Re: [PATCH v3] stash: infer "push" when push-specific options are given

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

Deveshi Dwivedi <deveshigurgaon@gmail.com> writes:

> +test_expect_success 'assume push when options imply push' '
> +	git reset --hard &&
> +	echo changed >file &&
> +	git add file &&
> +	git stash -m "implied push" file &&
> +	git stash pop &&
> +
> +	git add file &&
> +	git stash --staged file &&
> +	git stash pop &&
> +
> +	git add file &&
> +	git stash --keep-index file &&
> +	git stash pop &&
> +
> +	echo untracked >untracked-file &&
> +	git stash --include-untracked untracked-file &&
> +	test_path_is_missing untracked-file &&

A comment on these three lines.

> +	git stash pop &&
> +	rm -f untracked-file &&
> +	git reset --hard
> +'
> +

I suspect that they are meant to be "clean-up after we are done with
the test, to avoid interfering with the next test", but if so,
"clean-up at the very end" is not a very effective strategy to do
so.  Imagine that one of the previous steps fails, breaking all
later commands in the &&- cascade.  Sitting at the very end, your
clean-up sequence will not run.  Unless the tester is running this
test script with the "-i" option, the test will move on to the next
piece.  Installing clean-up handler with test_when_finished may be
a cleaner approach.

	test_expect_success 'do this test' '
		test_when_finished "git stash clear; git reset --hard" &&
		git reset --hard &&
		... do all the dirty things in the working tree ...

		test_when_finished "rm -f untracked-file" &&
                echo untracked >untracked-file &&
                git stash --include-untracked untracked-file &&
		test_path_is_missing untracked-file
	'

You can use more than one test_when_finished in a single test.  

It is often done to add an upfront blunt hammer at the beginning to
do a clean-up without worrying too much about where exactly in the
command sequence a breakage may happen (e.g., we may fail before we
run our first "git add", or "git stash", and "git reset --hard" or
"git stash clear" may be an unnecessary no-op, but we do not worry
too much about the clean-up step doing potentially unnecessary
things.

Or you would set up a clean-up handler immediately before you create
a thing that you want to make sure you clean up.  If the command
sequence fails before you echo the string into untracked-file to
create it, there is no point preparing to remove it when you are
done.

Both approaches are commmonly used.

>  test_expect_success 'stash --invalid-option' '
>  	echo bar5 >file &&
>  	echo bar6 >file2 &&
>
> base-commit: 2855562ca6a9c6b0e7bc780b050c1e83c9fcfbd0