DIII-D IDA Diagnostics missing in OMAS #380

I have sorted the diagnostics from easiest to most difficult:

RIP

RIP interferometer - provides line integrated density and Faraday angle
https://github.com/GA-IDA/ida_lite/blob/c1398c826b7a327d6629b5518c3219b8870436ce/D3D/synt_diags/RIP.py#L47

It does downsampling, basic fringe jump correction, estimate of the statistical uncertainty etc

Update: implemented here #382

REFLECTOMETERS
The easiest diagnostics is likely the reflectometer, because the ITER structure mostly matches the MDS+ structure. An example of fetching is here:

https://github.com/GA-IDA/ida_lite/blob/c1398c826b7a327d6629b5518c3219b8870436ce/D3D/synt_diags/REFL.py#L41

But instead of dphase_df, it will be saved to the database just the phase. It would also be beneficial to save the reconstructed density from legacy code.

Update: implemented here #383

Visible Bremstrahlung
A more difficult diagnostic is visible bremsstrahlung. Fortunately, it is included in IMAS, at least.
https://github.com/GA-IDA/ida_lite/blob/main/D3D/synt_diags/VB.py

The raw data from the filterscope array are fetched by this https://github.com/ORNL-Fusion/VB_d3d
and CER data are retrieved directly from the MDS+ database. Then some preprocessing is done.

Here is a list of required signals
brightness (nch, nt)
brightness_err (nch, nt)
brightness_sys_err (nch, nt) (Missing in IMAS)
R (2, nch)
Z (2, nch)
Phi (2, nch)
razor (nch) (Missing in IMAS)
label (nch)
nbi_on (nch, nt) (Missing in IMAS)
wavelength (nch)

Eventually, I would like to fetch the raw signal from the reflectometer and calculate the spectrogram instead of the phase. Phase is not measured; it is already a highly processed signal.

LLAMA
LLAMA has a lower priority, but it would highly benefit from IDA analysis, and the synthetic diagnostics are relatively easy.
It is fetched by this package
https://github.com/LLAMAGithub/LLAMA-analysis/

Here is an example:
https://github.com/GA-IDA/ida_lite/blob/c1398c826b7a327d6629b5518c3219b8870436ce/D3D/synt_diags/LLAMA.py#L18

Unfortunately, LLAMA is not in IMAS.

CER

CER is the most challenging, and I would like to be involved.
Unfortunately, many essential signals are missing in IMAS.

Intensity (nch, nt) (missing in IMAS)
Intensity error (nch, nt) (missing in IMAS)
Intensity systematic error (nch, nt) (missing in IMAS)

(effective means value matching the Gaussian shift or width, without atomic data corrections)
effective rotation (nch, nt) (projection in LOS direction) (missing in IMAS)
effective rotation error (nch, nt) (missing in IMAS)
effective rotation systematic error (nch, nt) (missing in IMAS)

effective temperature (nch, nt) (projection in LOS direction) (missing in IMAS)
effective temperature error (nch, nt) (missing in IMAS)
effective temperature systematic error (nch, nt) (missing in IMAS)

visible Bremstahlung (nch, nt) (missing in IMAS)
visible Bremstahlung error (nch, nt) (missing in IMAS)
visible Bremstahlung systematic error (nch, nt) (missing in IMAS)

#this will be used when user does not use internal model for impurity density calculation
IMPDEN impurity density (nch, nt) (projection in LOS direction)
IMPDEN impurity density error (nch, nt)
IMPDEN impurity density systematic error (nch, nt) (missing in IMAS)

R_lens - radial coordinate of lens ( nch)
Z_lens - z coordinate of lens ( nch)
PHI_lens - phi coordinate of lens ( nch)

R_plasma - radial coordinate in plasma (nbeam_side, nch)
Z_plasma - z coordinate in plasma(nbeam_side, nch)
PHI_plasma - phi coordinate in plasma ( nbeam_side, nch)

nbi_geom (nbeam_side, nch) - beam geometry factor (missing in IMAS)
#relative power can be different for each channel!! Based on subtraction, if no data, nbi_rel_pow_err is inf for correct slicing
nbi_rel_pow (nbeam_side, nch, ntime) (missing in IMAS)
nbi_rel_pow_err (nbeam_side, nch, ntime) (missing in IMAS)
time - (ntime) time in the middle of time integration
integration time (ntime) (missing in IMAS)
lineid (nch)
wavelength (nch)

CER in IMAS has option to store directly the raw spectra and this will be used eventually. But not in the first phase.
And not for in between shot analysis.