Ralf S. Engelschall wrote:
>
> I've tested it with gcc on dev.openssl.org (a Solaris 2.6 box)
> and we've two problems:
>
> First GCC discovered some type inconsitencies related to the ctype functions
> isxxx() which expect an int as the argument per definition/prototype while we
> usually pass a character. I had to apply the appended patch to make it quiet,
> but I'm sure Ben votes -1 for it. Please decide on your own what to do here,
> I'm not sure. At least I've not found a better soltion than casting. BTW, the
> warnings were:
Hmmm ... we've hit this one in Apache - the problem is that casting to
int doesn't actually fix it: if the char is -ve, then you get a -ve
offset, I seem to remember, causing Bad Things to happen. I'm still
thinking about the best solution, though probably it is the same one I
was going to suggest for the more general casting problem: write
functions that do the cast and call the "real" function (but only use
them when you can't fix the problem directly), and make them inline
where they can be (e.g. under gcc). This makes the casts a great deal
safer.
Cheers,
Ben.
--
http://www.apache-ssl.org/ben.html
"My grandfather once told me that there are two kinds of people: those
who work and those who take the credit. He told me to try to be in the
first group; there was less competition there."
- Indira Gandhi
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
