Make run_type in mixing scheme consistent with entries (#4255)

* Test that entries with run_type 'r2SCAN' would not be ignored; 'r2SCAN' as an input run_type is not case-sensitive.

* Fix to ensure r2SCAN entries would not be ignored in mixing scheme correction

---------

Co-authored-by: Shyue Ping Ong <[email protected]>

Author: peikai

src/pymatgen/entries/mixing_scheme.py

@@ -40,15 +40,15 @@ class MaterialsProjectDFTMixingScheme(Compatibility):
     may lead to unexpected results.
 
     This is the scheme used by the Materials Project to generate Phase Diagrams containing
-    a mixture of GGA(+U) and R2SCAN calculations. However in principle it can be used to
+    a mixture of GGA(+U) and r2SCAN calculations. However in principle it can be used to
     mix energies from any two functionals.
     """
 
     def __init__(
         self,
         structure_matcher: StructureMatcher | None = None,
         run_type_1: str = "GGA(+U)",
-        run_type_2: str = "R2SCAN",
+        run_type_2: str = "r2SCAN",
         compat_1: (Compatibility | None) = MaterialsProject2020Compatibility(),  # noqa: B008
         compat_2: Compatibility | None = None,
         fuzzy_matching: bool = True,
@@ -64,7 +64,7 @@ def __init__(
             run_type_1: The first DFT run_type. Typically this is the majority or run type or
                 the "base case" onto which the other calculations are referenced. Valid choices
                 are any run_type recognized by Vasprun.run_type, such as "LDA", "GGA", "GGA+U",
-                "PBEsol", "SCAN", or "R2SCAN". The class will ignore any entries that have a
+                "PBEsol", "SCAN", or "r2SCAN". The class will ignore any entries that have a
                 run_type different than run_type_1 or run_type_2.
 
                 The list of run_type_1 entries provided to process_entries MUST form a complete
@@ -103,8 +103,10 @@ def __init__(
             raise ValueError(f"run_type_1={run_type_2=}. The mixing scheme is meaningless unless run_types different")
         self.run_type_1 = run_type_1
         self.run_type_2 = run_type_2
-        self.valid_rtypes_1 = ["GGA", "GGA+U"] if self.run_type_1 == "GGA(+U)" else [self.run_type_1]
-        self.valid_rtypes_2 = ["GGA", "GGA+U"] if self.run_type_2 == "GGA(+U)" else [self.run_type_2]
+        """Valid run_type, allowing for archive R2SCAN entries"""
+        valid_rtype_dict = {"GGA(+U)": ["GGA", "GGA+U"], "R2SCAN": ["r2SCAN", "R2SCAN"]}
+        self.valid_rtypes_1 = valid_rtype_dict.get(run_type_1.upper(), [self.run_type_1])
+        self.valid_rtypes_2 = valid_rtype_dict.get(run_type_2.upper(), [self.run_type_2])
 
         self.compat_1 = compat_1
         self.compat_2 = compat_2
@@ -253,7 +255,7 @@ def process_entries(
 
     def get_adjustments(self, entry, mixing_state_data: pd.DataFrame | None = None):
         """Get the corrections applied to a particular entry. Note that get_adjustments is not
-        intended to be called directly in the R2SCAN mixing scheme. Call process_entries instead,
+        intended to be called directly in the r2SCAN mixing scheme. Call process_entries instead,
         and it will pass the required arguments to get_adjustments.
 
         Args:
@@ -326,7 +328,7 @@ def get_adjustments(self, entry, mixing_state_data: pd.DataFrame | None = None):
                 # For run_type_2 entries, there is no correction
                 return adjustments
 
-            # Discard GGA ground states whose structures already exist in R2SCAN.
+            # Discard GGA ground states whose structures already exist in r2SCAN.
             df_slice = mixing_state_data[(mixing_state_data["entry_id_1"] == entry.entry_id)]
 
             if df_slice["entry_id_2"].notna().item():
@@ -344,8 +346,8 @@ def get_adjustments(self, entry, mixing_state_data: pd.DataFrame | None = None):
                     f"because there is a matching {self.run_type_2} material."
                 )
 
