Uploaded image for project: 'Request For Comments'
  1. Request For Comments
  2. RFC-247

Move common datasets to obs_base

    Details

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

      Description

      The following datasets are extremely common among all the mappers. I propose that they be move to obs_base, with the differences per mapper kept in the Mapper.paf files.

      isr_metadata
      postISRCCD
      icSrc
      raw
      visitim
      wcs
      icExpBackground
      processCcd_metadata
      characterizeImage_metadata
      calibrate_metadata
      fcr
      forcedPhotCcd_metadata
      isr_config
      psf
      forced_src
      calexp
      src
      icMatch
      apCorr
      icExp
      calexpBackground
      srcMatch

        Attachments

          Issue Links

            Activity

            Hide
            jbosch Jim Bosch added a comment -

            jointcal does output a wcs item, but we probably will want to re-architect how that works in the future: it's just a dumping ground right now

            I think that means that this RFC shouldn't remove it from any mappers where it currently exists (Perry Gee's proposal to move most of it to obs_base is still fine). I think we'll want to re-architect exposure persistence more generally when Nate Pease's work on butler composites is complete. That will give us the capability to save overrides for exposure components more naturally as well as give us the ability to load individual exposure components more efficiently.

            Show
            jbosch Jim Bosch added a comment - jointcal does output a wcs item, but we probably will want to re-architect how that works in the future: it's just a dumping ground right now I think that means that this RFC shouldn't remove it from any mappers where it currently exists ( Perry Gee 's proposal to move most of it to obs_base is still fine). I think we'll want to re-architect exposure persistence more generally when Nate Pease 's work on butler composites is complete. That will give us the capability to save overrides for exposure components more naturally as well as give us the ability to load individual exposure components more efficiently.
            Hide
            Parejkoj John Parejko added a comment -

            I think that means that this RFC shouldn't remove it from any mappers where it currently exists (Perry Gee's proposal to move most of it to obs_base is still fine). I think we'll want to re-architect exposure persistence more generally when Nate Pease's work on butler composites is complete. That will give us the capability to save overrides for exposure components more naturally as well as give us the ability to load individual exposure components more efficiently.

            I agree on all points: move wcs to obs_base (which will be good for jointcal's current behavior), and then rethink things on a higher level later.

            Show
            Parejkoj John Parejko added a comment - I think that means that this RFC shouldn't remove it from any mappers where it currently exists (Perry Gee's proposal to move most of it to obs_base is still fine). I think we'll want to re-architect exposure persistence more generally when Nate Pease's work on butler composites is complete. That will give us the capability to save overrides for exposure components more naturally as well as give us the ability to load individual exposure components more efficiently. I agree on all points: move wcs to obs_base (which will be good for jointcal's current behavior), and then rethink things on a higher level later.
            Hide
            pgee Perry Gee added a comment -

            icMatch appears to be obsolete, but not srcMatch.

            visitim is still referred to in lsstSimMapper, but I don't see any code which uses it, except in examples in meas_astrom and meas_deblender. It is probably really obsolete.

            Show
            pgee Perry Gee added a comment - icMatch appears to be obsolete, but not srcMatch. visitim is still referred to in lsstSimMapper, but I don't see any code which uses it, except in examples in meas_astrom and meas_deblender. It is probably really obsolete.
            Hide
            pgee Perry Gee added a comment -

            I am going to now move this work to DM-8138, taking all of the suggestions stated here.

            I have no one defending visitim, and the only uses seem to be in examples in lsstSim.

            Show
            pgee Perry Gee added a comment - I am going to now move this work to DM-8138 , taking all of the suggestions stated here. I have no one defending visitim, and the only uses seem to be in examples in lsstSim.
            Hide
            pgee Perry Gee added a comment -

            The last described changes will be implement in DM-8138

            Show
            pgee Perry Gee added a comment - The last described changes will be implement in DM-8138

              People

              • Assignee:
                pgee Perry Gee
                Reporter:
                pgee Perry Gee
                Watchers:
                Jim Bosch, John Parejko, Perry Gee, Simon Krughoff, Tim Jenness
              • Votes:
                0 Vote for this issue
                Watchers:
                5 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:
                  Planned End:

                  Summary Panel