Re: [PATCH 5/5] t/t5520: test --[no-]autostash with pull.rebase=true

[Eric Sunshine <sunshine@sunshineco.com>]

On Tue, Mar 29, 2016 at 9:30 AM, Mehul Jain <mehul.jain2029@gmail.com> wrote:
> "--[no-]autostash" option for git-pull is only valid in rebase mode.
> That is, either --rebase is used or pull.rebase=true. Existing tests
> already check the cases when --rebase is used but fails to check for
> pull.rebase=true case.
>
> Add two new tests to check that --[no-]autostash option works with
> pull.rebase=true.
>
> Signed-off-by: Mehul Jain <mehul.jain2029@gmail.com>
> ---
> diff --git a/t/t5520-pull.sh b/t/t5520-pull.sh
> @@ -316,6 +316,26 @@ test_expect_success 'pull.rebase' '
> +test_expect_success 'pull --autostash & pull.rebase=true' '
> +       test_config pull.rebase true &&
> +       git reset --hard before-rebase &&
> +       echo dirty >new_file &&
> +       git add new_file &&
> +       git pull --autostash . copy &&
> +       test_cmp_rev HEAD^ copy &&
> +       test "$(cat new_file)" = dirty &&
> +       test "$(cat file)" = "modified again"
> +'

With the exception of the missing --rebase argument, this is exactly
the same code as in test_rebase_autostash(), right? Rather than
repeating this code yet again, it might be nice to augment that
function to accept a (possibly) optional argument controlling whether
--rebase is used.

> +
> +test_expect_success 'pull --no-autostash & pull.rebase=true' '
> +       test_config pull.rebase true &&
> +       git reset --hard before-rebase &&
> +       echo dirty >new_file &&
> +       git add new_file &&
> +       test_must_fail git pull --no-autostash . copy 2>err &&
> +       test_i18ngrep "Cannot pull with rebase: Your index contains uncommitted changes." err
> +'

Ditto with regard to test_rebase_no_autostash() (or
test_rebase_autostash_fail() as I suggested in my patch 4/5 review).