Minutes of OTFC Mtg 9/17/98

Starview & DADS (Allen Farris & Warren King) - Testing of DADS revealed a problem with the translation of logicals.

Action Item: Megan will talk to Bruce Mayhew about setting up a separate screen for wfpc2-otfc. Megan is working on the screen design. This screen will display the last-change dates of parameters and software for each dataset.

Steve asked whether there was an easy way to determine the date of calibration that can be used to compare with the last-change dates. Daryl determined the keyword name for the calibration date is DATE in the calibrated header.

OPUS (Mike Swam) - 4 9GB drives have arrived. Wfpc2 reference files (10 GB) and wfpc2 test data (3GB) can be stored on these drives. A first set of the reference files for wfpc2 was requested. This represents half of 1250 needed. Coding of the CDBS interface is done.

getref/upref (Bernie Simon) - Open issues:

  • Handling of NICMOS associations
  • Determining the appropriate subset of possible reference file names that should be put in a dataset for a given type of observation. Currently, getref/upref will put all types of reference files in the header, whether or not they are appropriate. This no harm in processing, but could cause confusion to a user. Eventually bestref will populate such values with N/A and getref will not list them. This is an issue for STIS.

    Change System (Lisa Gardner & Steve Lubow) -

    We have developed a plan to keep the test phase 0 system up to date. The goal is to make the system to be useful to internal users by having recent datasets and reference file available. It appears possible to do this by

    1. changing the wfpc2 ref table in dadsops to match the new design for otfc
    2. running bestswitch in dadsops
    3. nightly copying over the ref table data for wfpc2 to the test system
    4. enterring the names of the datasets with okr (ket) entries into the pending datasets table
    5. running the change script that applies the oke changes, bestswitch, and bestref changes on those pending datsets.

    Action Item: Steve to develop a test suite for CDBS changes that affect bestref.

    Action Item: Steve to add to check_cdbs a test for null otfc dates on past deliveries.

    Action Item: Bestref changes for otfc need to be made soon. Bernie Shiao is doing this.

    wfpc2 (Sylvia Baggett) - Datasets are being selected and tested for the regression test. We expect about 100 datasets in the test. So far 46 have been identified and run through the OTFC calibration.

    Associations (Ed Hopkins) - Associations involving multiple files like NICMOS can be handled by taking advantage of the current association table in DADS. It was determined that for NICMOS and ACS only the full association will be provided, i.e., no individual products such as background products will be returned alone. The reason is that only a small percentage of NICMOS datasets involve such associated data.

    There remain several issues including:

  • Should otfc allow the selective return of a subset of the processed and raw data?
  • How will exceptions be represented for associated data? How will getref/upref use them?
  • Should the association table be archived or should it be generated on the fly through the association table?

    Action Item: Ed to write up summary of discussion

    Action Item: Steve to ask NICMOS group for an otfc representative

    OTFC Meeting: (lubow@stsci.edu)