Re: [PATCH 1/2] t3404: demonstrate CHERRY_PICK_HEAD bug

> Johannes Schindelin <johannes.schindelin@gmx.de> writes:
> 
>> +test_expect_failure 'rebase --continue removes CHERRY_PICK_HEAD' '
>> +	git checkout -b commit-to-skip &&
>> +	for double in X 3 1
>> +	do
>> +		seq 5 | sed "s/$double/&&/" >seq &&
>> +		git add seq &&
>> +		test_tick &&
>> +		git commit -m seq-$double
>> +	done &&
>> +	git tag seq-onto &&
>> +	git reset --hard HEAD~2 &&
>> +	git cherry-pick seq-onto &&
>> +	test_must_fail git rebase -i seq-onto &&
> 
> Shouldn't this explicitly specify what fake editor is to be used,
> instead of relying on whatever the last test happened to have used?
> 
> I thought this deserved to go to older maintenance track, but the
> fake editor that was used last are different between branches, so
> "rebase -i" fails for a wrong reason (i.e. cannot spawn the editor)
> when queued on say maint-2.2.

>> +	test -d .git/rebase-merge &&
>> +	git rebase --continue &&
>> +	git diff seq-onto &&
> 
> I am puzzled with this "diff"; what is this about?  Is it a remnant
> from an earlier debugging session, or is it making sure seq-onto is
> a valid tree-ish?