Disable classes in sloppy mode unless --harmony-sloppy is set

Disable classes in sloppy mode unless --harmony-sloppy is set

Also clean up flag names a little.

Baseline: https://codereview.chromium.org/713413003/

R=arv@chromium.org, dslomov@chromium.org
BUG=

Committed: https://chromium.googlesource.com/v8/v8/+/1808badc2d83432af0e9fd1ba5bf49e40f279b39

[arv (Not doing code reviews), 2014-11-18 16:22:30]

We should introduce a new flag for supporting harmony-scoping-sloppy. We will
need that as we work on implementing that anyway and by adding it now we can
make sure that classes do not regress.

[rossberg, 2014-11-19 12:53:42]

On 2014/11/18 16:22:30, arv wrote:
> We should introduce a new flag for supporting harmony-scoping-sloppy. We will
> need that as we work on implementing that anyway and by adding it now we can
> make sure that classes do not regress.

Good point, done.

[rossberg, 2014-11-19 14:58:13]

Patchset #2 (id:20001) has been deleted

[rossberg, 2014-11-19 14:59:57]

rossberg@chromium.org changed reviewers:
+ dslomov@chromium.org

[arv (Not doing code reviews), 2014-11-19 16:51:04]

LGTM with minor issues.

https://codereview.chromium.org/722203006/diff/40001/test/cctest/test-parsing.cc
File test/cctest/test-parsing.cc (right):

https://codereview.chromium.org/722203006/diff/40001/test/cctest/test-parsing...
test/cctest/test-parsing.cc:1375: printf("SetParserFlags: %d\n",
flags.Contains(kAllowHarmonySloppy));
Remove?

https://codereview.chromium.org/722203006/diff/40001/test/mjsunit/harmony/blo...
File test/mjsunit/harmony/block-non-strict-errors.js (right):

https://codereview.chromium.org/722203006/diff/40001/test/mjsunit/harmony/blo...
test/mjsunit/harmony/block-non-strict-errors.js:33: CheckError("1, class {};");
I assume your intention was to test class expression here. But this is parsing a
statement so you need add parens around the expression.

[rossberg, 2014-11-19 17:23:43]

https://codereview.chromium.org/722203006/diff/40001/test/cctest/test-parsing.cc
File test/cctest/test-parsing.cc (right):

https://codereview.chromium.org/722203006/diff/40001/test/cctest/test-parsing...
test/cctest/test-parsing.cc:1375: printf("SetParserFlags: %d\n",
flags.Contains(kAllowHarmonySloppy));
On 2014/11/19 16:51:04, arv wrote:
> Remove?

Oops.

https://codereview.chromium.org/722203006/diff/40001/test/mjsunit/harmony/blo...
File test/mjsunit/harmony/block-non-strict-errors.js (right):

https://codereview.chromium.org/722203006/diff/40001/test/mjsunit/harmony/blo...
test/mjsunit/harmony/block-non-strict-errors.js:33: CheckError("1, class {};");
On 2014/11/19 16:51:04, arv wrote:
> I assume your intention was to test class expression here. But this is parsing
a
> statement so you need add parens around the expression.

No, the "1," in front causes it to be parsed as a member-expression inside a
comma expression inside an expression statement.

But I put it in parens for more clarity.

[arv (Not doing code reviews), 2014-11-19 17:43:09]

https://codereview.chromium.org/722203006/diff/40001/test/mjsunit/harmony/blo...
File test/mjsunit/harmony/block-non-strict-errors.js (right):

https://codereview.chromium.org/722203006/diff/40001/test/mjsunit/harmony/blo...
test/mjsunit/harmony/block-non-strict-errors.js:33: CheckError("1, class {};");
On 2014/11/19 17:23:43, rossberg wrote:
> On 2014/11/19 16:51:04, arv wrote:
> > I assume your intention was to test class expression here. But this is
parsing
> a
> > statement so you need add parens around the expression.
> 
> No, the "1," in front causes it to be parsed as a member-expression inside a
> comma expression inside an expression statement.

Ooops

The 1 was lost in the noise.
> But I put it in parens for more clarity.

[Dmitry Lomov (no reviews), 2014-11-19 18:22:34]

lgtm
you will have to rebase onto
https://chromium.googlesource.com/v8/v8/+/c32114f1dcb1102c685007a5da813c664de...

[rossberg, 2014-11-20 10:47:40]

On 2014/11/19 18:22:34, Dmitry Lomov (chromium) wrote:
> you will have to rebase onto
>
https://chromium.googlesource.com/v8/v8/+/c32114f1dcb1102c685007a5da813c664de...

That's already the baseline, see CL Description. :)

[rossberg, 2014-11-20 10:52:12]

Committed patchset #3 (id:60001) manually as
1808badc2d83432af0e9fd1ba5bf49e40f279b39 (presubmit successful).

[Dmitry Lomov (no reviews), 2014-11-20 23:28:44]

On 2014/11/20 10:47:40, rossberg wrote:
> On 2014/11/19 18:22:34, Dmitry Lomov (chromium) wrote:
> > you will have to rebase onto
> >
>
https://chromium.googlesource.com/v8/v8/+/c32114f1dcb1102c685007a5da813c664de...
> 
> That's already the baseline, see CL Description. :)

Ah ic, your last patchset was against the ToT I think