On Wed, Aug 19, 2009 at 22:28:47 +0200, Reinier Lamers wrote: > Here you are, sir...
Thanks. > Are replace patches actually useful, and not some historical curiosity? I > used to be somewhat afraid of them for some reason. I personally find replace patches useful because they generally commute more/reduce dependencies. Two gotchas: - Merging replace patches with different token specs will lead to a conflict, which isn't very nice. - Conflict marking for replace is virtually non-existent. That aside, I don't really think there is anything to be afraid of. Just remember that replace is extremely dumb. It just knows about 'things that fit my token spec' and 'everything else'. See http://lists.osuosl.org/pipermail/darcs-users/2009-August/020880.html for a discussion on this. You can shoot yourself in the foot by replacing more than you really intended, but I don't think it's any worse than using the search-replace-all functionality of your text editor. See also http://bugs.darcs.net/issue887 Even for people who like darcs replace, there are times where it's not clear whether to just use plain old hunk patches (no risk of replacing too many things there). For this kind of work, i.e. doing a bunch of boring camel casing, it's very clear-cut. > Will it do the right thing if someone later adds a comment 'this used > to be called xml_summary', and the comment-adding patch and the > replace patch are commuted back and forth? That would depend on the replace patch, so no commutation. -- Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow> PGP Key ID: 08AC04F9
pgpzhsPUQGisl.pgp
Description: PGP signature
_______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
