PromiseThen should ignore non-function parameters.

PromiseThen should ignore non-function parameters.

When non-function parameters are given, PromiseThen should work as if
undefined parameters were given.

BUG=347455
LOG=Y
R=rossberg@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=21025

[yhirano, 2014-02-27 07:43:33]

https://codereview.chromium.org/182713002/ includes the test for this issue.

[rossberg, 2014-02-27 08:42:57]

Hm, I actually think this is a mistake. I filed a bug against the spec:

https://bugs.ecmascript.org/show_bug.cgi?id=2566

Let's defer this CL until the bug is resolved one way or the other.

[rossberg, 2014-04-28 08:32:51]

Hi Yutaka, seems like the committee flipped its decision back to the original
semantics due to legacy issues. So we should land this CL. But please see my
comments. Also, there should probably be a test for this in harmony/promises.js
(maybe there is by now).

https://codereview.chromium.org/180723011/diff/60001/src/promise.js
File src/promise.js (right):

https://codereview.chromium.org/180723011/diff/60001/src/promise.js#newcode152
src/promise.js:152: typeof onResolve === 'function' ? onResolve :
PromiseIdResolveHandler;
We shouldn't change .chain, only .then

https://codereview.chromium.org/180723011/diff/60001/src/promise.js#newcode206
src/promise.js:206: typeof onResolve === 'function' ? onResolve :
PromiseIdResolveHandler;
Please use the IS_SPEC_FUNCTION macro instead.

[yhirano, 2014-04-28 12:35:40]

https://codereview.chromium.org/180723011/diff/60001/src/promise.js
File src/promise.js (right):

https://codereview.chromium.org/180723011/diff/60001/src/promise.js#newcode152
src/promise.js:152: typeof onResolve === 'function' ? onResolve :
PromiseIdResolveHandler;
On 2014/04/28 08:32:52, rossberg wrote:
> We shouldn't change .chain, only .then

Done.

https://codereview.chromium.org/180723011/diff/60001/src/promise.js#newcode206
src/promise.js:206: typeof onResolve === 'function' ? onResolve :
PromiseIdResolveHandler;
On 2014/04/28 08:32:52, rossberg wrote:
> Please use the IS_SPEC_FUNCTION macro instead.

Done.

[rossberg, 2014-04-28 12:37:41]

LGTM. Thanks!

[rossberg, 2014-04-28 15:19:27]

Committed patchset #5 manually as r21025 (presubmit successful).