Subject: Re: fixpath patch (rev 3)
From: Tim Van Holder <tim DOT van DOT holder AT pandora DOT be>
To: djgpp-workers AT delorie DOT com
In-Reply-To: <200110010921.LAA18740@lws256.lu.erisoft.se>
References: <200110010921 DOT LAA18740 AT lws256 DOT lu DOT erisoft DOT se>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Evolution/0.13.99+cvs.2001.09.11.22.18 (Preview Release)
Date: 01 Oct 2001 12:26:07 +0200
Message-Id: <1001931968.21287.12.camel@bender.falconsoft.be>
Mime-Version: 1.0
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

On Mon, 2001-10-01 at 11:21, Martin Stromberg wrote:
> > Not good: there could be a real directory "/bogus" on that drive.
> > 
> > "/dev/:bogus:" would be a better idea.
> 
> But we already disallow /dev/ (because of Unix compatibility).

Do we?  mkinstalldirs currently creates stray directories because
test -d /dev checks c:/dev (with C the current drive), and
test -d /dev/env checks ./env (for some reason).
I would be inclined to say that usurping /dev as our own virtual
tree would be a good thing, but I believe Eli doesn't agree.

> Plus we already have /dev/env. In what way would /dev/bogus be
> different?
It wouldn't be different in the sense that it also removes that
name from the user's namespace.

Note that I would prefer /dev/[bogus] or something similar, to avoid
any and all issues with colons (either as drive or path separator).