/layer.h"
On 2013/07/04 22:30:26, tfarina wrote:
> do you need this include? if you do, can't you forward declare it?

It's needed for the constant Layer::INVALID_ID, but I moved the constructor
implementation to the .cc file, some we no longer need it here.

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:22: float CurrentPos() const { return
current_pos_; }
On 2013/07/04 22:30:26, tfarina wrote:
> unix_hacker(), applies until SetMaximum

Done.

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:35: ScrollbarOrientation Orientation()
const { return orientation_; }
On 2013/07/04 22:30:26, tfarina wrote:
> unix_hacker

Done.

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:51: visible_to_total_length_ratio_(1.f),
On 2013/07/04 22:30:26, tfarina wrote:
> can you do all this initialization in the source file instead? I guess clang
> isn't complaining because they are all POD?

Done.

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:61: int scroll_layer_id_;
On 2013/07/04 22:30:26, tfarina wrote:
> these data member variables should be private.
> 
> See:
>
http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml?showone=Inheri...
> 
> You have exposed other data members in other class in protected section too.

Done.

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:71: };
On 2013/07/04 22:30:26, tfarina wrote:
> DISALLOW_ ....

Done.

https://codereview.chromium.org/18341009/diff/12002/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.h:74: #endif  //
CC_LAYERS_SCROLLBAR_LAYER_IMPL_BASE_H_
On 2013/07/04 22:30:26, tfarina wrote:
> tinynit: add blank line above.

Done.

[enne (OOO), 2013-08-12 23:15:01]

https://codereview.chromium.org/18341009/diff/25001/cc/layers/scrollbar_layer.cc
File cc/layers/scrollbar_layer.cc (right):

https://codereview.chromium.org/18341009/diff/25001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.cc:122: // wjm: remove this?
No.  Mac can have overlay scrollbars that are not solid color scrollbars.

https://codereview.chromium.org/18341009/diff/25001/cc/trees/tree_synchronize...
File cc/trees/tree_synchronizer.cc (right):

https://codereview.chromium.org/18341009/diff/25001/cc/trees/tree_synchronize...
cc/trees/tree_synchronizer.cc:152: ScrollbarLayerType* scrollbar_layer =
layer->ToScrollbarLayerBase();
I agree with jamesr that there's really not that much in ScrollbarLayerBase
going on that needs to be shared.

One suggestion would be to have a ScrollbarLayerInterface that exposed
Orientation() and ScrollLayerId() and have both ScrollbarLayer and
SolidColorScrollbarLayer implement that interface.  Then
SolidColorScrollbarLayer would no longer need to inherit from
ContentsScalingLayer, but you wouldn't have to complicate TreeSynchronizer with
different kinds of scrollbars.

https://codereview.chromium.org/18341009/diff/25001/webkit/renderer/composito...
File webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc (right):

https://codereview.chromium.org/18341009/diff/25001/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:17: #if
defined(OS_ANDROID)
Please no platform #ifdefs.  Checking the setting for what kind of scrollbars to
create seems preferable.

[wjmaclean, 2013-08-14 18:19:44]

PTAL

aelias@, can you confirm I'm correctly preserving Android's expectations for
solid color scrollbars.

https://codereview.chromium.org/18341009/diff/25001/cc/layers/scrollbar_layer.cc
File cc/layers/scrollbar_layer.cc (right):

https://codereview.chromium.org/18341009/diff/25001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.cc:122: // wjm: remove this?
On 2013/08/12 23:15:01, enne wrote:
> No.  Mac can have overlay scrollbars that are not solid color scrollbars.

Done.

https://codereview.chromium.org/18341009/diff/25001/cc/trees/tree_synchronize...
File cc/trees/tree_synchronizer.cc (right):

https://codereview.chromium.org/18341009/diff/25001/cc/trees/tree_synchronize...
cc/trees/tree_synchronizer.cc:152: ScrollbarLayerType* scrollbar_layer =
layer->ToScrollbarLayerBase();
On 2013/08/12 23:15:01, enne wrote:
> I agree with jamesr that there's really not that much in ScrollbarLayerBase
> going on that needs to be shared.
> 
> One suggestion would be to have a ScrollbarLayerInterface that exposed
> Orientation() and ScrollLayerId() and have both ScrollbarLayer and
> SolidColorScrollbarLayer implement that interface.  Then
> SolidColorScrollbarLayer would no longer need to inherit from
> ContentsScalingLayer, but you wouldn't have to complicate TreeSynchronizer
with
> different kinds of scrollbars.

Sure, that sounds good. I've converted to an interface as you suggest (let me
know if the way I've done this seems reasonable).

https://codereview.chromium.org/18341009/diff/25001/webkit/renderer/composito...
File webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc (right):

https://codereview.chromium.org/18341009/diff/25001/webkit/renderer/composito...
webkit/renderer/compositor_bindings/web_scrollbar_layer_impl.cc:17: #if
defined(OS_ANDROID)
On 2013/08/12 23:15:01, enne wrote:
> Please no platform #ifdefs.  Checking the setting for what kind of scrollbars
to
> create seems preferable.

