Path: utzoo!attcan!uunet!lll-winken!lll-lcc!ames!mailrus!nrl-cmf!ukma!rutgers!att!mtuxo!rolls!westmark!mole-end!mat
From: mat@mole-end.UUCP (Mark A Terribile)
Newsgroups: comp.lang.c
Subject: Re: Simple C Question (was: down)
Summary: { 0 1 } != { -( 2**N ) ... ( 2**N )-1 }
Message-ID: <128@mole-end.UUCP>
Date: 21 Dec 88 07:14:35 GMT
References: <gables.352@umigw.miami.edu> <192@broadway.UUCP> <6959@pyr.gatech.EDU>
Organization: mole-end--private system.  admin: mole-end!newtnews
Lines: 25

> I think I like one of the following two better:
> chcnt += (c == '\n') ? 2 : 1;
> chcnt += 1 + (c == '\n');
> 
> Any comments one which one of the ones we've seen is most efficient?

I certainly hope that the economy or practicality of your program does not
hinge on this!  What's more important, I think, is that unless you have good
reason for using tricks (e.g. you are writing a ``compiler'' that produces C)
your second form

	... + ( c == n )

works because [ 0 1 ] is the ordered set of things you want to
AND COINCIDENTALLY [ 0 1 ] is the range of the C equality operators.

Making your code depend on such a coincidence forces the reader to be mindful
of the coincidence.  If the code is human-written to be human-read, I believe
that it's very poor practice.

Besides, any respectable compiler should generate the same code for both,
modulo the assembly language label names (if such there be).
-- 

(This man's opinions are his own.)
From mole-end				Mark Terribile