Parse arrow functions at proper precedence level

Parse arrow functions at proper precedence level

BUG=v8:4211
LOG=Y
R=rossberg@chromium.org

Committed: https://crrev.com/79e74db3e0df3f481793f169d945505886700e55
Cr-Commit-Position: refs/heads/master@{#30373}

[wingo, 2015-08-25 08:21:19]

The CQ bit was checked by wingo@igalia.com to run a CQ dry run

[commit-bot: I haz the power, 2015-08-25 08:21:28]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1315823002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1315823002/1

[wingo, 2015-08-25 08:22:11]

A reapplication of https://codereview.chromium.org/1286383005, now that the
rebaseline patch landed on the Blink side.

[wingo, 2015-08-25 08:22:25]

wingo@igalia.com changed reviewers:
+ adamk@chromium.org

[commit-bot: I haz the power, 2015-08-25 08:24:57]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-08-25 08:24:58]

Dry run: Try jobs failed on following builders:
  v8_linux_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_rel/builds/8895)

[wingo, 2015-08-25 09:06:48]

The CQ bit was checked by wingo@igalia.com to run a CQ dry run

[commit-bot: I haz the power, 2015-08-25 09:06:58]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1315823002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1315823002/20001

[commit-bot: I haz the power, 2015-08-25 09:34:45]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-08-25 09:34:45]

Dry run: This issue passed the CQ dry run.

[wingo, 2015-08-25 10:13:46]

On 2015/08/25 09:34:45, commit-bot: I haz the power wrote:
> Dry run: This issue passed the CQ dry run.

I think things are OK but we should probably wait for the layout tests to cycle
green before applying.

[wingo, 2015-08-25 12:46:45]

On 2015/08/25 10:13:46, wingo wrote:
> On 2015/08/25 09:34:45, commit-bot: I haz the power wrote:
> > Dry run: This issue passed the CQ dry run.
> 
> I think things are OK but we should probably wait for the layout tests to
cycle
> green before applying.

Actually I dunno, it's clear which test needed rebaselining -- it was in the
logs, same one last time I changed this error, and
https://codereview.chromium.org/1311153002/ landed the rebaseline thing so it's
probably OK to apply now, which can also unblock Conrad's strict mode patch.

[rossberg, 2015-08-25 12:55:56]

https://codereview.chromium.org/1315823002/diff/20001/src/interpreter/bytecod...
File src/interpreter/bytecode-generator.cc (right):

https://codereview.chromium.org/1315823002/diff/20001/src/interpreter/bytecod...
src/interpreter/bytecode-generator.cc:339: UNIMPLEMENTED();
Nit: This should probably be UNREACHABLE (same for Spread)

https://codereview.chromium.org/1315823002/diff/20001/src/pattern-rewriter.cc
File src/pattern-rewriter.cc (right):

https://codereview.chromium.org/1315823002/diff/20001/src/pattern-rewriter.cc...
src/pattern-rewriter.cc:367: // TODO(dslomov): implement.
While you're here, can you change this to UNREACHABLE?

https://codereview.chromium.org/1315823002/diff/20001/src/preparser.h
File src/preparser.h (right):

https://codereview.chromium.org/1315823002/diff/20001/src/preparser.h#newcode...
src/preparser.h:2279: } else if (allow_harmony_rest_parameters() &&
Check(Token::ELLIPSIS)) {
I wonder, does rest actually need special treatment here? Can't it simply be
parsed as spread in the else case below, and later be rewritten?

https://codereview.chromium.org/1315823002/diff/20001/src/prettyprinter.cc
File src/prettyprinter.cc (right):

https://codereview.chromium.org/1315823002/diff/20001/src/prettyprinter.cc#ne...
src/prettyprinter.cc:854: Print("<empty-parentheses>");
I'd print "()" here.

[wingo, 2015-08-25 13:18:18]

Fixed nits.  See comment below about spread treatment.

https://codereview.chromium.org/1315823002/diff/20001/src/preparser.h
File src/preparser.h (right):

https://codereview.chromium.org/1315823002/diff/20001/src/preparser.h#newcode...
src/preparser.h:2279: } else if (allow_harmony_rest_parameters() &&
Check(Token::ELLIPSIS)) {
On 2015/08/25 12:55:56, rossberg wrote:
> I wonder, does rest actually need special treatment here? Can't it simply be
> parsed as spread in the else case below, and later be rewritten?

I think it does need special treatment.  All the other places that parse spread
(array literals, call arguments, comma expressions) parse spread specially.  We
could of course parse it anywhere as a prefix unary op, marking the production
invalid as an expression, but that is a riskier change if we are considering
uplifting this to M46.  WDYT?

[wingo, 2015-08-25 13:18:30]

The CQ bit was checked by wingo@igalia.com to run a CQ dry run

[commit-bot: I haz the power, 2015-08-25 13:18:35]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1315823002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1315823002/40001

[rossberg, 2015-08-25 13:26:42]

LGTM

https://codereview.chromium.org/1315823002/diff/20001/src/preparser.h
File src/preparser.h (right):

https://codereview.chromium.org/1315823002/diff/20001/src/preparser.h#newcode...
src/preparser.h:2279: } else if (allow_harmony_rest_parameters() &&
Check(Token::ELLIPSIS)) {
On 2015/08/25 13:18:18, wingo wrote:
> On 2015/08/25 12:55:56, rossberg wrote:
> > I wonder, does rest actually need special treatment here? Can't it simply be
> > parsed as spread in the else case below, and later be rewritten?
> 
> I think it does need special treatment.  All the other places that parse
spread
> (array literals, call arguments, comma expressions) parse spread specially. 
We
> could of course parse it anywhere as a prefix unary op, marking the production
> invalid as an expression, but that is a riskier change if we are considering
> uplifting this to M46.  WDYT?

SGTM. Let's leave that as a clean-up for a separate CL.

[commit-bot: I haz the power, 2015-08-25 14:12:55]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-08-25 14:12:55]

Dry run: This issue passed the CQ dry run.

[wingo, 2015-08-25 14:13:33]

The CQ bit was checked by wingo@igalia.com

[commit-bot: I haz the power, 2015-08-25 14:13:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1315823002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1315823002/40001

[commit-bot: I haz the power, 2015-08-25 14:15:55]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-08-25 14:15:56]

Try jobs failed on following builders:
  v8_presubmit on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_presubmit/builds/5202)

[wingo, 2015-08-25 15:39:14]

wingo@igalia.com changed reviewers:
+ mstarzinger@chromium.org

[wingo, 2015-08-25 15:39:14]

Michael would you mind a look, this is a reland

[Michael Starzinger, 2015-08-25 18:30:45]

LGTM (rubber-stamp on compiler).

[wingo, 2015-08-26 07:23:05]

The CQ bit was checked by wingo@igalia.com

[commit-bot: I haz the power, 2015-08-26 07:23:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1315823002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1315823002/40001

[commit-bot: I haz the power, 2015-08-26 09:36:44]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-08-26 09:36:52]

Patchset 3 (id:??) landed as
https://crrev.com/79e74db3e0df3f481793f169d945505886700e55
Cr-Commit-Position: refs/heads/master@{#30373}