[EMAIL PROTECTED] wrote:
Author: lnicoara Date: Thu Sep 6 11:04:23 2007 New Revision: 573328
Great, thanks. Now if you could also add the test case to the regression suite that would be swell :)
URL: http://svn.apache.org/viewvc?rev=573328&view=rev Log: 2007-09-06 Liviu Nicoara <[EMAIL PROTECTED]> * rw/_traits.h: added cast to unsigned char
You forgot the reference to the issue that this change fixes (in both Change Logs), and the name of the function you modified. I should have pointed that out (and you should have read the Patch Format section on the Bugs page: http://incubator.apache.org/stdcxx/bugs.html#patch_format). Here's what the Change Log should have looked like: 2007-09-06 Liviu Nicoara <[EMAIL PROTECTED]> STDCXX-541 * _traits.h (find): Added cast to unsigned char to prevent sign extension when char is a signed type. Martin
Modified: incubator/stdcxx/trunk/include/rw/_traits.h Modified: incubator/stdcxx/trunk/include/rw/_traits.h URL: http://svn.apache.org/viewvc/incubator/stdcxx/trunk/include/rw/_traits.h?rev=573328&r1=573327&r2=573328&view=diff ============================================================================== --- incubator/stdcxx/trunk/include/rw/_traits.h (original) +++ incubator/stdcxx/trunk/include/rw/_traits.h Thu Sep 6 11:04:23 2007 @@ -406,8 +406,9 @@ // cast to const void* used to get around a gcc 2.95 bug // that prevents a static_cast from void* --> const T* // (only occurs if memchr() isn't overloaded on const) - return _RWSTD_STATIC_CAST (const char_type*, - (const void*)_RWSTD_MEMCHR (__s, __c, __n)); + return _RWSTD_STATIC_CAST ( + const char_type*, (const void*)_RWSTD_MEMCHR ( + __s, _RWSTD_STATIC_CAST (unsigned char, __c), __n)); }static _RWSTD_SIZE_T length (const char_type *__s) {
