Bug 221153 – [iOS][FCR] Add new picker for select elements

• Status: RESOLVED
• Resolution: FIXED
• Component: Forms

[Aditya Keerthi]

...

[Aditya Keerthi]

<rdar://problem/73770389>

[Aditya Keerthi]

Created attachment 418768 [details]
Patch

[Tim Horton]

Comment on attachment 418768 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=418768&action=review

> Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm:8027
> +    // and for select elements.

I feel like these comments are a bit silly :) (I also probably r+'d them)

> Source/WebKit/UIProcess/ios/forms/WKFormSelectPicker.mm:473
> +@interface UIAction (IPI)
> +- (void)_performActionWithSender:(id)sender;
> +@end

This should be in a SPI header, no?

> Source/WebKit/UIProcess/ios/forms/WKFormSelectPicker.mm:476
> +    WKContentView *_view;

Should this be strong (or __weak)? What guarantees we don't outlive it (and maybe get something accidentally called on us?)?

Even worse, dealloc indirectly calls things on _view, so even just something accidentally extending our lifetime would be disastrous (and there are no guarantees with refcounted objects).

[Aditya Keerthi]

Created attachment 418867 [details]
Patch

[Aditya Keerthi]

(In reply to Tim Horton from comment #3)
> Comment on attachment 418768 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=418768&action=review
> 
> > Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm:8027
> > +    // and for select elements.
> 
> I feel like these comments are a bit silly :) (I also probably r+'d them)

Removed the comment.
 
> > Source/WebKit/UIProcess/ios/forms/WKFormSelectPicker.mm:473
> > +@interface UIAction (IPI)
> > +- (void)_performActionWithSender:(id)sender;
> > +@end
> 
> This should be in a SPI header, no?

Moved to an SPI header.
 
> > Source/WebKit/UIProcess/ios/forms/WKFormSelectPicker.mm:476
> > +    WKContentView *_view;
> 
> Should this be strong (or __weak)? What guarantees we don't outlive it (and
> maybe get something accidentally called on us?)?
> 
> Even worse, dealloc indirectly calls things on _view, so even just something
> accidentally extending our lifetime would be disastrous (and there are no
> guarantees with refcounted objects).

Updated to use __weak.

[EWS]

Committed r272334: <https://trac.webkit.org/changeset/272334>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 418867 [details].