[strong] Check super constructor calls

[strong] Check super constructor calls

R=marja@chromium.org
BUG=v8:3956
LOG=N

Committed: https://crrev.com/3fcb38f772efb8b4855e2edc8945721215baa284
Cr-Commit-Position: refs/heads/master@{#27188}

[marja, 2015-03-13 15:21:21]

LGTM

Hmm, so this disallows:

constructor() {
  if (sth)
    super(a, b, c);
  else
    super(d, e, f);
}

which might have some legit uses. Idk.

https://codereview.chromium.org/1002253002/diff/20001/test/mjsunit/strong/cla...
File test/mjsunit/strong/classes.js (right):

https://codereview.chromium.org/1002253002/diff/20001/test/mjsunit/strong/cla...
test/mjsunit/strong/classes.js:30: for (let b of bodies)
assertThrows(constructor(b), SyntaxError);
Nit: this style of writing tests will give less useful error messages than just
duplicating assertThrows (or a call to a helper function), since from the error
position you don't know which case failed.

[rossberg, 2015-03-13 15:43:11]

On 2015/03/13 15:21:21, marja wrote:
> LGTM
> 
> Hmm, so this disallows:
> 
> constructor() {
>   if (sth)
>     super(a, b, c);
>   else
>     super(d, e, f);
> }

Yes, that's intentional. You can usually rewrite to

  let [x, y, z] = sth ? [a, b, c] : [d, e, f]
  super(x, y, z)

or something along these lines. That's still far more flexibility than you have
in other common languages, say, C++ or Java.


>
https://codereview.chromium.org/1002253002/diff/20001/test/mjsunit/strong/cla...
> test/mjsunit/strong/classes.js:30: for (let b of bodies)
> assertThrows(constructor(b), SyntaxError);
> Nit: this style of writing tests will give less useful error messages than
just
> duplicating assertThrows (or a call to a helper function), since from the
error
> position you don't know which case failed.

Simplified

[rossberg, 2015-03-13 15:43:39]

The CQ bit was checked by rossberg@chromium.org

[rossberg, 2015-03-13 15:43:39]

The patchset sent to the CQ was uploaded after l-g-t-m from marja@chromium.org
Link to the patchset: https://codereview.chromium.org/1002253002/#ps40001
(title: "Comments marja")

[commit-bot: I haz the power, 2015-03-13 15:43:54]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1002253002/40001

[commit-bot: I haz the power, 2015-03-13 15:45:05]

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

[commit-bot: I haz the power, 2015-03-13 15:45:06]

Try jobs failed on following builders:
  v8_linux_arm64_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm64_rel/builds/3796)

[rossberg, 2015-03-13 15:47:50]

The CQ bit was checked by rossberg@chromium.org

[rossberg, 2015-03-13 15:47:50]

The patchset sent to the CQ was uploaded after l-g-t-m from marja@chromium.org
Link to the patchset: https://codereview.chromium.org/1002253002/#ps60001
(title: "Rebased")

[commit-bot: I haz the power, 2015-03-13 15:48:03]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1002253002/60001

[rossberg, 2015-03-13 16:02:37]

The CQ bit was unchecked by rossberg@chromium.org

[rossberg, 2015-03-13 16:06:12]

The CQ bit was checked by rossberg@chromium.org

[rossberg, 2015-03-13 16:06:12]

The patchset sent to the CQ was uploaded after l-g-t-m from marja@chromium.org
Link to the patchset: https://codereview.chromium.org/1002253002/#ps80001
(title: "Should be ReferenceError")

[commit-bot: I haz the power, 2015-03-13 16:06:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1002253002/80001

[commit-bot: I haz the power, 2015-03-13 16:26:45]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2015-03-13 16:27:06]

Patchset 5 (id:??) landed as
https://crrev.com/3fcb38f772efb8b4855e2edc8945721215baa284
Cr-Commit-Position: refs/heads/master@{#27188}

[Dmitry Lomov (no reviews), 2015-03-13 16:41:04]

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

[Dmitry Lomov (no reviews), 2015-03-13 16:41:04]

DBC

https://codereview.chromium.org/1002253002/diff/20001/src/messages.js
File src/messages.js (right):

https://codereview.chromium.org/1002253002/diff/20001/src/messages.js#newcode173
src/messages.js:173: strong_super_call_missing:     ["Please always invoke the
super constructor in subclasses in strong mode"],
Do we need to plead here?