There is some residual code that is still using (unsigned)-1 to express std::numeric_limits<unsigned>::max(). We should fix them.
Created attachment 248346 [details] the patch.
Comment on attachment 248346 [details] the patch. Clearing flags on attachment: 248346 Committed r181343: <http://trac.webkit.org/changeset/181343>
All reviewed patches have been landed. Closing bug.
Why not use UINT_MAX? That's much clearer.
(In reply to comment #4) > Why not use UINT_MAX? That's much clearer. I was misled by http://www.cplusplus.com/reference/climits/ which that UINT_MAX is 16 bit. I see from http://en.cppreference.com/w/cpp/types/climits that I was wrong. If you feel strongly about it, I can switch all the sites to UINT_MAX.