Support FrozenArray in idl_parser.py

Support FrozenArray in idl_parser.py

Spec: http://heycam.github.io/webidl/#idl-frozen-array
Almost the same as sequence<T>.

BUG=515920
Committed: https://crrev.com/cb5c1661b199801a9763919353702c9ea020be01
Cr-Commit-Position: refs/heads/master@{#344129}

[bashi, 2015-08-19 01:05:15]

bashi@chromium.org changed reviewers:
+ haraken@chromium.org, yukishiino@chromium.org

[bashi, 2015-08-19 01:05:15]

PTAL?

The WebIDL spec has introduced FrozenArray and removed Arrays. I don't think we
want to remove Arrays but some specs start using FrozenArray so I think we
should support FrozenArray.

Arrays, sequences and FrozenArrays have different semantics and I don't fully
understand the semantics, but in summary, arrays and FrozenArrays values are
references while sequences are passed by value.

That said, the blink's code generator doesn't handle the semantics correctly.
For Arrays and sequences, the code generator generates attribute getter
callbacks which always generate an array object. This means that we always pass
arrays or sequences by value.

My current plan is that support FrozenArray just the same as Arrays and
sequences. This doesn't follow the spec, but I think it's fine given that we
don't have serious interop problems so far.

[haraken, 2015-08-19 01:14:18]

LGTM

https://codereview.chromium.org/1295083004/diff/1/tools/idl_parser/idl_lexer.py
File tools/idl_parser/idl_lexer.py (right):

https://codereview.chromium.org/1295083004/diff/1/tools/idl_parser/idl_lexer....
tools/idl_parser/idl_lexer.py:83: 'FrozenArray' : 'FROZENARRAY',

Nit: This dictionary should be alphabetically sorted.

[bashi, 2015-08-19 01:19:04]

Thanks!

https://codereview.chromium.org/1295083004/diff/1/tools/idl_parser/idl_lexer.py
File tools/idl_parser/idl_lexer.py (right):

https://codereview.chromium.org/1295083004/diff/1/tools/idl_parser/idl_lexer....
tools/idl_parser/idl_lexer.py:83: 'FrozenArray' : 'FROZENARRAY',
On 2015/08/19 01:14:18, haraken wrote:
> 
> Nit: This dictionary should be alphabetically sorted.

Yeah, but this dict seems to use case-insensitive sorting. Not sure we want to
re-sort this.

[bashi, 2015-08-19 01:19:10]

The CQ bit was checked by bashi@chromium.org

[commit-bot: I haz the power, 2015-08-19 01:19:42]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1295083004/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1295083004/1

[commit-bot: I haz the power, 2015-08-19 02:11:34]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-08-19 02:12:12]

Patchset 1 (id:??) landed as
https://crrev.com/cb5c1661b199801a9763919353702c9ea020be01
Cr-Commit-Position: refs/heads/master@{#344129}