Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-960

Remove test of non-existant source table object

    Details

      Description

      An AP unittest failure was not caught and reported when it first occurred. Now that the unittest failures are correctly being caught, the packages with such failures are now being tracked down and repaired.

      Thies failure is one of them.

      Regarding this failure, Perry said:
      "... the one in AP might have been caused by a change Jim and I made 2 mos. ago. Has it been a long time since the tests on ap was examined? I don't think anything that I did recently could have caused this.

      I am going to suggest that we disable this assert, since the test runs fine without the one assert on lines 238-239 of $AP_DIR/tests/sourceClusterTable.cc. I just commented out this assert, and the test is otherwise OK. "

        Attachments

          Issue Links

            Activity

            Hide
            robyn Robyn Allsman [X] (Inactive) added a comment -

            Implemented Perry's fix, tested it in isolation using the latest successful master stack. All worked. merged to master.

            Show
            robyn Robyn Allsman [X] (Inactive) added a comment - Implemented Perry's fix, tested it in isolation using the latest successful master stack. All worked. merged to master.
            Hide
            robyn Robyn Allsman [X] (Inactive) added a comment -

            I removed the lines containing the reference to an undefined object in the unittest.

            Show
            robyn Robyn Allsman [X] (Inactive) added a comment - I removed the lines containing the reference to an undefined object in the unittest.
            Hide
            ktl Kian-Tat Lim added a comment -

            Shouldn't we at least understand why the number of names is different in the output and input catalogs, given that the latter is supposed to be a readback of the former?

            Show
            ktl Kian-Tat Lim added a comment - Shouldn't we at least understand why the number of names is different in the output and input catalogs, given that the latter is supposed to be a readback of the former?
            Hide
            pgee Perry Gee added a comment -

            This problem was caused by the subclassing of lsst::afw:table::SourceTable in ap, so that code which is supposed to remove the tag AFW_TABLE_VERSION when a fits table is read from disk got missed by the subclass.

            I have confirmed this is the problem with the debugger. It is safe to have commented out Serge's nameCount check. I will reinstate it when I check in DM-958, which will move the versioning persistance into the schema. I will leave myself a note about this.

            Show
            pgee Perry Gee added a comment - This problem was caused by the subclassing of lsst::afw:table::SourceTable in ap, so that code which is supposed to remove the tag AFW_TABLE_VERSION when a fits table is read from disk got missed by the subclass. I have confirmed this is the problem with the debugger. It is safe to have commented out Serge's nameCount check. I will reinstate it when I check in DM-958 , which will move the versioning persistance into the schema. I will leave myself a note about this.
            Hide
            robyn Robyn Allsman [X] (Inactive) added a comment -

            Annotated DM-958 so that unittest is re-inserted when this ticket is completed.

            Show
            robyn Robyn Allsman [X] (Inactive) added a comment - Annotated DM-958 so that unittest is re-inserted when this ticket is completed.

              People

              • Assignee:
                robyn Robyn Allsman [X] (Inactive)
                Reporter:
                robyn Robyn Allsman [X] (Inactive)
                Reviewers:
                Perry Gee
                Watchers:
                Kian-Tat Lim, Perry Gee
              • Votes:
                0 Vote for this issue
                Watchers:
                2 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:

                  Summary Panel

                    Time Tracking

                    Estimated:
                    Original Estimate - 2 minutes
                    2m
                    Remaining:
                    Remaining Estimate - 0 minutes
                    0m
                    Logged:
                    Time Spent - 1 day
                    1d