Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
Date: Thu, 6 Dec 2001 13:29:25 -0600
From: "Kim, Anthony" <Anthony DOT Kim AT vw DOT com>
To: cygwin AT cygwin DOT com, Gerald Villemure <GVillemure AT ik DOT ca>
Subject: Re: Windows 2000 Junction Points
Message-ID: <20011206192925.GD10907@nabokov.afc.vw.com>
Mail-Followup-To: cygwin AT cygwin DOT com, Gerald Villemure <GVillemure AT ik DOT ca>
References: <20011206155247 DOT GC8981 AT nabokov DOT afc DOT vw DOT com> <002201c17e75$4af6db90$0200000a AT dyn DOT ik DOT ca> <20011206173601 DOT GB10289 AT nabokov DOT afc DOT vw DOT com> <20011206191440 DOT A2441 AT cygbert DOT vinschen DOT de> <20011206184317 DOT GA10907 AT nabokov DOT afc DOT vw DOT com> <20011206200245 DOT B2441 AT cygbert DOT vinschen DOT de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20011206200245.B2441@cygbert.vinschen.de>
User-Agent: Mutt/1.3.23.2i

On Thu, Dec 06, 2001, Corinna Vinschen wrote:

> On Thu, Dec 06, 2001 at 12:43:17PM -0600, Kim, Anthony wrote:
> > 
> > A couple of comments: I agree the MS implementation is not
> > flexible.  However, if the reparse points do not cross file
> > systems, I believe they can be relative. I do this now and
> > again..
> > 
> > C:\some\dir\here\and\there\> linkd otherdir ..\..\otherdir
> 
> Linkd takes a relative path as argument but it's translated
> into an absolute path before storing it in the reparse point.

Oh, I did not know that.


> > The crappy part about the MS implementation is there doesn't seem
> > to exist an easy way to obtain the link destination. There's no
> > 'ls -l' equivalent.
> 
> It's no problem to read the reparse point content so a Cygwin
> implementation would have given that information.
> 
> No, reparse points are not useful for a POSIX emulation layer.
> Sure, it would be possible to add code to Cygwin which allows
> to treat reparse points as symlinks but I'm not going to do
> that in the near future and I'm not as interested in them as
> I was in early 2000.  At least it would again slow down the
> symlink evaluation code (the third method to look for).

OK, I got it. It was just a suggestion.


> > You're right about reparse points not working with files, but
> > hardlinks solve that issue. I was thinking in pseudo code:
> > 
> > -s flag given:
> >     if src == directory 
> >         create_junction()
> >     else
> >         create_shortcut()
> >     endif
> 
> But hardlinks are a completely different thing.  And they are
> actually supported in Cygwin for years on NTFS.  FAT can't have
> hardlinks.

Sorry, I mixed contexts. I meant to say, when I needed another
link to a file, I would normally just use a hardlink and if I
need the file to be linked across filesystems, I'd normally
junction point the parent directory - not that frequent I admit,
but hey people do some strange stuff.

Thanks for answering my query and all the hard work you've done.

Anthony


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/