Details
-
Bug
-
Status: Resolved
-
Normal
-
Resolution: Fixed
-
None
-
None
-
Correctness - Test Failure
-
Normal
-
Normal
-
Unit Test
Description
The implementation of PageAssertionMixin::assertEqualIgnoreOrder introduced in CASSANDRA-14134 can't be used to compare expected and actual results when the row data is represented by a dict. The underlying list_to_hashed_dict function expected lists of values and when it encounters a dict, it constructs its normalized list using only the keys. So the actual result values may be completely wrong, but as long as the field names are the same the equality check will pass. This affects only paging_tests.py and upgrade_tests/paging_test.py, and looks like it maybe a leftover from an earlier dev iteration, as some tests in the affected fixtures are already using the alternative (and correct) assertions.py::assert_lists_equal_ignoring_order.
Attachments
Issue Links
- relates to
-
CASSANDRA-14421 Reenable upgrade tests
- Resolved