X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f
Message-ID: <413ae452$0$174$cc7c7865@news.luth.se>
From: Martin Str|mberg <ams AT speedy DOT ludd DOT ltu DOT se>
Subject: Re: scanf: strange behavior?
Newsgroups: comp.os.msdos.djgpp
References: <20040902144152 DOT GH6858 AT webhome DOT cz> <41376655 DOT 2A1807EC AT yahoo DOT com> <a4tej0ha3ogninujdhurgel3jcn0r441m2 AT 4ax DOT com> <200409022030 DOT i82KUYUY004665 AT envy DOT delorie DOT com>
User-Agent: tin/1.4.6-20020816 ("Aerials") (UNIX) (NetBSD/1.6Q (alpha))
Date: 05 Sep 2004 10:02:58 GMT
Lines: 34
NNTP-Posting-Host: speedy.ludd.ltu.se
X-Trace: 1094378578 news.luth.se 174 130.240.16.13
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

DJ Delorie <dj AT delorie DOT com> wrote:

> IIRC this was discussed on comp.std.c recently, and the djgpp-workers
> list (er, last March).  We don't need to discuss it again, we just
> need to make sure the right way is implemented.

Not on comp.std.c if my memory and search is correct. What was
discussed there was what should be returned if there are suppressed
assignments in combination with input failure. Links:
<http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&frame=right&th=96bf8969887ac993&seekm=4057500f%240%2496976%24cc7c7865%40news.luth.se#s>
<http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&frame=right&th=410bb8eb9e4390ac&seekm=3f342e20%240%24165%24cc7c7865%40news.luth.se#s>

I see that I noticed there is a problem in libc regarding returning
EOF and the test program:
<http://www.delorie.com/djgpp/mail-archives/browse.cgi?p=djgpp-workers/2004/03/09/14:24:30>.
I don't remember but I think I corrected the underlying function (that
*scanf() call) to work according to the corrected test program.

So if anyone wants to verify that I did so: check the CVS log that I
did some correction regarding this, get CVS libc and test programs,
check the test programs that you agree and compile the lot and check
if it's right.

> IIRC it was a choice between "correct" or "compatible".  Either way,
> someone is going to be disappointed.

If the above is right, I suspect the correct behaviour hasn't gotten
around to the users yet. And I think that correct might be ==
compatible too.


Right,

						MartinS