David Kastrup <dak@xxxxxxx> wrote: > Eric Wong <normalperson@xxxxxxxx> writes: > > > Interesting. I didn't know about the --parents option before. > > Could this possibly be connected with some of the weirder commit > decisions of git-svn? Or is the alternative way of getting this > information up to now just as reliable? Not at all, these are completely unrelated. This change only affects users using grafts. The "weirder commit decisions" you describe only happen when users attempt to dcommit non-linear history (merges) to SVN. This is completely accidental, as the dcommit workflow was designed for a divergence-only (branching) history with no concept of convergence (merging): just like SVN. This is to simplify the mapping to SVN so all commits could be mapped 1:1 between the two systems. The current manpage advises against (non-squash) merge/pulls the CAVEATS section for exactly this reason. Maybe Lars's latest --first-parent patch makes everything wonderful for people using (non-squash) merges/pulls. Feedback on that is appreciated. -- Eric Wong - To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
