[SSO] Functionalities from fink-science

[JulienPeloton]

fink-science contains a lot of functionalities that should be here:

Move

from fink-science/fink_science/ssoft/processor.py to fink-utils/fink_utils/sso/spins.py

• angle_between_vectors
• extract_obliquity
• angular_separation

from fink-science/fink_science/ssoft/processor.py to fink-utils/fink_utils/sso/utils.py

• correct_ztf_mpc_names
• rockify

Units

• I/O homogeneisation: estimate_sso_params
• RA0 & DEC0 in degree in and out
• Period: hour or day, but should be consistent

Renaming

• query_miriade_epehemcc -> query_miriade_ephemcc
• func_hg1g2_with_spin -> func_shg1g2

Thanks @bcarry !

[bcarry]

Very dangerous absolute definition, ZTF-centric in fink_utils/sso/spins.py > func_sshg1g2
the reference epoch is now written as t0 = 2459580.5
which corresponds to Jan 1 2022 (2022-01-01)
(I did it, I take the responsibility and blame ;-))

I see two options:

1. Define t0 as a parameter of the func_sshg1g2 function. It is more flexible, it can be tailored for each SSO and/or dataset. But it should not be optimized by the estimate_sso_params function (there is a degeneracy between the parameters t0 and phi0 in the equations, so one must be fixed, and the reference time seems the most logical)
2. Definet0 as a general Fink constant. Something like "J2000".

I do not like option 2: the "only" usual time definition is indeed J2000, but between J2000 and LSST observations, there will be 20 to 30 years. So any tiny error on the rotation period will have an enormous impact.