Re: Merging back fixup commits
Greg Ward <[email protected]> Thu, 29 Oct 2009 09:39:08 -0400
| Newsgroups | gmane.comp.version-control.subversion.cvs2svn.devel |
|---|---|
| Message-ID | <[email protected]> |
On Wed, Oct 21, 2009 at 5:21 AM, Max Bowsher <[email protected]> wrote: > Greg Ward wrote: >> I just want to be able to specify "fewer synthetic changesets even if >> that makes history a tad dubious by making normal commits look like >> dummy merges" over "more synthetic changesets in order to avoid making >> normal commits look like dummy merges". > > Um. I'm afraid we have a disagreement here. My feeling is that cvs2svn > should assiduously avoid mixing its synthetic changes into > human-attributed changesets. Moreso, given how bafflingly configurable > cvs2svn is already, I seek to avoid adding options whenever possible. Fair enough. > Therefore, if you want to get your "eliminate synthetic changesets by > adding extra merges to real changesets which happen to be in the right > place in the tree" option into trunk, you'll have to Michael/Oswald that > it's a good idea. - Sorry :-/ OK. When I get some time, I'll see about refactoring HgOutputOption so that I can implement my desired policy in a subclass, and change the default. Unlikely to happen any time soon. Greg ------------------------------------------------------ http://cvs2svn.tigris.org/ds/viewMessage.do?dsForumId=1667&dsMessageId=2412624 To unsubscribe from this discussion, e-mail: [[email protected]].