Re-implement %Generator% intrinsic as an object

Re-implement %Generator% intrinsic as an object

From ES6 25.2.3 ("Properties of the GeneratorFunction Prototype
Object"):

> The GeneratorFunction prototype object is an ordinary object. It is
> not a function object and does not have an [[ECMAScriptCode]] internal
> slot or any other of the internal slots listed in Table 27 or Table
> 56.

Introduce one assertion for the value's type and additional tests for its
properties. Remove an invalid assertion that fails as a result of this
fix.

BUG=v8:3991
LOG=N

Committed: https://crrev.com/3b624a179682a6b8351eea8cba28cf21e2afcdcd
Cr-Commit-Position: refs/heads/master@{#27603}

[mike3, 2015-04-03 23:42:04]

mike@mikepennisi.com changed reviewers:
+ arv@chromium.org

[mike3, 2015-04-03 23:42:05]

Hi Erik--could you please take a look?

[arv (Not doing code reviews), 2015-04-06 15:16:25]

LGTM with nits

https://codereview.chromium.org/1062633002/diff/1/src/bootstrapper.cc
File src/bootstrapper.cc (right):

https://codereview.chromium.org/1062633002/diff/1/src/bootstrapper.cc#newcode...
src/bootstrapper.cc:2065:
factory()->InternalizeUtf8String("GeneratorFunctionPrototype"),
Wrong indentation. `git cl upload` should have complained.

https://codereview.chromium.org/1062633002/diff/1/src/bootstrapper.cc#newcode...
src/bootstrapper.cc:2070: factory()->InternalizeUtf8String("prototype"),
wrong indentation

https://codereview.chromium.org/1062633002/diff/1/test/mjsunit/es6/generators...
File test/mjsunit/es6/generators-runtime.js (right):

https://codereview.chromium.org/1062633002/diff/1/test/mjsunit/es6/generators...
test/mjsunit/es6/generators-runtime.js:84: GeneratorFunctionPrototype,
"constructor");
wrong indentation

[arv (Not doing code reviews), 2015-04-06 15:16:29]

LGTM with nits

https://codereview.chromium.org/1062633002/diff/1/src/bootstrapper.cc
File src/bootstrapper.cc (right):

https://codereview.chromium.org/1062633002/diff/1/src/bootstrapper.cc#newcode...
src/bootstrapper.cc:2065:
factory()->InternalizeUtf8String("GeneratorFunctionPrototype"),
Wrong indentation. `git cl upload` should have complained.

https://codereview.chromium.org/1062633002/diff/1/src/bootstrapper.cc#newcode...
src/bootstrapper.cc:2070: factory()->InternalizeUtf8String("prototype"),
wrong indentation

https://codereview.chromium.org/1062633002/diff/1/test/mjsunit/es6/generators...
File test/mjsunit/es6/generators-runtime.js (right):

https://codereview.chromium.org/1062633002/diff/1/test/mjsunit/es6/generators...
test/mjsunit/es6/generators-runtime.js:84: GeneratorFunctionPrototype,
"constructor");
wrong indentation

[mike3, 2015-04-06 16:24:11]

Hi Erik,

I've fixed the code style inconsistencies, so this should be ready for another
review.

[caitp (gmail), 2015-04-06 16:31:24]

caitpotter88@gmail.com changed reviewers:
+ caitpotter88@gmail.com

[caitp (gmail), 2015-04-06 16:31:26]

https://codereview.chromium.org/1062633002/diff/20001/src/bootstrapper.cc
File src/bootstrapper.cc (right):

https://codereview.chromium.org/1062633002/diff/20001/src/bootstrapper.cc#new...
src/bootstrapper.cc:2065:
factory()->InternalizeUtf8String("GeneratorFunctionPrototype"),
I think clang-format will complain about this indentation. The styles that it
prefers seems to be:

1) align everything with the ( paren, like this:

```
FunctionToCall(argument1,
                       
reallyReallyReallyReallyReallyReallyReallyReallyReallyReallyReallyLongArgument2)
```

or 2) start the first argument 4 spaces in:

```
FunctionToCall(
    argument1,
reallyReallyReallyReallyReallyReallyReallyReallyReallyReallyReallyLongArgument2)
```

If you run `git cl format src` it should fix it all up though (might want to
just target the specific files in your patch to avoid diff noise)

[arv (Not doing code reviews), 2015-04-06 16:32:33]

On 2015/04/06 16:31:26, caitp wrote:
> ... (might want to
> just target the specific files in your patch to avoid diff noise)

git cl format only formats the lines you changed.

