Path: utzoo!utgpu!water!watmath!clyde!cbosgd!osu-cis!tut!lvc
From: lvc@tut.cis.ohio-state.edu (Lawrence V. Cipriani)
Newsgroups: comp.lang.c
Subject: Re: (So-called) ANSI C
Message-ID: <4160@tut.cis.ohio-state.edu>
Date: 10 Jan 88 14:29:00 GMT
References: <4668@pyr.gatech.EDU> <495@xyzzy.UUCP> <9930@mimsy.UUCP> <6996@brl-smoke.ARPA> <10027@ut-sally.UUCP> <1821@bsu-cs.UUCP>
Organization: The Ohio State University Dept of Computer and Information Science
Lines: 33

Rahul Dhesi writes:
	...
>MSDOS :-), but I seem to recall that while Turbo C will complain about
>
>     if (a = getinput())
>
>(and I think it ought to), it will not complain about the equivalent
>
>     if ((a = getinput()) != 0)
>
>since you are making your intention clear, both to the human reader and
>to the compiler.  I think Datalight C makes the same distinction.
>-- 
>Rahul Dhesi         UUCP:  <backbones>!{iuvax,pur-ee,uunet}!bsu-cs!dhesi

Does it also restrict assignment in whiles and the test part of a for
loop?  What about:
	if ((a = getinput()))

I disagree very strongly.  This should not be restricted by the *compiler*.
This is a perfectly legitimate thing to do, you just have to be careful.

Use a separate program (or compile time option) to *check* for this.
There is the public domain ccheck and a syntax checker (inspect) that
I wrote (you'll have to buy it from AT&T through the Toolchest) that
will flag this (among other things).

Maybe X3J11 should have a representative from OSHA to make certain the
users of ANSI C don't hurt themselves stupidly!

-- 
	Larry Cipriani AT&T Network Systems at
	cbosgd!osu-cis!tut!lvc Ohio State University