Equivalent points

[samwaseda]

Most important changes

• All symmetry-related operations are now found in symmetry.py (which allowed me to reduce 200 lines in atoms.py)
• Initialization via symmetry = structure.get_symmetry(), after which the attributes of symmetry can be used for different purposed (in the end this follows @jan-janssen's suggestion)
• Original output of structure.get_symmetry() is still available (i.e. structure.get_symmetry()['rotations'] etc.)
• !!! Unit tests are still relatively weak

Original purpose of this PR (i.e. get_arg_equivalent_sites())

This is very much equivalent to structure.group_points_by_symmetry, although the old function didn't work in many cases. Also instead of grouping the sites by lists, the new function returns their ID's.

Example: Group octahedral interstitial sites according to the box symmetries:

n_repeat = 2
a_0 = 4.
structure = pr.create.structure.bulk('Al', cubic=True, a=a_0).repeat(n_repeat)
sites = structure.get_wrapped_coordinates(structure.positions+np.array([0, 0, 0.5*a_0]))
del structure[0]
print(structure.get_symmetry().get_arg_equivalent_sites(sites))

This returns:

[0 1 1 2 1 1 3 2 1 3 1 2 3 3 3 2 0 0 0 2 1 0 1 2 1 1 0 2 3 1 1 2]

This should be very much equivalent to the distances between the vacancy and these positions:

print(np.round(structure.get_distances_array(v_position, sites), decimals=2))

[0 2 2 1 2 2 3 1 2 3 2 1 3 3 3 1 0 0 0 1 2 0 2 1 2 2 0 1 3 2 2 1]

Comments

• There are some properties/functions with no DocString -> there I don't know what they exactly do, so I didn't add stuff
• spglib-output was repeated in every function in the former version. This is now unified in symmetry._get_spglib_cell(). However, for some reason structure.get_cell().T was used for most of the places except for get_spacegroup() (which is now a property, spacegroup). As I have the feeling that .T should be removed everywhere, I removed it -> if you know more about this, and if this presents a problem, please let me know.

[coveralls]

Pull Request Test Coverage Report for Build 1049774120

• 107 of 129 (82.95%) changed or added relevant lines in 2 files are covered.
• 74 unchanged lines in 2 files lost coverage.
• Overall coverage increased (+0.1%) to 68.017%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
pyiron_atomistics/atomistics/structure/atoms.py	11	18	61.11%
pyiron_atomistics/atomistics/structure/symmetry.py	96	111	86.49%

Files with Coverage Reduction	New Missed Lines	%
pyiron_atomistics/atomistics/structure/neighbors.py	14	97.01%
pyiron_atomistics/atomistics/job/atomistic.py	60	72.95%

Totals
Change from base Build 1037555151:	0.1%
Covered Lines:	10893
Relevant Lines:	16015

---

💛 - Coveralls

[pmrv]

Just briefly going over, I have to admit that I don't understand the uses cases get_symmetry() vs. get_equivalent_point() vs. group_points_by_symmetry(). I'll try to have a detailed look again tomorrow, but I do feel this is something the docstring should tell me without reading the source.