Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-8238

Fix check_astrometry in the stack demo for OS X

    Details

    • Type: Story
    • Status: Done
    • Resolution: Done
    • Fix Version/s: None
    • Component/s: lsst_dm_stack_demo
    • Labels:
      None

      Description

      The instructions for running the lsst_dm_stack_demo and checking the astrometric relative RMS with bin/check_astrometry output won't work for OS X El Capitan+ because of SIP. I think there is a known fix for this, if the issue is similar to https://jira.lsstcorp.org/browse/DM-4921.

        Attachments

          Activity

          Hide
          swinbank John Swinbank added a comment -

          I don't think the fix here can be quite the same as elsewhere. Usually, we can rely on "shebang-rewriting" as part of the scons build to fix our scripts to work around SIP (by changing #!/usr/bin/env python to an absolute path to Python). However, lsst_dm_stack_demo isn't built by scons, so this doesn't apply.

          A simple solution would be to simply invoke check_astrometry using Python directly (ie, I think python bin/check_astrometry would work).

          We could also question whether check_astrometry is required here: I suspect that it was developed as a prototype for the validate_drp packages, and may now be effectively obsolete. Since we're comparing stack outputs (including positions) against pre-generated, known-good reference outputs, that should presumably pick up on bad astrometry without a further check. Michael Wood-Vasey probably has opinions.

          Show
          swinbank John Swinbank added a comment - I don't think the fix here can be quite the same as elsewhere. Usually, we can rely on "shebang-rewriting" as part of the scons build to fix our scripts to work around SIP (by changing #!/usr/bin/env python to an absolute path to Python). However, lsst_dm_stack_demo isn't built by scons, so this doesn't apply. A simple solution would be to simply invoke check_astrometry using Python directly (ie, I think python bin/check_astrometry would work). We could also question whether check_astrometry is required here: I suspect that it was developed as a prototype for the validate_drp packages, and may now be effectively obsolete. Since we're comparing stack outputs (including positions) against pre-generated, known-good reference outputs, that should presumably pick up on bad astrometry without a further check. Michael Wood-Vasey probably has opinions.
          Hide
          mtpatter Maria Patterson [X] (Inactive) added a comment -

          check_astrometry is left out of the instructions in the pipelines guide, so maybe it isn't needed (https://pipelines.lsst.io/install/demo.html#check-the-demo-results). Can the demo README just be updated to remove it if it is not needed or to add python bin/check_astrometry, whichever folks prefer? I recently upgraded to Sierra and banged my head a few times thinking that the stack wasn't working while running the demo, maybe others would do the same.

          Show
          mtpatter Maria Patterson [X] (Inactive) added a comment - check_astrometry is left out of the instructions in the pipelines guide, so maybe it isn't needed ( https://pipelines.lsst.io/install/demo.html#check-the-demo-results ). Can the demo README just be updated to remove it if it is not needed or to add python bin/check_astrometry , whichever folks prefer? I recently upgraded to Sierra and banged my head a few times thinking that the stack wasn't working while running the demo, maybe others would do the same.
          Hide
          swinbank John Swinbank added a comment -

          Can the demo README just be updated to remove it if it is not needed...

          I'd vote for this, but I think it's Michael Wood-Vasey's call.

          Show
          swinbank John Swinbank added a comment - Can the demo README just be updated to remove it if it is not needed... I'd vote for this, but I think it's Michael Wood-Vasey 's call.
          Hide
          wmwood-vasey Michael Wood-Vasey added a comment -

          I've updated the README to list python bin/compare detected-sources.txt and python bin/check_astrometry output.

          I think the purpose and overall intended scope of lsst_dm_stack_demo remain unclear. I'll thus leave in check_astrometry, which as John Swinbank notes was developed in the course of prototyping for validate_drp.

          Comparing against a static reference and seeing if the performance of the pipeline are reasonable are separate. E.g., if someone wants to update the reference catalogs for lsst_dm_stack_demo, they may wish to have just a simple check that things didn't go crazy. Clearly, this wouldn't be comprehensive, but it's simple just to have something here with no additional dependencies.

          PR at
          https://github.com/lsst/lsst_dm_stack_demo/pull/12

          Show
          wmwood-vasey Michael Wood-Vasey added a comment - I've updated the README to list python bin/compare detected-sources.txt and python bin/check_astrometry output . I think the purpose and overall intended scope of lsst_dm_stack_demo remain unclear. I'll thus leave in check_astrometry , which as John Swinbank notes was developed in the course of prototyping for validate_drp . Comparing against a static reference and seeing if the performance of the pipeline are reasonable are separate. E.g., if someone wants to update the reference catalogs for lsst_dm_stack_demo , they may wish to have just a simple check that things didn't go crazy. Clearly, this wouldn't be comprehensive, but it's simple just to have something here with no additional dependencies. PR at https://github.com/lsst/lsst_dm_stack_demo/pull/12
          Hide
          tjenness Tim Jenness added a comment -

          I'll note that adding a little driver shell script may also help here (as is done in demo.sh).

          Show
          tjenness Tim Jenness added a comment - I'll note that adding a little driver shell script may also help here (as is done in demo.sh ).
          Hide
          mtpatter Maria Patterson [X] (Inactive) added a comment -

          Thanks. Looks good.

          Show
          mtpatter Maria Patterson [X] (Inactive) added a comment - Thanks. Looks good.
          Hide
          wmwood-vasey Michael Wood-Vasey added a comment -

          Merged to master.

          Show
          wmwood-vasey Michael Wood-Vasey added a comment - Merged to master.

            People

            • Assignee:
              wmwood-vasey Michael Wood-Vasey
              Reporter:
              mtpatter Maria Patterson [X] (Inactive)
              Reviewers:
              Maria Patterson [X] (Inactive)
              Watchers:
              John Swinbank, Maria Patterson [X] (Inactive), Michael Wood-Vasey, Tim Jenness
            • Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Summary Panel