New policy: Only use test_expect_failure for broken tests

[stgit] / t / README

diff --git a/t/README b/t/README
index d88bad2724954900bc80a71fc9f25fcd8ea97e12..77f0b6cf63e498291aac9b83bb2afb4165eea650 100644 (file)
--- a/t/README
+++ b/t/README
@@ -162,11 +162,10 @@ library for your script to use.
    This is the opposite of test_expect_success.  If <script>
    yields success, test is considered a failure.
 
-   Example:
-
-       test_expect_failure \
-           'git-update-index without --add should fail adding.' \
-           'git-update-index should-be-empty'
+   This should _not_ be used for tests that succeed when their
+   commands fail -- use test_expect_success and shell negation (!) for
+   that. test_expect_failure is for cases when a test is known to be
+   broken.
 
  - test_debug <script>