Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx> writes:
> Refactor the repository setup code for tests that test hooks the use
> of sub-shells when setting up the test repository and hooks.
>
> A subsequent commit will change the hook setup to use a helper that
> makes use of "test_when_finished", which cannot be used in
> sub-shells. Let's change that setup code, and the adjacent and similar
> "mk_empty" code.
Makes sense.
> diff --git a/t/t5516-fetch-push.sh b/t/t5516-fetch-push.sh
> index 1a20e54adc1..e4bb7581568 100755
> --- a/t/t5516-fetch-push.sh
> +++ b/t/t5516-fetch-push.sh
> @@ -24,12 +24,8 @@ D=$(pwd)
> mk_empty () {
> repo_name="$1"
> rm -fr "$repo_name" &&
> - mkdir "$repo_name" &&
> - (
> - cd "$repo_name" &&
> - git init &&
> - git config receive.denyCurrentBranch warn
> - )
> + git init "$repo_name" &&
> + git -C "$repo_name" config receive.denyCurrentBranch warn
> }
I thought of this exact change when I saw a change to this part in
this series in an earlier step. Excellent.
