Bug 59170 – Remove remaining uses of deprecated version of toJS

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Sam Weinig]

Remove remaining uses of deprecated version of toJS

[Sam Weinig]

Created attachment 90661 [details]
Patch

[Early Warning System Bot]

Attachment 90661 [details] did not build on qt:
Build output: http://queues.webkit.org/results/8496228

[Collabora GTK+ EWS bot]

Attachment 90661 [details] did not build on gtk:
Build output: http://queues.webkit.org/results/8494369

[Sam Weinig]

Committed r84641: <http://trac.webkit.org/changeset/84641>

[WebKit Review Bot]

http://trac.webkit.org/changeset/84641 might have broken Qt Linux Release minimal

[Eric Seidel (no email)]

Comment on attachment 90661 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=90661&action=review

> LayoutTests/ChangeLog:10
> +        failures are pure JS objects (Array, Boolean, String, Number, Function).

I think these aren't failures, and the test shoudl be updated.  Ask abarth.  But I think I was wrong when I wrote this test.  I think pure JS types always use the dynamic global object per the spec.

[Adam Barth]

Comment on attachment 90661 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=90661&action=review

>> LayoutTests/ChangeLog:10
>> +        * fast/dom/prototype-inheritance-2-expected.txt:
>> +        Add new passing results since we now use the correct global object. The remaining
>> +        failures are pure JS objects (Array, Boolean, String, Number, Function).
> 
> I think these aren't failures, and the test shoudl be updated.  Ask abarth.  But I think I was wrong when I wrote this test.  I think pure JS types always use the dynamic global object per the spec.

They always use the lexical global object, but, yeah, the test isn't correct for these primitive types.