Bug 64056

Summary: Fix WebKit2 expected results search paths for Mac and Qt under new-run-webkit-tests
Product: WebKit Reporter: Eric Seidel (no email) <eric>
Component: New BugsAssignee: Eric Seidel (no email) <eric>
Status: RESOLVED FIXED    
Severity: Normal CC: abarth, dpranke, ossy, webkit.review.bot
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Bug Depends on:    
Bug Blocks: 56729    
Attachments:
Description Flags
Patch abarth: review+, eric: commit-queue+

Description Eric Seidel (no email) 2011-07-06 21:17:03 PDT
Fix WebKit2 expected results search paths for Mac and Qt under new-run-webkit-tests
Comment 1 Eric Seidel (no email) 2011-07-06 23:39:34 PDT
Created attachment 99943 [details]
Patch
Comment 2 Adam Barth 2011-07-06 23:47:17 PDT
Comment on attachment 99943 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=99943&action=review

> Tools/Scripts/webkitpy/layout_tests/port/mac_unittest.py:77
> -        port = mac.MacPort()
> +        port = MacPort(filesystem=MockFileSystem(), user=MockUser(), executive=MockExecutive())

This all seems very copy/pasted.

> Tools/Scripts/webkitpy/layout_tests/port/mac_unittest.py:137
> +        # FIXME: Is this really right?  Should mac-leopard fallback to mac-snowleopard?

Yes.  The fallback logic somewhat convoluted and revolves around the expected migration path for baselines when adding new version of a platform.

> Tools/Scripts/webkitpy/layout_tests/port/qt.py:52
> +        if platform.startswith('linux'):
> +            return "linux"
> +        elif platform in ('win32', 'cygwin'):
> +            return "win"
> +        elif platform == 'darwin':
> +            return "mac"
> +        return None

This feels more general than its current location.
Comment 3 Eric Seidel (no email) 2011-07-06 23:48:50 PDT
Comment on attachment 99943 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=99943&action=review

>> Tools/Scripts/webkitpy/layout_tests/port/qt.py:52
>> +        return None
> 
> This feels more general than its current location.

Yeah, but it's specific to how Qt names its ports.  Eventually we will standardize our port naming and share more of this logic.
Comment 4 Eric Seidel (no email) 2011-07-06 23:50:38 PDT
(In reply to comment #2)
> (From update of attachment 99943 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=99943&action=review
> 
> > Tools/Scripts/webkitpy/layout_tests/port/mac_unittest.py:77
> > -        port = mac.MacPort()
> > +        port = MacPort(filesystem=MockFileSystem(), user=MockUser(), executive=MockExecutive())
> 
> This all seems very copy/pasted.

Yup.  Eventually we'll need to share more here.  But these are strictly better than before (which were talking to your real filesystem).

> > Tools/Scripts/webkitpy/layout_tests/port/mac_unittest.py:137
> > +        # FIXME: Is this really right?  Should mac-leopard fallback to mac-snowleopard?
> 
> Yes.  The fallback logic somewhat convoluted and revolves around the expected migration path for baselines when adding new version of a platform.

Yeah, eventually I'd like to standarize fallback across all port flavors.

> > Tools/Scripts/webkitpy/layout_tests/port/qt.py:52
> > +        if platform.startswith('linux'):
> > +            return "linux"
> > +        elif platform in ('win32', 'cygwin'):
> > +            return "win"
> > +        elif platform == 'darwin':
> > +            return "mac"
> > +        return None
> 
> This feels more general than its current location.

That's exactly the way its supposed to feel.  Right now we can't really share this, but eventually we will when we standardize naming.
Comment 5 Eric Seidel (no email) 2011-07-06 23:51:10 PDT
I'll try to watch the Qt bots to make sure this doesn't break them.
Comment 6 Eric Seidel (no email) 2011-07-06 23:56:54 PDT
Committed r90543: <http://trac.webkit.org/changeset/90543>