Bug 204967 – Make it possible to add regression tests for blocked XPC services

• Component: Tools / Tests
• Status: RESOLVED FIXED

[Per Arne Vollan]

On macOS and iOS, it should be possible to add regression tests for blocked mach services.

[Per Arne Vollan]

rdar://problem/57414291

[Per Arne Vollan]

Created attachment 385054 [details]
Patch

[Per Arne Vollan]

Created attachment 385058 [details]
Patch

[Alexey Proskuryakov]

Comment on attachment 385058 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=385058&action=review

> Source/WebCore/ChangeLog:9
> +        to Internals, which checks whether the WebContent process has access to mach lookup for a given global

Good idea! I do not see why this should be limited to WebContent - sandbox_check can check other processes too. May be worth adding a universal method right away.

[Per Arne Vollan]

Created attachment 385167 [details]
Patch

[Per Arne Vollan]

(In reply to Alexey Proskuryakov from comment #4)
> Comment on attachment 385058 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=385058&action=review
> 
> > Source/WebCore/ChangeLog:9
> > +        to Internals, which checks whether the WebContent process has access to mach lookup for a given global
> 
> Good idea! I do not see why this should be limited to WebContent -
> sandbox_check can check other processes too. May be worth adding a universal
> method right away.

Credits go to Geoff and Darin :)

I have added a process name parameter to the method.

Thanks for reviewing!

[Brent Fulgham]

Comment on attachment 385167 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=385167&action=review

> Source/WebCore/testing/Internals.cpp:5337
> +    if (process == "com.apple.WebKit.WebContent")

What about WebContent.Development?

[Per Arne Vollan]

(In reply to Brent Fulgham from comment #7)
> Comment on attachment 385167 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=385167&action=review
> 
> > Source/WebCore/testing/Internals.cpp:5337
> > +    if (process == "com.apple.WebKit.WebContent")
> 
> What about WebContent.Development?

This string will be passed from the layout test html file, and will be the same for both debug and release builds, so I don't think we need to include a check for WebContent.Development in this case.

Thanks for reviewing!

[Brent Fulgham]

Comment on attachment 385167 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=385167&action=review

R=me

>>> Source/WebCore/testing/Internals.cpp:5337
>>> +    if (process == "com.apple.WebKit.WebContent")
>> 
>> What about WebContent.Development?
> 
> This string will be passed from the layout test html file, and will be the same for both debug and release builds, so I don't think we need to include a check for WebContent.Development in this case.
> 
> Thanks for reviewing!

Very good!

[Per Arne Vollan]

(In reply to Brent Fulgham from comment #9)
> Comment on attachment 385167 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=385167&action=review
> 
> R=me
> 
> >>> Source/WebCore/testing/Internals.cpp:5337
> >>> +    if (process == "com.apple.WebKit.WebContent")
> >> 
> >> What about WebContent.Development?
> > 
> > This string will be passed from the layout test html file, and will be the same for both debug and release builds, so I don't think we need to include a check for WebContent.Development in this case.
> > 
> > Thanks for reviewing!
> 
> Very good!

Thanks for reviewing :)

[WebKit Commit Bot]

Comment on attachment 385167 [details]
Patch

Clearing flags on attachment: 385167

Committed r253353: <https://trac.webkit.org/changeset/253353>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.