No longer OOM on invalid string length.

No longer OOM on invalid string length.

R=ishell@chromium.org
BUG=v8:3060
LOG=Y

Committed: https://code.google.com/p/v8/source/detail?r=20202

[Yang, 2014-03-21 12:32:17]

https://codereview.chromium.org/207613005/diff/1/src/api.cc
File src/api.cc (right):

https://codereview.chromium.org/207613005/diff/1/src/api.cc#newcode5423
src/api.cc:5423: CHECK(!result.is_null());
Not using CHECK_NOT_EMPTY_HANDLE because this could be called after an exception
has already been thrown.

[Igor Sheludko, 2014-03-24 13:43:16]

lgtm with nits:

https://codereview.chromium.org/207613005/diff/1/src/ast.cc
File src/ast.cc (right):

https://codereview.chromium.org/207613005/diff/1/src/ast.cc#newcode1182
src/ast.cc:1182: return res;
As the allocation size is small and you are not going to check the result this
looks like a spurious change.

https://codereview.chromium.org/207613005/diff/1/src/json-parser.h
File src/json-parser.h (right):

https://codereview.chromium.org/207613005/diff/1/src/json-parser.h#newcode798
src/json-parser.h:798: RETURN_IF_EMPTY_HANDLE_VALUE(isolate(), result,
Handle<String>());
Do we really need both ASSERT and RETURN_IF_EMPTY_HANDLE_VALUE here?

https://codereview.chromium.org/207613005/diff/1/src/runtime.cc
File src/runtime.cc (right):

https://codereview.chromium.org/207613005/diff/1/src/runtime.cc#newcode4112
src/runtime.cc:4112: 
ASSERT(!answer.is_null()); ?

[Yang, 2014-03-24 15:01:31]

Committed patchset #2 manually as r20202 (presubmit successful).