Merge pull request #73 from paulsaxe/main

Handle uppercase X, Y, Z in strings for symmetry operators

HISTORY.rst

@@ -1,6 +1,14 @@
 =======
 History
 =======
+2024.3.13 -- Handle uppercase X, Y, Z in strings for symmetry operators
+    * the Crystallographic Open Database CIF files seems to use upper case X, Y, Z in
+      explicit symmetry operators. These need to be lowercased in the code.
+
+2023.12.5 -- Bugfixes for symmetry
+    * Fixed issue #72, where symmetry was not correctly handled for trigonal and
+      hexagonal cells where atoms had coordinates of 1/3 or 2/3.
+
 2023.11.19 -- Bugfixes in symmetry and CIF files
     * Reading CIF files could fail if the symmetry operators were given
     * The symmetry handling did not recognize hexagonal spacegroups without :H. Changed

molsystem/cif.py

@@ -17,6 +17,7 @@
 import io
 import json
 import logging
+import re
 
 import CifFile
 
@@ -113,9 +114,15 @@ def to_cif_text(self):
         # Get the chemical formula
         formula, empirical_formula, Z = self.formula
 
-        _id = empirical_formula.replace(" ", "")
+        system_name = self.system.name
+        conf_name = self.name
+        if system_name == "" or conf_name == "":
+            _id = empirical_formula.replace(" ", "")
+        else:
+            _id = f"SEAMM:{system_name}/{conf_name}"
+            _id = _id.replace(" ", "")
 
-        # And created the file, line-by-line
+        # And create the file, line-by-line
         lines = []
         lines.append("# Generated by MolSSI SEAMM")
         lines.append(f"data_{_id}")
@@ -130,7 +137,7 @@ def to_cif_text(self):
             if spgname == "":
                 pass
             elif symmetry.n_symops == 1:
-                lines.append("space_group_name_H-M_full   'P 1'")
+                lines.append("_space_group_name_H-M_full   'P 1'")
             else:
                 spgname_system = symmetry.spacegroup_names_to_system[spgname]
                 lines.append(f"_space_group_{spgname_system}       '{spgname}'")
@@ -447,9 +454,42 @@ def from_cif_text(self, text):
                 symbol = "H"
 
             # May have uncertainties in ()
-            x = float(x.split("(")[0])
-            y = float(y.split("(")[0])
-            z = float(z.split("(")[0])
+            x = x.split("(")[0]
+            y = y.split("(")[0]
+            z = z.split("(")[0]
+
+            # Check for 1/3 and 2/3
+            if "." in x:
+                xx, dx = x.split(".")
+                if re.fullmatch(r"333+", dx):
+                    x = float(xx) - 1.0 / 3.0 if "-" in xx else float(xx) + 1.0 / 3.0
+                elif re.fullmatch(r"66+(6|7)?", dx):
+                    x = float(xx) - 2.0 / 3.0 if "-" in xx else float(xx) + 2.0 / 3.0
+                else:
+                    x = float(x)
+            else:
+                x = float(x)
+            if "." in y:
+                yy, dy = y.split(".")
+                if re.fullmatch(r"333+", dy):
+                    y = float(yy) - 1.0 / 3.0 if "-" in yy else float(yy) + 1.0 / 3.0
+                elif re.fullmatch(r"66+(6|7)?", dy):
+                    y = float(yy) - 2.0 / 3.0 if "-" in yy else float(yy) + 2.0 / 3.0
+                else:
+                    y = float(y)
+            else:
+                y = float(y)
+            if "." in z:
+                zz, dz = z.split(".")
+                if re.fullmatch(r"333+", dz):
+                    z = float(zz) - 1.0 / 3.0 if "-" in zz else float(zz) + 1.0 / 3.0
+                elif re.fullmatch(r"66+(6|7)?", dz):
+                    z = float(zz) - 2.0 / 3.0 if "-" in zz else float(zz) + 2.0 / 3.0
+                else:
+                    z = float(z)
+            else:
+                z = float(z)
+
             logger.debug(f"xyz = {x:7.3f} {y:7.3f} {z:7.3f}")
             if self.periodicity == 3:
                 if not have_fractionals:

molsystem/openbabel.py

@@ -77,13 +77,15 @@ def from_OBMol(self, ob_mol, properties="all"):
         Xs = []
         Ys = []
         Zs = []
+        qs = []
         for ob_atom in ob.OBMolAtomIter(ob_mol):
             atno = ob_atom.GetAtomicNum()
             atnos.append(atno)
             Xs.append(ob_atom.x())
             Ys.append(ob_atom.y())
             Zs.append(ob_atom.z())
             logger.debug(f"atom {atno} {ob_atom.x()} {ob_atom.z()} {ob_atom.z()}")
+            qs.append(ob_atom.GetFormalCharge())
 
         Is = []
         Js = []
@@ -105,7 +107,12 @@ def from_OBMol(self, ob_mol, properties="all"):
             self.charge = ob_mol.GetTotalCharge()
             self.spin_multiplicity = ob_mol.GetTotalSpinMultiplicity()
 
