Csse layout reorg rb2025 (#468)

* first tidying

* fix pkg_resources error?

* misc

* Mol v3

* trajectory workings

* fix var

* post-Christmas opt/td

* use qcel next2025

.github/workflows/CI.yml

@@ -163,12 +163,12 @@ jobs:
         # note: any activate/deactivate use the conda cmd. other cmds use mamba cmd.
 
     - name: Special Config - NWChem
-      if: "(matrix.cfg.label == 'NWChem70')"
+      if: matrix.cfg.label == 'NWChem70'
       run: |
         sudo apt-get -y install nwchem
 
     - name: Special Config - QCore
-      if: "(matrix.cfg.label == 'QCore')"
+      if: matrix.cfg.label == 'QCore'
       run: |
         qcore --accept-license
 
@@ -180,18 +180,18 @@ jobs:
       #if: false
       run: |
         conda remove qcelemental --force
-        python -m pip install 'git+https://github.com/loriab/QCElemental.git@csse_layout_536d' --no-deps
+        python -m pip install 'git+https://github.com/loriab/QCElemental.git@csse_layout_536f_rb2025_take2' --no-deps
 
       # note: conda remove --force, not mamba remove --force b/c https://github.com/mamba-org/mamba/issues/412
       #       alt. is micromamba but not yet ready for setup-miniconda https://github.com/conda-incubator/setup-miniconda/issues/75
     - name: Special Config - QCEngine Dep
-      if: "(startsWith(matrix.cfg.label, 'Psi4')) || (matrix.cfg.label == 'ADCC') || (matrix.cfg.label == 'optimization-dispersion')"
+      if: (startsWith(matrix.cfg.label, 'Psi4')) || (matrix.cfg.label == 'ADCC') || (matrix.cfg.label == 'optimization-dispersion')
       run: |
         conda remove qcengine --force
 
       # QCEngine CI and Psi4 are circularly dependent, so a hack is in order
     - name: Special Config - Faux Pydantic Upgrade
-      if: "((matrix.cfg.label == 'Psi4-1.6') || (matrix.cfg.label == 'optimization-dispersion')) && (runner.os != 'Windows')"
+      if: ((matrix.cfg.label == 'Psi4-1.6') || (matrix.cfg.label == 'optimization-dispersion')) && (runner.os != 'Windows')
       run: |
         sed -i s/from\ pydantic\ /from\ pydantic.v1\ /g ${CONDA_PREFIX}/lib/python${{ matrix.cfg.python-version }}/site-packages/psi4/driver/*py
 
@@ -203,7 +203,7 @@ jobs:
         python -m pip install 'git+https://github.com/MolSSI/QCElemental.git@next2025' --no-deps
 
     - name: Special Config - Forced Interface Upgrade
-      if: "(matrix.cfg.label == 'Psi4-1.6')"
+      if: matrix.cfg.label == 'Psi4-1.6'
       run: |
         grep -r "local_options" ${CONDA_PREFIX}/lib/python${{ matrix.cfg.python-version }}/site-packages/psi4/driver/
         sed -i "s/local_options/task_config/g" ${CONDA_PREFIX}/lib/python${{ matrix.cfg.python-version }}/site-packages/psi4/driver/procrouting/*py
@@ -221,7 +221,7 @@ jobs:
         git describe
 
     - name: QCEngineRecords
-      if: "(matrix.cfg.label != 'Psi4-1.6')"
+      if: matrix.cfg.label != 'Psi4-1.6'
       run: |
         qcengine info
         export QCER_VER=`python -c "import qcengine.testing; print(qcengine.testing.QCENGINE_RECORDS_COMMIT)"`
@@ -269,7 +269,7 @@ jobs:
       #if: false
       run: |
         conda remove qcelemental --force
-        python -m pip install 'git+https://github.com/loriab/QCElemental.git@csse_layout_536d' --no-deps
+        python -m pip install 'git+https://github.com/loriab/QCElemental.git@csse_layout_536f_rb2025_take2' --no-deps
 
     - name: Environment Information
       run: |

docs/source/changelog.rst

@@ -70,10 +70,16 @@ Misc.
 
