[iOS] Add SPI for internal clients to consult whether or not viewport quirks should be enabled

• Component: Platform
• Status: RESOLVED FIXED

[Wenson Hsieh]

rdar://80397679

[Wenson Hsieh]

Created attachment 433906 [details]
Patch

[EWS]

Committed r280119 (239833@main): <https://commits.webkit.org/239833@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 433906 [details].

Can you explain here why clients can’t use the _needsSiteSpecificQuirks property of WKPreferences?

[Wenson Hsieh]

(In reply to mitz from comment #3)
> Can you explain here why clients can’t use the _needsSiteSpecificQuirks
> property of WKPreferences?

The rationale is that this value will change out from underneath the WebKit client, so  we'd want to avoid situations where a client that sets the `_needsSiteSpecificQuirks` property later reads back a different value than what it previously set.

More generally, this is a hint from WebKit to the client, whereas WKPreferences is a way for clients to provide hints to WebKit.

The naming collision is a bit unfortunate, though. I think that ideally, the preference would be called something like `_prefersSiteSpecificQuirks` to indicate a difference between the "hint" and "effective" value.

(In reply to Wenson Hsieh from comment #4)
> (In reply to mitz from comment #3)
> > Can you explain here why clients can’t use the _needsSiteSpecificQuirks
> > property of WKPreferences?
> 
> The rationale is that this value will change out from underneath the WebKit
> client, so  we'd want to avoid situations where a client that sets the
> `_needsSiteSpecificQuirks` property later reads back a different value than
> what it previously set.
> 
> More generally, this is a hint from WebKit to the client, whereas
> WKPreferences is a way for clients to provide hints to WebKit.
> 
> The naming collision is a bit unfortunate, though. I think that ideally, the
> preference would be called something like `_prefersSiteSpecificQuirks` to
> indicate a difference between the "hint" and "effective" value.

Thank you. I think I now understand this new property’s purpose. Seems like adding “viewport” to the name would have helped with both disambiguation and clarifying the intent.

[Wenson Hsieh]

(In reply to mitz from comment #5)
> (In reply to Wenson Hsieh from comment #4)
> > (In reply to mitz from comment #3)
> > > Can you explain here why clients can’t use the _needsSiteSpecificQuirks
> > > property of WKPreferences?
> > 
> > The rationale is that this value will change out from underneath the WebKit
> > client, so  we'd want to avoid situations where a client that sets the
> > `_needsSiteSpecificQuirks` property later reads back a different value than
> > what it previously set.
> > 
> > More generally, this is a hint from WebKit to the client, whereas
> > WKPreferences is a way for clients to provide hints to WebKit.
> > 
> > The naming collision is a bit unfortunate, though. I think that ideally, the
> > preference would be called something like `_prefersSiteSpecificQuirks` to
> > indicate a difference between the "hint" and "effective" value.
> 
> Thank you. I think I now understand this new property’s purpose. Seems like
> adding “viewport” to the name would have helped with both disambiguation and
> clarifying the intent.

The patch adopting this in the WebKit client has not landed yet, so we can (easily) fix this!

[Wenson Hsieh]

Reopening to attach new patch.

[Wenson Hsieh]

Created attachment 433912 [details]
Renaming

[Wenson Hsieh]

Created attachment 433913 [details]
Renaming

Comment on attachment 433913 [details]
Renaming

Thank you!

[Wenson Hsieh]

Comment on attachment 433913 [details]
Renaming

Thanks for the review!

[EWS]

Committed r280122 (239834@main): <https://commits.webkit.org/239834@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 433913 [details].