Support [TypeChecking] on dependent/partial interfaces.

Support [TypeChecking] on dependent/partial interfaces.

Have the [TypeChecking] extended attribute on a partial interface
distribute down to its methods and attributes. This avoids having the
interface that it extends to declare the [TypeChecking] behavior that
the partial interface methods&attributes require.

The attribute's value is appended to any [TypeChecking] attribute
values already present on the partial interface members.

R=haraken,jl@opera.com,bashi
BUG=460550
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=190639

[sof, 2015-02-21 17:45:39]

sigbjornf@opera.com changed reviewers:
+ bashi@chromium.org, haraken@chromium.org, jl@opera.com

[sof, 2015-02-21 17:45:39]

Please take a look.

Seems like a useful little extension, needed for
https://codereview.chromium.org/925953003/.

afaict, this is the only extended attribute that distributes down from the
interface to the method/attribute like this. If accepted, I'll follow up and
update
http://www.chromium.org/developers/design-documents/idl-compiler#TOC-Dependen...

[sof, 2015-02-21 21:02:35]

If the case for this extension isn't sufficiently convincing, I'm fine with that
& will address the upstream differently. Up to you.

[haraken, 2015-02-21 22:39:48]

LGTM

https://codereview.chromium.org/938403007/diff/1/Source/bindings/tests/idls/c...
File Source/bindings/tests/idls/core/TestPartialInterface4.idl (right):

https://codereview.chromium.org/938403007/diff/1/Source/bindings/tests/idls/c...
Source/bindings/tests/idls/core/TestPartialInterface4.idl:6:
TypeChecking=Interface,

Just to confirm: This affects methods/attributes of TestPartialInterface4 but
doesn't affect methods/attributes of TestInterfaceDocument, right?

[Jens Widell, 2015-02-21 22:52:15]

LGTM

[sof, 2015-02-22 14:04:19]

New patchsets have been uploaded after l-g-t-m from
haraken@chromium.org,jl@opera.com

[sof, 2015-02-22 14:05:16]

https://codereview.chromium.org/938403007/diff/1/Source/bindings/tests/idls/c...
File Source/bindings/tests/idls/core/TestPartialInterface4.idl (right):

https://codereview.chromium.org/938403007/diff/1/Source/bindings/tests/idls/c...
Source/bindings/tests/idls/core/TestPartialInterface4.idl:6:
TypeChecking=Interface,
On 2015/02/21 22:39:48, haraken wrote:
> 
> Just to confirm: This affects methods/attributes of TestPartialInterface4 but
> doesn't affect methods/attributes of TestInterfaceDocument, right?

Correct. I've altered the partial interface to make verifying that easier.

Also noticed that the merging of these attributes was a bit
unexpected&surprising: the interface-level extended attribute value would
override and replace any local ones. To continue this "principle of least
astonishment"-like effort if nothing else, I tidied that up. pta(nother)l?

[bashi, 2015-02-23 01:54:11]

LGTM

https://codereview.chromium.org/938403007/diff/20001/Source/bindings/scripts/...
File Source/bindings/scripts/interface_dependency_resolver.py (right):

https://codereview.chromium.org/938403007/diff/20001/Source/bindings/scripts/...
Source/bindings/scripts/interface_dependency_resolver.py:283: def toList(x):
nit: toList() -> to_list()

[haraken, 2015-02-23 03:28:16]

LGTM

[sof, 2015-02-23 06:07:03]

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

[sof, 2015-02-23 06:08:55]

https://codereview.chromium.org/938403007/diff/20001/Source/bindings/scripts/...
File Source/bindings/scripts/interface_dependency_resolver.py (right):

https://codereview.chromium.org/938403007/diff/20001/Source/bindings/scripts/...
Source/bindings/scripts/interface_dependency_resolver.py:283: def toList(x):
On 2015/02/23 01:54:11, bashi1 wrote:
> nit: toList() -> to_list()

thanks, done.

[sof, 2015-02-23 06:09:00]

The CQ bit was checked by sigbjornf@opera.com

[commit-bot: I haz the power, 2015-02-23 06:09:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/938403007/40001

[Jens Widell, 2015-02-23 07:07:52]

LGTM

This CL affects code generation for various overloads of
Window.createImageBitmap(). In that case it doesn't really matter though, since
there's an overload resolution method doing type-checking before invoking the
specific overloads.

[commit-bot: I haz the power, 2015-02-23 07:22:21]

Committed patchset #3 (id:40001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=190639

[sof, 2015-02-23 07:34:56]

https://sites.google.com/a/chromium.org/dev/developers/design-documents/idl-c...
updated.