Path: csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Keith Thompson <kst-u@mib.org>
Newsgroups: comp.lang.c
Subject: Re: while loop vs do while loop reading from the file issue
Date: Sun, 11 Dec 2016 12:56:10 -0800
Organization: None to speak of
Lines: 32
Message-ID: <ln4m2atcb9.fsf@kst-u.example.com>
References: <52f1c174-6af1-4fda-9ff8-a5e2cf507b5f@googlegroups.com> <o29hq3$c7p$1@dont-email.me> <o29i1r$cfv$1@dont-email.me> <o29id9$egc$1@dont-email.me> <ihng4clp2cjofv1ggvdfu0oqqiglsu9u6f@4ax.com> <o29ocj$519$1@dont-email.me> <5d94897b-fedc-4a28-81d0-3579e9bf754d@googlegroups.com> <o2b2dn$on2$1@dont-email.me> <4d53f1c5-b622-4425-9dfe-694c42676cae@googlegroups.com> <o2dvo8$j82$1@dont-email.me> <5d68da29-653c-42f5-8dbe-c11d3b304322@googlegroups.com> <o2gsv2$qqp$1@dont-email.me> <o2jp4d$ffe$2@dont-email.me> <o2jr4i$nao$1@dont-email.me> <o2judn$a99$3@dont-email.me> <o2jvie$uvi$1@dont-email.me>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Info: mx02.eternal-september.org; posting-host="d0d5c56ca22fa2e528cedb366e6a556f"; logging-data="4039"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/uvflfq7sM/jkcRgiVS3nv"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:oA/LebYPeq/44G1tVpP+TGmoMAE= sha1:01o8h5n1qUWgyZNxjVBaJyj7mcM=
Xref: csiph.com comp.lang.c:98892

BartC <bc@freeuk.com> writes:
> On 11/12/2016 16:22, Richard Heathfield wrote:
[...]
>> Do you really think that reading data from a stream is isomorphic to
>> filling a series of glasses?
>
> So what /is/ it like then? You blindly keep on doing something until it 
> breaks and using that to tell you've finished (and it's time to pick up 
> the pieces).

Yes.

That's not as sensible solution if you're backing up a car or filling a
series of glasses, because in those contexts a failure in one step can
have harmful side effects.

An input function that fails because there's no more data to read will
report failure *without* doing any further damage.  It is therefore
sensible to attempt to read input and then check whether the attempt
succeeded.

> Sometimes there is no choice, but here there is.

Sure, there's a choice.  Checking for end-of-file at the top of the
loop, given the way C's input functions are designed, is not the best
one.

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Working, but not speaking, for JetHead Development, Inc.
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"