On Oct 15,  8:13am, "Rakhesh Sasidharan" wrote:
} 
} >      That is the master site, so nowhere else will have a fresher
} > copy.  To use CVS, just 'cd /usr/pkgsrc' (or wherever you unpacked the
} > tarball) and type 'cvs up -dP'.
} 
} Thank you John. I was under the impression if I want to keep updating
} with CVS, then I'll have to start all over with CVS. As in, (1) delete
} the current /usr/pkgsrc (since it was created from the tarball), (2)
} then fetch things with a CVS checkout, and (3) then keep updating
} using "cvs upd -dP". From your post, however, I understand that I can
} just use "cvs upd -dP" on whatever I currently have (extracted from
} the tarball), and that should work fine. Right? Cool.

     The tarball contains the metainformation needed by CVS, so yes,
you can use it as a starting point.

} I have one more question. In the pkgsrc users' guide
} (http://www.netbsd.org/Documentation/pkgsrc/getting.html), I see these
} two paras that I don't quite understand:
} * Please also note that it is possible to have multiple copies of the
} pkgsrc hierarchy in use at any one time - all work is done relatively
} within the pkgsrc tree.

     You can create /usr/pkgsrc1 and /usr/pkgsrc2 and everything will
just work.  However, very few people will have a need to do this
(primarily people doing pkgsrc development work).

} * If your copy of pkgsrc contains a lot of CVS directories, you can
} update it using the cvs(1) program. First, cd to the top level
} directory of pkgsrc. Then run cvs -q update -dP, and you're done.

     If you see CVS subdirectories all over the place under /usr/pkgsrc
then you can use cvs to update it.  The "(1)" in "cvs(1)" tells you
what section of the manpages (type 'man man') the command is documented
in.  The "-q" just tells cvs to be quieter (i.e. don't give as much
information as it usually does).

}-- End of excerpt from "Rakhesh Sasidharan"