Mercurial > public > mercurial-scm > hg-stable
diff hgext/rebase.py @ 16696:d1afbf03e69a
rebase: allow collapsing branches in place (issue3111)
We allow rebase plus collapse, but not collapse only? I imagine people would
rebase first then collapse once they are sure the rebase is correct and it is
the right time to finish it.
I was reluctant to submit this patch for reasons detailed below, but it
improves rebase --collapse usefulness so much it is worth the ugliness.
The fix is ugly because we should be fixing the collapse code path rather than
the merge. Collapsing by merging changesets repeatedly is inefficient compared
to what commit --amend does: commitctx(), update, strip. The problem with the
latter is, to generate the synthetic changeset, copy records are gathered with
copies.pathcopies(). copies.pathcopies() is still implemented with merging in
mind and discards information like file replaced by the copy of another,
criss-cross copies and so forth. I believe this information should not be lost,
even if we decide not to interpret it fully later, at merge time.
The second issue with improving rebase --collapse is the option should not be
there to begin with. Rebasing and collapsing are orthogonal and a dedicated
command would probably enable a better, simpler ui. We should avoid advertizing
rebase --collapse, but with this fix it becomes the best shipped solution to
collapse changesets.
And for the record, available techniques are:
- revert + commit + strip: lose copies
- mq/qfold: repeated patching() (mostly correct, fragile)
- rebase: repeated merges (mostly correct, fragile)
- collapse: revert + tag rewriting wizardry, lose copies
- histedit: repeated patching() (mostly correct, fragile)
- amend: copies.pathcopies() + commitctx() + update + strip
| author | Patrick Mezard <patrick@mezard.eu> |
|---|---|
| date | Thu, 03 May 2012 15:14:58 +0200 |
| parents | ae6dddffe4f1 |
| children | 38caf405d010 |
line wrap: on
line diff
--- a/hgext/rebase.py Sat May 12 14:00:51 2012 +0200 +++ b/hgext/rebase.py Thu May 03 15:14:58 2012 +0200 @@ -214,7 +214,7 @@ % repo[root], hint=_('see hg help phases for details')) else: - result = buildstate(repo, dest, rebaseset, detachf) + result = buildstate(repo, dest, rebaseset, detachf, collapsef) if not result: # Empty state built, nothing to rebase @@ -265,7 +265,7 @@ else: try: ui.setconfig('ui', 'forcemerge', opts.get('tool', '')) - stats = rebasenode(repo, rev, p1, state) + stats = rebasenode(repo, rev, p1, state, collapsef) if stats and stats[3] > 0: raise util.Abort(_('unresolved conflicts (see hg ' 'resolve, then hg rebase --continue)')) @@ -383,7 +383,7 @@ repo.dirstate.invalidate() raise -def rebasenode(repo, rev, p1, state): +def rebasenode(repo, rev, p1, state, collapse): 'Rebase a single revision' # Merge phase # Update to target and merge it with local @@ -397,7 +397,9 @@ base = None if repo[rev].rev() != repo[min(state)].rev(): base = repo[rev].p1().node() - return merge.update(repo, rev, True, True, False, base) + # When collapsing in-place, the parent is the common ancestor, we + # have to allow merging with it. + return merge.update(repo, rev, True, True, False, base, collapse) def defineparents(repo, rev, target, state, targetancestors): 'Return the new parent relationship of the revision that will be rebased' @@ -589,7 +591,7 @@ repo.ui.warn(_('rebase aborted\n')) return 0 -def buildstate(repo, dest, rebaseset, detach): +def buildstate(repo, dest, rebaseset, detach, collapse): '''Define which revisions are going to be rebased and where repo: repo @@ -617,9 +619,9 @@ raise util.Abort(_('source is ancestor of destination')) if commonbase == dest: samebranch = root.branch() == dest.branch() - if samebranch and root in dest.children(): - repo.ui.debug('source is a child of destination\n') - return None + if not collapse and samebranch and root in dest.children(): + repo.ui.debug('source is a child of destination\n') + return None # rebase on ancestor, force detach detach = True if detach: