Refactoring of parse_subarray_map_step and test

[astrobc1]

@zonca Could you take a look at my progress on ParseSubarrayMapStep? It fails in this loop:

liger_iris_pipeline/liger_iris_pipeline/parse_subarray_map/parse_subarray_map_step.py

Lines 49 to 50 in d1cd12c

	for each in parse_subarray_map(result["subarr_map"]):
	result.meta.subarray_map.append(each)

result is a LigerIrisImage and result.meta has no attribute subarr_map to store a list of xstart, ystart, xsize, etc. The schemas are consistent with this. There is result.meta.subarray, but it is designed for one subarray; result.meta.subarray = each works for a single dict.

Can a single LigerIrisImage contain multiple subarrays at this stage? Eventually we should consider that subarrays are only relevant for IRIS, but maybe not just yet.

[zonca]

@astrobc1 reading https://oirlab.github.io/iris-pipeline/parse_subarray_map/index.html it seems like the purpose of the step is to create those fields.

we definitely had subarray map at:

liger_iris_pipeline/iris_pipeline/datamodels/schemas/iris_image.schema.yaml

Lines 17 to 21 in bce4533

	subarr_map:
	title: Subarray map
	fits_hdu: SUBARR_MAP
	default: 0
	datatype: uint16

my bad, I lost that field in the refactoring, it defintely was there, see

liger_iris_pipeline/iris_pipeline/datamodels/schemas/tmt_core.schema.yaml

Lines 1979 to 1996 in bce4533

	subarray_map:
	title: Subarray map
	type: array
	items:
	type: object
	properties:
	xstart:
	title: Starting pixel in axis 1 direction
	type: integer
	ystart:
	title: Starting pixel in axis 2 direction
	type: integer
	xsize:
	title: Number of pixels in axis 1 direction
	type: integer
	ysize:
	title: Number of pixels in axis 2 direction
	type: integer

I now restored it.

[zonca]

I think we were able to have multiple subarrays, but I don't remember exactly how it worked, let me look into that and report back here.

[zonca]

@astrobc1 actually no, each file only has 1 subarray,
see for example https://github.com/oirlab/liger_iris_pipeline/blob/master/liger_iris_pipeline/tests/test_merge_subarrays.ipynb

so I think subarray map is defined in the "Full frame" to keep track of the subarrays, but the subarrays are not actually in that file but in separate fits files.

[zonca]

example of subarray is at https://figshare.com/articles/dataset/TMT-IRIS_raw_science_frame_subarray/11953512

[zonca]

@astrobc1 there are some examples and explanation around page 60 of the design document