Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!lll-winken!lll-crg.llnl.gov
From: berry@lll-crg.llnl.gov (Berry Kercheval)
Newsgroups: comp.lang.c
Subject: Re: lint won't verify printf formatting against variable types??
Keywords: lint, cc
Message-ID: <27762@lll-winken.LLNL.GOV>
Date: 29 Jun 89 20:17:10 GMT
References: <328@tree.UUCP> <417@siswat.UUCP> <441a30e5.8f55@fiero.engin.umich.edu> <27729@lll-winken.LLNL.GOV> <2271@trantor.harris-atd.com>
Sender: usenet@lll-winken.LLNL.GOV
Reply-To: berry@lll-crg.llnl.gov (Berry Kercheval)
Organization: Lawrence Livermore National Laboratory
Lines: 24
In-reply-to: bbadger@x102c.harris-atd.com (Badger BA 64810)

In article <2271@trantor.harris-atd.com>, bbadger@x102c (Badger BA 64810) writes:
>In article <27729@lll-winken.LLNL.GOV> berry@lll-crg.llnl.gov (Berry Kercheval) writes without thinking....:
>>constructs.  This is NOT an "obvious" semantic error.  
>
>Give the guy a break -- he didn't say he expected ``cc'' to catch it, he 
>just hoped that ``lint'' would.  That's lint's job: finding *legal* constructs
>which aren't portable or probably aren't what was intended.


Yeah, you're right.  I focused on the 'useless null statement' and
didn't consider when it's useful and when it isn't.  Of course, I
figured this out on the way home last night when it was too late...

Just to save myself anymore angry mail telling me how stupid I was, let me
summarize what we seem to agree on:

	"if( expr );" is useless, but legal, and a case can be made for
	issuing a warning if it's detected, but it should not be forbidden.

OK?

  --berry