Format mis-matching INCAR tags warning for many defects

doped/analysis.py

@@ -1230,19 +1230,32 @@ def _call_multiple_corrections_tolerance_warning(correction_errors, type="FNV"):
         # within the charge correction functions (with self._check_if_multiple_finite_size_corrections())
 
         mismatching_INCAR_warnings = [
-            (name, defect_entry.calculation_metadata.get("mismatching_INCAR_tags"))
+            (name, set(defect_entry.calculation_metadata.get("mismatching_INCAR_tags")))
             for name, defect_entry in self.defect_dict.items()
             if defect_entry.calculation_metadata.get("mismatching_INCAR_tags", True) is not True
         ]
         if mismatching_INCAR_warnings:
+            # group by the mismatching tags, so we can print them together:
+            mismatching_tags_name_list_dict = {
+                tuple(sorted(mismatching_set)): [
+                    name
+                    for name, other_mismatching_set in mismatching_INCAR_warnings
+                    if other_mismatching_set == mismatching_set
+                ]
+                for mismatching_set in [mismatching for name, mismatching in mismatching_INCAR_warnings]
+            }
             joined_info_string = "\n".join(
-                [f"{name}: {mismatching}" for name, mismatching in mismatching_INCAR_warnings]
+                [
+                    f"{defect_list}:\n{list(mismatching)}"
+                    for mismatching, defect_list in mismatching_tags_name_list_dict.items()
+                ]
             )
             warnings.warn(
                 f"There are mismatching INCAR tags for (some of) your bulk and defect calculations which "
                 f"are likely to cause errors in the parsed results (energies). Found the following "
                 f"differences:\n"
-                f"(in the format: (INCAR tag, value in bulk calculation, value in defect calculation)):"
+                f"(in the format: 'Defects: (INCAR tag, value in bulk calculation, value in defect "
+                f"calculation))':"
                 f"\n{joined_info_string}\n"
                 f"In general, the same INCAR settings should be used in all final calculations for these "
                 f"tags which can affect energies!"

doped/utils/parsing.py

@@ -802,21 +802,28 @@ def check_atom_mapping_far_from_defect(
         )[0]
     ]
 
-    bulk_species_outside_near_ws_coord_dict = {}
-    defect_species_outside_ws_coord_dict = {}
     for species in bulk_supercell.composition.elements:  # divide and vectorise calc for efficiency
-        bulk_species_outside_near_ws_coord_dict[species.name] = get_coords_and_idx_of_species(
+        bulk_species_outside_near_ws_coords = get_coords_and_idx_of_species(
             bulk_sites_outside_or_at_ws_radius, species.name
         )[0]
-        defect_species_outside_ws_coord_dict[species.name] = get_coords_and_idx_of_species(
+        defect_species_outside_ws_coords = get_coords_and_idx_of_species(
             defect_sites_outside_wigner_radius, species.name
         )[0]
-        vecs, d_2 = pbc_shortest_vectors(
-            bulk_supercell.lattice,
-            defect_species_outside_ws_coord_dict[species.name],
-            bulk_species_outside_near_ws_coord_dict[species.name],
-            return_d2=True,
+        if (
+            min(
+                len(bulk_species_outside_near_ws_coords),
+                len(defect_species_outside_ws_coords),
+            )
+            == 0
+        ):
+            continue  # if no sites of this species outside the WS radius, skip
+
+        subset, superset = (  # supa-set
+            (defect_species_outside_ws_coords, bulk_species_outside_near_ws_coords)
+            if len(defect_species_outside_ws_coords) < len(bulk_species_outside_near_ws_coords)
+            else (bulk_species_outside_near_ws_coords, defect_species_outside_ws_coords)
         )
+        vecs, d_2 = pbc_shortest_vectors(bulk_supercell.lattice, subset, superset, return_d2=True)
         site_matches = LinearAssignment(d_2).solution  # matching superset indices, of len(subset)
         matching_vecs = vecs[np.arange(len(site_matches)), site_matches]
         displacements = np.linalg.norm(matching_vecs, axis=1)

tests/test_analysis.py

@@ -193,8 +193,8 @@ def _check_default_CdTe_DefectsParser_outputs(
                 i in str(warn.message)
                 for i in [
                     "There are mismatching INCAR tags for (some of)",
-                    "in the format: (INCAR tag, value in bulk calculation, value in defect",
-                    "Int_Te_3_Unperturbed_1: [('ADDGRID', True, False)]",
+                    "in the format: 'Defects: (INCAR tag, value in bulk calculation, value in defect",
+                    "['Int_Te_3_Unperturbed_1']:\n[('ADDGRID', True, False)]",
                     "In general, the same INCAR settings should be used",
                 ]
             )
@@ -968,9 +968,7 @@ def test_SrTiO3_diff_ISYM_bulk_defect_and_concentration_funcs(self):
             i in str(w[0].message)
             for i in [
                 "There are mismatching INCAR tags",
-                "vac_O_0: [('LASPH', False, True)]",
-                "vac_O_1: [('LASPH', False, True)]",
-                "vac_O_2: [('LASPH', False, True)]",
+                "['vac_O_1', 'vac_O_0', 'vac_O_2']:\n[('LASPH', False, True)]",
             ]
         )
 
@@ -1149,13 +1147,10 @@ def test_ZnS_non_diagonal_NKRED_mismatch(self):
             i in str(w[0].message)
             for i in [
                 "There are mismatching INCAR tags",
-                "vac_1_Zn_0: [('NKRED', 2, 1)]",
-                "vac_1_Zn_-2: [('NKRED', 2, 1)]",
-                "vac_2_S_2: [('NKRED', 2, 1)]",
-                "inter_29_Al_3: [('NKRED', 2, 1)]",
-                "sub_1_Al_on_Zn_-1: [('NKRED', 2, 1)]",
+                ":\n[('NKRED', 2, 1)]\nIn",
             ]
         )
+        assert str(w[0].message).count(":\n[('NKRED', 2, 1)]\nIn") == 1  # only once
 
         assert len(dp.defect_dict) == 17
         self._check_DefectsParser(dp)