CVS update: samba/source/client

[Alexandre Oliva]

On Mar 26, 1999, Andrew Tridgell <tridge at samba.org> wrote:

> I don't we want the long long support in our snprintf code.

Should I revert it, or it is ok to just leave it in?

> We only want to use our snprintf code on systems where we really have
> to.

I see; I had assumed it would always be used.  Sorry about that.

> I think the correct patch for dir_total is to use %.0f and cast to
> double. It is portable and has plenty of bits.

Assuming we have long long, yep, that's correct.  But if we don't,
we'll overflow on 32bits.  Should we document this somewhere (or is it
documented already?) and recommend the use of gcc if the native
compiler does not support long longs?  Should this be a warning
printed by the configure script?

Do we care for compilers that support long long but not unsigned long
long?

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists