Larry Jones wrote:
> Ingolf Steinbach writes:
> >
> > Do you have a pointer to information on which commands are affected? We
> > have used this method for a while on a busy CVS server without encountering
> > any porblems. Would it make sense to modify these assertions for future CVS
> > releases such that symbolic links work?
>
> It seems to me that it's mostly the r* commands, but I'm not sure. The
> problem is that CVS tends to change directory inside the repository and
> then do pwd to find out where it is. If the root is a symlink, the
> result of the pwd doesn't start with the CVSROOT and CVS gets terminally
> confused.
Donald Sharp proposed what I think is the proper fix to this some time ago on
bug-cvs:
http://www.mail-archive.com/[email protected]/msg00852.html
Unfortunately he never implemented it. Patches will be happily accepted. :)
Derek
--
Derek Price CVS Solutions Architect ( http://CVSHome.org )
mailto:[EMAIL PROTECTED] CollabNet ( http://collab.net )
--
Bugs come in through open Windows.
_______________________________________________
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
