[bindings] support legacyiterable<> WebIDL syntax

[bindings] support legacyiterable<> WebIDL syntax

Add basic support for legacyiterable<> in order to implement the iterable
HTMLCollection.

BUG=538558
LOG=N
R=jsbell@chromium.org, phillipj@opera.com, jl@opera.com

[caitp (gmail), 2015-10-03 16:57:15]

PTAL when you get the chance. Adds the underlying WebIDL support needed to
support iterable HTMLCollection.

Python isn't my strongest language, but based on the test results it seems to do
the right thing, so I expect this is all that is needed (other than possibly
some additional IDL validation stuff).

If it's necessary to add support for iterable HTMLCollection at the same time as
this feature, just ask --- I think this can be safely tied to the intent thread
for iterable NodeLists at
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/VscIPcVtKrc

[caitp (gmail), 2015-10-03 16:59:54]

On 2015/10/03 16:57:15, caitp wrote:
> PTAL when you get the chance. Adds the underlying WebIDL support needed to
> support iterable HTMLCollection.
> 
> Python isn't my strongest language, but based on the test results it seems to
do
> the right thing, so I expect this is all that is needed (other than possibly
> some additional IDL validation stuff).
> 
> If it's necessary to add support for iterable HTMLCollection at the same time
as
> this feature, just ask --- I think this can be safely tied to the intent
thread
> for iterable NodeLists at
> https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/VscIPcVtKrc

This might not really be needed anyways though, I think a second attempt at this
is better.

[haraken, 2015-10-04 03:04:16]

haraken@chromium.org changed reviewers:
+ haraken@chromium.org

[haraken, 2015-10-04 03:04:17]

The implementation looks good.

https://codereview.chromium.org/1383093002/diff/1/third_party/WebKit/Source/b...
File third_party/WebKit/Source/bindings/tests/idls/core/TestInterface4.idl
(right):

https://codereview.chromium.org/1383093002/diff/1/third_party/WebKit/Source/b...
third_party/WebKit/Source/bindings/tests/idls/core/TestInterface4.idl:16:
[RuntimeEnabled=FeatureName] legacyiterable<any>;

Can we add this to any of existing test interface? (I just don't want to bloat
the size of test result files.)

[caitp (gmail), 2015-10-04 12:27:45]

On 2015/10/04 03:04:17, haraken wrote:
> The implementation looks good.
> 

There are  some making list  threads saying legacyiterable isn't really needed,
since you get the same behavior for integer length attribute + indexed property
getter. So, it's more invasive to implement that because there are more
interfaces like that, but maybe it's the right thing to do? I've started working
on an approach like that. Does that seem like a better approach?

>
https://codereview.chromium.org/1383093002/diff/1/third_party/WebKit/Source/b...
> File third_party/WebKit/Source/bindings/tests/idls/core/TestInterface4.idl
> (right):
> 
>
https://codereview.chromium.org/1383093002/diff/1/third_party/WebKit/Source/b...
> third_party/WebKit/Source/bindings/tests/idls/core/TestInterface4.idl:16:
> [RuntimeEnabled=FeatureName] legacyiterable<any>;
> 
> Can we add this to any of existing test interface? (I just don't want to bloat
> the size of test result files.)