[XPS] support multiple core levels for the same element

[superstar54]

workchain support multiple core levels for the same elements

for example, 1s, 2s, 2p for Al:

GUI to allow user specific the atom indices to be considered.

[superstar54]

Hi @PNOGillespie , if you have time to review this PR today, that would be great. Otherwise, would it be okay for me to merge it today so that a user can start testing this new feature?

[PNOGillespie]

Hi @superstar54. The feature here does seem useful, however there are some issues regarding input validation which need resolving before this will be good to go.

I've left some comments on the various changes, but the main things which need changing are:

1. There is no input validation for core_levels or core_hole_pseudos. I understand that this automatically done when loading in your pseudo_demo pseudopotentials - however when working with other pseudopotentials, the lack of validation can result in the WorkChain excepting only at the end of a run due to simple spelling/formatting errors.
2. Docstrings (at least for the workflows) should be updated to include these new parameters, in addition to something showing how they need to be formatted.

I will have another look at this once these two points have been addressed.

[superstar54]

Hi @PNOGillespie , thanks for the review. I have implemented suggestions, in the commit: c6c9ef7, also with tests.

I have added the following validation:

• keys of core_levels should be a subset of the structure elements and core_hole_pseudos
• atom_indices should be in the range of the number of atoms in the structure
• all the elements corresponding to the atom indices should be in the core_hole_pseudos

[PNOGillespie]

Hi @superstar54. Thanks for addressing some of the changes and adding tests. I have one last point which I think maybe wasn't made clear in the previous review.

We can probably go ahead and merge after addressing that last point.

[PNOGillespie]

Hi @superstar54. Please take a look at the comment for this review - I don't think we're quite there yet with input validation.

Please let me know if anything about it is unclear.

[superstar54]

Hi @PNOGillespie, thanks for your feedback!

I've added more input validation along with corresponding tests in this commit: f4ea8f3.

These improvements should make the input validation more robust and help catch potential issues early. Let me know if you have any further suggestions!

[PNOGillespie]

Thanks for the changes @superstar54. The input scheme is now a lot more robust for workflow users.

Just need to update the help string for core_hole_pseudos and I think we're done here.

[PNOGillespie]

Since this part has been removed, and the gipaw_pseudos has been merged into core_hole_pseudos, the help string for core_hole_pseudos needs to be updated for the new input format.

Doesn't need to be extensive, just something like:

'Dynamic namespace for ground-state and core-hole pseudopotentials for each absorbing'
' element. Must use the mapping "{element}" : {"gipaw" : {Upf}, "1s" : {Upf}}".'

[superstar54]

Done, in commit: 3b52fcb

[PNOGillespie]

Good work @superstar54! Go ahead and merge this.

[superstar54]

@PNOGillespie Thank you for the review!

[superstar54]

Hi @PNOGillespie , I think it's good to make another release (v0.1.1) to include the new functionality. I will do it now.