On Tue, Jun 07, 2005 at 10:52:59AM -0500, John Goerzen wrote: > On Tue, Jun 07, 2005 at 12:33:54PM -0300, [EMAIL PROTECTED] wrote: > > > > The commands like unrecord and amend-record are inevitable consequences of > > decentralized revision control. If the commands did not exist, you could > > easily achieve the same effect by creating a second repository with all > > patches except the offending patch. That achieves unrecord. Then you > > could > > apply the contents of the offending patch with "diff | patch", alter the > > contents, and record the newly altered patch. That achieves amend-record. > > True, but that still doesn't explain their utility... I guess I struggle > to see why it's easier to do this than to just check in a patch that > adjusts things to the way they should be, as one would so in most any > other VC system.
As others have said, it's nice to have a clean history. A dirty history is all right when it's just you working on the project, but "never mind" patches lead to conflicts with other developers. I agree that if you fix a bug then realize the bugfix is wrong, probably you shouldn't amend-record that patch. But if you add a new feature, but forget to darcs add the relevant file, it makes perfect sense to amend-record the patch. I think of darcs more as a means of communication than a means of storing history, and clean patches make it much easier to read and review changes. -- David Roundy _______________________________________________ darcs-users mailing list [email protected] http://www.abridgegame.org/mailman/listinfo/darcs-users