[caitp (gmail), 2015-04-06 16:33:36]

On 2015/04/06 16:32:33, arv wrote:
> On 2015/04/06 16:31:26, caitp wrote:
> > ... (might want to
> > just target the specific files in your patch to avoid diff noise)
> 
> git cl format only formats the lines you changed.

That doesn't seem to be true when I run it, for instance it will frequently
change the position of backslashes in macros I didn't touch

[arv (Not doing code reviews), 2015-04-06 16:34:34]

On 2015/04/06 16:33:36, caitp wrote:
> On 2015/04/06 16:32:33, arv wrote:
> > On 2015/04/06 16:31:26, caitp wrote:
> > > ... (might want to
> > > just target the specific files in your patch to avoid diff noise)
> > 
> > git cl format only formats the lines you changed.
> 
> That doesn't seem to be true when I run it, for instance it will frequently
> change the position of backslashes in macros I didn't touch

Yes, it will reformat the whole macro if needed.

[mike3, 2015-04-06 16:59:38]

I've included the changes suggested by `git cl format`. Sorry for the noise

[arv (Not doing code reviews), 2015-04-06 17:26:16]

LGTM

[caitp (gmail), 2015-04-06 17:27:06]

On 2015/04/06 17:26:16, arv wrote:
> LGTM

LGTM, but need to fix the `checkConstructor` test in test/mjsunit/builtins.js

[mike3, 2015-04-06 18:03:33]

Thanks, Cait! I've fixed the failing test. There are a lot of ways to do that,
so I'm all ears if you think my approach could be improved.

[caitp (gmail), 2015-04-06 18:05:05]

That approach looks pretty solid (so long as the list of exceptions doesn't
change too much =)

https://codereview.chromium.org/1062633002/diff/60001/test/mjsunit/builtins.js
File test/mjsunit/builtins.js (right):

https://codereview.chromium.org/1062633002/diff/60001/test/mjsunit/builtins.j...
test/mjsunit/builtins.js:51: // TODO: Update the following values to the
%IteratorPrototype% intrinsic
TODO(name): ...

[caitp (gmail), 2015-04-06 18:07:04]

https://codereview.chromium.org/1062633002/diff/60001/test/mjsunit/builtins.js
File test/mjsunit/builtins.js (right):

https://codereview.chromium.org/1062633002/diff/60001/test/mjsunit/builtins.j...
test/mjsunit/builtins.js:53: // Issue 3568: Generator Prototype should have an
object between itself
bug numbers might be iffy, especially if v8 has to migrate to a new issue
tracker once code.google.com is closed for good (which may not need to happen?
not sure)

[arv (Not doing code reviews), 2015-04-06 18:15:14]

https://codereview.chromium.org/1062633002/diff/60001/test/mjsunit/builtins.js
File test/mjsunit/builtins.js (right):

https://codereview.chromium.org/1062633002/diff/60001/test/mjsunit/builtins.j...
test/mjsunit/builtins.js:53: // Issue 3568: Generator Prototype should have an
object between itself
On 2015/04/06 18:07:03, caitp wrote:
> bug numbers might be iffy, especially if v8 has to migrate to a new issue
> tracker once http://code.google.com is closed for good (which may not need to
happen?
> not sure)

Bug numbers are fine. We have them all over. It might be better to have the full
bug url here but it doesn't matter.

https://code.google.com/p/v8/issues/detail?id=3568

[mike3, 2015-04-06 19:03:09]

Alrighty, I've added a TODO contact and the URL to the bug.

[mike3, 2015-04-06 20:10:35]

The CQ bit was unchecked by mike@mikepennisi.com

[mike3, 2015-04-06 20:10:36]

The CQ bit was checked by mike@mikepennisi.com

[mike3, 2015-04-06 20:10:36]

The patchset sent to the CQ was uploaded after l-g-t-m from
caitpotter88@gmail.com, arv@chromium.org
Link to the patchset: https://codereview.chromium.org/1062633002/#ps80001
(title: "Add TODO contact and bug URL")

[commit-bot: I haz the power, 2015-04-06 20:10:48]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1062633002/80001

[commit-bot: I haz the power, 2015-04-06 21:04:49]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2015-04-06 21:04:59]

Patchset 5 (id:??) landed as
https://crrev.com/3b624a179682a6b8351eea8cba28cf21e2afcdcd
Cr-Commit-Position: refs/heads/master@{#27603}