Fixed.

I now check an existing flag, --enable-overlay-scrollbars, which in ToT seems to
do what we want (though it's plumbed to Blink but doesn't seem to be used there,
so maybe a separate flag would be better?). I moved it's definition to
cc::switches so it would be available to web_scrollbar_layer_impl, let me know
if this is ok.

https://codereview.chromium.org/18341009/diff/32001/content/public/common/con...
File content/public/common/content_switches.cc (right):

https://codereview.chromium.org/18341009/diff/32001/content/public/common/con...
content/public/common/content_switches.cc:8: 
Ooops, forgot to remove this, it will be gone in the next upload.

[wjmaclean, 2013-08-14 20:48:26]

https://codereview.chromium.org/18341009/diff/32001/content/renderer/gpu/rend...
File content/renderer/gpu/render_widget_compositor.cc (right):

https://codereview.chromium.org/18341009/diff/32001/content/renderer/gpu/rend...
content/renderer/gpu/render_widget_compositor.cc:277: #elif !defined(OS_MACOSX)
This should be a separate #if instead of an #elif now.

[enne (OOO), 2013-08-14 21:01:09]

I like this direction better.  Can you update the description to match what
you're doing here?

jamesr: Does this address your concerns about ScrollbarLayer inheritance
messiness?

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer.cc
File cc/layers/scrollbar_layer.cc (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.cc:73: return orientation_;
Does this need a separate variable? What about just scrollbar_->Orientation()?

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.cc:281: // WJM: is this specific to solid color
scrollbars? probably!!
This is in ScrollbarLayer, so no longer has anything to do with solid color
scrollbars?

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer.h
File cc/layers/scrollbar_layer.h (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.h:34: virtual int scroll_layer_id() const OVERRIDE;
style nit: don't use hacker_style for virtual functions.  Here and elsewhere.

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_impl.h (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl.h:62: virtual int ThumbThickness() const
OVERRIDE;
Comment about what class's functions these are overriding?

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl.h:77: ScrollbarOrientation orientation_;
Is this shadowing ScrollbarLayerImplBase::orientation_? Same for
scroll_layer_id_ and vertical_adjust_?

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_unittest.cc (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_unittest.cc:230: const int thumb_thickness = 3;
style nit: constants in chromium are kThumbThickness.

[aelias_OOO_until_Jul13, 2013-08-14 21:26:07]

The approach looks fine in general.  You'll need to rebase on another big
scrollbar change that just landed: https://codereview.chromium.org/21917004/ . 
But it's largely orthogonal to what you're doing so the conflicts should be easy
to resolve.

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer.h
File cc/layers/scrollbar_layer.h (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.h:20: class CC_EXPORT ScrollbarLayer : public
ScrollbarLayerInterface,
This name sounds like it's a base class of all scrollbars.  How about renaming
it to PaintedScrollbarLayer?

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_impl.cc (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl.cc:70: if (draw_mode ==
DRAW_MODE_RESOURCELESS_SOFTWARE)
This "resourceless software" mode only arises in Android WebView which should
always fall into solid color scrollbar forcing, so let's delete this early out
entirely.

https://codereview.chromium.org/18341009/diff/32001/content/renderer/gpu/rend...
File content/renderer/gpu/render_widget_compositor.cc (right):

https://codereview.chromium.org/18341009/diff/32001/content/renderer/gpu/rend...
content/renderer/gpu/render_widget_compositor.cc:269:
settings.solid_color_scrollbar_color =
This will fail to compile.  Android WebView needs a way to hide scrollbars.

[danakj, 2013-08-14 21:31:30]

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_impl.cc (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl.cc:70: if (draw_mode ==
DRAW_MODE_RESOURCELESS_SOFTWARE)
On 2013/08/14 21:26:08, aelias wrote:
> This "resourceless software" mode only arises in Android WebView which should
> always fall into solid color scrollbar forcing, so let's delete this early out
> entirely.

Maybe DCHECK it?

[jamesr, 2013-08-14 22:29:06]

Can you update the patch description to reflect the new class hierarchy?  It's
hard to extract that from the patch.

[wjmaclean, 2013-08-15 20:09:39]

PTAL.

I've updated the description as requested.

This CL has been through two major re-bases in the last 24 hours, so there may
be minor tidyness issues to fix (ordering of functions in .h/.cc files for
example).

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer.cc
File cc/layers/scrollbar_layer.cc (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.cc:73: return orientation_;
On 2013/08/14 21:01:09, enne wrote:
> Does this need a separate variable? What about just scrollbar_->Orientation()?

