[Webkit-unassigned] [Bug 167963] [ESnext] Implement Object Spread

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=167963

Yusuke Suzuki <utatane.tea at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #304214|review?                     |review+
              Flags|                            |

--- Comment #3 from Yusuke Suzuki <utatane.tea at gmail.com> ---
Comment on attachment 304214
  --> https://bugs.webkit.org/attachment.cgi?id=304214
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=304214&action=review

r=me with comments.

> Source/JavaScriptCore/builtins/BuiltinNames.h:73
> +    macro(getOwnPropertySymbols) \

I think it is not necessary. The reason will be described in the following comments :).

> Source/JavaScriptCore/builtins/GlobalOperations.js:90
>          return target;

How about just using `== null`?

> Source/JavaScriptCore/builtins/GlobalOperations.js:92
>      let from = @toObject(source);

I think we should use `@Object(source)` here. And we can drop `@toObject`.
`@Object` is nice because it is well handled in DFG :) (Like, if source is likely an object, we emit the fixup edge and drop @Object call!).

> Source/JavaScriptCore/builtins/GlobalOperations.js:94
> +    let propertyNames = @Object. at getOwnPropertyNames(from);
> +    let keys = propertyNames. at concat(@Object. at getOwnPropertySymbols(from));

You can use @Reflect. at ownKeys().

> Source/JavaScriptCore/bytecompiler/NodesCodegen.cpp:4238
> +    HashSet<UniquedStringImpl*> excludedSet;

BTW, who guarantees this UniquedStringImpl*'s lifetime? Is HashSet<Ref<UniquedStringImpl>> correct?

> Source/JavaScriptCore/parser/NodeConstructors.h:239
> +        : m_name(0)

Use nullptr.

> Source/JavaScriptCore/runtime/ArrayPrototype.cpp:93
> +    JSC_BUILTIN_FUNCTION_WITHOUT_TRANSITION(vm.propertyNames->builtinNames().concatPrivateName(), arrayPrototypeConcatCodeGenerator, DontEnum | DontDelete | ReadOnly);

So I think it is not necessary.

> Source/JavaScriptCore/runtime/ObjectConstructor.cpp:108
> +    JSC_NATIVE_FUNCTION_WITHOUT_TRANSITION(vm.propertyNames->builtinNames().getOwnPropertySymbolsPrivateName(), objectConstructorGetOwnPropertySymbols, DontEnum, 1);

Ditto.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20170314/cb9e9920/attachment.html>