lm_img_cal_distortion naming is inconsistent

[astronomyk]

pyesorex metis_lm_img_cal_distortion "${SOF_DIR}/metis_lm_img_distortion.sof"

• The recipe is called img_cal, whereas the SOF file is only called img.
• The DRLD lists the recipe as metis_lm_img_calibrate

The recipe also generates ILM_DISTORTION_TABLE, which is accepted by metis_lm_img_calibrate - the extra "I" in "ILM" is at least consistent between the two recipes, but needs to be corrected

HOWEVER - this has raised the question - should this recipe be considered a CAL or IMG recipe, as it is only needed every couple of months to do a long-term (almost static) calibration for the IMG distortion table. If the ILM_DISTORTION_TABLE is present in the calibration database, then the recipe doesn't need to be run and the calibration data will not be present in the observer's current directory.

[oczoske]

Not sure I understand what the problem is (or why I was assigned to it?). I guess I would have called the recipes metis_lm_img_distortion (or the other way round, not my Baustelle) and metis_lm_img_calibrate (the former measures the distortion, the latter corrects for it - putting cal in the former might clarify that).
I put metis_cal_chophome in the cal category (only member so far) because its output is not specific to any instrument mode but calibrates a particular instrument component - it uses IMG_LM to measure that but the result is more general. Pupil imaging is maybe similar - that was put into an instrument folder but that might be merged with the cal folder.
By contrast, the result of the distortion recipe applies to the particular mode with which is was measured and should be used to correct data taken with just that mode. There should be such a recipe for each mode, which could be used to argue for it to be put into the img folder. Another criterion might be whether the input data for the recipe use the WCU (cal) or are taken on sky, but that would put the spectroscopic rsrf recipes where you do not want them.