Done.

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.cc:281: // WJM: is this specific to solid color
scrollbars? probably!!
On 2013/08/14 21:01:09, enne wrote:
> This is in ScrollbarLayer, so no longer has anything to do with solid color
> scrollbars?

Yeah, picked this one off in a recent rebase ... gone!

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer.h
File cc/layers/scrollbar_layer.h (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.h:20: class CC_EXPORT ScrollbarLayer : public
ScrollbarLayerInterface,
On 2013/08/14 21:26:08, aelias wrote:
> This name sounds like it's a base class of all scrollbars.  How about renaming
> it to PaintedScrollbarLayer?

I agree.

Since that's going to touch a lot of files, and this CL is already getting
pretty big, can we do that in a follow-on CL?

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.h:34: virtual int scroll_layer_id() const OVERRIDE;
On 2013/08/14 21:01:09, enne wrote:
> style nit: don't use hacker_style for virtual functions.  Here and elsewhere.

Done.

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_impl.cc (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl.cc:70: if (draw_mode ==
DRAW_MODE_RESOURCELESS_SOFTWARE)
On 2013/08/14 21:26:08, aelias wrote:
> This "resourceless software" mode only arises in Android WebView which should
> always fall into solid color scrollbar forcing, so let's delete this early out
> entirely.

Done.

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl.cc:70: if (draw_mode ==
DRAW_MODE_RESOURCELESS_SOFTWARE)
On 2013/08/14 21:31:30, danakj wrote:
> On 2013/08/14 21:26:08, aelias wrote:
> > This "resourceless software" mode only arises in Android WebView which
should
> > always fall into solid color scrollbar forcing, so let's delete this early
out
> > entirely.
> 
> Maybe DCHECK it?

Done.

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_impl.h (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl.h:62: virtual int ThumbThickness() const
OVERRIDE;
On 2013/08/14 21:01:09, enne wrote:
> Comment about what class's functions these are overriding?

Done.

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl.h:77: ScrollbarOrientation orientation_;
On 2013/08/14 21:01:09, enne wrote:
> Is this shadowing ScrollbarLayerImplBase::orientation_? Same for
> scroll_layer_id_ and vertical_adjust_?

Done. Good catch!

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_unittest.cc (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_unittest.cc:230: const int thumb_thickness = 3;
On 2013/08/14 21:01:09, enne wrote:
> style nit: constants in chromium are kThumbThickness.

Done.

https://codereview.chromium.org/18341009/diff/32001/content/public/common/con...
File content/public/common/content_switches.cc (right):

https://codereview.chromium.org/18341009/diff/32001/content/public/common/con...
content/public/common/content_switches.cc:8: 
On 2013/08/14 18:19:44, wjmaclean wrote:
> Ooops, forgot to remove this, it will be gone in the next upload.

Done.

https://codereview.chromium.org/18341009/diff/32001/content/renderer/gpu/rend...
File content/renderer/gpu/render_widget_compositor.cc (right):

https://codereview.chromium.org/18341009/diff/32001/content/renderer/gpu/rend...
content/renderer/gpu/render_widget_compositor.cc:269:
settings.solid_color_scrollbar_color =
On 2013/08/14 21:26:08, aelias wrote:
> This will fail to compile.  Android WebView needs a way to hide scrollbars.

OK. I've fixed this by re-implementing the --solid-color-scrollbar-color switch
we considered adding in the other patch, and removing the --hide-scrollbars
switch.

https://codereview.chromium.org/18341009/diff/32001/content/renderer/gpu/rend...
content/renderer/gpu/render_widget_compositor.cc:277: #elif !defined(OS_MACOSX)
On 2013/08/14 20:48:27, wjmaclean wrote:
> This should be a separate #if instead of an #elif now.

Done.

[aelias_OOO_until_Jul13, 2013-08-15 21:22:31]

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer.h
File cc/layers/scrollbar_layer.h (right):

https://codereview.chromium.org/18341009/diff/32001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer.h:20: class CC_EXPORT ScrollbarLayer : public
ScrollbarLayerInterface,
On 2013/08/15 20:09:40, wjmaclean wrote:
> On 2013/08/14 21:26:08, aelias wrote:
> > This name sounds like it's a base class of all scrollbars.  How about
renaming
> > it to PaintedScrollbarLayer?
> 
> I agree.
> 
> Since that's going to touch a lot of files, and this CL is already getting
> pretty big, can we do that in a follow-on CL?