 MUST (Unmerged)
 +++++++++++++++
-torsiondrive rewritten in v2.
-berny harness rewritten in v2. optking and geometric natively speak v1, so adapted as well as can be.
-allow nwchemdriver w/o driver=energy. provenance now nwchemdriver not nwchemrelax
- Optking now fills in ``v2.OptimizationResult.stdout``. Through v2, once can alter gradient protocols in an optimization.
+- (:pr:`462`) 
+- (:pr:`462`) adapt harnesses for TD.initial_molecules -> TD.initial_molecule and TD.optimization_history -> TD.scan_results
+- (:pr:`462`) rdkit, store ``AtomicResult.properties.return_gradient`` and ``calcinfo_natom``. mrchem, store ``AtomicResult.properties.return_gradient``.
+- (:pr:`462`) lightly Adapt harnesses for Mol v3
+- (:pr:`462`) Use packaging instead of setuptools to provide version parsing
+- (:pr:`462`) torsiondrive now accepts protocols. use ``protocols={"scan_results": "all"}`` if going to be converted to v1.
+- (:pr:`461`) torsiondrive rewritten in v2.
+- (:pr:`461`) berny harness rewritten in v2. optking and geometric natively speak v1, so adapted as well as can be.
+- (:pr:`461`) allow nwchemdriver w/o driver=energy. provenance now nwchemdriver not nwchemrelax
+- (:pr:`461`)  Optking now fills in ``v2.OptimizationResult.stdout``. Through v2, once can alter gradient protocols in an optimization.
 - (:pr:`460`) integrate ``AtomicInput.specification`` into harnesses and show what new inputs look like in tests
 - (:pr:`459`) gcp, mp2d several got properties.return_energy, retunr_gradient
 - (:pr:`460`) If you're missing something from AtomicResult.extras, check AtomicResult.input_data.extras in case it was passed in on input

qcengine/procedures/berny.py

@@ -5,7 +5,7 @@
 from typing import Any, ClassVar, Dict, Union
 
 import numpy as np
-from qcelemental.models.v2 import FailedOperation, OptimizationInput, OptimizationResult
+from qcelemental.models.v2 import FailedOperation, Molecule, OptimizationInput, OptimizationResult
 from qcelemental.util import which_import
 
 import qcengine
@@ -85,11 +85,14 @@ def compute(
         except Exception:
             error = {"error_type": "unknown", "error_message": f"Berny error:\n{traceback.format_exc()}"}
         else:
+            final_molecule = trajectory[-1]["molecule"]
             output = {
                 "input_data": input_model,
-                "final_molecule": trajectory[-1]["molecule"],
+                "final_molecule": final_molecule,
                 "properties": {
+                    "nuclear_repulsion_energy": Molecule(**final_molecule).nuclear_repulsion_energy(),
                     "return_energy": trajectory[-1]["properties"]["return_energy"],
+                    "return_gradient": trajectory[-1]["properties"]["return_gradient"],
                     "optimization_iterations": len(trajectory),
                 },
                 "trajectory_results": trajectory,

qcengine/procedures/optking.py

@@ -1,14 +1,17 @@
 import logging
 import sys
 from io import StringIO
-from typing import Any, ClassVar, Dict, Union
+from typing import TYPE_CHECKING, Any, ClassVar, Dict, Union
 
 from qcelemental.models.v1 import OptimizationResult
 from qcelemental.models.v2 import OptimizationInput
 from qcelemental.util import safe_version, which_import
 
 from .model import ProcedureHarness
 
+if TYPE_CHECKING:
+    from ..config import TaskConfig
+
 
 class OptKingProcedure(ProcedureHarness):
 

qcengine/procedures/torsiondrive.py

@@ -54,8 +54,8 @@ def _compute(self, input_model: "TorsionDriveInput", config: "TaskConfig"):
         state = torsiondrive.td_api.create_initial_state(
             dihedrals=dihedrals,
             grid_spacing=grid_spacing,
-            elements=input_model.initial_molecules[0].symbols,
-            init_coords=[molecule.geometry.flatten().tolist() for molecule in input_model.initial_molecules],
+            elements=input_model.initial_molecule[0].symbols,
+            init_coords=[molecule.geometry.flatten().tolist() for molecule in input_model.initial_molecule],
             dihedral_ranges=dihedral_ranges,
             energy_upper_limit=energy_upper_limit,
             energy_decrease_thresh=energy_decrease_thresh,
@@ -128,7 +128,7 @@ def _compute(self, input_model: "TorsionDriveInput", config: "TaskConfig"):
             output_data["final_energies"][grid_point] = final_energy
             output_data["final_molecules"][grid_point] = final_molecule
 
-        output_data["optimization_history"] = optimization_results
+        output_data["scan_results"] = optimization_results
 
         if error is not None:
             output_data["error"] = error
@@ -185,7 +185,7 @@ def _spawn_optimization(
 
         from qcengine import compute
 
-        input_molecule = input_model.initial_molecules[0].copy(deep=True).dict()
+        input_molecule = input_model.initial_molecule[0].copy(deep=True).dict()
         input_molecule["geometry"] = np.array(job).reshape(len(input_molecule["symbols"]), 3)
         input_molecule = Molecule.from_data(input_molecule)
 

qcengine/programs/cfour/harvester.py

@@ -997,7 +997,7 @@ def harvest_outfile_pass(outtext):
         psivar_coord = Molecule(
             validate=False,
             **qcel.molparse.to_schema(
-                qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=2
+                qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=3
             ),
         )
 
@@ -1015,7 +1015,7 @@ def harvest_outfile_pass(outtext):
         psivar_coord = Molecule(
             validate=False,
             **qcel.molparse.to_schema(
-                qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=2
+                qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=3
             ),
         )
 
