Fix V8 bug 1084

Fix V8 bug 1084
Allow "\0" in strict mode as valid escape sequence.
http://code.google.com/p/v8/issues/detail?id=1084


BUG=http://code.google.com/p/v8/issues/detail?id=1084
TEST=test/mjsunit/strict_mode.js

[Lasse Reichstein, 2011-01-31 06:05:19]

LGTM.

http://codereview.chromium.org/6386014/diff/1/src/scanner-base.cc
File src/scanner-base.cc (right):

http://codereview.chromium.org/6386014/diff/1/src/scanner-base.cc#newcode100
src/scanner-base.cc:100: // ECMA-262. Other JS VMs support them.
Add comment here (or, preferably, at the declaration) saying that it also
accepts the decimal escape "\0".

http://codereview.chromium.org/6386014/diff/1/src/scanner-base.cc#newcode112
src/scanner-base.cc:112: // Allow \0 only. Rest are octal escapes, illegal in
strict mode.
We "allow" everything (we don't reject anything here), it's just '\0' that isn't
really an octal escape.
Make comment something like "Anything except '\0' is an ocatal escape, which are
illegal in strict mode. Remember position to be able to give a better error
message.".

http://codereview.chromium.org/6386014/diff/1/test/mjsunit/strict-mode.js
File test/mjsunit/strict-mode.js (right):

http://codereview.chromium.org/6386014/diff/1/test/mjsunit/strict-mode.js#new...
test/mjsunit/strict-mode.js:158: 
If the preparser marks this function as lazy (which it will), the body won't be
parsed by the parser at all. In that case, even if there was a strict-mode
syntax error, I'm not sure it would be thrown.
You need to call the function to be sure that it's parsed.

(We should probably add checks to the preparser too, to ensure that we make it
an early error).

[Martin Maly, 2011-01-31 22:44:05]

Done.

http://codereview.chromium.org/6386014/diff/1/src/scanner-base.cc
File src/scanner-base.cc (right):

http://codereview.chromium.org/6386014/diff/1/src/scanner-base.cc#newcode100
src/scanner-base.cc:100: // ECMA-262. Other JS VMs support them.
On 2011/01/31 06:05:19, Lasse Reichstein wrote:
> Add comment here (or, preferably, at the declaration) saying that it also
> accepts the decimal escape "\0".

Done.

http://codereview.chromium.org/6386014/diff/1/src/scanner-base.cc#newcode112
src/scanner-base.cc:112: // Allow \0 only. Rest are octal escapes, illegal in
strict mode.
On 2011/01/31 06:05:19, Lasse Reichstein wrote:
> We "allow" everything (we don't reject anything here), it's just '\0' that
isn't
> really an octal escape.
> Make comment something like "Anything except '\0' is an ocatal escape, which
are
> illegal in strict mode. Remember position to be able to give a better error
> message.".

Done.

http://codereview.chromium.org/6386014/diff/1/test/mjsunit/strict-mode.js
File test/mjsunit/strict-mode.js (right):

http://codereview.chromium.org/6386014/diff/1/test/mjsunit/strict-mode.js#new...
test/mjsunit/strict-mode.js:158: 
On 2011/01/31 06:05:19, Lasse Reichstein wrote:
> If the preparser marks this function as lazy (which it will), the body won't
be
> parsed by the parser at all. In that case, even if there was a strict-mode
> syntax error, I'm not sure it would be thrown.
> You need to call the function to be sure that it's parsed.
> 
> (We should probably add checks to the preparser too, to ensure that we make it
> an early error).

Done. The pereparser work is on my todo list.