Re: [PATCH v2 3/4] Add 'promisor-remote' capability to protocol v2

[Junio C Hamano <gitster@pobox.com>]

Christian Couder <christian.couder@gmail.com> writes:

>> > +     git -C server -c repack.writebitmaps=false repack -a -d \
>> > +         --filter=blob:limit=5k --filter-to="$(pwd)" &&
>>
>> This --filter-to="$(pwd)" expands to $(TRASH_DIRECTORY), which is
>> "..../t/trash-directory.t5710-promisor-remote-capability".  I think
>> that is the cause for two extra trash files that are _OUTSIDE_ the
>> trash directory, which is an absolute no-no for tests to be safely
>> runnable.  Next to the trash directory, this ends up creating three
>> files
>>
>> trash directory.t5710-...-980d3ff591aae1651cdd52f7dfad4fb6319ee3c2.idx
>> trash directory.t5710-...-980d3ff591aae1651cdd52f7dfad4fb6319ee3c2.pack
>> trash directory.t5710-...-980d3ff591aae1651cdd52f7dfad4fb6319ee3c2.rev
>
> Yeah, right. It should be --filter-to="$(pwd)/pack"

It would create "pack-*" next to (not inside) "server", both as
immediate children of the "$TRASH_DIRECTORY".  I guess it is fine as
long as they are not created inside server/objects/ ;-)

>> The first failure in leak check seems to be
>>
>> not ok 5 - fetch with promisor.advertise set to 'false'
>> #
>> #               git -C server config promisor.advertise false &&
>> #
>> #               # Clone from server to create a client
>> #               GIT_NO_LAZY_FETCH=0 git clone -c remote.server2.promisor=true \
>> #                       -c remote.server2.fetch="+refs/heads/*:refs/remotes/server2/*" \
>> #                       -c remote.server2.url="file://$(pwd)/server2" \
>> #                       -c promisor.acceptfromserver=All \
>> #                       --no-local --filter="blob:limit=5k" server client &&
>> #               test_when_finished "rm -rf client" &&
>> #
>> #               # Check that the largest object is not missing on the server
>> #               check_missing_objects server 0 "" &&
>> #
>> #               # Reinitialize server so that the largest object is missing again
>> #               initialize_server
>>
>> but I didn't dig further.  Can you take a look?  I'll eject the
>> topic from 'seen' in the meantime to unblock the CI.
>
> No problem with ejecting the topic from 'seen'. I hope to send a new
> version with a design doc hopefully next week.

Thanks.