Fix bug with using enums as default values in mojom. We were previously

Fix bug with using enums as default values in mojom. We were previously
generating invalid JavaScript bindings. We now translate an enum value like
FOO_VALUE in a mojom file to proper_namespace.EnumName.FOO_VALUE in JS.

I also fixed enum declarations in JS so an enum value can be defined in terms
of itself, e.g.
  enum Foo { FOO_1, FOO_2 = FOO_1 * 2 };

BUG=320082
R=darin@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=251417

[darin (slow to review), 2014-02-13 22:55:21]

LGTM

https://codereview.chromium.org/164873002/diff/1/mojo/apps/js/bindings/sample...
File mojo/apps/js/bindings/sample_service_unittests.js (right):

https://codereview.chromium.org/164873002/diff/1/mojo/apps/js/bindings/sample...
mojo/apps/js/bindings/sample_service_unittests.js:124:
expect(full.thing.color).toBe(imported2.Color.COLOR_BLACK);
Note: it seems like it would be nice if we generated "Color.BLACK" here instead.
Having to type the word "color" twice seems of no value.

Maybe we could address this by changing the way we generate the enum values in
C++. For C++ we could add the prefix so we don't have to figure out how to
subtract it for JS?

For example the following mojom:

  enum Shape {
    CIRCLE,
    ...
  };

Could generate the following C++:

  enum Shape {
    SHAPE_CIRCLE,
    ...
  };

And then in JS it would be:

  Shape.CIRCLE

Looks good?

[Matt Perry, 2014-02-14 02:01:06]

The CQ bit was checked by mpcomplete@chromium.org

[commit-bot: I haz the power, 2014-02-14 02:03:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/mpcomplete@chromium.org/164873002/220001

[commit-bot: I haz the power, 2014-02-14 03:20:20]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-02-14 03:20:21]

Try jobs failed on following builders: linux_chromium_rel

[Matt Perry, 2014-02-14 19:43:42]

https://codereview.chromium.org/164873002/diff/1/mojo/apps/js/bindings/sample...
File mojo/apps/js/bindings/sample_service_unittests.js (right):

https://codereview.chromium.org/164873002/diff/1/mojo/apps/js/bindings/sample...
mojo/apps/js/bindings/sample_service_unittests.js:124:
expect(full.thing.color).toBe(imported2.Color.COLOR_BLACK);
On 2014/02/13 22:55:21, darin wrote:
> Note: it seems like it would be nice if we generated "Color.BLACK" here
instead.
> Having to type the word "color" twice seems of no value.
> 
> Maybe we could address this by changing the way we generate the enum values in
> C++. For C++ we could add the prefix so we don't have to figure out how to
> subtract it for JS?
> 
> For example the following mojom:
> 
>   enum Shape {
>     CIRCLE,
>     ...
>   };
> 
> Could generate the following C++:
> 
>   enum Shape {
>     SHAPE_CIRCLE,
>     ...
>   };
> 
> And then in JS it would be:
> 
>   Shape.CIRCLE
> 
> Looks good?

That's definitely doable. It feels a bit.. dirty to me, somehow. I'll file a bug
on it and maybe I'll warm up to it.

[Matt Perry, 2014-02-14 19:45:08]

The CQ bit was checked by mpcomplete@chromium.org

[commit-bot: I haz the power, 2014-02-14 19:45:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/mpcomplete@chromium.org/164873002/440001

[commit-bot: I haz the power, 2014-02-14 20:34:18]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-02-14 20:34:18]

Retried try job too often on linux_chromeos for step(s) interactive_ui_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_chro...

[Matt Perry, 2014-02-14 21:08:40]

Committed patchset #3 manually as r251417 (presubmit successful).