Re: [PATCH 1/2] tests: make sure nested lazy prereqs work reliably

From

Jeff King <peff@peff.net>

Date

Nov 19, 2020, 15:58 UTC

Message-ID

<20201119155824.GB25426@coredump.intra.peff.net>

In-Reply-To

<20201118190414.32616-1-szeder.dev@gmail.com>

On Wed, Nov 18, 2020 at 08:04:13PM +0100, SZEDER Gábor wrote:

> So to prevent nested prereqs from interfering with each other let's
> evaluate each prereq in its own dedicated directory by appending the
> prereq's name to the directory name, e.g. 'prereq-test-dir-SYMLINKS'.
> In the test we check not only that the prereq test dir is still there,
> but also that the inner prereq can't mess with the outer prereq's
> files.

I.e., your test passes equally well with just:

diff --git a/t/test-lib-functions.sh b/t/test-lib-functions.sh
index 59bbf75e83..f5dc6801d9 100644
--- a/t/test-lib-functions.sh
+++ b/t/test-lib-functions.sh
@@ -474,15 +474,15 @@ test_lazy_prereq () {
 
 test_run_lazy_prereq_ () {
 	script='
-mkdir -p "$TRASH_DIRECTORY/prereq-test-dir" &&
+mkdir -p "prereq-test-dir" &&
 (
-	cd "$TRASH_DIRECTORY/prereq-test-dir" &&'"$2"'
+	cd "prereq-test-dir" &&'"$2"'
 )'
 	say >&3 "checking prerequisite: $1"
 	say >&3 "$script"
 	test_eval_ "$script"
 	eval_ret=$?
-	rm -rf "$TRASH_DIRECTORY/prereq-test-dir"
+	rm -rf "prereq-test-dir"
 	if test "$eval_ret" = 0; then
 		say >&3 "prerequisite $1 ok"
 	else

though I guess it is not really much simpler (it avoids the funky
quoting around $1 in the embedded script, but we already have that for
$2). And perhaps debugging is easier with a more predictable and
descriptive directory name.

> +test_lazy_prereq NESTED_INNER '
> +	>inner &&
> +	rm -f outer
> +'
> +test_lazy_prereq NESTED_PREREQ '
> +	>outer &&
> +	test_have_prereq NESTED_INNER &&
> +	echo "can create new file in cwd" >file &&
> +	test -f outer &&
> +	test ! -f inner
> +'
> +test_expect_success NESTED_PREREQ 'evaluating nested lazy prereqs dont interfere with each other' '
> +	nestedworks=yes
> +'
> +
> +if test -z "$GIT_TEST_FAIL_PREREQS_INTERNAL" && test "$nestedworks" != yes
> +then
> +	say 'bug in test framework: nested lazy prerequisites do not work'
> +	exit 1
> +fi

I was surprised to see this bare exit, because I know we have some
functions (run_sub_test_*) to help with testing the framework itself. It
looks like the other prereq tests don't use it either, though. I wonder
if there is a technical reason, or if they were simply added at a
different time. (Either way, I am OK for your new test to match the
surrounding ones like you have here).

-Peff