Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!mouse
From: mouse@thunder.mcrcim.mcgill.edu (der Mouse)
Newsgroups: comp.lang.c
Subject: Re: strcmp
Message-ID: <1991Jun22.135109.26901@thunder.mcrcim.mcgill.edu>
Date: 22 Jun 91 13:51:09 GMT
References: <2695@m1.cs.man.ac.uk> <1991Jun18.074029.12226@panix.uucp>
Organization: McGill Research Centre for Intelligent Machines
Lines: 23

In article <1991Jun18.074029.12226@panix.uucp>, yanek@panix.uucp (Yanek Martinson) writes:
> For intcmp(a,b) why not just use a-b ?

In a word, overflow.

> That will tell you if they are equal and if not, which is greater.

Only when a-b doesn't overflow.  On a machine with 16-bit
two's-complement ints, consider a = -32000 and b = 32000.  Then a < b
(I trust nobody disagrees with this :-), so we want a negative result.
Unfortunately a-b overflows.  If it actually produces a result (as
opposed to a crash at run-time), the most likely thing to get, I would
say, would be 1536, which is not exactly negative.

> For strcmp(s,t): while(*s++==*t++&&*s&&*t); return *s-*t;

I think this has been sufficiently dissected already, so I'll skip it.

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu