Re: What's cooking in git.git (Aug 2016, #03; Mon, 8)

[Stefan Beller <sbeller@google.com>]

On Wed, Aug 10, 2016 at 10:30 AM, Lars Schneider
<larsxschneider@gmail.com> wrote:
>
>>
>> * sb/submodule-update-dot-branch (2016-08-03) 7 commits
>>  (merged to 'next' on 2016-08-04 at 47bff41)
>> + submodule update: allow '.' for branch value
>> + submodule--helper: add remote-branch helper
>> + submodule-config: keep configured branch around
>> + submodule--helper: fix usage string for relative-path
>> + submodule update: narrow scope of local variable
>> + submodule update: respect depth in subsequent fetches
>> + t7406: future proof tests with hard coded depth
>>
>> A few updates to "git submodule update".
>>
>> Will merge to 'master'.
>
> I think "t7406: future proof tests with hard coded depth"
> breaks the tests on OSX:
>
> https://travis-ci.org/git/git/jobs/150779244
>
> Cheers,
> Lars
>


error: pathspec '4' did not match any file(s) known to git.

not ok 46 - submodule update clone shallow submodule

#
# test_when_finished "rm -rf super3" &&
# first=$(git -C cloned submodule status submodule |cut -c2-41) &&
# second=$(git -C submodule rev-parse HEAD) &&
# commit_count=$(git -C submodule rev-list $first^..$second | wc -l) &&
# git clone cloned super3 &&
# pwd=$(pwd) &&
# (
# cd super3 &&
# sed -e "s#url = ../#url = file://$pwd/#" <.gitmodules >.gitmodules.tmp &&
# mv -f .gitmodules.tmp .gitmodules &&
# git submodule update --init --depth=$commit_count &&
# test 1 = $(git -C submodule log --oneline | wc -l)
# )
#


Is it possible that the "wc -l" produces  SP <NUMBER> on OSX,
such that the

# git submodule update --init --depth=$commit_count

contains "--depth= 4" which means empty depth and 4 as the pathspec
for the update command?