`Doscar` does not handle ISPIN = 1 calculations correctly.

[alexsquires]

Doscar.tdos dataframe is mislabeled (possibly incorrect) when setting ispin=1. For the same vasp calculation with ISPIN = 2 and ISPIN = 1, these are the dataframes returned by vasppy

[bjmorgan]

The same calculation, or different calculations on the same system (one spin-polarised, the other non-spin-polarised)?

[alexsquires]

two different calculations on the same system (LiF).

[bjmorgan]

Is it possible to attach the two DOSCARs? (should these give the same result in both cases? In the output above the energies and integrated DOS appear different?)

[alexsquires]

Seems like they are too big. I could rerun with a coarse DOS?

[alexsquires]

(should these give the same result in both cases? In the output above the energies and integrated DOS appear different?)

I would have assumed so (the only thing that changed between the two calculations was ISPIN)

[bjmorgan]

Would be nice to have a minimal example that I can e.g. write some proper tests against.

[alexsquires]

DOSCAR_non_spin_polarised.txt DOSCAR_spin_polarised.txt

[bjmorgan]

@alexsquires I think this is fixed now in ecefd14b3f2de8658130f5f7d370f69099c3dd86 (currently only in the doscar_nsp branch). Can you please test?

[alexsquires]

Sorry for the late response, this does fix the initial issue.