On Wed, Apr 11, 2007 at 11:40:31AM +0200, Matthias Saou wrote: > Christian Iseli wrote : > > > It was noticed that the cvs-import script gets used by many folks, not > > only for actually importing a new package into CVS, but also for > > importing new versions of an existing package. > > > > This latter usage led to a few cases where changes made in CVS got > > silently lost when the updated package was imported. > > > > After some discussions (like disallowing the use of cvs-import > > for anything else than initial import) it was decided to modify the > > script to at least show a diff of what's currently in CVS with what's > > going to be imported (so that there is a chance the maintainer will > > notice changes about to be undone), and will ask the maintainer to > > enter a commit message. > > > > Thanks to Jens Petersen for implementing the changes. > > This is definitely a good thing. > > FWIW, I'm not even using that script anymore : Now that the CVS > branches need to be "manually" created by the CVS admins, I simply do a > "cvs update -d new-package-name" and then copy all of my local files > (spec, patches, sources) to the new directory's devel sub-directory, > use "make new-sources FILES="foo1 foo2" there, test a local build, then > add/commit/tag and finally build :-) > > All this to say that cvs-import.sh isn't even actually _required_ > anymore :-) In fact I thought that "not required" meant "not allowed" anymore, since only the CVS admins should import packages until a new mechanism is put in place. Did some packagers cheat? :) -- Axel.Thimm at ATrpms.net
Attachment:
pgpdLQEXMSiEu.pgp
Description: PGP signature
-- Fedora-maintainers mailing list Fedora-maintainers@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-maintainers
-- Fedora-maintainers-readonly mailing list Fedora-maintainers-readonly@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-maintainers-readonly
