GeomFunc todo & tidy-up about epsproc HOT 6 OPEN

Need to check Euler/RX coord handling and sub-selection. Seem to be issues here with QN assignment and wignerD routine, likely due to both assumptions about data array.

Tested quick fixes by mods to sphCalc.wDcalc() function, but not yet working, and may not be the best way to handle this in any case!

(Ran into this during fitting routine development, for more see http://127.0.0.1:8888/lab/tree/dev/PEMtk/fitting/fitting_routines_dev_PEMtk_300421.ipynb)

from epsproc.

Fixed accidental pol geometry issue with AF calcs - due to this option remaining in AF code from MF code conversion/base functions.

By definition, there is no frame rotation in the AF case! For cases where E-field and alignment distribution are rotated, set AKQS in rotated frame (see frame rotation notes).

See 5b5fbcf

from epsproc.

Xarray version issues in geom code

Testing on Jake/Ubuntu20 today, some new issues... Note this is currently an issue on Jake only, maybe other env/dependencies problem

• v0.17 breaks in-place assignments (on Linux only, possibly)? E.g. da['mu'] = -1 * da['mu'] is OK, but da['mu'] *= -1 no longer working (was OK for v0.15). Possibly also Panadas-related changes.
• Quick test with v0.19 gives other issues to - to be addressed - and v0.15 also not working (although was previously OK, at least on Win).
• v0.13 is working in same env.,
• All tested via conda install xarray=XX, and each env backed up for reference. (conda list --revisions to check env changes)

from epsproc.

Xarray and Python version issues

Fixed in #51

UPDATE: also some Xarray coord assignment issues outstanding.

• Fixed for MF calcs in b2b80f5

from epsproc.

Testing for 3D alignment/AFPAD case July 2023

• Dev branch https://github.com/phockett/ePSproc/tree/3d-AFPAD-dev
• 79abfeb implements phase fixes, this gives correct M!=0 terms for ePS matrix elements in testing with OCS and C2H4 cases. Fix here was setting phaseCons['afblmCons']['negM'] = not(phaseCons['genMatEcons']['negm']), although it's possible this introduces inconsistencies elsewhere.
• 6a17868 allows for phaseCons['afblmCons'] for different cases. For S case, note that fix above removes M!=0 terms from DeltaTerm. This is incorrect!!!! And may also be incorrect above... TBC.

NOTE: this version used for QM3 book builds, corresponding archive at https://github.com/phockett/ePSproc/releases/tag/v1.3.2-dev-3d-QM3-310723

from epsproc.

Polarization states and pol geoms Feb/March 2024

• Dev on current 3d branch, https://github.com/phockett/ePSproc/tree/3d-AFPAD-dev
• Basics in place (e914b16, intro docs https://epsproc.readthedocs.io/en/3d-afpad-dev/demos/Epol_class_demo_docs_030324.html), but note some odd Blms in some cases in testing, maybe related to phase issues above, particularly for degen states and/or cases where the cross-section is small.
• TODO: more careful testing, and docs for pol geoms generally.

UPDATE 14/03/24: Basics in place, including docs and testing, as of f126bbb

from epsproc.