Re: should abs() be 'uint'?

Article View: comp.std.c

Article #2169

#2169

From: Henry Spencer
Date: Wed, 30 Sep 1992 19:35

13 lines
1160 bytes

In article <1992Sep30.040719.2852@waggen.twuug.com> alpha@waggen.twuug.com (Joe Wright) writes:
>Aren't we missing the point here?  The whole point of abs() is
>to take a signed value and return a positive result...
>If the tested value can actually range to -32768 (16 bits) we are
>using the wrong type.  Use the next larger type (long?) instead.

Either making the returned value unsigned or using a larger signed type
would make sense *if we were starting from scratch*.  It is not feasible
to change the type of the existing abs() function, however; the mistake
was made long ago and it is too late to fix it.
-- 
There is nothing wrong with making      | Henry Spencer @ U of Toronto Zoology
mistakes, but... make *new* ones. -D.Sim|  henry@zoo.toronto.edu  utzoo!henry

Technical Details

Message-ID: <BvEp3G.540@zoo.toronto.edu>

Path: arc1-ca.pugleaf.net!sparky!uunet!stanford.edu!rutgers!utcsri!utzoo!henry

References: <BvAzvK.EFx@zoo.toronto.edu> <1992Sep30.040719.2852@waggen.twuug.com>

Show Full Headers

Complete Article Headers

Path: arc1-ca.pugleaf.net!sparky!uunet!stanford.edu!rutgers!utcsri!utzoo!henry
From: henry@zoo.toronto.edu (Henry Spencer)
Newsgroups: comp.std.c
Subject: Re: should abs() be 'uint'?
Message-ID: <BvEp3G.540@zoo.toronto.edu>
Date: 30 Sep 92 19:35:37 GMT
References: <BvAzvK.EFx@zoo.toronto.edu> <1992Sep30.040719.2852@waggen.twuug.com>
Organization: U of Toronto Zoology
Lines: 13

Reply to Parent Message