Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-19373

Add defect list support to butler gen 3

    XMLWordPrintable

    Details

      Description

      Investigate storing and retrieving of a defects list calibration product from gen 3 butler.

      It's not yet clear where the code for doing this would end up. Special formatters will need to be written to ensure that "list of defect" is returned and not an afw Catalog. Need to decide whether it's meas_algorithms or obs_base.

        Attachments

          Issue Links

            Activity

            Hide
            tjenness Tim Jenness added a comment -

            This ticket is expanding a little in that I have created a Defects class for collecting a list of Defect and adding readFits and writeFits methods that are suitable for use in butler. As part of this work it is clear that there is much code in ip_isr that would be cleaned up if it was moved to Defects.

            Show
            tjenness Tim Jenness added a comment - This ticket is expanding a little in that I have created a Defects class for collecting a list of Defect and adding readFits and writeFits methods that are suitable for use in butler. As part of this work it is clear that there is much code in ip_isr that would be cleaned up if it was moved to Defects.
            Hide
            tjenness Tim Jenness added a comment -

            Christopher Waters thanks for agreeing to review. I added DefectsList support to gen 3 and ended up doing that by creating a new Defects class. The changes in the other packages are to switch them to using this new class. There are additional changes in ip_isr to remove afwGeom usage that should be using geom.

            One thing I would like you to consider is whether I should move the "mask this image using these defects" into the Defects class as a method.

            This work is needed to support Simon Krughoff in DM-18739.

            Show
            tjenness Tim Jenness added a comment - Christopher Waters thanks for agreeing to review. I added DefectsList support to gen 3 and ended up doing that by creating a new Defects class. The changes in the other packages are to switch them to using this new class. There are additional changes in ip_isr to remove afwGeom usage that should be using geom. One thing I would like you to consider is whether I should move the "mask this image using these defects" into the Defects class as a method. This work is needed to support Simon Krughoff in DM-18739 .
            Hide
            czw Christopher Waters added a comment - - edited

            Looks good to me.  This seems like a much simpler solution than before.

            Just to double check, no other obs_ package needs updating, right?

            As for moving the "mask this image with these defects", I am agnostic.  meas_algorithms makes sense from the perspective of doing it downstream, but ideally all defects are masked by ip_isr.

            Show
            czw Christopher Waters added a comment - - edited Looks good to me.  This seems like a much simpler solution than before. Just to double check, no other obs_ package needs updating, right? As for moving the "mask this image with these defects", I am agnostic.  meas_algorithms makes sense from the perspective of doing it downstream, but ideally all defects are masked by ip_isr.
            Hide
            tjenness Tim Jenness added a comment -

            obs_base, obs_test, and obs_monocam have the almost the same code that obs_cfht has but ip_isr can handle those by updating the DefectBase to Defect and they are going to be fixed by Simon Krughoff on his revamp of the obs packages.

            Show
            tjenness Tim Jenness added a comment - obs_base, obs_test, and obs_monocam have the almost the same code that obs_cfht has but ip_isr can handle those by updating the DefectBase to Defect and they are going to be fixed by Simon Krughoff on his revamp of the obs packages.
            Hide
            czw Christopher Waters added a comment -

            After writing that, I think I lean a bit towards putting it in meas_algorithms.  I would hope all defects are masked by ip_isr, but I can imagine cases where masking with a defect list could be done downstream (like handling bright stars once they have known positions after the astrometry solution is found).

            Show
            czw Christopher Waters added a comment - After writing that, I think I lean a bit towards putting it in meas_algorithms.  I would hope all defects are masked by ip_isr, but I can imagine cases where masking with a defect list could be done downstream (like handling bright stars once they have known positions after the astrometry solution is found).
            Hide
            tjenness Tim Jenness added a comment -

            Ok, I've moved the defects-from-mask and mask-image-from-defect methods into Defects and tweaked the calling code in ip_isr and pipe_tasks.

            Show
            tjenness Tim Jenness added a comment - Ok, I've moved the defects-from-mask and mask-image-from-defect methods into Defects and tweaked the calling code in ip_isr and pipe_tasks.
            Hide
            tjenness Tim Jenness added a comment -

            Thanks for the review. All PRs merged.

            Show
            tjenness Tim Jenness added a comment - Thanks for the review. All PRs merged.

              People

              Assignee:
              tjenness Tim Jenness
              Reporter:
              tjenness Tim Jenness
              Reviewers:
              Christopher Waters
              Watchers:
              Christopher Waters, Jim Bosch, Simon Krughoff, Tim Jenness
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

                Dates

                Created:
                Updated:
                Resolved:

                  CI Builds

                  No builds found.