Header inconsistency of LM imaging simulation

[chyan26]

I used lmg.yaml to run a set of simulation data with following setting

LM_IMAGE_SCI_RAW1:
  do.catg: LM_IMAGE_SCI_RAW
  mode: "img_lm"
  source:
    name: star_field
    kwargs: {}
  properties:
    dit: 0.25
    ndit: 4
    filter_name: "Lp"
    catg: "SCIENCE"
    tech: "IMAGE,LM"
    type: "OBJECT"
    tplname: "METIS_img_lm_obs_GenericOffset"
    nObs: 3

LM_IMAGE_SCI_RAW2:
  do.catg: LM_IMAGE_SKY_RAW
  mode: "img_lm"
  source:
    name: "empty_sky"
    kwargs: {}
  properties:
    dit: 0.25
    ndit: 4
    filter_name: "Lp"
    catg: "SCIENCE"
    tech: "IMAGE,LM"
    type: "SKY"
    tplname: "METIS_img_lm_obs_GenericOffset"
    nObs: 3

However, the FITS header DPR CATG of the first frame of sky background is not the same as reset. This caused problem for EDPS to pick the sky background image.

[hugobuddel]

I'm not sure I understand the problem you face. There seems to be a typo in your sentence, as the word 'reset' seems out of place.

That said, there is at least one problem with the yaml file, because the second one is called LM_IMAGE_SCI_RAW2 but the do.catg is LM_IMAGE_SKY_RAW, which is unusual.

I'm going to assume that the do.catgs are true. In that case we have two different raw data items: an LM_IMAGE_SCI_RAW and an LM_IMAGE_SKY_RAW. That is, a science and a sky observations. Since these raws are used in different ways in the pipeline, they should have different DPR keywords.

I'm not sure what these DPR keywords should be; I personally would think the sky raws should have DPR.CATG of CALIB, but maybe we agreed on SCIENCE; I don't recall.

Given the information above, is there still a problem?

[chyan26]

When I run the code, this setting produces one file with catg:"CALIB" two files with catg: "SCIENCE". That is what I mean.

[JenniferKarr]

The label for each block can be anything as long as it's unique; it's just the tag for the dictionary to loop over. It started out as the DO.CATG, but I had to add that explicitly, because a set of simulations can easily involve more than one product with the same DO.CATG.