Bug 178842 – Fix issues with WebView subframe painting

• Component: WebKit Misc.
• Status: RESOLVED FIXED

[Simon Fraser (smfr)]

Fix issues with WebView subframe painting

[Simon Fraser (smfr)]

Created attachment 324940 [details]
Patch

[Daniel Bates]

Comment on attachment 324940 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=324940&action=review

> Source/WebKitLegacy/mac/ChangeLog:11
> +        WebHTMLView overrides some NSView internal methods to make sure that AppKit doesn't paint
> +        subframes (WebKit controls subframe painting). The method signature of one of these changed in macOS
> +        High Sierra, so match the new signature.

Is there an equivalent to C++11's override keyword to help us catch such changes in the future and/or can we write a unit test to ensure this does not regress?

[Simon Fraser (smfr)]

(In reply to Daniel Bates from comment #2)
> Comment on attachment 324940 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=324940&action=review
> 
> > Source/WebKitLegacy/mac/ChangeLog:11
> > +        WebHTMLView overrides some NSView internal methods to make sure that AppKit doesn't paint
> > +        subframes (WebKit controls subframe painting). The method signature of one of these changed in macOS
> > +        High Sierra, so match the new signature.
> 
> Is there an equivalent to C++11's override keyword to help us catch such
> changes in the future and/or can we write a unit test to ensure this does
> not regress?

The only think I can think of is a runtime assertion that NSView implements certain functions.

[WebKit Commit Bot]

Comment on attachment 324940 [details]
Patch

Clearing flags on attachment: 324940

Committed r224040: <https://trac.webkit.org/changeset/224040>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Radar WebKit Bug Importer]

<rdar://problem/35568031>