This is the mail archive of the
cygwin
mailing list for the Cygwin project.
Re: chown with not existing user/group
- From: Corinna Vinschen <corinna-cygwin at cygwin dot com>
- To: cygwin at cygwin dot com
- Date: Tue, 4 Mar 2008 14:56:01 +0100
- Subject: Re: chown with not existing user/group
- References: <loom.20080228T135447-901@post.gmane.org> <008b01c87a16$308a0540$2e08a8c0@CAM.ARTIMI.COM> <loom.20080229T091201-103@post.gmane.org> <20080229092205.GQ9539@calimero.vinschen.de> <loom.20080304T132211-852@post.gmane.org> <47CD5140.5090306@byu.net>
- Reply-to: cygwin at cygwin dot com
On Mar 4 06:40, Eric Blake wrote:
> | I was thinking cygwin goal was to emulate as much as possible posix
> | spec.
>
> Yes, and cygwin's behavior in this case is still POSIX compliant.
Right. The error code in question is this:
[EINVAL]
The owner or group ID supplied is not a value supported by the
implementation.
We *could* create non-existant SIDs in the security descriptors of
files, but I don't know what the sense would be. These SIDs would
never, on no machine, have a real user or group representation. It's
not quite comparable with using non-existant uids on a Unix box. These
uids can exist on another machine where they make sense. That's not how
it is with SIDs since SIDs are bound to a computer or domain. Fake SIDs
are bound to nothing at all and are never correctly recognized.
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/