It's fine with me to delay the change.

https://codereview.chromium.org/18341009/diff/46001/cc/trees/layer_tree_setti...
File cc/trees/layer_tree_settings.cc (right):

https://codereview.chromium.org/18341009/diff/46001/cc/trees/layer_tree_setti...
cc/trees/layer_tree_settings.cc:33:
solid_color_scrollbar_color(SkColorSetARGB(128, 128, 128, 128)),
Please set this back to SK_ColorWHITE so that platforms don't rely on the
default (in case desired color between platforms starts to diverge at some
point).

https://codereview.chromium.org/18341009/diff/46001/content/browser/android/c...
File content/browser/android/content_startup_flags.cc (right):

https://codereview.chromium.org/18341009/diff/46001/content/browser/android/c...
content/browser/android/content_startup_flags.cc:65:
parsed_command_line->AppendSwitch(cc::switches::kEnableOverlayScrollbars);
Please add the solid-color flag with 0x80808080 here in case the default
changes.

[aelias_OOO_until_Jul13, 2013-08-15 21:23:40]

[+joth for android_webview/ OWNERS rubberstamp]

[enne (OOO), 2013-08-15 21:29:26]

https://codereview.chromium.org/18341009/diff/46001/content/renderer/gpu/rend...
File content/renderer/gpu/render_widget_compositor.cc (right):

https://codereview.chromium.org/18341009/diff/46001/content/renderer/gpu/rend...
content/renderer/gpu/render_widget_compositor.cc:77: bool StringToSkColor(const
std::string& str, SkColor& color) {
Why does scrollbar color need to move to be a command line flag and not just a
LayerTreeSetting as it was before? The previous code seemed more straightforward
than parsing hex colors.

[joth, 2013-08-15 21:40:07]

lgtm % enne's comment on using command line.

https://codereview.chromium.org/18341009/diff/46001/android_webview/lib/main/...
File android_webview/lib/main/aw_main_delegate.cc (right):

https://codereview.chromium.org/18341009/diff/46001/android_webview/lib/main/...
android_webview/lib/main/aw_main_delegate.cc:61:
cl->AppendSwitch(hide_scrollbar_switch.c_str());
nit: c_str() not needed

[jamesr, 2013-08-15 22:34:23]

NAK on the switches changes.  I'm not sure what you are trying to accomplish
with them.

I haven't looked at the rest closely yet.

https://codereview.chromium.org/18341009/diff/46001/content/child/runtime_fea...
File content/child/runtime_features.cc (right):

https://codereview.chromium.org/18341009/diff/46001/content/child/runtime_fea...
content/child/runtime_features.cc:140: if
(command_line.HasSwitch(cc::switches::kEnableOverlayScrollbars))
this is a bad change.  cc::switches should only change cc's behavior, not the
behavior of Blink

https://codereview.chromium.org/18341009/diff/46001/content/public/common/con...
File content/public/common/content_switches.cc (left):

https://codereview.chromium.org/18341009/diff/46001/content/public/common/con...
content/public/common/content_switches.cc:427: // Enables overlay scrollbars on
Aura or Linux. Does nothing on Mac.
why are you deleting this? enabling overlay scrollbars on aura/linux is a
recently added and useful flag.

https://codereview.chromium.org/18341009/diff/46001/content/renderer/gpu/rend...
File content/renderer/gpu/render_widget_compositor.cc (right):

https://codereview.chromium.org/18341009/diff/46001/content/renderer/gpu/rend...
content/renderer/gpu/render_widget_compositor.cc:298: #if !defined(OS_MACOSX)
this change doesn't seem right. the logic before this patch was:

If on android
  set pile of flags
otherwise, if not on mac (i.e. on windows/cros/linux)
  if have overlay scrollbar flags
    set different set of flags

The new logic will check for the kEnableOverlayScrollbars even on android, which
doesn't make a lot of sense - android always uses overlay scrollbars so it
shouldn't be conditional on a switch

[tony, 2013-08-15 22:52:45]

To provide some context, I'm experimenting with adding overlay scrollbars for
use on ChromeOS.  The current work (not much so far) is behind the
--enable-overlay-scrollbars command line switch.

I've opened https://code.google.com/p/chromium/issues/detail?id=274010 to track
my progress.

Sorry for not mentioning this more broadly.

[wjmaclean, 2013-08-16 13:34:15]

PTAL

jamesr@ - As I hope this CL is converging, can you take a more detailed look
now? Please let me know which proposed alternative to moving the
--enable-overlay-scrollbars switch you prefer.

enne@ - Please let me know if you're ok with my explanation of why I propose
setting the solid scrollbar color via a flag. If not, let's gchat about what
you'd prefer to see.

https://codereview.chromium.org/18341009/diff/46001/android_webview/lib/main/...
File android_webview/lib/main/aw_main_delegate.cc (right):

https://codereview.chromium.org/18341009/diff/46001/android_webview/lib/main/...
android_webview/lib/main/aw_main_delegate.cc:61:
cl->AppendSwitch(hide_scrollbar_switch.c_str());
On 2013/08/15 21:40:07, joth wrote:
> nit: c_str() not needed

Did not know that, thanks!

