Re: [PATCH 2/2] config: allow specifying config entries via envvar pairs

On Fri, Nov 13 2020, Patrick Steinhardt wrote:

While not document, it is currently possible to specify config entries

"While not documented..."

+		strbuf_addf(&envvar, "GIT_CONFIG_KEY_%d", i);
+		if ((key = getenv(envvar.buf)) == NULL)
+			break;

The convention in git.git is to avoid explicit NULL checks. So maybe
something like this, which also avoids the assignment inside an "if"

    key = getenv(envvar.buf);
    if (!key)
        break;

+test_expect_success 'git config handles environment config pairs' '
+	GIT_CONFIG_KEY_1="pair.one" GIT_CONFIG_VALUE_1="foo" \
+		GIT_CONFIG_KEY_2="pair.two" GIT_CONFIG_VALUE_2="bar" \
+		GIT_CONFIG_KEY_4="pair.four" GIT_CONFIG_VALUE_4="not-parsed" \
+		git config --get-regexp "pair.*" >actual &&
+	cat >expect <<-EOF &&
+	pair.one foo
+	pair.two bar
+	EOF
+	test_cmp expect actual
+'
+
+test_expect_success 'git config copes with missing config pair value' '
+	GIT_CONFIG_KEY_1="pair.one" git config --get-regexp "pair.*" >actual &&
+	echo pair.one >expect &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git config fails with invalid config pair key' '
+	test_must_fail env GIT_CONFIG_KEY_1= git config --list &&
+	test_must_fail env GIT_CONFIG_KEY_1=missing-section git config --list
+'
+
 test_expect_success 'git config --edit works' '
 	git config -f tmp test.value no &&
 	echo test.value=yes >expect &&

I think we should have a bit more complete tests of what happens if you
clobber existing config keys, and testing that this is set last after
system/global/local config.