X-Authentication-Warning: delorie.com: mailnull set sender to djgpp-workers-bounces using -f
From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <10201061745.AA17960@clio.rice.edu>
Subject: Re: bash's 'test' is inconsistent on XP (causing autoconf testsuite failures)
To: eliz AT is DOT elta DOT co DOT il
Date: Sun, 6 Jan 2002 11:45:09 -0600 (CST)
Cc: djgpp-workers AT delorie DOT com
In-Reply-To: <7458-Sun06Jan2002182007+0200-eliz@is.elta.co.il> from "Eli Zaretskii" at Jan 06, 2002 06:20:08 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

> > We should track down which call isn't behaving properly and see if
> > we can fix it in the libc first.
> 
> Yes, I agree.
> 
> The two methods used by lstat to detect devices are a bit in
> attribute byte returned by findfirst and the special form in which
> _truename returns the normalized string for devices.  IIRC, you've
> found that none of them is emulated correctly by W2K under LFN.
> 
> What we need is some system call--any system call--that can
> distinguish between files and devices.

What we did for fstat() was to recognize the os_trueversion, return
value looking like a file but no valid other information (time/length)
and say it's a device.  I hope there is some combination in stat
which looks the same.

Hopefully a test in the right place will work OK.