[cvsnt] problems with edit/commit -c
jerzyk at wndtabs.com
Tue Sep 17 16:15:31 BST 2002
>'commit -c' just checks to see if you're a valid editor of the file.
It appears that it doesn't actually, and looks like a bug to me. It should
not allow to commit (if used the commit -c) if you don't have a valid edit
on the file, just like it doesn't allow to "edit -c" if someone else did
Of course there is always "-f" to force commit and edit, but general "-c"
should be reserved as it claims by it's name ;)
----- Original Message -----
From: "Tony Hoyle" <tmh at nodomain.org>
To: <cvsnt at cvsnt.org>
Sent: Tuesday, September 17, 2002 4:48 PM
Subject: Re: [cvsnt] problems with edit/commit -c
> On Mon, 16 Sep 2002 19:44:40 +0000, Kim Sparrow wrote:
> > The 'edit -c' part seems to work fine. When the other person tries to
> > do an 'edit -c', it says "cvs [edit aborted]: files being edited!"
> > Good. When the other tries to do a plain 'edit', the CVS\fileattr file
> > in the repository correctly notes that both are now editors of the file.
> > But when only one of us has a reserved edit and the other tries a
> > 'commit -c' the file gets committed as if it wasn't locked, no warning
> > message or anything. From
> > http://www.devguy.com/fp/cfgmgmt/tools/install.htm I think I understand
> > how it's supposed to work.. and yet what I'm observing doesn't quite
> > match up. I don't need to do any scripting in the CVSROOT directory
> > like in the old 'admin -l' days do I? I'm using 57i.
> CVS doesn't distinguish between a 'reserved' edit and a 'plain' edit, so
> you can circumvent the 'edit -c' system in the way you describe. 'commit
> -c' just checks to see if you're a valid editor of the file. If you want
> to use reserved edits all the time you can tell CVS to always use '-c' by
> using a .cvsrc file in the home directory containing the default options.
> cvsnt mailing list
> cvsnt at cvsnt.org
More information about the cvsnt