Code review comment for lp:~al-maisan/bzr-builddeb/merge-package

James Westby wrote:
> Excerpts from Muharem Hrnjadovic's message of Mon Aug 24 10:57:08 UTC 2009:
>> thank you very much for your suggestions above. They are all good. I
>> hope the following text is more suitable.
>>
>> {{{
>> The upstream branches for the merge source and target have diverged.
>> Unfortunately, the attempt to fix this problem resulted in conflicts.
>> Please resolve these and re-run the "merge-package" command to finish.
>> Alternatively, you can restore the original merge target state by
>> running "bzr revert".
>> }}}
>
> That's better. I think "resolve these and commit" is more explicit, and
> I would prefer something other than "original merge target state", as
> that sounds a bit jargonish, perhaps 'Alternatively, until you commit
> you can use "bzr revert" to restore the state of the unmerged branch.'
>
> It's not as clear what is going on, but I think that makes it easier
> to understand :-)

Thank you very much for your help with this :)

>>> Writing this mail also interested me in whether you ever get any
>>> conflicts in the case when merging the faked upstream merge
>>> in to the packaging branch in the case where your upstream
>>> version was newer than the other upstream version.
>> That would be the test_debian_upstream_older() test, and, no, I did not
>> get any conflicts.
>
> Sure, there's one test, and I wouldn't expect any conflicts in that
> test, I was thinking more of testing other situations though. Not
> as unit tests, but one-off testing to investigate whether you can
> provoke conflicts. If it is possible then we may want to investigate
> ways to avoid that.

I can think of a way to set up such a situation. There would need to
be a file that's shared and modified in all four branches: source
upstream, source packaging, target upstream and target packaging.
Furthermore the target upstream branch would have to be more recent one.

I will try to come up with this.

[..]

Best regards

--
Muharem Hrnjadovic <email address hidden>
Public key id : B2BBFCFC
Key fingerprint : A5A3 CC67 2B87 D641 103F 5602 219F 6B60 B2BB FCFC