1135560 - Number.{parseInt, parseFloat} should be the same functions as global ones

Status: RESOLVED FIXED

(Core :: JavaScript: Standard Library, defect)

Description

[ziyunfei]

js> Number.parseInt == parseInt
false // should be true
js> Number.parseFloat == parseFloat
false // should be true

Comment 1

[ziyunfei]

Attachment: bug-1135560-v1.patch

Comment 2

[Till Schneidereit [:till]]

Comment on attachment 8568400 [details] [diff] [review]
bug-1135560-v1.patch

Review of attachment 8568400 [details] [diff] [review]:
-----------------------------------------------------------------

Great! f+ instead of r+ because I'd like to see a second version with the comment below addressed.

::: js/src/jsnum.cpp
@@ +1194,5 @@
>          return nullptr;
>      if (!JS_DefineFunctions(cx, global, number_functions))
>          return nullptr;
> +    /* Number.parseInt should be the same function object as global parseInt. */
> +    JSFunction *parseInt = JS_DefineFunction(cx, global, "parseInt", num_parseInt, 2, 0);

I think it'd be better to use the non-jsapi versions of JS_DefineFunction and JS_DefineProperty here. That way we don't have to atomize the names at all: they're already in the commonNames list. To do this, get the id like so:
RootedId id(cx, NameToId(cx->runtime()->commonNames->parseInt));

For DefineProperty, pass nullptr for the get and set op parameters.

Comment 3

[ziyunfei]

Attachment: bug-1135560-v2.patch

Comment 4

[Till Schneidereit [:till]]

Comment on attachment 8568534 [details] [diff] [review]
bug-1135560-v2.patch

Review of attachment 8568534 [details] [diff] [review]:
-----------------------------------------------------------------

Excellent!

Comment 5

[ziyunfei]

https://treeherder.mozilla.org/#/jobs?repo=try&revision=fcf297b1c462

Comment 6

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/617d8a7292f1

Comment 7

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/617d8a7292f1