Re: Expected behaviour for pathspecs matching attributes in subdirectories

[Matthew Hughes <mhughes@uw.co.uk>]

On Thu, 6 Jul 2023 at 18:35, Junio C Hamano <gitster@pobox.com> wrote:

> I wonder if this serves a good addition to the tests?

Good idea, that would help clear up what's expected.

What about some tests for the case of attributes defined in a subdirectory? I'm
still trying to understand what's expected in that case, specifically the
test_must_be_empty case (the 'rm' at the end is just to avoid polluting any
other tests with extra attributes, not sure if there's a more standard way of
doing this):

diff --git a/t/t6135-pathspec-with-attrs.sh b/t/t6135-pathspec-with-attrs.sh
index 457cc167c7..7a7502a6eb 100755
--- a/t/t6135-pathspec-with-attrs.sh
+++ b/t/t6135-pathspec-with-attrs.sh
@@ -87,6 +87,23 @@ test_expect_success 'check specific set attr (2)' '
     test_cmp expect actual
 '

+test_expect_success 'check specific set attr nested .gitattributes' '
+    cat <<-\EOF >sub/.gitattributes &&
+    fileSetLabel otherLabel
+    EOF
+    test_write_lines sub/fileSetLabel >expect &&
+    git ls-files ":(attr:otherLabel)" >actual &&
+    test_cmp expect actual &&
+
+    git ls-files ":(attr:otherLabel)sub" >actual &&
+    test_cmp expect actual &&
+
+    git ls-files ":(attr:otherLabel)sub/" >actual &&
+    test_must_be_empty actual &&
+
+    rm sub/.gitattributes
+'
+
 test_expect_success 'check specific unset attr' '
     cat <<-\EOF >expect &&
     fileUnsetLabel