-            # If a GGA is not present in R2SCAN, correct its energy to give the same
-            # e_above_hull on the R2SCAN hull that it would have on the GGA hull
+            # If a GGA is not present in r2SCAN, correct its energy to give the same
+            # e_above_hull on the r2SCAN hull that it would have on the GGA hull
             hull_energy_1 = df_slice["hull_energy_1"].iloc[0]
             hull_energy_2 = df_slice["hull_energy_2"].iloc[0]
             correction = (hull_energy_2 - hull_energy_1) * entry.composition.num_atoms

tests/entries/test_mixing_scheme.py

@@ -1844,10 +1844,10 @@ def test_fuzzy_matching(self, ms_complete):
                 with pytest.raises(CompatibilityError, match="ground state"):
                     compat.get_adjustments(entry, mixing_state)
             elif entry.entry_id == "r2scan-3":
-                with pytest.raises(CompatibilityError, match="and no R2SCAN ground state"):
+                with pytest.raises(CompatibilityError, match="and no r2SCAN ground state"):
                     compat.get_adjustments(entry, mixing_state)
             elif entry.entry_id in ["gga-2", "gga-5", "gga-6", "gga-7"]:
-                with pytest.raises(CompatibilityError, match="there is a matching R2SCAN"):
+                with pytest.raises(CompatibilityError, match="there is a matching r2SCAN"):
                     compat.get_adjustments(entry, mixing_state)
             elif entry.parameters["run_type"] == "GGA":
                 assert not compat.get_adjustments(entry, mixing_state)
@@ -1861,7 +1861,7 @@ def test_fuzzy_matching(self, ms_complete):
             if entry.parameters["run_type"] == "GGA":
                 assert entry.correction == 0
             elif entry.entry_id in ["r2scan-2", "r2scan-7"]:
-                assert "Replace R2SCAN energy with GGA" in entry.energy_adjustments[0].description
+                assert "Replace r2SCAN energy with GGA" in entry.energy_adjustments[0].description
             else:
                 assert "onto the GGA(+U) hull" in entry.energy_adjustments[0].description
 
@@ -1906,7 +1906,7 @@ def test_compat_args(self, ms_complete):
                 ):
                     compat.get_adjustments(entry, state_data)
             else:
-                with pytest.raises(CompatibilityError, match="there is a matching R2SCAN"):
+                with pytest.raises(CompatibilityError, match="there is a matching r2SCAN"):
                     assert not compat.get_adjustments(entry, state_data)
 
         for entry in ms_complete.scan_entries:
@@ -1952,7 +1952,7 @@ def test_multiple_matching_structures(self, mixing_scheme_no_compat, ms_complete
                 ):
                     mixing_scheme_no_compat.get_adjustments(entry, ms_complete_duplicate_structs.state_data)
             else:
-                with pytest.raises(CompatibilityError, match="there is a matching R2SCAN"):
+                with pytest.raises(CompatibilityError, match="there is a matching r2SCAN"):
                     mixing_scheme_no_compat.get_adjustments(entry, ms_complete_duplicate_structs.state_data)
 
         # process_entries should discard all GGA entries and return all R2SCAN
@@ -1996,10 +1996,10 @@ def test_alternate_structure_matcher(self, ms_complete):
                 ):
                     compat.get_adjustments(entry, ms_complete.state_data)
             else:
-                with pytest.raises(CompatibilityError, match="there is a matching R2SCAN"):
+                with pytest.raises(CompatibilityError, match="there is a matching r2SCAN"):
                     compat.get_adjustments(entry, ms_complete.state_data)
 
-        # process_entries should discard all GGA entries except gga-6 and return all R2SCAN
+        # process_entries should discard all GGA entries except gga-6 and return all r2SCAN
         # entries unmodified. gga-6 should be corrected to the R2SCAN hull
         entries = compat.process_entries(ms_complete.all_entries)
         assert len(entries) == 8
@@ -2014,6 +2014,14 @@ def test_processing_entries_inplace(self):
         MaterialsProjectDFTMixingScheme().process_entries(entries, inplace=False)
         assert all(e.correction == e_copy.correction for e, e_copy in zip(entries, entries_copy, strict=True))
 
+    def test_run_type_variations(self):
+        """Test that entries with run_type 'r2SCAN' would not be ignored in mixing scheme correction."""
+        scheme = MaterialsProjectDFTMixingScheme(run_type_1="GGA(+U)", run_type_2="r2SCAN")
+        assert scheme.valid_rtypes_2 == ["r2SCAN", "R2SCAN"]
+        """Test that 'r2SCAN' as an input run_type is not case-sensitive."""
+        scheme = MaterialsProjectDFTMixingScheme(run_type_1="GGA(+U)", run_type_2="R2SCAN")
+        assert scheme.valid_rtypes_2 == ["r2SCAN", "R2SCAN"]
+
     def test_check_potcar(self, ms_complete):
         """Entries with invalid or missing POTCAR raise error by default but should be ignored if
         check_potcar=False in MaterialsProjectDFTMixingScheme.
@@ -2058,7 +2066,7 @@ def test_state_complete_entries(self, mixing_scheme_no_compat, ms_complete):
                 ):
                     mixing_scheme_no_compat.get_adjustments(entry, ms_complete.state_data)
             else:
-                with pytest.raises(CompatibilityError, match="there is a matching R2SCAN"):
+                with pytest.raises(CompatibilityError, match="there is a matching r2SCAN"):
                     mixing_scheme_no_compat.get_adjustments(entry, ms_complete.state_data)
 
