Update Q-Chem schema (#1393)

Use Emmet's Qchem TaskDoc. Closes #956.

Requires: materialsproject/emmet#929,
materialsproject/emmet#930

---------

Co-authored-by: deepsource-autofix[bot] <62050782+deepsource-autofix[bot]@users.noreply.github.com>

pyproject.toml

@@ -30,7 +30,7 @@ dependencies = [
     "ase>3.22.1", # for Atoms object and calculators
     "cclib>=1.8", # for I/O parsing of molecular DFT codes
     "custodian>=2023.6.5", # for automated error corrections
-    "emmet-core>=0.69.6", # for pre-made schemas
+    "emmet-core>=0.77.1", # for pre-made schemas
     "maggma>=0.57.0", # for database handling
     "monty>=2023.9.25", # miscellaneous Python utilities
     "numpy>=1.25.0", # for array handling

src/quacc/calculators/qchem/io.py

@@ -6,10 +6,10 @@
 from typing import TYPE_CHECKING
 
 from ase import units
+from emmet.core.qc_tasks import TaskDoc
 from emmet.core.tasks import _parse_custodian
 from pymatgen.io.qchem.inputs import QCInput
 from pymatgen.io.qchem.outputs import (
-    QCOutput,
     gradient_parser,
     hessian_parser,
     orbital_coeffs_parser,
@@ -68,24 +68,14 @@ def read_qchem(directory: Path | str = ".") -> tuple[Results, list[float]]:
     """
     directory = Path(directory)
 
-    qc_input = QCInput.from_file(directory / "mol.qin")
-    qc_output = QCOutput(directory / "mol.qout").data
+    task_doc = TaskDoc.from_directory(directory).model_dump()
 
     results: Results = {
-        "energy": qc_output["final_energy"] * units.Hartree,
-        "qc_output": qc_output,
-        "qc_input": qc_input.as_dict(),
+        "energy": task_doc["output"]["final_energy"] * units.Hartree,
+        "taskdoc": task_doc,
         "custodian": _parse_custodian(directory),
     }
 
-    # Parse thermo properties
-    if "total_enthalpy" in qc_output:
-        results["enthalpy"] = qc_output["total_enthalpy"] * (units.kcal / units.mol)
-    if "total_entropy" in qc_output:
-        results["entropy"] = qc_output["total_entropy"] * (
-            0.001 * units.kcal / units.mol
-        )
-
     # Read the gradient scratch file in 8 byte chunks
     grad_scratch = directory / "131.0"
     if grad_scratch.exists() and grad_scratch.stat().st_size > 0:
@@ -96,9 +86,7 @@ def read_qchem(directory: Path | str = ".") -> tuple[Results, list[float]]:
     # Read Hessian scratch file in 8 byte chunks
     hessian_scratch = directory / "132.0"
     if hessian_scratch.exists() and hessian_scratch.stat().st_size > 0:
-        reshaped_hess = hessian_parser(
-            hessian_scratch, n_atoms=len(qc_output["species"])
-        )
+        reshaped_hess = hessian_parser(hessian_scratch, n_atoms=task_doc["natoms"])
         results["hessian"] = reshaped_hess * (units.Hartree / units.Bohr**2)
 
     # Read orbital coefficients scratch file in 8 byte chunks

src/quacc/calculators/qchem/qchem.py

@@ -22,14 +22,7 @@ class Results(TypedDict, total=False):
         energy: float  # electronic energy in eV
         forces: NDArray  # forces in eV/A
         hessian: NDArray  # Hessian in eV/A^2/amu
-        enthalpy: float  # total enthalpy in eV
-        entropy: float  # total entropy in eV/K
-        qc_output: dict[
-            str, Any
-        ]  # Output from `pymatgen.io.qchem.outputs.QCOutput.data`
-        qc_input: dict[
-            str, Any
-        ]  # Input from `pymatgen.io.qchem.inputs.QCInput.as_dict()`
+        taskdoc: dict[str, Any]  # Output from `emmet.core.qc_tasks.TaskDoc`
         custodian: dict[str, Any]  # custodian.json file metadata
 
 
@@ -40,10 +33,7 @@ class QChem(FileIOCalculator):
         "energy",
         "forces",
         "hessian",
-        "enthalpy",
-        "entropy",
-        "qc_output",
-        "qc_input",
+        "taskdoc",
         "custodian",
     ]
     results: ClassVar[Results] = {}

tests/core/calculators/qchem/test_qchem.py

@@ -9,6 +9,10 @@
 
 from quacc.calculators.qchem import QChem
 
+try:
+    import openbabel as ob
+except ImportError:
+    ob = None
 FILE_DIR = Path(__file__).parent
 
 
@@ -191,6 +195,7 @@ def test_qchem_write_input_freq(tmp_path, monkeypatch, test_atoms):
     assert qcinp.as_dict() == ref_qcinp.as_dict()
 
 
+@pytest.mark.skipif(ob is None, reason="openbabel needed")
 def test_qchem_read_results_basic_and_write_53(tmp_path, monkeypatch, test_atoms):
     calc = QChem(
         test_atoms,
@@ -217,6 +222,7 @@ def test_qchem_read_results_basic_and_write_53(tmp_path, monkeypatch, test_atoms
     assert qcinp.rem.get("scf_guess") == "read"
 
 
+@pytest.mark.skipif(ob is None, reason="openbabel needed")
 def test_qchem_read_results_intermediate(tmp_path, monkeypatch, test_atoms):
     monkeypatch.chdir(tmp_path)
     calc = QChem(test_atoms)
@@ -228,6 +234,7 @@ def test_qchem_read_results_intermediate(tmp_path, monkeypatch, test_atoms):
     assert calc.prev_orbital_coeffs is not None
 