Done.

https://codereview.chromium.org/18341009/diff/46001/cc/trees/layer_tree_setti...
File cc/trees/layer_tree_settings.cc (right):

https://codereview.chromium.org/18341009/diff/46001/cc/trees/layer_tree_setti...
cc/trees/layer_tree_settings.cc:33:
solid_color_scrollbar_color(SkColorSetARGB(128, 128, 128, 128)),
On 2013/08/15 21:22:32, aelias wrote:
> Please set this back to SK_ColorWHITE so that platforms don't rely on the
> default (in case desired color between platforms starts to diverge at some
> point).

Done.

https://codereview.chromium.org/18341009/diff/46001/content/browser/android/c...
File content/browser/android/content_startup_flags.cc (right):

https://codereview.chromium.org/18341009/diff/46001/content/browser/android/c...
content/browser/android/content_startup_flags.cc:65:
parsed_command_line->AppendSwitch(cc::switches::kEnableOverlayScrollbars);
On 2013/08/15 21:22:32, aelias wrote:
> Please add the solid-color flag with 0x80808080 here in case the default
> changes.

Done.

https://codereview.chromium.org/18341009/diff/46001/content/child/runtime_fea...
File content/child/runtime_features.cc (right):

https://codereview.chromium.org/18341009/diff/46001/content/child/runtime_fea...
content/child/runtime_features.cc:140: if
(command_line.HasSwitch(cc::switches::kEnableOverlayScrollbars))
On 2013/08/15 22:34:24, jamesr wrote:
> this is a bad change.  cc::switches should only change cc's behavior, not the
> behavior of Blink

OK. Is it preferable to add another, CC-specific flag? Or, should I add a static
function in some neutral location (i.e. that WebScrollbarLayerImpl can see
without a layering violation) that can see this flag when it is moved back to
content switches.

*** I will hold off changing this until you let me know your preference.

https://codereview.chromium.org/18341009/diff/46001/content/public/common/con...
File content/public/common/content_switches.cc (left):

https://codereview.chromium.org/18341009/diff/46001/content/public/common/con...
content/public/common/content_switches.cc:427: // Enables overlay scrollbars on
Aura or Linux. Does nothing on Mac.
On 2013/08/15 22:34:24, jamesr wrote:
> why are you deleting this? enabling overlay scrollbars on aura/linux is a
> recently added and useful flag.

Of course it's useful ... I was just proposing moving the flag's location. Given
your comments elsewhere I know propose we find another way that doesn't require
moving this flag out of content_switches.cc.

https://codereview.chromium.org/18341009/diff/46001/content/renderer/gpu/rend...
File content/renderer/gpu/render_widget_compositor.cc (right):

https://codereview.chromium.org/18341009/diff/46001/content/renderer/gpu/rend...
content/renderer/gpu/render_widget_compositor.cc:77: bool StringToSkColor(const
std::string& str, SkColor& color) {
On 2013/08/15 21:29:26, enne wrote:
> Why does scrollbar color need to move to be a command line flag and not just a
> LayerTreeSetting as it was before? The previous code seemed more
straightforward
> than parsing hex colors.

Actually, it will remain in LayerTreeSettings, the question is just a matter of
how to initialise that value. At present only Android uses these scrollbars, but
that's about to change. The current initialisation is done inside an #if
defined(OS_ANDROID). Adding a flag for setting the color gives us a way to
customise the color to be used across platforms, and at runtime (so we can get
rid of another flag, --hide-scrollbars, used by android-WebView. And, it removes
a few lines of code from the #if block.

I'm open to alternative ways of doing this, but this idea had positive feedback
in the previous (now abandoned) IsSolidColor() patch
(https://codereview.chromium.org/17550008/).

https://codereview.chromium.org/18341009/diff/46001/content/renderer/gpu/rend...
content/renderer/gpu/render_widget_compositor.cc:298: #if !defined(OS_MACOSX)
On 2013/08/15 22:34:24, jamesr wrote:
> this change doesn't seem right. the logic before this patch was:
> 
> If on android
>   set pile of flags
> otherwise, if not on mac (i.e. on windows/cros/linux)
>   if have overlay scrollbar flags
>     set different set of flags
> 
> The new logic will check for the kEnableOverlayScrollbars even on android,
which
> doesn't make a lot of sense - android always uses overlay scrollbars so it
> shouldn't be conditional on a switch

OK, fair enough. I knew this was a trade off, but thought I would propose it to
see what people thought. I can remove the setting of the switch from Androids
startup flags, and add the code back into the #if defined(OS_ANDROID) block.

Done.

[wjmaclean, 2013-08-28 01:03:13]

Sorry for the delay ... the rebasing to account for the re-naming was a little
ugly. :-)

Please take a look ... everything runs, but I may do a little more tidying
(function ordering may have bounced around a bit during the rebase).

