Avoid implicit changes to ict and space counts in thermal calibration

[carloshorn]

To avoid an implicit change of the input arrays, I would move the interpolation of missing/wrong data in the ICT and space counts to the POD/KLM specific reader method get_telemetry or use a copy instead of changing the input in-place.

pygac/pygac/calibration.py

Lines 409 to 425 in 8a7f11e

	# Thresholds to flag missing/wrong data for interpolation
	ict_threshold = 100
	space_threshold = 100
	if channel == 3:
	zeros = ict < ict_threshold
	nonzeros = np.logical_not(zeros)
	ict[zeros] = np.interp((zeros).nonzero()[0],
	(nonzeros).nonzero()[0],
	ict[nonzeros])
	zeros = space < space_threshold
	nonzeros = np.logical_not(zeros)
	space[zeros] = np.interp((zeros).nonzero()[0],
	(nonzeros).nonzero()[0],
	space[nonzeros])

[sfinkens]

I like the idea to move this to the POD/KLM specific get_telemetry methods. There are probably a number of methods in pygac which change data in-place and would need a refactoring. This one's a good start!

[mraspaud]

Sounds good, go ahead!