httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jim Jagielski <>
Subject Re: pre-patch: struct sockaddr_in error messages
Date Sat, 25 Oct 1997 00:36:32 GMT
Dean Gaudet wrote:
> On Fri, 24 Oct 1997, Jim Jagielski wrote:
> > Ben Laurie wrote:
> > > 
> > > 
> > > Not convinced. The argument about our snprintf being broken but the
> > > existing one not is spurious. If any of our code is broken, the user is
> > > scuppered.
> > > 
> > 
> > I think Marc's argument was that if our snprintf() was broken,
> > the user was hosed. By only ``concern'' was that as it stands,
> > the current ap_snprintf() function is a drop-in replacement
> > for the ANSI-standard version. If we add the 2 extensions, that
> > stops being the case. That's not a plus or a minus, but it _is_
> > a change from the current baseline.
> You mean a replacement for the POSIX version... and I'm not even sure it
> is a replacement for the POSIX version, for example we don't use it for
> floating point much so that may not even work.
> I'm actually advocating more here -- don't think of it as snprintf any
> longer, think of it as apache_printf or whatever you want.  At any rate
> it's not something I want to fight hard for, so it's really up to you
> guys. 

I think it's a neat feature and I'm not non-advocating it. The only
think I'm saying is that if someone currently uses their working
snprintf() instead of our version, Apache will still work.

What we're doing is providing an _extended_ snprintf() implementation,
one which we will _require_, not a ``replacement'' version for those
OSs that lack one. That's cool with me, but we should all understand
that (and make sure the developers realize that too :) ).

      Jim Jagielski            |       jaguNET Access Services           |
            "Look at me! I'm wearing a cardboard belt!"

View raw message