-        ids = self.atoms.append(x=Xs, y=Ys, z=Zs, atno=atnos)
+        if any([i != 0.0 for i in qs]):
+            if "formal_charge" not in self.atoms:
+                self.atoms.add_attribute("formal_charge", coltype="int", default=0)
+            ids = self.atoms.append(x=Xs, y=Ys, z=Zs, atno=atnos, formal_charge=qs)
+        else:
+            ids = self.atoms.append(x=Xs, y=Ys, z=Zs, atno=atnos)
         i = [ids[x - 1] for x in Is]
         j = [ids[x - 1] for x in Js]
         self.bonds.append(i=i, j=j, bondorder=BondOrders)

molsystem/symmetry.py

@@ -296,8 +296,10 @@ def symops(self, ops):
             text = ""
         else:
             text = " | ".join(ops)
-
-        self.db.execute("UPDATE symmetry SET symops = ? WHERE id = ?", (text, self.id))
+        # Ensure the lower case letters are used! x, y, z not X, Y, Z
+        self.db.execute(
+            "UPDATE symmetry SET symops = ? WHERE id = ?", (text.lower(), self.id)
+        )
         # Unset the group
         self.db.execute('UPDATE symmetry SET "group" = "" WHERE id = ?', (self.id,))
         self.db.commit()
@@ -348,6 +350,95 @@ def spacegroup_numbers_to_hall(self):
 
     @property
     def spacegroup_names_to_hall(self):
+        """Dictionary of Hall number for spacegroup names."""
+        if _Symmetry.spgname_to_hall is None:
+            system_name = {
+                "international_full": "name_H-M_full",
+                "international": "name_H-M_alt",
+                "international_short": "name_H-M_short",
+                "hall_symbol": "name_Hall",
+            }
+            # Initialize the symmetry data
+            _Symmetry.spgname_to_hall = {}
+            _Symmetry.hall_to_spgname = {}
+            _Symmetry.hall_to_hall_symbol = {}
+            _Symmetry.hall_to_IT_number = {}
+            if _Symmetry.spgname_to_system is None:
+                _Symmetry.spgname_to_system = {}
+            for hall in range(1, 530):
+                data = spglib.get_spacegroup_type(hall)
+                # pws pprint.pprint(data)
+                choice = data["choice"]
+                _Symmetry.hall_to_hall_symbol[hall] = data["hall_symbol"]
+                _Symmetry.hall_to_IT_number[hall] = data["number"]
+
+                # Handle Hall to spacegroup using the full H-M name
+                key = "international_full"
+                name = data[key]
+
+                # Default setting if there are multiple, leave unadorned
+                if choice in ("2", "b", "b1", "H"):
+                    if (
+                        hall in _Symmetry.hall_to_spgname
+                        and name != _Symmetry.hall_to_spgname[hall]
+                    ):
+                        raise RuntimeError(
+                            f"{hall=} {key} --> {name} exists: "
+                            f"{_Symmetry.hall_to_spgname[hall]}"
+                        )
+                    else:
+                        _Symmetry.hall_to_spgname[hall] = name
+                else:
+                    # Add other settings to the spacegroup name
+                    if choice != "":
+                        name += f" :{choice}"
+                    if hall not in _Symmetry.hall_to_spgname:
+                        # pws if choice != "":
+                        # pws print(f"{hall} = {name}   QQQQQQQQQQQQQQQQQQQQQQ")
+                        _Symmetry.hall_to_spgname[hall] = name
+
+                # Now handle all the rest
+                for key in (
+                    "international_full",
+                    "international",
+                    "international_short",
+                    "hall_symbol",
+                ):
+                    name = data[key]
+
+                    # SPGLib encodes the international name: 'C 2/c = B 2/n 1 1',
+                    if key == "international":
+                        name = name.split("=")[0].strip()
+
+                    # Default setting if there are multiple, leave unadorned
+                    if choice in ("2", "b", "b1", "H"):
+                        _Symmetry.spgname_to_hall[name] = hall
+                        _Symmetry.spgname_to_system[name] = system_name[key]
+                        for txt in ("_", " "):
+                            tmp = name.replace(txt, "")
+                            _Symmetry.spgname_to_hall[tmp] = hall
+                            _Symmetry.spgname_to_system[tmp] = system_name[key]
+                            if tmp[-2:] == ":H":
+                                tmp = tmp[:-2].strip()
+                                _Symmetry.spgname_to_hall[tmp] = hall
+                                _Symmetry.spgname_to_system[tmp] = system_name[key]
+
+                    if key in ("international", "international_short") and choice != "":
+                        name += f" :{choice}"
+
+                    _Symmetry.spgname_to_hall[name] = hall
+                    _Symmetry.spgname_to_system[name] = system_name[key]
+                    for txt in ("_", " "):
+                        tmp = name.replace(txt, "")
+                        _Symmetry.spgname_to_hall[tmp] = hall
+                        _Symmetry.spgname_to_system[tmp] = system_name[key]
+                        if tmp[-2:] == ":H":
+                            tmp = tmp[:-2].strip()
+                            _Symmetry.spgname_to_hall[tmp] = hall
+                            _Symmetry.spgname_to_system[tmp] = system_name[key]
+        return _Symmetry.spgname_to_hall
+
+    def old_spacegroup_names_to_hall(self):
         """Dictionary of Hall number for spacegroup names."""
         if _Symmetry.spgname_to_hall is None:
             system_name = {