Test that trailing commas in object literals are allowed

Test that trailing commas in object literals are allowed

ES6 will allow trailing commas in object literals.  It turns out that V8
already allowed it, too, as does JSC and SpiderMonkey.

R=marja@chromium.org
BUG=

Committed: https://code.google.com/p/v8/source/detail?r=22005

[wingo, 2014-06-25 09:59:54]

It seems that ES5 allows a trailing comma as well: http://es5.github.io/#x11.1.5

[marja, 2014-06-25 10:04:01]

lgtm; a nice catch!

https://codereview.chromium.org/350353002/diff/1/src/preparser.h
File src/preparser.h (right):

https://codereview.chromium.org/350353002/diff/1/src/preparser.h#newcode1677
src/preparser.h:1677: // TODO(1240767): Consider allowing trailing comma.
Pls remove this TODO.

[wingo, 2014-06-25 10:13:17]

Committed patchset #1 manually as r22005 (presubmit successful).

[marja, 2014-06-25 10:24:58]

On 2014/06/25 10:13:17, wingo wrote:
> Committed patchset #1 manually as r22005 (presubmit successful).

... you could've removed the comment I pointed out tho...

[wingo, 2014-06-25 10:43:07]

On 2014/06/25 10:24:58, marja wrote:
> On 2014/06/25 10:13:17, wingo wrote:
> > Committed patchset #1 manually as r22005 (presubmit successful).
> 
> ... you could've removed the comment I pointed out tho...

I'm sorry, I thought I did.  Clearly juggling too many patches today; will take
a look.

[wingo, 2014-06-25 10:45:59]

On 2014/06/25 10:43:07, wingo wrote:
> On 2014/06/25 10:24:58, marja wrote:
> > On 2014/06/25 10:13:17, wingo wrote:
> > > Committed patchset #1 manually as r22005 (presubmit successful).
> > 
> > ... you could've removed the comment I pointed out tho...
> 
> I'm sorry, I thought I did.  Clearly juggling too many patches today; will
take
> a look.

Yes, I believe I got this fix in.  Should I re-upload for these sorts of l.g.t.m
with nits feedback?

[marja, 2014-06-25 11:08:48]

On 2014/06/25 10:45:59, wingo wrote:
> On 2014/06/25 10:43:07, wingo wrote:
> > On 2014/06/25 10:24:58, marja wrote:
> > > On 2014/06/25 10:13:17, wingo wrote:
> > > > Committed patchset #1 manually as r22005 (presubmit successful).
> > > 
> > > ... you could've removed the comment I pointed out tho...
> > 
> > I'm sorry, I thought I did.  Clearly juggling too many patches today; will
> take
> > a look.
> 
> Yes, I believe I got this fix in.  Should I re-upload for these sorts of
l.g.t.m
> with nits feedback?

I'd say you should always upload the patch set which you commit. Otherwise it
gets very confusing!

Also, the commit message says "committed patch set #1 manually", so I thought
you committed patch set 1... which is apparently not the case?

[marja, 2014-06-25 11:10:57]

Yes, this seems to be the case; checked the actual commit. So it was the message
which led me think you didn't.

Thx for fixing anyhow :)

[wingo, 2014-06-25 11:27:22]

On 2014/06/25 11:08:48, marja wrote:
> I'd say you should always upload the patch set which you commit. Otherwise it
> gets very confusing!

ACK :)

> Also, the commit message says "committed patch set #1 manually", so I thought
> you committed patch set 1... which is apparently not the case?

I use "git cl dcommit", which apparently sets this as the message.  It produces
an understandable confusion though!  Will upload before dcomitting in the
future.

[arv (Not doing code reviews), 2014-06-25 17:19:27]

FWIW, ES5 added the trailing commas since it was required for the web.

I'm surprised that test-262 does not cover this?