From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=0.2 required=5.0 tests=BAYES_00,INVALID_MSGID,
	REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII
X-Google-Thread: 1108a1,9a0ff0bffdf63657
X-Google-Attributes: gid1108a1,public
X-Google-Thread: fac41,9a0ff0bffdf63657
X-Google-Attributes: gidfac41,public
X-Google-Thread: f43e6,9a0ff0bffdf63657
X-Google-Attributes: gidf43e6,public
X-Google-Thread: 103376,4b06f8f15f01a568
X-Google-Attributes: gid103376,public
From: Mats Weber <Mats.Weber@elca-matrix.ch>
Subject: Re: Software landmines (loops)
Date: 1998/09/10
Message-ID: <35F74C47.178B2B89@elca-matrix.ch>#1/1
X-Deja-AN: 389627367
Content-Transfer-Encoding: 8bit
References: <6sbuod$fra$1@hirame.wwa.com> <35f51e53.48044143@
 <6t6l4n$rep@jbm.nada.kth.se>
Content-Type: text/plain; charset=iso-8859-1
Organization: Swisscom AG, the blue window
Mime-Version: 1.0
Reply-To: Mats.Weber@elca-matrix.ch
Newsgroups: comp.lang.eiffel,comp.object,comp.software-eng,comp.lang.ada
Date: 1998-09-10T00:00:00+00:00
List-Id: <comp.lang.ada>

Jonas M�ls� wrote:

>         Also, I have never encountered a real situation
>         where I have needed an extra flag in the predicate.
>         One can actually do fine without them. As an example,
>         I hereby present to you three variations of str_index.
> 
>         The first one strictly exits at the top. The other two
>         in the middle. The first one is much easier to reason
>         about, and to understan, IMHO. Anyway, introducing an
>         extra flag in the predicate is very often a sign of
>         "patching the loop". Done right, one does not need it.
> 
> 
> char *
> str_index (
>     char   *s,                  /* string to search           */
>     char    c)                  /* character to search for    */
> {
>     while (*s != '\0'  &&  *s != c)
>         ++s;
> 
>     return   (*s != c)  ?  NULL : s;
> }

The problem with this variant is that you are comparing *s with c once more
than in the other two variants. With characters, this is perfectly OK, but
this could become a problem if the comparison for equality is more expensive,
for instance if you are comparing larger structures or have a more complicated
equality function.

In fact, you are using "*s == c" as a flag to check why you exited form your
loop. Try to reprogram your example doing each comparison only once: you will
need a flag if you don't use return in the loop.

(I find the other two variants more readable, but that is a matter of taste).