[enne (OOO), 2013-08-28 18:22:08]

https://codereview.chromium.org/18341009/diff/63001/android_webview/lib/main/...
File android_webview/lib/main/aw_main_delegate.cc (left):

https://codereview.chromium.org/18341009/diff/63001/android_webview/lib/main/...
android_webview/lib/main/aw_main_delegate.cc:59:
cl->AppendSwitch(switches::kHideScrollbars);
I don't understand what changing kHideScrollbars to kSolidColorScrollbarColor
gets you, other than churn and an untested StringToSkColor function.  I prefer
the previous version, which also keeps all the color-setting code in one place.

https://codereview.chromium.org/18341009/diff/63001/cc/layers/painted_scrollb...
File cc/layers/painted_scrollbar_layer.h (right):

https://codereview.chromium.org/18341009/diff/63001/cc/layers/painted_scrollb...
cc/layers/painted_scrollbar_layer.h:77: int thumb_thickness_;
Can you initialize these in the ctor?

https://codereview.chromium.org/18341009/diff/63001/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_impl_base.cc (right):

https://codereview.chromium.org/18341009/diff/63001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.cc:34: // Don't intersect with the bounds as
in layerRectToContentRect() because
layerRect => LayerRect

https://codereview.chromium.org/18341009/diff/63001/cc/layers/solid_color_scr...
File cc/layers/solid_color_scrollbar_layer_impl.h (right):

https://codereview.chromium.org/18341009/diff/63001/cc/layers/solid_color_scr...
cc/layers/solid_color_scrollbar_layer_impl.h:22: virtual scoped_ptr<LayerImpl>
CreateLayerImpl(LayerTreeImpl* tree_impl)
Comment about what class this is overriding.

https://codereview.chromium.org/18341009/diff/63001/cc/layers/solid_color_scr...
cc/layers/solid_color_scrollbar_layer_impl.h:36: virtual int ThumbThickness()
const OVERRIDE;
Comment about what class this is overriding.

[wjmaclean, 2013-08-28 20:51:46]

All comments addressed, PTAL.

https://codereview.chromium.org/18341009/diff/63001/android_webview/lib/main/...
File android_webview/lib/main/aw_main_delegate.cc (left):

https://codereview.chromium.org/18341009/diff/63001/android_webview/lib/main/...
android_webview/lib/main/aw_main_delegate.cc:59:
cl->AppendSwitch(switches::kHideScrollbars);
On 2013/08/28 18:22:08, enne wrote:
> I don't understand what changing kHideScrollbars to kSolidColorScrollbarColor
> gets you, other than churn and an untested StringToSkColor function.  I prefer
> the previous version, which also keeps all the color-setting code in one
place.

Done.

Removed. If there's any desire to control this color via a switch that can
happen in a separate CL.

https://codereview.chromium.org/18341009/diff/63001/cc/layers/painted_scrollb...
File cc/layers/painted_scrollbar_layer.h (right):

https://codereview.chromium.org/18341009/diff/63001/cc/layers/painted_scrollb...
cc/layers/painted_scrollbar_layer.h:77: int thumb_thickness_;
On 2013/08/28 18:22:08, enne wrote:
> Can you initialize these in the ctor?

Done.

https://codereview.chromium.org/18341009/diff/63001/cc/layers/scrollbar_layer...
File cc/layers/scrollbar_layer_impl_base.cc (right):

https://codereview.chromium.org/18341009/diff/63001/cc/layers/scrollbar_layer...
cc/layers/scrollbar_layer_impl_base.cc:34: // Don't intersect with the bounds as
in layerRectToContentRect() because
On 2013/08/28 18:22:08, enne wrote:
> layerRect => LayerRect

Done.

https://codereview.chromium.org/18341009/diff/63001/cc/layers/solid_color_scr...
File cc/layers/solid_color_scrollbar_layer_impl.h (right):

https://codereview.chromium.org/18341009/diff/63001/cc/layers/solid_color_scr...
cc/layers/solid_color_scrollbar_layer_impl.h:22: virtual scoped_ptr<LayerImpl>
CreateLayerImpl(LayerTreeImpl* tree_impl)
On 2013/08/28 18:22:08, enne wrote:
> Comment about what class this is overriding.

Done.

https://codereview.chromium.org/18341009/diff/63001/cc/layers/solid_color_scr...
cc/layers/solid_color_scrollbar_layer_impl.h:36: virtual int ThumbThickness()
const OVERRIDE;
On 2013/08/28 18:22:08, enne wrote:
> Comment about what class this is overriding.

Done.

[enne (OOO), 2013-08-28 21:09:08]

lgtm

In the future, could you upload your rebases separately from your changes?  For
large patches, it's really hard to view the diff otherwise.

[wjmaclean, 2013-08-28 21:11:27]

On 2013/08/28 21:09:08, enne wrote:
> lgtm
> 
> In the future, could you upload your rebases separately from your changes? 
For
> large patches, it's really hard to view the diff otherwise.

