Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!wuarchive!sdd.hp.com!spool.mu.edu!news.nd.edu!mentor.cc.purdue.edu!purdue!haven!umd5!jupiter.astro.umd.edu!jjk
From: jjk@jupiter.astro.umd.edu (Jim Klavetter)
Newsgroups: comp.lang.c
Subject: followup to Re: unbelievable "simple" question/bug?
Summary: why do I need a cast?
Message-ID: <8333@umd5.umd.edu>
Date: 28 Mar 91 23:19:24 GMT
References: <8331@umd5.umd.edu> <17989@crdgw1.crd.ge.com>
Sender: news@umd5.umd.edu
Reply-To: jjk@astro.umd.edu (Jim Klavetter)
Distribution: na
Organization: U of Md., Astronomy Program
Lines: 26

>
>Are you sure they are all ints? If one of either a or b or c is really
>an UNSIGNED int, then a-b-c would be some big positive number, which
>would explain the first example. If temp is an int, then putting that
>big positive integer back in temp would do what you want, since temp 
>would then have the value -1.                                 
>==================
>Chris Volpe

Thanks to Chris and the others who e-mailed me, this was indeed the
case.  I took the original expression and added a cast to make all
types "int" and it worked as expected.  According to PROGRAMMING IN
ANSI C by Kochan, there are 6 rules for data type conversions.
Unfortunately, nowhere in those rules does the question of int vs
unsigned (int) enter.  Upon reflection, I think this is a mistake, or
at least shortcoming, of this text.  Does K&R2 explicitely mention
this?  

Also, can someone post or send the full set of conversion rules?  I
would like to know exactly what I was comparing.  Thanks again.

jjk@astro.umd.edu also for Athabasca and Reudi
Jim Klavetter
Astronomy
UMD
College Park, MD  20742