SpaceGroup changes (#3859)

* Replaced SpaceGroup symbol attribute with its Hermann-Mauguin symbol, corrected Schoenflies point group attribute, changed handling of rhombohedral space group type settings by adding a SpaceGroup hexagonal bool attribute, modified and added tests.

* Removed crystal class key from symm_ops.json.

* Test for correct setting of hexagonal attribute when instantiating from int number.

* Noted change and replacement option for SpaceGroup symbol in compatibility.md

* Added from_space_group class method to PointGroup, added tests

* Added mapping to standard setting in PointGroup.from_space_group(), modified symm_ops.json and symm_data.json (documented in dev_scripts/update_space_group_data.py) to have same notation, added point group and short Hermann Mauguin symbol to symm_ops, fixed some typos, fixed rhombohedral space group type orbit issue.

* Updated core/test_surface.py to assign lattice as in SpaceGroup is_compatible().

* Modified databases and SpaceGroup init to ensure compatibility with non-underscore space group notations.

* Added tests for issue #3862, modified full_symbol and point_group attribute setting.

* Modified PointGroup.from_space_group() to also handle symbols with identity blickrichtungen and missed underscores, added warning to SpaceGroup init if full symbol is not available (for non-standard settings), added tests.

* Added test for warning if SpaceGroup.full_symbol is not available.

* Removed warning test.

* tweak incompat warning

* add test_full_symbol_warning

* add author + date to dev_scripts/update_spacegroup_data.py

* typos

* warning occurs only once, move test_full_symbol_warning up as workaround to annoying test pollution from side effects to std lib warnings registry

* Updated compatibility.md to also handle old symbol replacement of P2_12_12_1 and I2_12_12_1.

* pre-commit auto-fixes

* Updated dev script path to new src layout.

---------

Signed-off-by: Katharina Ueltzen <94910364+kaueltzen@users.noreply.github.com>
Co-authored-by: Janosh Riebesell <janosh.riebesell@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: J. George <JaGeo@users.noreply.github.com>

dev_scripts/update_spacegroup_data.py

@@ -0,0 +1,95 @@
+"""Script to update symm_ops.json and symm_data.yaml in symmetry module due to issues #3845 and #3862.
+symm_ops.json:
+- adds Hermann_mauguin point group key and short Hermann Mauguin space group symbol
+- converts screw axis notation to symm_data standard
+symm_data.json
+- removes mapping of rhombohedral space group types onto symbol + appended H
+- replaces I/P2_12_121 key with I/P2_12_12_1
+"""
+
+from __future__ import annotations
+
+import sys
+
+from monty.serialization import dumpfn, loadfn
+from pymatgen.symmetry.groups import PointGroup
+
+__author__ = "Katharina Ueltzen @kaueltzen"
+__date__ = "2024-06-06"
+
+SYMM_OPS = loadfn("../src/pymatgen/symmetry/symm_ops.json")
+SYMM_DATA = loadfn("../src/pymatgen/symmetry/symm_data.json")
+
+
+def convert_symmops_to_sg_encoding(symbol: str) -> str:
+    """
+    Utility function to convert SYMMOPS space group type symbol notation
+    into SYMM_DATA["space_group_encoding"] key notation with underscores before
+    translational part of screw axes.
+    Args:
+        symbol (str): "hermann_mauguin" or "universal_h_m" key of symmops.json
+    Returns:
+        symbol in the format of SYMM_DATA["space_group_encoding"] keys
+    """
+    symbol_representation = symbol.split(":")
+    representation = ":" + "".join(symbol_representation[1].split(" ")) if len(symbol_representation) > 1 else ""
+
+    blickrichtungen = symbol_representation[0].split(" ")
+    blickrichtungen_new = []
+    for br in blickrichtungen:
+        if len(br) > 1 and br[0].isdigit() and br[1].isdigit():
+            blickrichtungen_new.append(br[0] + "_" + br[1:])
+        else:
+            blickrichtungen_new.append(br)
+    return "".join(blickrichtungen_new) + representation
+
+
+def remove_identity_from_full_hermann_mauguin(symbol: str) -> str:
+    """
+    Utility function to remove identity along blickrichtung (except in P1).
+    Args:
+        symbol (str): "hermann_mauguin" key of symmops.json
+    Returns:
+        short "hermann_mauguin" key
+    """
+    if symbol in ("P 1", "C 1", "P 1 "):
+        return symbol
+    blickrichtungen = symbol.split(" ")
+    blickrichtungen_new = []
+    for br in blickrichtungen:
+        if br != "1":
+            blickrichtungen_new.append(br + " ")
+    return "".join(blickrichtungen_new)
+
+
+new_symm_data = {}
+for k, v in SYMM_DATA["space_group_encoding"].items():
+    if k.endswith("H"):
+        new_symm_data[k.removesuffix("H")] = v
+    elif k == "I2_12_121":
+        new_symm_data["I2_12_12_1"] = v
+    elif k == "P2_12_121":
+        new_symm_data["P2_12_12_1"] = v
+    else:
+        new_symm_data[k] = v
+
+SYMM_DATA["space_group_encoding"] = new_symm_data
+
+for spg_idx, spg in enumerate(SYMM_OPS):
+    if "(" in spg["hermann_mauguin"]:
+        SYMM_OPS[spg_idx]["hermann_mauguin"] = spg["hermann_mauguin"].split("(")[0]
+
+    short_h_m = remove_identity_from_full_hermann_mauguin(SYMM_OPS[spg_idx]["hermann_mauguin"])
+    SYMM_OPS[spg_idx]["short_h_m"] = convert_symmops_to_sg_encoding(short_h_m)
+    SYMM_OPS[spg_idx]["hermann_mauguin_u"] = convert_symmops_to_sg_encoding(spg["hermann_mauguin"])
+
+for spg_idx, spg in enumerate(SYMM_OPS):
+    try:
+        pg = PointGroup.from_space_group(spg["short_h_m"])
+    except AssertionError as e:
+        print(spg, str(e))
+        sys.exit(1)
+    SYMM_OPS[spg_idx]["point_group"] = pg.symbol
+
+dumpfn(SYMM_DATA, "../src/pymatgen/symmetry/symm_data.json")
+dumpfn(SYMM_OPS, "../src/pymatgen/symmetry/symm_ops.json")

