summaryrefslogtreecommitdiff
path: root/t/test-lib.sh
diff options
context:
space:
mode:
authorJunio C Hamano <gitster@pobox.com>2011-04-12 11:23:23 -0700
committerJunio C Hamano <gitster@pobox.com>2011-04-12 11:23:23 -0700
commitcabdee2c5ed55aa42784b37c1249829fca9d71b8 (patch)
tree9f7463177d927e53c289bf51d635bfd569ab700e /t/test-lib.sh
parent2c050e017a5e7727a7448620d872d46ba22675e9 (diff)
downloadgit-cabdee2c5ed55aa42784b37c1249829fca9d71b8.tar.gz
i18n: do not overuse C_LOCALE_OUTPUT
It is too coarse-grained way that led to artificial splitting of a logically single test case into "do" and "check only without poison". As the majority of check is done by comparing expected and actual output stored in a file with test_cmp anyway, just introduce test_i18ncmp that pretends the actual output matched the expected one when gettext-poison is in effect. Signed-off-by: Junio C Hamano <gitster@pobox.com>
Diffstat (limited to 't/test-lib.sh')
-rw-r--r--t/test-lib.sh8
1 files changed, 8 insertions, 0 deletions
diff --git a/t/test-lib.sh b/t/test-lib.sh
index abc47f3abc..3e7c2bb726 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -1077,6 +1077,14 @@ else
test_set_prereq C_LOCALE_OUTPUT
fi
+# Use this instead of test_cmp to compare files that contain expected and
+# actual output from git commands that can be translated. When running
+# under GETTEXT_POISON this pretends that the command produced expected
+# results.
+test_i18ncmp () {
+ test -n "$GETTEXT_POISON" || test_cmp "$@"
+}
+
# test whether the filesystem supports symbolic links
ln -s x y 2>/dev/null && test -h y 2>/dev/null && test_set_prereq SYMLINKS
rm -f y