[Webkit-unassigned] [Bug 43077] check-webkit-style false positive on "new uint32_t"

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=43077





--- Comment #8 from Kenichi Ishibashi <bashi at google.com>  2010-08-03 08:56:06 PST ---
(In reply to comment #7)
Hi Stephen,

> Just FYI, I originally used int32_t* buf = new int32_t[size], which gave no warnings from check-webkit-style.  That might be because of the brackets, or some other parsing issue, but switching to OwnArrayPtr<uint32_t> definitely made it start happening.

Switching to OwnArrayPtr<uint32_t> would involve introducing parentheses on the same line and such parentheses are the cause of this false positive. Since parsing declarations correctly is complicate process, check-webkit-style simply assumes that the line would include function arguments when an open parenthesis follows after an identifier. As a result, check-webkit-style could evaluate 'new' as a type and 'uint32_t' as an identifier for the case mentioned in the description of this issue.

Another way to fix this issue is treating uint32_t and it's family as special case like 'const_iterator'. What do you think which way is appropriate? I'd like to ask your opinion.

Thanks,

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.