 
+@pytest.mark.skipif(ob is None, reason="openbabel needed")
 def test_qchem_read_results_advanced(tmp_path, monkeypatch, test_atoms):
     monkeypatch.chdir(tmp_path)
     calc = QChem(test_atoms)
@@ -240,7 +247,9 @@ def test_qchem_read_results_advanced(tmp_path, monkeypatch, test_atoms):
     assert calc.results.get("hessian") is None
 
 
+@pytest.mark.skipif(ob is None, reason="openbabel needed")
 def test_qchem_read_results_freq(tmp_path, monkeypatch, test_atoms):
+    monkeypatch.chdir(tmp_path)
     calc = QChem(test_atoms, job_type="freq")
     monkeypatch.chdir(FILE_DIR / "examples" / "freq")
     calc.read_results()
@@ -250,8 +259,8 @@ def test_qchem_read_results_freq(tmp_path, monkeypatch, test_atoms):
     assert calc.prev_orbital_coeffs is not None
     assert len(calc.results["hessian"]) == 42
     assert len(calc.results["hessian"][0]) == 42
-    assert calc.results["qc_output"]["frequencies"][0] == -340.2
-    assert len(calc.results["qc_output"]["frequencies"]) == 36
-    assert len(calc.results["qc_output"]["frequency_mode_vectors"]) == 36
-    assert len(calc.results["qc_output"]["frequency_mode_vectors"][0]) == 14
-    assert len(calc.results["qc_output"]["frequency_mode_vectors"][0][0]) == 3
+    assert calc.results["taskdoc"]["output"]["frequencies"][0] == -340.2
+    assert len(calc.results["taskdoc"]["output"]["frequencies"]) == 36
+    assert len(calc.results["taskdoc"]["output"]["frequency_modes"]) == 36
+    assert len(calc.results["taskdoc"]["output"]["frequency_modes"][0]) == 14
+    assert len(calc.results["taskdoc"]["output"]["frequency_modes"][0][0]) == 3

tests/core/recipes/qchem_recipes/test_qchem_recipes.py

@@ -130,7 +130,7 @@ def test_static_job_v1(monkeypatch, tmp_path, test_atoms):
     qcin = QCInput.from_file("mol.qin.gz")
     ref_qcin = QCInput.from_file(str(QCHEM_DIR / "mol.qin.basic"))
     qcinput_nearly_equal(qcin, ref_qcin)
-    qcinput_nearly_equal(ref_qcin, QCInput.from_dict(output["results"]["qc_input"]))
+    assert output["results"]["taskdoc"]
 
 
 def test_static_job_v2(monkeypatch, tmp_path, test_atoms):
@@ -160,7 +160,7 @@ def test_static_job_v2(monkeypatch, tmp_path, test_atoms):
     qcin = QCInput.from_file("mol.qin.gz")
     ref_qcin = QCInput.from_file(str(QCHEM_DIR / "mol.qin.intermediate"))
     qcinput_nearly_equal(qcin, ref_qcin)
-    qcinput_nearly_equal(ref_qcin, QCInput.from_dict(output["results"]["qc_input"]))
+    assert output["results"]["taskdoc"]
 
 
 def test_static_job_v3(monkeypatch, tmp_path, test_atoms):
@@ -187,7 +187,7 @@ def test_static_job_v3(monkeypatch, tmp_path, test_atoms):
     qcin = QCInput.from_file("mol.qin.gz")
     ref_qcin = QCInput.from_file(str(QCHEM_DIR / "mol.qin.alternate"))
     qcinput_nearly_equal(qcin, ref_qcin)
-    qcinput_nearly_equal(ref_qcin, QCInput.from_dict(output["results"]["qc_input"]))
+    assert output["results"]["taskdoc"]
 
 
 def test_static_job_v4(monkeypatch, tmp_path, os_atoms):
@@ -236,7 +236,7 @@ def test_relax_job_v1(monkeypatch, tmp_path, test_atoms):
     qcin = QCInput.from_file("mol.qin.gz")
     ref_qcin = QCInput.from_file(str(QCHEM_DIR / "mol.qin.basic.sella_opt_iter1"))
     qcinput_nearly_equal(qcin, ref_qcin)
-    assert len(output["results"]["qc_input"]) > 1
+    assert len(output["results"]["taskdoc"]["input"]) > 1
 
 
 def test_relax_job_v2(monkeypatch, tmp_path, test_atoms):
@@ -326,16 +326,7 @@ def test_freq_job_v1(monkeypatch, tmp_path, test_atoms):
     assert output["parameters"]["spin_multiplicity"] == 2
     assert output["results"]["energy"] == pytest.approx(-605.6859554019 * units.Hartree)
     assert output["results"].get("hessian") is not None
-    assert output["results"]["enthalpy"] == pytest.approx(
-        output["results"]["qc_output"]["total_enthalpy"] * (units.kcal / units.mol)
-    )
-    assert (
-        output["results"]["entropy"]
-        == output["results"]["qc_output"]["total_entropy"]
-        * 0.001
-        * units.kcal
-        / units.mol
-    )
+    assert output["results"]["taskdoc"]["output"]["enthalpy"] is not None
 
 
 def test_ts_job_v1(monkeypatch, tmp_path, test_atoms):

tests/requirements.txt

@@ -1,12 +1,12 @@
 git+https://gitlab.com/ase/ase.git
 cclib==1.8
 custodian==2023.10.9
-emmet-core==0.77.0
+emmet-core==0.77.1
 maggma==0.62.1
 monty==2024.2.2
 numpy==1.26.3
 psutil==5.9.8
 pydantic==2.6.1
 pydantic-settings==2.1.0
 pymatgen==2023.12.18
-typer[all]==0.9.0
+typer[all]==0.9.0