Bug 46523 – Use WTF::StringHasher in WTF::CaseFoldingHash

• Status: RESOLVED
• Resolution: FIXED
• Component: Web Template Framework

[Patrick R. Gansterer]

see patch

[Patrick R. Gansterer]

Created attachment 68762 [details]
Patch

[Early Warning System Bot]

Attachment 68762 [details] did not build on qt:
Build output: http://queues.webkit.org/results/4038129

[Patrick R. Gansterer]

Created attachment 69977 [details]
Patch

[WebKit Commit Bot]

Comment on attachment 69977 [details]
Patch

Clearing flags on attachment: 69977

Committed r69472: <http://trac.webkit.org/changeset/69472>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[WebKit Review Bot]

http://trac.webkit.org/changeset/69472 might have broken Chromium Win Release

[Evan Martin]

This fails on our Clang buildbots.

 CXX(target) out/Debug/obj.target/wtf/third_party/WebKit/JavaScriptCore/wtf/WTFThreadData.o
In file included from
third_party/WebKit/JavaScriptCore/wtf/text/AtomicString.cpp:25:
third_party/WebKit/JavaScriptCore/wtf/text/StringHash.h:107:52:error:
'foldCase' is a private member of 'WTF::CaseFoldingHash'
           return StringHasher::createHash<UChar, foldCase<UChar>
>(data, length);
                                                  ^~~~~~~~~~~~~~~
In file included from
third_party/WebKit/JavaScriptCore/wtf/text/AtomicString.cpp:23:
In file included from
third_party/WebKit/JavaScriptCore/wtf/text/AtomicString.h:24:
In file included from
third_party/WebKit/JavaScriptCore/wtf/text/AtomicStringImpl.h:24:
In file included from
third_party/WebKit/JavaScriptCore/wtf/text/StringImpl.h:32:
third_party/WebKit/JavaScriptCore/wtf/StringHashFunctions.h:91:69:
note: while substituting explicitly-specified template arguments into
function template 'createHash'
   template<typename T, UChar Converter(T)> static inline unsigned
createHash(const T* data, unsigned length)
                                                                   ^

Clang is much more picky about C++ language laws.  I think since you really are using this function from outside of the class, the function should maybe be public.

[Adam Barth]

> This fails on our Clang buildbots.

How hard is it to set up a clank build of WebKit?  We could either set up a buildbot or an EWS bot to catch these things automatically.

[Nico Weber]

abarth: It's pretty easy, see http://code.google.com/p/chromium/wiki/Clang

[Patrick R. Gansterer]

(In reply to comment #7)
> Clang is much more picky about C++ language laws.  I think since you really are using this function from outside of the class, the function should maybe be public.

Does changing foldCase to public fix the problem?

(In reply to comment #8)
> How hard is it to set up a clank build of WebKit?  We could either set up a buildbot or an EWS bot to catch these things automatically.
+1

[Nico Weber]

(In reply to comment #10)
> (In reply to comment #7)
> > Clang is much more picky about C++ language laws.  I think since you really are using this function from outside of the class, the function should maybe be public.
> 
> Does changing foldCase to public fix the problem?

Yes.

[Patrick R. Gansterer]

(In reply to comment #11)
> (In reply to comment #10)
> > (In reply to comment #7)
> > > Clang is much more picky about C++ language laws.  I think since you really are using this function from outside of the class, the function should maybe be public.
> > 
> > Does changing foldCase to public fix the problem?
> 
> Yes.

Committed r69535: <http://trac.webkit.org/changeset/69535>