[JSC] REGRESSION(r122912): CodeGeneratorJS.pm should not implicitly assume ScriptExecutionContext for static attributes

• Component: WebCore JavaScript
• Status: RESOLVED FIXED

[Kentaro Hara]

r122912 implemented static attributes in CodeGeneratorJS.pm. The generated code implicitly assumes that a static attribute always requires ScriptExecutionContext, which is wrong. If ScriptExecutionContext is needed, we should specify [CallWith=ScriptExecutionContext].

I will upload a patch after bug 91764 is fixed.

[Kentaro Hara]

Created attachment 153942 [details]
Patch

[Kentaro Hara]

Comment on attachment 153942 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=153942&action=review

> Source/WebCore/bindings/scripts/CodeGeneratorJS.pm:-1887
>                      }
> -                }

Indentation is wrong here. Let me fix it in a follow-up patch (because the indentation fix will mess up this patch).

> Source/WebCore/bindings/scripts/test/JS/JSTestInterface.cpp:248
> +    JSValue result = jsNumber(TestSupplemental::supplementalStaticReadOnlyAttr(impl));

Passing 'impl' is wrong. This will be fixed in bug 91925.

[Adam Barth]

Thanks for keeping the diff clean.  rs=me to fix the indent.

[Kentaro Hara]

Created attachment 153967 [details]
fix indentation, patch for landing

[WebKit Review Bot]

Comment on attachment 153967 [details]
fix indentation, patch for landing

Clearing flags on attachment: 153967

Committed r123433: <http://trac.webkit.org/changeset/123433>

[Jon Lee]

This patch does not appropriately call the static attribute function with ScriptExecutionContext if [CallWith=ScriptExecutionContext] is specified.

[Jon Lee]

(In reply to comment #6)
> This patch does not appropriately call the static attribute function with ScriptExecutionContext if [CallWith=ScriptExecutionContext] is specified.

Sorry I misspoke, I put the attribute in the wrong place.

[Radar WebKit Bug Importer]

<rdar://problem/11947922>