Thanks ... I will upload re-bases separately in future.

[wjmaclean, 2013-08-28 21:33:13]

[email protected]: Please review changes in 

content/{browser|public} please?

[jam, 2013-08-29 05:10:13]

what's the purpose of kUseSolidColorScrollbars? it's always being set in the
browser, so why not just hard code setting that preference to true for android?

[wjmaclean, 2013-08-29 13:58:18]

>what's the purpose of kUseSolidColorScrollbars? it's always 
>being set in the browser, so why not just hard code setting 
>that preference to true for android?

Removed, PTAL?

[jam, 2013-08-29 16:22:07]

On 2013/08/29 13:58:18, wjmaclean wrote:
> >what's the purpose of kUseSolidColorScrollbars? it's always 
> >being set in the browser, so why not just hard code setting 
> >that preference to true for android?
> 
> Removed, PTAL?

thanks. now you don't need me for review since no more changes in
content/public, right?

[wjmaclean, 2013-08-29 16:26:07]

On 2013/08/29 16:22:07, jam wrote:
> On 2013/08/29 13:58:18, wjmaclean wrote:
> > >what's the purpose of kUseSolidColorScrollbars? it's always 
> > >being set in the browser, so why not just hard code setting 
> > >that preference to true for android?
> > 
> > Removed, PTAL?
> 
> thanks. now you don't need me for review since no more changes in
> content/public, right?

Are you able to do content/browser, or should I ask Sky?

[wjmaclean, 2013-08-29 16:50:15]

On 2013/08/29 16:22:07, jam wrote:
> On 2013/08/29 13:58:18, wjmaclean wrote:
> > >what's the purpose of kUseSolidColorScrollbars? it's always 
> > >being set in the browser, so why not just hard code setting 
> > >that preference to true for android?
> > 
> > Removed, PTAL?
> 
> thanks. now you don't need me for review since no more changes in
> content/public, right?

Actually, there's still one line in
content/public/common/common_param_traits_macros.h

[sky, 2013-08-29 16:51:16]

content/browser LGTM

[jamesr, 2013-08-29 18:40:42]

content/renderer/ and webkit/ lgtm

[Chris Evans, 2013-08-29 18:45:04]

On 2013/08/29 18:40:42, jamesr wrote:
> content/renderer/ and webkit/ lgtm

LGTM

[commit-bot: I haz the power, 2013-08-29 18:48:13]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/18341009/88001

[wjmaclean, 2013-08-29 18:58:51]

joi@ Could you look at https://codereview.chromium.org/18341009/ for 

common_param_traits_macro.h 

please?

[jam, 2013-08-29 23:29:28]

On 2013/08/29 18:58:51, wjmaclean wrote:
> joi@ Could you look at https://codereview.chromium.org/18341009/ for 
> 
> common_param_traits_macro.h 
> 
> please?

lgtm

[commit-bot: I haz the power, 2013-08-30 01:07:25]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/18341009/88001

[commit-bot: I haz the power, 2013-08-30 01:07:31]

Failed to apply patch for cc/layers/painted_scrollbar_layer.cc:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file cc/layers/painted_scrollbar_layer.cc
  Hunk #4 FAILED at 110.
  Hunk #5 succeeded at 147 (offset 9 lines).
  Hunk #6 succeeded at 177 (offset 9 lines).
  Hunk #7 succeeded at 203 (offset 9 lines).
  Hunk #8 succeeded at 227 (offset 9 lines).
  1 out of 8 hunks FAILED -- saving rejects to file
cc/layers/painted_scrollbar_layer.cc.rej

Patch:       cc/layers/painted_scrollbar_layer.cc
Index: cc/layers/painted_scrollbar_layer.cc
diff --git a/cc/layers/painted_scrollbar_layer.cc
b/cc/layers/painted_scrollbar_layer.cc
index
7c590d004fdb4c39b1f2f8c34eb43f72f64e5c08..4cc7bf8f250c768e93c7adcafed080615cf5bdb3
100644
--- a/cc/layers/painted_scrollbar_layer.cc
+++ b/cc/layers/painted_scrollbar_layer.cc
@@ -37,13 +37,19 @@ PaintedScrollbarLayer::PaintedScrollbarLayer(
     scoped_ptr<Scrollbar> scrollbar,
     int scroll_layer_id)
     : scrollbar_(scrollbar.Pass()),
-      scroll_layer_id_(scroll_layer_id) {
+      scroll_layer_id_(scroll_layer_id),
+      thumb_thickness_(scrollbar_->ThumbThickness()),
+      thumb_length_(scrollbar_->ThumbLength()) {
   if (!scrollbar_->IsOverlay())
     SetShouldScrollOnMainThread(true);
 }
 
