}
END_TEST
+START_TEST(test_ns_duplicate_hashes)
+{
+ /* The hash of an attribute is calculated as the hash of its URI
+ * concatenated with a space followed by its name (after the
+ * colon). We wish to generate attributes with the same hash
+ * value modulo the attribute table size so that we can check that
+ * the attribute hash table works correctly. The attribute hash
+ * table size will be the smallest power of two greater than the
+ * number of attributes, but at least eight. There is
+ * unfortunately no programmatic way of getting the hash or the
+ * table size at user level, but the test code coverage percentage
+ * will drop if the hashes cease to point to the same row.
+ *
+ * The cunning plan is to have few enough attributes to have a
+ * reliable table size of 8, and have the single letter attribute
+ * names be 8 characters apart, producing a hash which will be the
+ * same modulo 8.
+ */
+ const char *text =
+ "<doc xmlns:a='http://example.org/a'\n"
+ " a:a='v' a:i='w' />";
+ if (_XML_Parse_SINGLE_BYTES(parser, text, strlen(text),
+ XML_TRUE) == XML_STATUS_ERROR)
+ xml_failure(parser);
+}
+END_TEST
+
/* Regression test for SF bug #695401: unbound prefix. */
START_TEST(test_ns_unbound_prefix_on_attribute)
{
tcase_add_test(tc_namespace, test_ns_unbound_prefix);
tcase_add_test(tc_namespace, test_ns_default_with_empty_uri);
tcase_add_test(tc_namespace, test_ns_duplicate_attrs_diff_prefixes);
+ tcase_add_test(tc_namespace, test_ns_duplicate_hashes);
tcase_add_test(tc_namespace, test_ns_unbound_prefix_on_attribute);
tcase_add_test(tc_namespace, test_ns_unbound_prefix_on_element);
tcase_add_test(tc_namespace, test_ns_parser_reset);