Bug 51303 – check-webkit-style should detect function declarations (and trivial functions).

• Status: RESOLVED
• Resolution: FIXED
• Component: Tools / Tests

[David Levin]

This is needed for bug 49394 (which examines declarations).

[David Levin]

Created attachment 76955 [details]
Patch

[Shinichiro Hamaji]

Comment on attachment 76955 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=76955&action=review

Looks good, putting some nitpicks on test. Please feel free to ignore some of (or even all of) them if they don't make sense.

> Tools/Scripts/webkitpy/style/checkers/cpp_unittest.py:240
> +        if not function_information:

I'd add a trivial test case which passes this case. Not sure, but "if (foobar) { something... }" or "for (foobar;...) { ... }" would be good candidates?

> Tools/Scripts/webkitpy/style/checkers/cpp_unittest.py:2818
> +        self.assert_pass_ptr_check(

s/pass_ptr/pass_ref_ptr/ ?

> Tools/Scripts/webkitpy/style/checkers/cpp_unittest.py:2823
> +        self.assert_pass_ptr_check(

s/pass_ptr/pass_ref_ptr/ ?

> Tools/Scripts/webkitpy/style/checkers/cpp_unittest.py:2837
> +            '    PassRefPtr<Type1> m_other;\n'

I guess RefPtr would be more realistic?

[David Levin]

(In reply to comment #2)
> (From update of attachment 76955 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=76955&action=review
> 
> Looks good, putting some nitpicks on test. Please feel free to ignore some of (or even all of) them if they don't make sense.
> 
> > Tools/Scripts/webkitpy/style/checkers/cpp_unittest.py:240
> > +        if not function_information:
> 
> I'd add a trivial test case which passes this case. Not sure, but "if (foobar) { something... }" or "for (foobar;...) { ... }" would be good candidates?

Done.

> > Tools/Scripts/webkitpy/style/checkers/cpp_unittest.py:2818
> > +        self.assert_pass_ptr_check(
> 
> s/pass_ptr/pass_ref_ptr/ ?

I left this as is because the test is about Pass*Ptr. (It checks Pass.*Ptr so that it catches PassRefPtr and PassOwnPtr. Other things named the same way should follow the same rule.)


> > Tools/Scripts/webkitpy/style/checkers/cpp_unittest.py:2837
> > +            '    PassRefPtr<Type1> m_other;\n'
> 
> I guess RefPtr would be more realistic?

Done. (Good point :).)

[David Levin]

Committed as http://trac.webkit.org/changeset/74356