Julian Foad <julianf...@btopenworld.com> writes:
> Philip Martin
>
>> "Bert Huijben" <b...@qqmail.nl> writes:
>>>> + # The incoming "move" creates a tree-conflict as an incoming change
>>>> + # in a local move. We don't yet track moves on the server so we
>>>> + # don't recognise the incoming change as a move.
>>>> expected_output = svntest.wc.State(wc_dir, {
>>>> - 'A/B/E2/alpha' : Item(status='A '),
>>>> + 'A/B/E' : Item(status=' ', treeconflict='C'),
>>>> + 'A/B/E/alpha' : Item(status=' ', treeconflict='D'),
>>>
>>> treeconflict='D' ?
>>>
>>> I'm surprised that it works that way.
>>
>> The update produces the output:
>>
>> Updating 'svn-test-work/working_copies/update_tests-64':
>> C svn-test-work/working_copies/update_tests-64/A/B/E
>> D svn-test-work/working_copies/update_tests-64/A/B/E/alpha
>> A svn-test-work/working_copies/update_tests-64/A/B/F/alpha
>> Updated to revision 2.
>> Summary of conflicts:
>> Tree conflicts: 1
>>
>> Do you think it should be something else?
>
> I've never heard of us printing a 'D' in the tree conflict column, and
> the help for 'up' and 'merge' don't mention it where they describe
> this output, so yes it looks wrong or else is some bit of new design
> that hasn't been completed.
I believe Bert introduced this notification in r1089779:
More preparation for allowing more kinds of shadowed updates: Show
notifications of shadowed updates in the tree conflict column, to
make it more obvious that they are special and to allow verifying
their behavior in the test suite.
--
Certified & Supported Apache Subversion Downloads:
http://www.wandisco.com/subversion/download
