On Wednesday 08 August 2012, Gregg Smith wrote: > 1 warning > crypt_blowfish.c > .\crypto\crypt_blowfish.c(894) : warning C4244: '=' : conversion > from 'unsigned long' to 'char', possible loss of data
That's a false positive. "count" can only be between 4 and 31 at this point. IMHO the compiler should be able to recognize that. > I have a single patch (attached) to deal with all these issues. > This is same in > apr-util 1.4.1 and patch should work for both 1.4.x & 1.5.x I cannot say anything about the other issues you mentioned. Has anyone who is familiar with the Windows toolchain looked at the patch?
