I have an ase Atom class defined as such:
atoms = ase.Atoms(elements, np.array(flat_AP), cell=rand_box_dim, pbc=True)
Written as an ext-xyz file, this gives:
9
Lattice="2.172266536730066 0.0 0.0 0.0 2.497767468610632 0.0 0.0 0.0 2.6921538042558764" Properties=species:S:1:pos:R:3 pbc="T T T"
C 0.00000000 0.00000000 0.00000000
H 0.00000000 -0.04846279 -0.98881311
H 0.00000000 -0.83210588 0.53637656
C 1.08613327 0.68275479 0.35079970
H 1.08613327 1.51486066 -0.18557686
H 1.08613327 0.73121757 1.33961280
C 2.17226654 0.00000000 0.00000000
H 2.17226654 -0.83210588 0.53637656
H 2.17226654 -0.04846279 -0.98881311
I can introduce a new column for a property 'molid' as such:
atoms.arrays['molid'] = np.ones(len(atoms),dtype='int')
Which gives me:
9
Lattice="2.172266536730066 0.0 0.0 0.0 2.497767468610632 0.0 0.0 0.0 2.6921538042558764" Properties=species:S:1:pos:R:3:molid:I:1 pbc="T T T"
C 0.00000000 0.00000000 0.00000000 1
H 0.00000000 -0.04846279 -0.98881311 1
H 0.00000000 -0.83210588 0.53637656 1
C 1.08613327 0.68275479 0.35079970 1
H 1.08613327 1.51486066 -0.18557686 1
H 1.08613327 0.73121757 1.33961280 1
C 2.17226654 0.00000000 0.00000000 1
H 2.17226654 -0.83210588 0.53637656 1
H 2.17226654 -0.04846279 -0.98881311 1
Note that it adds the property 'molid' to the Properties= section of the file.
My question is - how would I be able to do the same thing for the property 'type', e.g. ['CT', 'HC', 'HC', 'CT', 'HC', 'HC', 'CT', 'HC', 'HC'], and add this to the ext-xyz file as a column?
I have tried this:
types = []
[types.append("CT") if x%3 == 0 else types.append("HC") for x in range(len(atoms))]
atoms.arrays['types'] = types
Which returns this error:
AttributeError: 'list' object has no attribute 'dtype'
Solved: it turns out numpy arrays are able to take strings as input!
[types.append("CT") if x%3 == 0 else types.append("HC") for x in range(len(atoms))]
types = np.array(types)
atoms.arrays['type'] = types