-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Dan White <[EMAIL PROTECTED]> writes:
> On Thu, 29 Jul 2004, Mark D. Baushke wrote:
>
> > Dan White <[EMAIL PROTECTED]> writes:
> >
> > > On Thu, 29 Jul 2004, Mark D. Baushke wrote:
> > >
> > > > Dan White <[EMAIL PROTECTED]> writes:
> > > >
> > > > > I envisioned a cron script, run from the
> > > > > build manager account that does a mass
> > > > > checkout/watch on/watch add. But that
> > > > > strikes me as criminally kludgy.
> > > > >
> > > > > Is there a better way to do it ?
> > > >
> > > > You probably only need to add new 'watch'
> > > > statements as new files are added to the
> > > > system. So, something that is wired into the
> > > > 'loginfo' to look for new files and have them
> > > > watched might be the way to go.
> > > >
> > > > I don't understand why a cron script would be
> > > > needed...
> > >
> > > I'll explore that. Cron was considered as a
> > > brute-force approach.
> >
> > Hmmm... the point is that I do not understand what
> > you intended to do with your cron job.
>
> Something like this:
>
> cd scratchdir
>
> foreach project-in-the-repository
> cvs checkout project
> cvs watch on project
> cvs watch add project
> end
Actually, it may be that 'cvs watch on' may
be inherited, but 'cvs watch add' that is not.
In any case, your crontab job might be more
efficient if it does the entire repository
at once...
rm -fr scratchdir
mkdir scratchdir
cd scratchdir
cvs checkout .
cvs watch add
cvs watch on
yes | cvs release *
Your mileage may vary...
-- Mark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (FreeBSD)
iD8DBQFBCU+I3x41pRYZE/gRAuL/AKDCGJFAfSmGYbeBx0dPJP21Qb8ePwCfVY1n
f/T+kXspOhknI4t71ks04Wg=
=IbmD
-----END PGP SIGNATURE-----
_______________________________________________
Info-cvs mailing list
[EMAIL PROTECTED]
http://lists.gnu.org/mailman/listinfo/info-cvs