 PaintedScrollbarLayer::~PaintedScrollbarLayer() {}
 
+int PaintedScrollbarLayer::ScrollLayerId() const {
+  return scroll_layer_id_;
+}
+
 void PaintedScrollbarLayer::SetScrollLayerId(int id) {
   if (id == scroll_layer_id_)
     return;
@@ -56,7 +62,7 @@ bool PaintedScrollbarLayer::OpacityCanAnimateOnImplThread()
const {
   return scrollbar_->IsOverlay();
 }
 
-ScrollbarOrientation PaintedScrollbarLayer::Orientation() const {
+ScrollbarOrientation PaintedScrollbarLayer::orientation() const {
   return scrollbar_->Orientation();
 }
 
@@ -66,9 +72,6 @@ int PaintedScrollbarLayer::MaxTextureSize() {
 }
 
 float PaintedScrollbarLayer::ClampScaleToMaxTextureSize(float scale) {
-  if (layer_tree_host()->settings().solid_color_scrollbars)
-    return scale;
-
   // If the scaled content_bounds() is bigger than the max texture size of the
   // device, we need to clamp it by rescaling, since content_bounds() is used
   // below to set the texture size.
@@ -107,14 +110,9 @@ void PaintedScrollbarLayer::PushPropertiesTo(LayerImpl*
layer) {
   PaintedScrollbarLayerImpl* scrollbar_layer =
       static_cast<PaintedScrollbarLayerImpl*>(layer);
 
-  if (layer_tree_host() &&
-      layer_tree_host()->settings().solid_color_scrollbars) {
-    scrollbar_layer->SetThumbThickness(scrollbar_->ThumbThickness());
-  } else {
-    scrollbar_layer->SetThumbThickness(thumb_thickness_);
-  }
+  scrollbar_layer->SetThumbThickness(thumb_thickness_);
   scrollbar_layer->SetThumbLength(thumb_length_);
-  if (Orientation() == HORIZONTAL) {
+  if (orientation() == HORIZONTAL) {
     scrollbar_layer->SetTrackStart(
         track_rect_.x() - location_.x());
     scrollbar_layer->SetTrackLength(track_rect_.width());
@@ -135,7 +133,7 @@ void PaintedScrollbarLayer::PushPropertiesTo(LayerImpl*
layer) {
   needs_push_properties_ = true;
 }
 
-PaintedScrollbarLayer* PaintedScrollbarLayer::ToScrollbarLayer() {
+ScrollbarLayerInterface* PaintedScrollbarLayer::ToScrollbarLayer() {
   return this;
 }
 
@@ -165,7 +163,7 @@ gfx::Rect
PaintedScrollbarLayer::ScrollbarLayerRectToContentRect(
 
 gfx::Rect PaintedScrollbarLayer::OriginThumbRect() const {
   gfx::Size thumb_size;
-  if (Orientation() == HORIZONTAL) {
+  if (orientation() == HORIZONTAL) {
     thumb_size =
         gfx::Size(scrollbar_->ThumbLength(), scrollbar_->ThumbThickness());
   } else {
@@ -191,8 +189,7 @@ bool PaintedScrollbarLayer::Update(ResourceUpdateQueue*
queue,
   gfx::Rect scaled_track_rect = ScrollbarLayerRectToContentRect(
       gfx::Rect(location_, bounds()));
 
-  if (layer_tree_host()->settings().solid_color_scrollbars ||
-      track_rect_.IsEmpty() || scaled_track_rect.IsEmpty())
+  if (track_rect_.IsEmpty() || scaled_track_rect.IsEmpty())
     return false;
 
   {
@@ -216,7 +213,6 @@ bool PaintedScrollbarLayer::Update(ResourceUpdateQueue*
queue,
 scoped_refptr<UIResourceBitmap> PaintedScrollbarLayer::RasterizeScrollbarPart(
     gfx::Rect rect,
     ScrollbarPart part) {
-  DCHECK(!layer_tree_host()->settings().solid_color_scrollbars);
   DCHECK(!rect.size().IsEmpty());
 
   scoped_refptr<UIResourceBitmap> bitmap =

[Jói, 2013-08-30 08:07:13]

Removing myself from reviewers list as jam@ has already given you the l-g-t-m
you need.

[commit-bot: I haz the power, 2013-08-30 13:57:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/18341009/107001

[commit-bot: I haz the power, 2013-08-30 14:21:07]

Sorry for I got bad news for ya.
Compile failed with a clobber build on android_clang_dbg.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android_cl...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2013-08-30 14:54:58]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/18341009/74001

[commit-bot: I haz the power, 2013-08-30 15:52:05]

Sorry for I got bad news for ya.
Compile failed with a clobber build on win.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win&number...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2013-08-30 16:07:06]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/18341009/143001

[commit-bot: I haz the power, 2013-08-30 20:25:35]

Change committed as 220637