)]}'
{"tests/unit/test_timer_driver.py":[{"author":{"_account_id":4146,"name":"Clark Boylan","email":"cboylan@sapwetik.org","username":"cboylan"},"change_message_id":"e2dd37407bb0118b110a88983713b538498a39e4","unresolved":true,"context_lines":[{"line_number":35,"context_line":""},{"line_number":36,"context_line":"        # Verify that the right job ran in the right tenant"},{"line_number":37,"context_line":"        self.assertEqual(tenant_one_projects, {\u0027org/project1\u0027})"},{"line_number":38,"context_line":"        self.assertEqual(tenant_two_projects, {\u0027org/project2\u0027})"}],"source_content_type":"text/x-python","patch_set":2,"id":"579c38a4_952b53dc","line":38,"updated":"2021-09-28 22:30:24.000000000","message":"Does it make sense to create a test-job1 and test-job2 for the respective tenants and projects then also check that the jobs don\u0027t leak across? I think this is sufficient to prevent regressions like the current bug we are fixing, but wonder if this would be good belts and suspenders against potential other issues.","commit_id":"0a29ddcc1693de5c5ecc94d416f2689294f963f8"},{"author":{"_account_id":1,"name":"James E. Blair","email":"jim@acmegating.com","username":"corvus"},"change_message_id":"0077c71c0c65dc29857293026b005b0f68892d58","unresolved":false,"context_lines":[{"line_number":35,"context_line":""},{"line_number":36,"context_line":"        # Verify that the right job ran in the right tenant"},{"line_number":37,"context_line":"        self.assertEqual(tenant_one_projects, {\u0027org/project1\u0027})"},{"line_number":38,"context_line":"        self.assertEqual(tenant_two_projects, {\u0027org/project2\u0027})"}],"source_content_type":"text/x-python","patch_set":2,"id":"9b4a0c33_c1c3dd14","line":38,"updated":"2021-09-28 22:36:06.000000000","message":"We could, but I think/hope that we have \"pipeline runs the right job\" covered elsewhere.  I tried to give Zuul as much rope as possible here to get it wrong (by ensuring that we don\u0027t do something based on the job before we get to the point where an incorrect change object would cause an error).  If we added other jobs, we\u0027d have to at least run 12 jobs in the test (instead of 4) in order to allow for all error possibilities.","commit_id":"0a29ddcc1693de5c5ecc94d416f2689294f963f8"}]}