src/pymatgen/symmetry/groups.py

@@ -112,6 +112,7 @@ class PointGroup(SymmetryGroup):
 
     def __init__(self, int_symbol: str) -> None:
         """Initialize a Point Group from its international symbol.
+        Please note that only the 32 crystal classes are supported right now.
 
         Args:
             int_symbol (str): International or Hermann-Mauguin Symbol.
@@ -165,6 +166,54 @@ def get_orbit(self, p: ArrayLike, tol: float = 1e-5) -> list[np.ndarray]:
                 orbit.append(pp)
         return orbit
 
+    @classmethod
+    def from_space_group(cls, sg_symbol: str) -> PointGroup:
+        """Instantiate one of the 32 crystal classes from a space group symbol in
+        Hermann Mauguin notation (int symbol or full symbol).
+
+        Args:
+            sg_symbol: space group symbol in Hermann Mauguin notation.
+
+        Raises:
+            AssertionError if a valid crystal class cannot be created
+        Returns:
+            crystal class in Hermann-Mauguin notation.
+        """
+        abbrev_map = {
+            "2/m2/m2/m": "mmm",
+            "4/m2/m2/m": "4/mmm",
+            "-32/m": "-3m",
+            "6/m2/m2/m": "6/mmm",
+            "2/m-3": "m-3",
+            "4/m-32/m": "m-3m",
+        }
+        non_standard_map = {
+            "m2m": "mm2",
+            "2mm": "mm2",
+            "-4m2": "-42m",  # technically not non-standard
+            "-62m": "-6m2",  # technically not non-standard
+        }
+        symbol = re.sub(r" ", "", sg_symbol)
+
+        symm_ops = loadfn(os.path.join(os.path.dirname(__file__), "symm_ops.json"))  # get short symbol if possible
+        for spg in symm_ops:
+            if symbol in [spg["hermann_mauguin"], spg["universal_h_m"], spg["hermann_mauguin_u"]]:
+                symbol = spg["short_h_m"]
+
+        assert symbol[0].isupper(), f"Invalid sg_symbol {sg_symbol}"
+        assert not symbol[1:].isupper(), f"Invalid sg_symbol {sg_symbol}"
+
+        symbol = symbol[1:]  # Remove centering
+        symbol = symbol.translate(str.maketrans("abcden", "mmmmmm"))  # Remove translation from glide planes
+        symbol = re.sub(r"_.", "", symbol)  # Remove translation from screw axes
+        symbol = abbrev_map.get(symbol, symbol)
+        symbol = non_standard_map.get(symbol, symbol)
+
+        assert (
+            symbol in SYMM_DATA["point_group_encoding"]
+        ), f"Could not create a valid crystal class ({symbol}) from sg_symbol {sg_symbol}"
+        return cls(symbol)
+
 
