Merge pull request #411 from arosen93/cclib-fix

Fix encoding of input `Molecule` coordinates in cclib `TaskDocument`

src/atomate2/common/schemas/cclib.py

@@ -5,7 +5,6 @@
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Tuple, Type, TypeVar, Union
 
-import numpy as np
 from emmet.core.structure import MoleculeMetadata
 from monty.dev import requires
 from monty.json import jsanitize
@@ -163,9 +162,9 @@ def from_logfile(
             cclib_obj.metadata.get("coord_type", None) == "xyz"
             and cclib_obj.metadata.get("coords", None) is not None
         ):
-            coords_obj = np.array(cclib_obj.metadata["coords"])
-            input_species = [Element(e) for e in coords_obj[:, 0]]
-            input_coords = coords_obj[:, 1:].tolist()
+            coords_obj = cclib_obj.metadata["coords"]
+            input_species = [Element(row[0]) for row in coords_obj]
+            input_coords = [row[1:] for row in coords_obj]
             input_molecule = Molecule(
                 input_species,
                 input_coords,

tests/common/schemas/test_cclib.py

@@ -56,11 +56,13 @@ def test_cclib_taskdoc(test_dir):
     assert "Could not parse" in str(e.value)
 
     # Test a population analysis
-    doc = TaskDocument.from_logfile(p, ".out", analysis="MBO").dict()
+    doc = TaskDocument.from_logfile(p, "psi_test.out", analysis="MBO").dict()
     assert doc["attributes"]["mbo"] is not None
 
     # Let's try with two analysis (also check case-insensitivity)
-    doc = TaskDocument.from_logfile(p, ".out", analysis=["mbo", "density"]).dict()
+    doc = TaskDocument.from_logfile(
+        p, "psi_test.out", analysis=["mbo", "density"]
+    ).dict()
     assert doc["attributes"]["mbo"] is not None
     assert doc["attributes"]["density"] is not None
 
@@ -75,7 +77,7 @@ def test_cclib_taskdoc(test_dir):
         p / "psi_test.cube", "wb"
     ) as f_out:
         shutil.copyfileobj(f_in, f_out)
-    doc = TaskDocument.from_logfile(p, ".out", analysis=["Bader"]).dict()
+    doc = TaskDocument.from_logfile(p, "psi_test.out", analysis=["Bader"]).dict()
     os.remove(p / "psi_test.cube")
     assert doc["attributes"]["bader"] is not None
 
@@ -89,6 +91,10 @@ def test_cclib_taskdoc(test_dir):
     doc = TaskDocument.from_logfile(p, ".log", additional_fields={"test": "hi"})
     assert doc.dict()["test"] == "hi"
 
+    # Test that the dict printing works
+    task = TaskDocument.from_logfile(p, "orca.out")
+    task.dict()
+
     # test document can be jsanitized
     d = jsanitize(doc, enum_values=True)