Subject: Re: Non-compliant strxfrm
MIME-Version: 1.0
Content-Type: text/plain;
	charset="Windows-1252"
Date: Tue, 2 Sep 2003 09:37:41 +0100
content-class: urn:content-classes:message
X-MimeOLE: Produced By Microsoft Exchange V6.0.4417.0
Message-ID: <F1A9C8D7A58D1B45A9C16FE7E3DA83D702188F@server.HME.hme.ltd.uk>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: Re: Non-compliant strxfrm
Thread-Index: AcNxLPt4hwwr5d0hEdeGDgBQvzMKIA==
From: "Melvin Curran" <Melvin AT HME DOT Ltd DOT uk>
To: <djgpp-workers AT delorie DOT com>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by delorie.com id h828bsI01725
Reply-To: djgpp-workers AT delorie DOT com

Right, below is the new version of the function which I believe matches
the
linux output (a few mods to the original function were easiest).

The only thing bothering me now is the the statement in the standard
that says "If n is zero, dst is permitted to be a null pointer."
Does this imply that if n isn't zero then dst must point to something?
If so, what should we do about it? What does linux/glibc do?
At the moment, I am ignoring it because the standard doesn't explicitly
say anything. I should probably take this question to comp.std.c.


size_t
strxfrm(dst, src, n)
     char *dst;
     const char *src;
     size_t n;
{
  size_t r = 0;
  int c;

  while ((c = *src++) != 0)
  {
    r++;
    if (n < 2)
    {
      while (*src++ != 0)
        r++;
      break;
    }
    --n;
    if (dst)
      *dst++ = c;
  }
  if (dst && n)
    *dst = 0;
  return r;
}

-- 
||  ||   |||||| | HME Ltd.          |
||  ||| |||     |                   |
|||||| | |||||  |                   |
||  ||   ||     | Melvin Curran     |
||  ||   |||||| | melvin AT hme DOT ltd DOT uk |