Minor fixes - Githubissues

nomad-coe / nomad-simulations

A NOMAD plugin containing base sections for simulations.

https://nomad-coe.github.io/nomad-simulations/

Apache License 2.0

6 stars 1 forks source link

Minor fixes #126

Open JosePizarro3 opened 2 months ago

JosePizarro3 commented 2 months ago

While I was debugging using the Wannier90 parser, I got some errors to be fixed (see nomad-parser-wannier90#3):

[ ] Change order of normalization for 1st KSpace and then its subsections (KMesh and KLinePath).
[x] Fix warning in spectral_profile.py adding condition for if self.name == 'ElectronicDensityOfStates' for total DOS.
[ ] Overwrite FermiLevel and call for ElectronicBandStructure.normalize() once DFT+TB workflow is resolved.
[ ] No idea why, but in the frontend, the SinglePoint.inputs and outputs section are not shown.
[x] Fix README
[x] Fix equality operator
[x] Move testing files to subfolders to mimic the src structure
[x] Delete copyright at the beginning of each file

JosePizarro3 commented 2 months ago

@ladinesa (once you feel better): can you tell me your thoughts on the first todo point regarding the order of calling normalize? I can explain more in detail (but you can also read the issue in the parser).

Also, the last point: I do not know why, but the SinglePoint links for inputs and outputs, the reference section is not showing on the front end:

Screenshot from 2024-09-23 16-10-28

It only happens for Input Model System and Outputs, but for Input Model Method is properly showing, see:

Screenshot from 2024-09-23 16-11-18

Any ideas what is going on wrong? I can share some branch with you to test.

ladinesa commented 2 months ago

Any ideas what is going on wrong? I can share some branch with you to test.

Did you generate the gui artifacts? it could be that the definition is simply not loaded.

JosePizarro3 commented 2 months ago

Yeah I did, but I can double check again (or maybe do before ctrl+f5...). ModelMethod seems to be working.

ladinesa commented 2 months ago

@ladinesa (once you feel better): can you tell me your thoughts on the first todo point regarding the order of calling normalize? I can explain more in detail (but you can also read the issue in the parser).

You could also try to normalize 'reciprocal_lattice_vectors' in the sub-section, and put an if in the main section if this is already normalized. You can isolate the normalize function for the reciprocal lattice vector so you can easily call it in the child section, something like self.m_parent.normalize_reciprocal_lattice_vector(..). The current normalizer level applies only for the sub sections in a section but I am not sure how we can do this to order generally and if this makes sense.