From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <10108201920.AA12044@clio.rice.edu>
Subject: Re: Fseek on STDIN problem on Win 2K
To: eliz AT is DOT elta DOT co DOT il
Date: Mon, 20 Aug 2001 14:20:51 -0500 (CDT)
Cc: acottrel AT ihug DOT com DOT au, djgpp-workers AT delorie DOT com
In-Reply-To: <2593-Mon20Aug2001213300+0300-eliz@is.elta.co.il> from "Eli Zaretskii" at Aug 20, 2001 09:33:01 PM
X-Mailer: ELM [version 2.5 PL2]
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> But I'm actually more bothered by your saying that _get_dev_info
> returns zero for redirected stdin.  I thought you tested that several
> days ago and found preconnected handles to be in good shape as far as
> IOCTL calls were concerned, no?  If that's not true, we might have
> problems with isatty which is used _very_ heavily in many
> applications.

When stdin is CON the device info is correct.  I did not try it on 
a redirected handle.  If we are redirected from a file we don't get
the device of the file correct, but that's probably not a huge problem
since it does not appear to be a device.

Why I tested for this and brought it up was not that it will break
isatty - but that it appears to be a lfn handle - and we know that
they behave poorly for some calls like IOCTL that might break other
things.  The huge problems were IOCTL on device handles that looked like
files - file handles that look like device A: instead of C: is less of
a problem.

For example, we know that the set time calls fail on LFN handles but
work on SFN handles; now it appears that seek may be flkey on them also.