On Wed, 21 Jul 2004, Brian Gough wrote:
"Robert P. J. Day" <[EMAIL PROTECTED]> writes:
is there an accepted way of setting up a CVS repo for a number of developers, where the repo itself is updated from yet another repo? (this might be the equivalent of BK's clone of a clone.)
There is no standard way to transfer individual commit log entries etc between repositories, the most obvious method is to simply do a import onto a vendor branch on a regular basis. See the chapter on "Tracking third-party sources" in the manual for details.
yup, that looks like what i want, but there are a couple things that strike me as a bit odd.
first, if i want, as a starting point, the current contents of the vendor's CVS tree, it looks like i first have to grab it and remove all CVS info from it, effectively to turn it into a regular directory structure as if i created it, then turn around and import it to turn it *back* into a CVS repo, albeit this time as if i created it. is that right?
and as for the release tag, i guess the easiest thing is just to use a representation of the date if the vendor updates their repo fairly frequently.
sound right?
rday
_______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/info-cvs
