Suppose you have this topology, and you are trying to make an octopus
across A, B and C (you are at C and merging A and B into your branch).
The protoccol between "git merge" and merge strategies is for the former
to pass common ancestor(s), '--' and then commits being merged.
git-merge-octopus does not produce the final merge in one-go. It
iteratively produces pairwise merges. So the first step might be to come
up with a merge between B and C:
o---o---o---o---C
/ :
/ o---o---o---B..(M)
/ /
---1---2---o---o---o---A
and for that, "1" is used as the merge base, not because it is the base
across A, B and C but because it is the base between B and C. For this
merge, A does not matter.
I drew M in parentheses and lines between B and C to it in dotted line
because we actually do _not_ create a real commit --- the only thing we
need is a tree object, in order to proceed to the next step.
Then the final merge result is obtained by merging tree of (M) and A using
their common ancestor. For that, we _could_ still use "1" as the merge
base.
But if you imagine a case where you started from A and M, you would _not_
pick "1" as the merge base; you would rather use "2" which is a better
base for this merge.
That is why git-merge-octopus ignores the merge base given by "merge" but
computes its own.
The comment at the end of git-merge-octopus talks about "merge reference
commit", that we used to update it to common found in this round, and that
that updating was pointless. After the first round of merge to produce
the tree for M (but without actually creating the commit object M itself),
in order to figure out the merge base used to merge that with A in the
second round, we used to use A and "1" (which was merge base between B and
C). That was pointless --- "merge-base A 1" is guaranteed to give a base
that is no better than either "merge-base A B" or "merge-base A C". So
the current code keeps using the original head (iow, MRC=C, because in
this case we are starting from C and merging B and then A into it).
This trickerly was necessary only because we avoided creating the extra
merge commit object M.
Side note. An alternative implementation could have been to
actually record it as a real merge commit M, and then let the
two-commit merge-base compute the base between A and M when
merging A to the result of the previous round, but we avoided
creating M, at the expense of potentially using suboptimal base in
the later rounds.
But we do not have to be that pessimistic. We can instead accumulate the
commits we have merged so far in MRC, and have merge_bases_many() compute
the merge base for the new head being merged and the heads we have
processed so far, which can give a better base than what we currently do.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A new flag --aggressive resolves what we traditionally resolved
with external git-merge-one-file inside index while read-tree
3-way merge works.
git-merge-octopus and git-merge-resolve use this flag before
running git-merge-index with git-merge-one-file.
Signed-off-by: Junio C Hamano <junkio@cox.net>
We rejected multi-base merge situations even though we used the
same underlying multi-base git-read-tree as the resolve strategy
uses. This was unneeded and did not add much to ensure the
merge to be truly trivial, so remove this restriction and be
more similar to what resolve does.
Also when the merge did not trivially resolve, we rejected
without stating that octopus strategy does not handle the
situation.
Signed-off-by: Junio C Hamano <junkio@cox.net>
As a fallout from not using git-sh-setup in scripts that can
operate from a subdirectory, we lost definition of die() from
them. It might make sense to do some cleanup to consolidate
them back again, but this should suffice for now.
Signed-off-by: Junio C Hamano <junkio@cox.net>
When doing an octopus, we incorrectly used the previous merge
base as the reference to compute next merge base. This was
unnecessary, because that can never be better than using the
original HEAD. And that is far simpler as well ;-).
Signed-off-by: Junio C Hamano <junkio@cox.net>
The new command 'git merge' takes the current head and one or more
remote heads, with the commit log message for the automated case.
If the heads being merged are simple fast-forwards, it acts the
same way as the current 'git resolve'. Otherwise, it tries
different merge strategies and takes the result from the one that
succeeded auto-merging, if there is any.
If no merge strategy succeeds auto-merging, their results are
evaluated for number of paths needed for hand resolving, and the
one with the least number of such paths is left in the working
tree. The user is asked to resolve them by hand and make a
commit manually.
The calling convention from the 'git merge' driver to merge
strategy programs is very simple:
- A strategy program is to be called 'git-merge-<strategy>'.
- They take input of this form:
<common1> <common2> ... '--' <head> <remote1> <remote2>...
That is, one or more the common ancestors, double dash, the
current head, and one or more remote heads being merged into
the current branch.
- Before a strategy program is called, the working tree is
matched to the current <head>.
- The strategy program exits with status code 0 when it
successfully auto-merges the given heads. It should do
update-cache for all the merged paths when it does so -- the
index file will be used to record the merge result as a
commit by the driver.
- The strategy program exits with status code 1 when it leaves
conflicts behind. It should do update-cache for all the
merged paths that it successfully auto-merged, and leave the
cache entry in the index file as the same as <head> for paths
it could not auto-merge, and leave its best-effort result
with conflict markers in the working tree when it does so.
- The strategy program exists with status code other than 0 or
1 if it does not handle the given merge at all.
As examples, this commit comes with merge strategies based on
'git resolve' and 'git octopus'.
Signed-off-by: Junio C Hamano <junkio@cox.net>
