Use get_atomic_numbers instead of atomic_numbers

[samwaseda]

Waiting for pyiron/pyiron_atomistics#1057

[jan-janssen]

Is not the parent=True option pyiron specific? The goal for the structuretoolkitis to be completely based on ASE Atoms.

[samwaseda]

Is not the parent=True option pyiron specific? The goal for the structuretoolkitis to be completely based on ASE Atoms.

Ah good point. Let me think about it.

[samwaseda]

Do you know whether it's actually possible to change the chemical symbols in ASE? I take a quick look at it and I couldn't find it.

[jan-janssen]

Do you know whether it's actually possible to change the chemical symbols in ASE? I take a quick look at it and I couldn't find it.

I guess the only special element they have is X:
https://gitlab.com/ase/ase/-/blob/master/ase/data/__init__.py#L12

from ase.atom import Atom
from ase.atoms import Atoms
structure = Atoms([Atom("X")])
print(structure.get_positions(), structure.cell)

[jan-janssen]

For changing elements you can use:
https://github.com/pyiron/structuretoolkit/blob/main/structuretoolkit/common/helper.py#L130

structure.symbols = element_lst 

[samwaseda]

That means the fundamental problem is the fact that we allow any element name then, which is then used in get_chemical_symbols.

[jan-janssen]

That's why I use pyiron_to_ase() in pyiron/pyiron_atomistics#1053 but I agree pyiron to ASE currently just ignores extra elements.

[jan-janssen]

I have to admit I was not aware that ASE had an get_atomic_numbers() function. But in this case I guess we can remove:

def get_atomic_numbers(structure):
    return structure.get_atomic_numbers()

And instead replace the get_atomic_numbers(structure) calls with structure.get_atomic_numbers().

[samwaseda]

I have to admit I was not aware that ASE had an get_atomic_numbers() function. But in this case I guess we can remove:

def get_atomic_numbers(structure):
    return structure.get_atomic_numbers()

And instead replace the get_atomic_numbers(structure) calls with structure.get_atomic_numbers().

I wasn't either :D but I guess then we found the optimal solution

[samwaseda]

sorry a mistake, just a moment

[jan-janssen]

sorry a mistake, just a moment

I am happy - at least the tests failed ;-)

[jan-janssen]

Looks good to me - feel free to merge it once the tests pass