Re: HOST_NAME_MAX

[Simon Josefsson]

Bruno Haible <address@hidden> writes:

>> It seems clear that FreeBSD isn't POSIX compliant here since
>> HOST_NAME_MAX needs to be provided, as far as I can tell.
>
> So, FreeBSD is POSIX compliant. See [1]: HOST_NAME_MAX is inside a section
> that begins with:
>   "A definition of one of the symbolic constants in the following list
>    shall be omitted from <limits.h> on specific implementations where the
>    corresponding value is equal to or greater than the stated minimum, but
>    is unspecified.
>    This indetermination might depend on the amount of available memory space
>    on a specific instance of a specific implementation. The actual value
>    supported by a specific instance shall be provided by the sysconf()
>    function."

Hmm, I didn't notice that.  Then I'm less sure that it makes sense for
gnulib's limit.h to define HOST_NAME_MAX -- programs written against
POSIX should not assume that symbol exists.

The rationale for adding the symbol was to make the self-test
test-gethostname.c build.  It uses HOST_NAME_MAX, which is my fault.
The proper fix appears to use sysconf if it is available, and otherwise
fall back on HOST_NAME_MAX, and otherwise just fail.  What do you think?

Do sysconf work under mingw?  Gnulib could provide a replacement if not.

>> I guess looking into FreeBSD libc would answer this.
>
> [2] simply makes a system call. In the kernel, it's really an access of
> an array of size MAXHOSTNAMELEN [3][4].

Thanks, so at least adding it wouldn't hurt, but it doesn't seem like
The Right Thing anyway.

/Simon

> [1] http://www.opengroup.org/onlinepubs/9699919799/basedefs/limits.h.html
> [2] http://www.freebsd.org/cgi/cvsweb.cgi/src/lib/libc/gen/gethostname.c
> [3] http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/kern/kern_mib.c
> [4] http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/sys/jail.h