Update HTMLFormControlsCollection's named getter to behave according to spec

Update HTMLFormControlsCollection's named getter to behave according to spec

Update HTMLFormControlsCollection's named getter to behave according to spec:
http://www.whatwg.org/specs/web-apps/current-work/multipage/common-dom-interfaces.html#htmlformcontrolscollection-0

1. namedItem() should be marked as named getter in IDL as per the spec.
   This also means we can get rid of the anonymous named getter in our
   IDL.
2. The named getter should be enumerable as per Web IDL, meaning that
   elements id / names should be enumerated. The supported property names
   are spec'd at:
   http://www.whatwg.org/specs/web-apps/current-work/multipage/common-dom-interfaces.html#htmlformcontrolscollection-0
   Firefox 26 and IE11 both show the names while enumerating. However, IE11
   does not show the ids when enumerating, which is incorrect.
3. The argument to namedItem() named getter should be mandatory. This is
   consistent with the spec, Firefox 26 and IE11.

This CL also gets rid of the custom bindings code needed for the named getter
by improving support for union types in the bindings generator.

R=haraken, arv, tkent
BUG=341269
TEST=fast/dom/htmlformcontrolscollection-enumerated-properties.html

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=166688

[Nils Barth (inactive), 2014-02-07 01:40:42]

https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
File Source/core/html/HTMLFormControlsCollection.idl (right):

https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
Source/core/html/HTMLFormControlsCollection.idl:27: [ImplementedAs=namedGetter]
legacycaller getter (RadioNodeList or Element) namedItem(DOMString name); //
shadows inherited namedItem()
legacycaller is currently a nop;
do you want to omit this from this CL,
and instead add and implement in a followup?

[Inactive, 2014-02-07 01:52:38]

https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
File Source/core/html/HTMLFormControlsCollection.idl (right):

https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
Source/core/html/HTMLFormControlsCollection.idl:27: [ImplementedAs=namedGetter]
legacycaller getter (RadioNodeList or Element) namedItem(DOMString name); //
shadows inherited namedItem()
On 2014/02/07 01:40:43, Nils Barth wrote:
> legacycaller is currently a nop;
> do you want to omit this from this CL,
> and instead add and implement in a followup?

I know it is a noop but the parser doesn't mind it and it is present in the
spec. I figured it didn't hurt to add it. That said, I can remove it before
landing if you think it is not worth adding until it is implemented.

[Nils Barth (inactive), 2014-02-07 01:54:55]

https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
File Source/core/html/HTMLFormControlsCollection.idl (right):

https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
Source/core/html/HTMLFormControlsCollection.idl:27: [ImplementedAs=namedGetter]
legacycaller getter (RadioNodeList or Element) namedItem(DOMString name); //
shadows inherited namedItem()
On 2014/02/07 01:52:38, Chris Dumez wrote:
> On 2014/02/07 01:40:43, Nils Barth wrote:
> > legacycaller is currently a nop;
> > do you want to omit this from this CL,
> > and instead add and implement in a followup?
> 
> I know it is a noop but the parser doesn't mind it and it is present in the
> spec. I figured it didn't hurt to add it. That said, I can remove it before
> landing if you think it is not worth adding until it is implemented.

I'd rather not include undocumented nops, since it can be misleading.
It's fine to include if you want to document it though,
preferably with a link to a bug, like:
// FIXME: legacycaller currently nop: http://crbug.com/???

How about this?

[Inactive, 2014-02-07 02:09:40]

On 2014/02/07 01:54:55, Nils Barth wrote:
>
https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
> File Source/core/html/HTMLFormControlsCollection.idl (right):
> 
>
https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
> Source/core/html/HTMLFormControlsCollection.idl:27:
[ImplementedAs=namedGetter]
> legacycaller getter (RadioNodeList or Element) namedItem(DOMString name); //
> shadows inherited namedItem()
> On 2014/02/07 01:52:38, Chris Dumez wrote:
> > On 2014/02/07 01:40:43, Nils Barth wrote:
> > > legacycaller is currently a nop;
> > > do you want to omit this from this CL,
> > > and instead add and implement in a followup?
> > 
> > I know it is a noop but the parser doesn't mind it and it is present in the
> > spec. I figured it didn't hurt to add it. That said, I can remove it before
> > landing if you think it is not worth adding until it is implemented.
> 
> I'd rather not include undocumented nops, since it can be misleading.
> It's fine to include if you want to document it though,
> preferably with a link to a bug, like:
> // FIXME: legacycaller currently nop: http://crbug.com/???
> 
> How about this?

I prefer to remove it at this point. I don't know if we want to implement
legacycaller (I have no plan on doing so).
It is anyway not needed for this CL.

[Nils Barth (inactive), 2014-02-07 02:11:41]

On 2014/02/07 02:09:40, Chris Dumez wrote:
> I prefer to remove it at this point. I don't know if we want to implement
> legacycaller (I have no plan on doing so).
> It is anyway not needed for this CL.

Got it, and agree -- if we're not planning on implementing legacycaller,
let's avoid putting it in IDLs.
Thanks!

[Inactive, 2014-02-07 02:14:31]

On 2014/02/07 01:54:55, Nils Barth wrote:
>
https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
> File Source/core/html/HTMLFormControlsCollection.idl (right):
> 
>
https://codereview.chromium.org/157013006/diff/80001/Source/core/html/HTMLFor...
> Source/core/html/HTMLFormControlsCollection.idl:27:
[ImplementedAs=namedGetter]
> legacycaller getter (RadioNodeList or Element) namedItem(DOMString name); //
> shadows inherited namedItem()
> On 2014/02/07 01:52:38, Chris Dumez wrote:
> > On 2014/02/07 01:40:43, Nils Barth wrote:
> > > legacycaller is currently a nop;
> > > do you want to omit this from this CL,
> > > and instead add and implement in a followup?
> > 
> > I know it is a noop but the parser doesn't mind it and it is present in the
> > spec. I figured it didn't hurt to add it. That said, I can remove it before
> > landing if you think it is not worth adding until it is implemented.
> 
> I'd rather not include undocumented nops, since it can be misleading.

I wouldn't say that it is undocumented since it is part of Web IDL :)
http://heycam.github.io/webidl/#idl-legacy-callers

[haraken, 2014-02-07 02:17:37]

LGTM.

[Nils Barth (inactive), 2014-02-07 02:19:10]

On 2014/02/07 01:54:55, Nils Barth wrote:
> I'd rather not include undocumented nops, since it can be misleading.

On 2014/02/07 02:14:31, Chris Dumez wrote:
> I wouldn't say that it is undocumented since it is part of Web IDL :)
> http://heycam.github.io/webidl/#idl-legacy-callers

Understood that the keyword itself is documented (^.-).

However, it's a nop for current CG, so documenting this fact
("We're matching the spec but not doing anything with it")
in the .idl file
...is useful, no?

[arv (Not doing code reviews), 2014-02-07 02:59:02]

LGTM

[tkent, 2014-02-07 04:12:56]

public API change lgtm.

[Inactive, 2014-02-07 04:26:27]

The CQ bit was checked by ch.dumez@samsung.com

[commit-bot: I haz the power, 2014-02-07 04:26:44]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ch.dumez@samsung.com/157013006/80002

[commit-bot: I haz the power, 2014-02-07 07:47:22]

Change committed as 166688

[Nils Barth (inactive), 2014-02-10 06:41:31]

Python side at:
IDL compiler: sync Python to r166688 (and clean up special operations+union
code)
https://codereview.chromium.org/158663002/

Thanks Chris!