 @cached_class
 class SpaceGroup(SymmetryGroup):
@@ -194,7 +243,7 @@ class SpaceGroup(SymmetryGroup):
         v["full_symbol"]: k for k, v in SYMM_DATA["space_group_encoding"].items()
     }
 
-    def __init__(self, int_symbol: str) -> None:
+    def __init__(self, int_symbol: str, hexagonal: bool = True) -> None:
         """Initialize a Space Group from its full or abbreviated international
         symbol. Only standard settings are supported.
 
@@ -209,9 +258,26 @@ def __init__(self, int_symbol: str) -> None:
                 possible settings for a spacegroup, use the get_settings()
                 classmethod. Alternative origin choices can be indicated by a
                 translation vector, e.g. 'Fm-3m(a-1/4,b-1/4,c-1/4)'.
+            hexagonal (bool): For rhombohedral groups, whether to handle as in
+                hexagonal setting (default) or rhombohedral setting.
+                If the int_symbol of a rhombohedral spacegroup is given with the
+                setting ("(:)H"/"(:)R"), this parameter is overwritten accordingly
+                (please note that the setting is not contained in the symbol
+                attribute anymore).
         """
         from pymatgen.core.operations import SymmOp
 
+        if int_symbol.endswith("H"):
+            self.hexagonal = True
+            if not int_symbol.endswith(":H"):
+                int_symbol = int_symbol[:-1] + ":H"
+        elif int_symbol.endswith("R"):
+            self.hexagonal = False
+            if not int_symbol.endswith(":R"):
+                int_symbol = int_symbol[:-1] + ":R"
+        else:
+            self.hexagonal = hexagonal
+
         int_symbol = re.sub(r" ", "", int_symbol)
         if int_symbol in SpaceGroup.abbrev_sg_mapping:
             int_symbol = SpaceGroup.abbrev_sg_mapping[int_symbol]
@@ -221,15 +287,22 @@ def __init__(self, int_symbol: str) -> None:
         self._symmetry_ops: set[SymmOp] | None
 
         for spg in SpaceGroup.SYMM_OPS:
-            if int_symbol in [spg["hermann_mauguin"], spg["universal_h_m"]]:
+            if int_symbol in [spg["hermann_mauguin"], spg["universal_h_m"], spg["hermann_mauguin_u"]]:
                 ops = [SymmOp.from_xyz_str(s) for s in spg["symops"]]
-                self.symbol = re.sub(r":", "", re.sub(r" ", "", spg["universal_h_m"]))
+                self.symbol = spg["hermann_mauguin_u"]
                 if int_symbol in SpaceGroup.sg_encoding:
                     self.full_symbol = SpaceGroup.sg_encoding[int_symbol]["full_symbol"]
                     self.point_group = SpaceGroup.sg_encoding[int_symbol]["point_group"]
+                elif self.symbol in SpaceGroup.sg_encoding:
+                    self.full_symbol = SpaceGroup.sg_encoding[self.symbol]["full_symbol"]
+                    self.point_group = SpaceGroup.sg_encoding[self.symbol]["point_group"]
                 else:
-                    self.full_symbol = re.sub(r" ", "", spg["universal_h_m"])
-                    self.point_group = spg["schoenflies"]
+                    self.full_symbol = spg["hermann_mauguin_u"]
+                    warnings.warn(
+                        f"Full symbol not available, falling back to short Hermann Mauguin symbol "
+                        f"{self.symbol} instead"
+                    )
+                    self.point_group = spg["point_group"]
                 self.int_number = spg["number"]
                 self.order = len(ops)
                 self._symmetry_ops = {*ops}
@@ -399,7 +472,7 @@ def check(param, ref, tolerance):
         if crys_system == "hexagonal" or (
             crys_system == "trigonal"
             and (
-                self.symbol.endswith("H")
+                self.hexagonal
                 or self.int_number
                 in [143, 144, 145, 147, 149, 150, 151, 152, 153, 154, 156, 157, 158, 159, 162, 163, 164, 165]
             )
@@ -531,7 +604,8 @@ def sg_symbol_from_int_number(int_number: int, hexagonal: bool = True) -> str:
             hexagonal setting (default) or rhombohedral setting.
 
