Uploaded image for project: 'Request For Comments'
  1. Request For Comments
  2. RFC-450

Write meas_mosaic and jointcal WCS using FitsCatalogStorage

    Details

    • Type: RFC
    • Status: Implemented
    • Resolution: Done
    • Component/s: DM
    • Labels:
      None

      Description

      We currently write the "wcs" dataset (output by meas_mosaic and jointcal) as a tiny Exposure, which means it needs to be loaded with something like the following:

      wcs = butler.get("wcs", ...).getWcs()

      instead of just

      wcs = butler.get("wcs", ...)

      I'd like to change this, by writing these datasets instead using FitsCatalogStorage.  This will save them as FITS binary tables, with no standard FITS WCS representation in the headers.  Given that there's no actual imaged in these files, I don't think that's a concern, and it could also serve as a workaround (or at least a significant mitigation) for DM-13539.

      This change would break code that reads the files in their current form (implementing this RFC would involve fixing any such code in the stack), and after being updated, that code would not be able to read WCSs written before this change.  To address that, I'm planning to add a new dataset with the same template as the old one and a new name, and have any code that utilizes these datasets fall back to that backwards-compatibility dataset when "wcs" cannot be found.

       

        Attachments

          Issue Links

            Activity

            Hide
            Parejkoj John Parejko added a comment -

            Can you file the same ticket for jointcal, please? I can give you a working implementation immediately, but I won't be able to merge or do anything with it until after the 24th (vacation and out of communications).

            Show
            Parejkoj John Parejko added a comment - Can you file the same ticket for jointcal, please? I can give you a working implementation immediately, but I won't be able to merge or do anything with it until after the 24th (vacation and out of communications).
            Hide
            jbosch Jim Bosch added a comment -

            I'm happy to just make the jointcal changes on DM-11138 (or for you to do so, and then I can merge them when the rest of the ticket is merged). I do think it makes sense to merge them all at once.

            Show
            jbosch Jim Bosch added a comment - I'm happy to just make the jointcal changes on DM-11138 (or for you to do so, and then I can merge them when the rest of the ticket is merged). I do think it makes sense to merge them all at once.
            Hide
            rowen Russell Owen added a comment -

            I really like your new for the data products (jointcal_wcs and jointcal_photoCalib) because they are far more descriptive than the old names.

            Show
            rowen Russell Owen added a comment - I really like your new for the data products (jointcal_wcs and jointcal_photoCalib) because they are far more descriptive than the old names.
            Hide
            jbosch Jim Bosch added a comment -

            There has also been some related discussion on DM-12796 on what the templates for the new jointcal_* datasets should be.  The current approach is to maximize the similarity in the templates across cameras and with the typical templates for coadd data products, by putting them in  jointcal-results/%(filter)s/%(tract)04d.  Details can be seen on the obs_* PRS for DM-11138.

            I'll keep this RFC open another day to make sure anyone interested has a chance to take a look at those.

             

            Show
            jbosch Jim Bosch added a comment - There has also been some related discussion on DM-12796 on what the templates for the new jointcal_* datasets should be.  The current approach is to maximize the similarity in the templates across cameras and with the typical templates for coadd data products, by putting them in  jointcal-results/%(filter)s/%(tract)04d .  Details can be seen on the obs_* PRS for DM-11138 . I'll keep this RFC open another day to make sure anyone interested has a chance to take a look at those.  
            Hide
            tjenness Tim Jenness added a comment -

            All the work associated with this RFC has been completed. Can it be marked as implemented?

            Show
            tjenness Tim Jenness added a comment - All the work associated with this RFC has been completed. Can it be marked as implemented?

              People

              • Assignee:
                jbosch Jim Bosch
                Reporter:
                jbosch Jim Bosch
                Watchers:
                Jim Bosch, John Parejko, Paul Price, Russell Owen, Tim Jenness
              • Votes:
                0 Vote for this issue
                Watchers:
                5 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:
                  Planned End:

                  Summary Panel