Fix render bounds of iframes on devices with a pixel ratio != 1.

Fix render bounds of iframes on devices with a pixel ratio != 1.

BUG=
R=abarth@chromium.org

Committed: https://chromium.googlesource.com/external/mojo/+/c42ef2ebecd17d6ed78e9fc04416127e416b4d60

[tonyg, 2015-02-04 16:13:34]

tonyg@chromium.org changed reviewers:
+ abarth@chromium.org

[tonyg, 2015-02-04 16:16:06]

A couple of questions on my first sky change:

- Do we have a pattern for testing this?
- The contentView's DPR isn't set, so I got it from the root View. This works
fine, but I'm wondering whether it's important to make sure DPR populates to all
Views and then get scaleFactor directly from the contentView.

[abarth-chromium, 2015-02-04 16:41:11]

https://codereview.chromium.org/896233002/diff/1/sky/engine/core/rendering/Re...
File sky/engine/core/rendering/RenderIFrame.cpp (right):

https://codereview.chromium.org/896233002/diff/1/sky/engine/core/rendering/Re...
sky/engine/core/rendering/RenderIFrame.cpp:37: const float scaleFactor =
scaleFactor -> devicePixelRatio

[abarth-chromium, 2015-02-04 16:44:02]

On 2015/02/04 at 16:16:06, tonyg wrote:
> A couple of questions on my first sky change:
> 
> - Do we have a pattern for testing this?

We don't have a good way to test this.  We can add support to sky/tools/tester
for dumping the mojo::View hierarchy, including bounds.  It's something we want
eventually, but I'm not going to make you build all that for this CL.

> - The contentView's DPR isn't set, so I got it from the root View. This works
fine, but I'm wondering whether it's important to make sure DPR populates to all
Views and then get scaleFactor directly from the contentView.

That's probably a bug in the view_manager.  When we create a view in the client
library, we should probably initialize the viewport_metrics to match the parent
view given that they're going to be in the same viewport.

[abarth-chromium, 2015-02-04 16:44:06]

lgtm

[tonyg, 2015-02-04 16:55:27]

New patchsets have been uploaded after l-g-t-m from abarth@chromium.org

[tonyg, 2015-02-04 16:56:27]

On 2015/02/04 16:44:02, abarth wrote:
> On 2015/02/04 at 16:16:06, tonyg wrote:
> > A couple of questions on my first sky change:
> > 
> > - Do we have a pattern for testing this?
> 
> We don't have a good way to test this.  We can add support to sky/tools/tester
> for dumping the mojo::View hierarchy, including bounds.  It's something we
want
> eventually, but I'm not going to make you build all that for this CL.
> 

Ack

> > - The contentView's DPR isn't set, so I got it from the root View. This
works
> fine, but I'm wondering whether it's important to make sure DPR populates to
all
> Views and then get scaleFactor directly from the contentView.
> 
> That's probably a bug in the view_manager.  When we create a view in the
client
> library, we should probably initialize the viewport_metrics to match the
parent
> view given that they're going to be in the same viewport.

Makes sense, I added a FIXME.

[tonyg, 2015-02-04 16:56:37]

https://codereview.chromium.org/896233002/diff/1/sky/engine/core/rendering/Re...
File sky/engine/core/rendering/RenderIFrame.cpp (right):

https://codereview.chromium.org/896233002/diff/1/sky/engine/core/rendering/Re...
sky/engine/core/rendering/RenderIFrame.cpp:37: const float scaleFactor =
On 2015/02/04 16:41:11, abarth wrote:
> scaleFactor -> devicePixelRatio

Done.

[tonyg, 2015-02-04 16:57:13]

Committed patchset #2 (id:20001) manually as
c42ef2ebecd17d6ed78e9fc04416127e416b4d60 (presubmit successful).