Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-28475

Make deepCoadd_obj use band labels, not pseudo-physical_filter labels

    XMLWordPrintable

    Details

    • Type: Story
    • Status: Won't Fix
    • Resolution: Done
    • Fix Version/s: None
    • Component/s: pipe_tasks
    • Labels:
      None
    • Epic Link:
    • Team:
      Data Release Production
    • Urgent?:
      No

      Description

      It looks like the deepCoadd_obj tables are using the old pseudo-physical_filter labels like "HSC-I -> HSC-I OR HSC-I2" (at least as of recent weeklies) instead of "band". We need to fix even the Gen2 logic that creates these in order this to get the Gen3 analysis code that uses them working properly in Gen3, because we may be using the Gen2->Gen3 converted versions of these datasets for a while.

        Attachments

          Issue Links

            Activity

            Hide
            jbosch Jim Bosch added a comment - - edited

            John Parejko, Krzysztof Findeisen : do you know if your FilterLabel changes have involved or will involve changing the "filter" data ID values passed to CmdLineTasks that run on coadds and coadd-based datasets (from "pseudo physical_filter" to band, in particular)?  I'm guessing no, because that sounds like it's deep in the bowels of Gen2, and I'm guessing it'll be easier to avoid that and instead make as-needed changes to Gen2 tasks so that any in-dataset filter values they write ( (like those relevant to this ticket)) are band, as they will naturally be in Gen3.  But if you're already going down that path, it will certainly make tickets like this one easier.

            Show
            jbosch Jim Bosch added a comment - - edited John Parejko , Krzysztof Findeisen : do you know if your FilterLabel changes have involved or will involve changing the "filter" data ID values passed to CmdLineTasks that run on coadds and coadd-based datasets (from "pseudo physical_filter" to band, in particular)?  I'm guessing no, because that sounds like it's deep in the bowels of Gen2, and I'm guessing it'll be easier to avoid that and instead make as-needed changes to Gen2 tasks so that any in-dataset filter values they write ( (like those relevant to this ticket)) are band, as they will naturally be in Gen3.  But if you're already going down that path, it will certainly make tickets like this one easier.
            Hide
            Parejkoj John Parejko added a comment -

            Watch for a Community post about this once I merge DM-27170, but I think the TLDR is that gen2 dataIds will not change at all, just what values are stored in persisted output. I think that the change you requested is even being done as part of DM-27170 (it's at least related to what Yusra AlSayyad and I worked on on Monday, if not exactly the same), but I'm not positive. One of the things I've tried to do is replace uses of dataId["filter"] with e.g. dataRef.get('something_filterLabel') and then using either bandLabel or physicalLabel as appropriate, but I doubt I caught all such uses.

            Show
            Parejkoj John Parejko added a comment - Watch for a Community post about this once I merge DM-27170 , but I think the TLDR is that gen2 dataIds will not change at all, just what values are stored in persisted output. I think that the change you requested is even being done as part of DM-27170 (it's at least related to what Yusra AlSayyad and I worked on on Monday, if not exactly the same), but I'm not positive. One of the things I've tried to do is replace uses of dataId ["filter"] with e.g. dataRef.get('something_filterLabel') and then using either bandLabel or physicalLabel as appropriate, but I doubt I caught all such uses.
            Hide
            jbosch Jim Bosch added a comment -

            Thanks for the update, John Parejko. I think we can wait on that to land before considering this ticket as "let's see if it's already fixed" ticket. Lauren MacArthur, please go ahead with DM-28389 with some TODO comments, and we'll fix that code up later as part of this ticket as well.

            Show
            jbosch Jim Bosch added a comment - Thanks for the update, John Parejko . I think we can wait on that to land before considering this ticket as "let's see if it's already fixed" ticket. Lauren MacArthur , please go ahead with DM-28389 with some TODO comments, and we'll fix that code up later as part of this ticket as well.
            Hide
            tjenness Tim Jenness added a comment -

            This ticket seems to be about improving gen2 to support the transition to gen3. Does this mean the ticket can now be closed?

            Show
            tjenness Tim Jenness added a comment - This ticket seems to be about improving gen2 to support the transition to gen3. Does this mean the ticket can now be closed?
            Hide
            jbosch Jim Bosch added a comment -

            Yup, closing as Won't Fix.

            Show
            jbosch Jim Bosch added a comment - Yup, closing as Won't Fix.

              People

              Assignee:
              Unassigned Unassigned
              Reporter:
              jbosch Jim Bosch
              Watchers:
              Jim Bosch, John Parejko, Krzysztof Findeisen, Lauren MacArthur, Tim Jenness
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

                Dates

                Created:
                Updated:
                Resolved:

                  Jenkins

                  No builds found.