When using macros it isn't inconceivable to have an attribute
being set by a macro, and then being reset explicitly.
Signed-off-by: Henrik Grubbström <grubba@xxxxxxxxxx>
---
NB: Currently the tests in the testsuite patch will have the
opposite meaning, which is probably not what the user expects,
and is contrary to the documentation.
attr.c | 2 +-
t/t0003-attributes.sh | 6 ++++++
2 files changed, 7 insertions(+), 1 deletions(-)
diff --git a/attr.c b/attr.c
index 5c6464e..968fb8b 100644
--- a/attr.c
+++ b/attr.c
@@ -599,7 +599,7 @@ static int fill_one(const char *what, struct match_attr *a, int rem)
struct git_attr_check *check = check_all_attr;
int i;
- for (i = 0; 0 < rem && i < a->num_attr; i++) {
+ for (i = a->num_attr - 1; 0 < rem && 0 <= i; i--) {
struct git_attr *attr = a->state[i].attr;
const char **n = &(check[attr->attr_nr].value);
const char *v = a->state[i].setto;
diff --git a/t/t0003-attributes.sh b/t/t0003-attributes.sh
index 1c77192..bd9c8de 100755
--- a/t/t0003-attributes.sh
+++ b/t/t0003-attributes.sh
@@ -22,6 +22,8 @@ test_expect_success 'setup' '
(
echo "f test=f"
echo "a/i test=a/i"
+ echo "onoff test -test"
+ echo "offon -test test"
) >.gitattributes &&
(
echo "g test=a/g" &&
@@ -44,6 +46,8 @@ test_expect_success 'attribute test' '
attr_check b/g unspecified &&
attr_check a/b/h a/b/h &&
attr_check a/b/d/g "a/b/d/*"
+ attr_check onoff unset
+ attr_check offon set
'
@@ -58,6 +62,8 @@ a/b/g: test: a/b/g
b/g: test: unspecified
a/b/h: test: a/b/h
a/b/d/g: test: a/b/d/*
+onoff: test: unset
+offon: test: set
EOF
sed -e "s/:.*//" < expect | git check-attr --stdin test > actual &&
--
1.7.0.3.316.g33b5e
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html