         # process_entries should discard all GGA entries and return all R2SCAN
@@ -2153,7 +2161,7 @@ def test_state_gga_1_scan_plus_novel(self, mixing_scheme_no_compat, ms_gga_1_sca
                 assert mixing_scheme_no_compat.get_adjustments(entry, ms_gga_1_scan_novel.state_data) == []
 
         for entry in ms_gga_1_scan_novel.scan_entries:
-            with pytest.raises(CompatibilityError, match="no R2SCAN ground states at this composition"):
+            with pytest.raises(CompatibilityError, match="no r2SCAN ground states at this composition"):
                 mixing_scheme_no_compat.get_adjustments(entry, ms_gga_1_scan_novel.state_data)
 
         entries = mixing_scheme_no_compat.process_entries(ms_gga_1_scan_novel.all_entries)
@@ -2190,7 +2198,7 @@ def test_state_gga_2_scan_same(self, mixing_scheme_no_compat, ms_gga_2_scan_same
                 ):
                     mixing_scheme_no_compat.get_adjustments(entry, ms_gga_2_scan_same.state_data)
             elif entry.entry_id == "gga-6":
-                with pytest.raises(CompatibilityError, match="there is a matching R2SCAN"):
+                with pytest.raises(CompatibilityError, match="there is a matching r2SCAN"):
                     mixing_scheme_no_compat.get_adjustments(entry, ms_gga_2_scan_same.state_data)
             else:
                 assert mixing_scheme_no_compat.get_adjustments(entry, ms_gga_2_scan_same.state_data) == []
@@ -2239,7 +2247,7 @@ def test_state_gga_2_scan_diff_match(self, mixing_scheme_no_compat, ms_gga_2_sca
                 ):
                     mixing_scheme_no_compat.get_adjustments(entry, ms_gga_2_scan_diff_match.state_data)
             elif entry.entry_id == "gga-7":
-                with pytest.raises(CompatibilityError, match="there is a matching R2SCAN"):
+                with pytest.raises(CompatibilityError, match="there is a matching r2SCAN"):
                     mixing_scheme_no_compat.get_adjustments(entry, ms_gga_2_scan_diff_match.state_data)
             else:
                 assert mixing_scheme_no_compat.get_adjustments(entry, ms_gga_2_scan_diff_match.state_data) == []
@@ -2273,7 +2281,7 @@ def test_state_gga_2_scan_diff_nomatch(self, mixing_scheme_no_compat, ms_gga_2_s
             if entry.entry_id == "r2scan-8":
                 # there is no matching GGA structure for r2scan-8, so there's no way
                 # to adjust its energy onto the GGA hull.
-                with pytest.raises(CompatibilityError, match="entry and no R2SCAN ground state"):
+                with pytest.raises(CompatibilityError, match="entry and no r2SCAN ground state"):
                     mixing_scheme_no_compat.get_adjustments(entry, ms_gga_2_scan_diff_no_match.state_data)
             elif entry.entry_id == "r2scan-4":
                 # r2scan-4 energy is -7 eV/atom. Needs to be adjusted to -6 eV/atom (3 atoms)
@@ -2366,7 +2374,7 @@ def test_state_all_gga_scan_gs(self, mixing_scheme_no_compat, ms_all_gga_scan_gs
 
         for entry in entries:
             if entry.parameters["run_type"] == "GGA":
-                assert "onto the R2SCAN hull" in entry.energy_adjustments[0].description
+                assert "onto the r2SCAN hull" in entry.energy_adjustments[0].description
                 assert_allclose(pd_mixed.get_e_above_hull(entry), gga_hull_e[entry.entry_id])
             else:
                 assert entry.correction == 0
@@ -2408,7 +2416,7 @@ def test_state_novel_scan_comp(self, mixing_scheme_no_compat, ms_all_gga_scan_gs
 
         for entry in entries:
             if entry.parameters["run_type"] == "GGA":
-                assert "onto the R2SCAN hull" in entry.energy_adjustments[0].description
+                assert "onto the r2SCAN hull" in entry.energy_adjustments[0].description
                 assert_allclose(pd_mixed.get_e_above_hull(entry), gga_hull_e[entry.entry_id])
             else:
                 assert entry.correction == 0