R=abarth@chromium.org, abarth

Review URL: https://codereview.chromium.org/1176053002

TBR=jackson

Review URL: https://codereview.chromium.org/1178523005

We can use const [] instead and we'll be sure we don't mutate this constant.

R=ianh@google.com

Review URL: https://codereview.chromium.org/1178623002.

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/1176763006

Our current static view_configuration design cannot work.

https://github.com/domokit/mojo/issues/231

R=ianh@google.com

Review URL: https://codereview.chromium.org/1169813009.

The app looks so much better with this on.

R=jackson@google.com, abarth@chromium.org

Review URL: https://codereview.chromium.org/1172413002.

Tweaked the layout in Scaffold a little: the Scaffold
accounts for the (Android) notification bar's height,
instead of the ToolBar. I think it would be best if
moving the app below the notification area were handled
at a lower level. Doing it in Scaffold seemed OK for the
moment, since one could put a ToolBar component anywhere
in an app.

Closes #208

R=ianh@google.com

Review URL: https://codereview.chromium.org/1171173003.

This also removes one bit of magic to make it more obvious what on is
going on during a sync, which should hopefully help.

Components have to decide if they support being stateful or not. If
they do, then they must implement syncFields() and have mutable
fields; if they don't, then they must have final fields. This isn't
particularly enforced, though.

This also renames _willSync() to _retainStatefulNodeIfPossible(), for
clarity, and fixes some minor style issues and one typo that was
breaking the drawer.

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/1174023003

Remove the usage of base::MessageLoopProxy since it has been deprecated.

BUG=465354
R=viettrungluu@chromium.org

Review URL: https://codereview.chromium.org/1174063002.

Patch from skyostil <skyostil@chromium.org>.

Renders interactive_flex and stocks demos fine. Could use more testing,
so I'm going to work on that after I land this.

R=abarth@chromium.org, abarth

Review URL: https://codereview.chromium.org/1173493003

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/1175763002.

Previously, the ink splashes would disappear immediately because the button
highlight would change the structure of the component hierarchy and we weren't
smart enough to retain the underlying RenderInkSplash. This CL is a point fix
for that issue by avoiding changing the structure of the component hierarchy. A
more complete fix would be to make Container smarter about how it syncs its
subcomponents.

R=eseidel@chromium.org, ianh@google.com

Review URL: https://codereview.chromium.org/1168323003.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1177563004.

This CL replaces the (non-working) components2 InkWell with some code based on
the ink_well example. There are at least two issues with the implementation:

1) The ink splash always starts at the center of the well because we don't have
   a facility for converting from global to local coordinates, which means we
   can't tell where the tap occurred in the local coordinates we need to use
   for painting.

2) When used inside a MenuItem, the in splash disappears shortly after
   starting, presumably because the button starts highlighting, which causes a
   component rebuild and somehow we lose the RenderInkWell instance.

I plan to address these issues in subsequent CLs.

R=ianh@google.com

Review URL: https://codereview.chromium.org/1172033003.

I also fixed image_cache to not even bother
to call ImageDecoder in the 404 case, but I also
fixed CanvasImageDecoder not to blow up if the
buffer is empty and its not able to allocate
an ImageDecoder.

Closes https://github.com/domokit/mojo/issues/229

R=abarth@chromium.org, jackson@google.com

Review URL: https://codereview.chromium.org/1170413002.

Update the native_viewport interface to allow specification of the surface configuration, currently only needed for and used by EGL on Android.

This also fixes an issue where eglChooseConfig was
only being called in InitializeOneOff, which is
only called once per process.  This CL makes
choosing the config happen once per GLSurface
instead, which will ultimately permit apps to
create multiple native_viewports with different
surface configurations on the same display.  The
eglDisplay object is still a global, though.

R=abarth@chromium.org, viettrungluu@chromium.org, jamesr@chromium.org

Review URL: https://codereview.chromium.org/1168993002.

StockRows are now a relatively simple application of flex layout.

The up/down arrows are now rendered within a circle, like
the original version. The arrows are a little bigger,
because.

R=ianh@google.com

Review URL: https://codereview.chromium.org/1165223005.

I accidentally removed two lines of code in https://codereview.chromium.org/1172173002 when merging upsteam before landing.

TBR=abarth

Review URL: https://codereview.chromium.org/1171163005

 - constructors first
 - have either a blank line at the top and bottom of each class, or no blank line in either place (it's still inconsistent about which we should do)
 - hide clamp()

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/1172173002

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/1171273002

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/1174683005

This would occur when bringing SkyDemo to the front after
having viewed a SkyView based demo and then paused
the SkyActivity (by going to the Android Launcher, etc.)

I also fixed deploy_domokit_site.py to ignore .gitignore files
found in packages.  We should actually probably ignore these
when creating the packages in the first place, but
this check seems valid regardless.  The .gitignore for
sky/lib/assets was causing the assets not to get pushed
to the demo site.

R=abarth@chromium.org, johnmccutchan@google.com

Review URL: https://codereview.chromium.org/1163323007.

TBR=abarth

Review URL: https://codereview.chromium.org/1173783002

Instead of triggering layout synchronously during AppView construction, we now
schedule the initial layout using a magic bootstrapping function.

TBR=ianh@google.com

Review URL: https://codereview.chromium.org/1172123003.

This CL implements a basic version of screen rotation where we just relayout at
the new size. In the future, we'll want a fancy version where we transition
between the two layouts.

R=ianh@google.com
BUG=

Review URL: https://codereview.chromium.org/1177563002.

Works like a charm, and is now much simpler thanks to John's work.

R=abarth@chromium.org, johnmccutchan@google.com

Review URL: https://codereview.chromium.org/1174683003.

Add asserts to prevent people from setting box.parentData.position or box.size except during layout.

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/1176623002

R=abarth@chromium.org, eseidel@chromium.org

Review URL: https://codereview.chromium.org/1170963003.

Also added a helper class to Sky framework for building shadow DrawLoopers.

R=abarth@chromium.org, eseidel@chromium.org, ianh@google.com

Review URL: https://codereview.chromium.org/1175683002.

Also considered "AnnotationNode", "LabelNode", "BadgeNode", but "TagNode" seemed simplest.

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/1169853008

This matches the material design spec.

R=ianh@google.com, jackson@chromium.org

Review URL: https://codereview.chromium.org/1169873003.

This would cause them to forget what the most recent child they'd been given was, with disastrous effects later.

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/1173683003

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1166223004.

TBR=ianh@google.com

Review URL: https://codereview.chromium.org/1169253002.

We already know how to talk to the network_service from Dart
via fetch.dart.  Might as well use that for Image loading
as well insetad of having ImageLoader do it.

As part of this I've renamed *ImageLoader to *ImageDecoder
and moved all the image loading logic into Dart.  This required
me to teach the idl system about mojo handles so that I could
pass the resulting MojoHandle from fetch.dart up through to
ImageDecoder.

R=abarth@chromium.org, jackson@google.com, hansmuller@google.com

Review URL: https://codereview.chromium.org/1173703002.

Also, I've fixed the mask color on the drawer.

R=ianh@google.com

Review URL: https://codereview.chromium.org/1168203004.

more ColorFilter types.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1163373007.

The FlexExpandingChild flex parameter is now a keyword as
well, which adds a small readability benefit.

R=ianh@google.com

Review URL: https://codereview.chromium.org/1172503004.

TBR=ianh@google.com

Review URL: https://codereview.chromium.org/1174623002.