     Returns:
-        str: Spacegroup symbol
+        str: Spacegroup symbol / Space group symbol + "H" if group is
+            rhombohedral and hexagonal=True
     """
     syms = []
     for n, v in SYMM_DATA["space_group_encoding"].items():

tests/analysis/test_piezo.py

@@ -18,7 +18,7 @@
 
 class TestPiezo(PymatgenTest):
     def setUp(self):
-        self.piezo_struc = self.get_structure("BaNiO3")
+        self.piezo_struct = self.get_structure("BaNiO3")
         self.voigt_matrix = np.array(
             [
                 [0.0, 0.0, 0.0, 0.0, 0.03839, 0.0],

tests/core/test_surface.py

@@ -379,9 +379,9 @@ def test_get_slab(self):
             if sg.crystal_system == "hexagonal" or (
                 sg.crystal_system == "trigonal"
                 and (
-                    sg.symbol.endswith("H")
+                    sg.hexagonal
                     or sg.int_number
-                    in [143, 144, 145, 147, 149, 150, 151, 152, 153, 154, 156, 157, 158, 159, 162, 163, 164, 165]
+                    in (143, 144, 145, 147, 149, 150, 151, 152, 153, 154, 156, 157, 158, 159, 162, 163, 164, 165)
                 )
             ):
                 lattice = Lattice.hexagonal(5, 10)

tests/files/.pytest-split-durations

@@ -2017,7 +2017,7 @@
     "tests/io/lobster/test_inputsenv.py::TestLobsterNeighbors::test_get_nn_info": 0.23225333401933312,
     "tests/io/lobster/test_inputsenv.py::TestLobsterNeighbors::test_get_plot_label": 0.2305897069745697,
     "tests/io/lobster/test_inputsenv.py::TestLobsterNeighbors::test_get_structure_environments": 0.2838197909295559,
-    "tests/io/lobster/test_inputsenv.py::TestLobsterNeighbors::test_get_strucuture_environments_further_tests": 0.2533352089812979,
+    "tests/io/lobster/test_inputsenv.py::TestLobsterNeighbors::test_get_structure_environments_further_tests": 0.2533352089812979,
     "tests/io/lobster/test_inputsenv.py::TestLobsterNeighbors::test_get_sum_icohps_between_neighbors_of_atom": 0.24758987501263618,
     "tests/io/lobster/test_inputsenv.py::TestLobsterNeighbors::test_molecules_allowed": 0.23277270805556327,
     "tests/io/lobster/test_inputsenv.py::TestLobsterNeighbors::test_order_parameter": 0.2319688760326244,

tests/io/lobster/test_lobsterenv.py

@@ -687,7 +687,7 @@ def test_get_structure_environments(self):
         lse2 = self.chem_env_lobster1.get_light_structure_environment()
         assert lse2.coordination_environments[0][0]["ce_symbol"] == "O:6"
 
-    def test_get_strucuture_environments_further_tests(self):
+    def test_get_structure_environments_further_tests(self):
         lse = self.chem_env_lobster1_second.get_light_structure_environment()
         lse.as_dict()
         lse.get_statistics()

tests/io/test_pwscf.py

@@ -376,7 +376,7 @@ def test_read_str(self):
         assert pw_in.sections["system"]["smearing"] == "cold"
 
 
-class TestPWOuput(PymatgenTest):
+class TestPWOutput(PymatgenTest):
     def setUp(self):
         self.pw_out = PWOutput(f"{TEST_DIR}/Si.pwscf.out")
 

tests/io/vasp/test_sets.py

@@ -728,12 +728,12 @@ def test_init(self):
             vis = self.set(vis.structure, user_incar_settings={"LUSE_VDW": True, "LASPH": False})
             vis.incar.items()
         with pytest.warns(BadInputSetWarning, match=r"LASPH"):
-            dummy_struc = Structure(
+            dummy_struct = Structure(
                 lattice=[[0, 2, 2], [2, 0, 2], [2, 2, 0]],
                 species=["Fe", "O"],
                 coords=[[0, 0, 0], [0.5, 0.5, 0.5]],
             )
-            vis = self.set(dummy_struc, user_incar_settings={"LDAU": True, "LASPH": False})
+            vis = self.set(dummy_struct, user_incar_settings={"LDAU": True, "LASPH": False})
             vis.incar.items()
 
     def test_user_incar_kspacing(self):