Considering there are many related bugs in spider, and there seem to be more bugs than we can work on simultaneously, I have the following idea.
- Create a testcase (e.g. under a testsuite "spider/unfixed") expecting failure as described in the bug, so that the test passes on expected failure. Push the testcase.
- Work on the issue or any other issue
- If while working on another issue, the testcase fails, check what the failure means
- if the failure indicates the bug may have been fixed (e.g. instead of expected failure, the test exhibits expected behaviour), check whether it is indeed fixed
- in any case, update the jira issue and the testcase based on the new info if needed
Failure in the suite should not be blocking a push, in any case.
Tickets we can add unfixed tests: