Skip to:

  1. Skip to Jira Navigation
  2. Skip to Side Navigation
  3. Skip to Main Content

Filter look-up for in isrTask is broken in a few places

Description

In updating the isr code to use filterLabel, a bug was introduced in the filter look up, whereby the check is being made on:

which will always evaluate to False. Rather, this should be:

This affects the setting of a rough zeropoint when config overrides are provided for the filters defined config.isr.fluxMag0T1, illumination correction for any filters defined in config.isr.illumFilters and a validation check for fringe correction for any filters defined in fringe.config.filters (but the fringe correction does indeed run as the check within the subtask is correct).

For example, when running on HSC data (with a few extra debugging prints):

Checklist

Lucidchart Diagrams

Issue Matrix

hide

Activity

Show:
Lauren MacArthur
June 12, 2021 at 11:29 PM

Thanks Chris.  I rebased and reran Jenkins + ci_hsc + ci_cpp_gen3 (still happy). Out of an abundance of paranoia (and as an excuse to try out running this), I also successfully ran:

so I think we're good. Merged and done!

Christopher Waters
June 8, 2021 at 8:43 PM

A few comments on the PR, mostly about using isrFunctions.checkFilter which handles the case of a config file having bands instead of physical filters (and does the right thing while complaining).

Lauren MacArthur
June 7, 2021 at 4:02 AM

Jenkins + ci_hsc + ci_cpp_gen3 is happy.

And I can confirm that the proper fluxMag0T1 values are being picked up in HSC processing.

Done
Pinned fields
Click on the next to a field label to start pinning.

Details

Assignee

Reporter

Reviewers

Christopher Waters

Story Points

RubinTeam

Components

Sprint

None

Parent

Checklist

Created June 5, 2021 at 9:55 PM
Updated July 22, 2021 at 12:36 AM
Resolved June 12, 2021 at 11:29 PM