From: krhohio AT my-deja DOT com
Newsgroups: comp.os.msdos.djgpp
Subject: Re: seekg() bug
Date: Wed, 09 Aug 2000 02:14:19 GMT
Organization: Deja.com - Before you buy.
Lines: 36
Message-ID: <8mqepq$s50$1@nnrp1.deja.com>
References: <8mg9j5$fis3 AT imsp212 DOT netvigator DOT com> <38nnos876cfdjok9gbkn9sn7968nnjh054 AT 4ax DOT com>
NNTP-Posting-Host: 12.23.19.70
X-Article-Creation-Date: Wed Aug 09 02:14:19 2000 GMT
X-Http-User-Agent: Mozilla/4.05 [en] (WinNT; I)
X-Http-Proxy: 1.0 x68.deja.com:80 (Squid/1.1.22) for client 12.23.19.70
X-MyDeja-Info: XMYDJUIDkrhohio
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

I seem to be having the same problem using Microsoft VC++ 6.0
to compile a "console application".  I called "bad()" after
detecting "eof()" and it returns false (as expected).
Do I need to add "clear()" after eof is detected?

Thanks,
Kent


In article <38nnos876cfdjok9gbkn9sn7968nnjh054 AT 4ax DOT com>,
  Jason Green <news AT jgreen4 DOT fsnet DOT co DOT uk> wrote:
> "David Lee" <keeyu AT poetic DOT com> wrote:
>
> > I opened a fstream object and tried to position the file pointer
with seekg
> > ().
> >
> > I found that seekg () always worked as long as eof () was not
encountered.
> > Once eof () holds, it would appear to work as usual, but subsequent
reading
> > from the fstream object were wrong.
> >
> > Any idea?
>
> You are not checking the stream for errors.  The stream state can be
> tested with (f), f.good(), f.eof(), f.fail() & f.bad(), which each
> have a specific meaning.
>
> My C++ reference says that if fail() is true the next operation will
> fail.  The state bits can be reset with f.clear().
>


Sent via Deja.com http://www.deja.com/
Before you buy.