@@ -1034,7 +1034,7 @@ def harvest_outfile_pass(outtext):
         psivar_coord = Molecule(
             validate=False,
             **qcel.molparse.to_schema(
-                qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=2
+                qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=3
             ),
         )
 
@@ -1374,7 +1374,7 @@ def harvest_GRD(grd):
     mol = Molecule(
         validate=False,
         **qcel.molparse.to_schema(
-            qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=2
+            qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=3
         ),
     )
 

qcengine/programs/gamess/harvester.py

@@ -544,7 +544,7 @@ def harvest_outfile_pass(outtext):
             qcvar_coord = Molecule(
                 validate=False,
                 **qcel.molparse.to_schema(
-                    qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=2
+                    qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"], dtype=3
                 ),
             )
 

qcengine/programs/mace.py

@@ -136,7 +136,7 @@ def compute(self, input_data: "AtomicInput", config: "TaskConfig") -> Union["Ato
         ret_data["input_data"] = input_data
         ret_data["molecule"] = input_data.molecule
         ret_data["provenance"] = Provenance(creator="mace", version=mace.__version__, routine="mace")
-        ret_data["schema_name"] = "qcschema_atomic_output"
+        ret_data["schema_name"] = "qcschema_atomic_result"
         ret_data["success"] = True
 
         # Form up a dict first, then sent to BaseModel to avoid repeat kwargs which don't override each other

qcengine/programs/molpro.py

@@ -443,7 +443,7 @@ def parse_output(self, outfiles: Dict[str, str], input_model: "AtomicInput") ->
         # Final output_data assignments needed for the AtomicResult object
         output_data["properties"] = properties
         output_data["extras"] = extras
-        output_data["schema_name"] = "qcschema_atomic_output"
+        output_data["schema_name"] = "qcschema_atomic_result"
         output_data["stdout"] = outfiles["dispatch.out"]
         output_data["success"] = True
         output_data["provenance"] = input_model.provenance  # TODO better stamp?

qcengine/programs/mrchem.py

@@ -101,7 +101,7 @@ def compute(self, input_model: "AtomicInput", config: "TaskConfig") -> "AtomicRe
         input_data = copy.deepcopy(job_input["mrchem_json"])
 
         output_data = {
-            "schema_name": "qcschema_atomic_output",
+            "schema_name": "qcschema_atomic_result",
             "schema_version": 2,
             "input_data": input_model,
             "molecule": input_model.molecule,
@@ -170,9 +170,9 @@ def compute(self, input_model: "AtomicInput", config: "TaskConfig") -> "AtomicRe
                 if input_model.specification.driver == "energy":
                     output_data["return_result"] = mrchem_output["properties"]["scf_energy"]["E_tot"]
                 elif input_model.specification.driver == "gradient":
-                    output_data["return_result"] = mrchem_output["properties"]["geometric_derivative"]["geom-1"][
-                        "total"
-                    ]
+                    grad = mrchem_output["properties"]["geometric_derivative"]["geom-1"]["total"]
+                    output_data["return_result"] = grad
+                    output_data["properties"]["return_gradient"] = grad
                 elif input_model.specification.driver == "properties":
                     output_data["return_result"] = {
                         f"{ks[1]}": {f"{ks[2]}": _nested_get(mrchem_output, ks)} for ks in computed_rsp_props

qcengine/programs/nwchem/harvester.py

@@ -2,14 +2,17 @@
 import logging
 import re
 from decimal import Decimal
-from typing import Optional, Tuple
+from typing import TYPE_CHECKING, Optional, Tuple
 
 import numpy as np
 import qcelemental as qcel
 from qcelemental.molparse import regex
 
 from ..util import PreservingDict
 
+if TYPE_CHECKING:
+    from qcelemental.models.v2 import Molecule
+
 logger = logging.getLogger(__name__)
 
 
@@ -845,7 +848,7 @@ def harvest_outfile_pass(outtext):
                     validate=False,
                     **qcel.molparse.to_schema(
                         qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"],
-                        dtype=2,
+                        dtype=3,
                     ),
                 )
 
@@ -867,7 +870,7 @@ def harvest_outfile_pass(outtext):
                     validate=False,
                     **qcel.molparse.to_schema(
                         qcel.molparse.from_string(molxyz, dtype="xyz+", fix_com=True, fix_orientation=True)["qm"],
-                        dtype=2,
+                        dtype=3,
                     ),
                 )
 
@@ -921,6 +924,7 @@ def harvest_outfile_pass(outtext):
             logger.debug("matched total dipole")
 
             # UNIT = DEBYE(S)
+            d2au = Decimal(qcel.constants.conversion_factor("debye", "e * bohr"))
             psivar[f"CURRENT DIPOLE"] = d2au * np.array([mobj.group(7), mobj.group(8), mobj.group(9)])
             # total?
 

qcengine/programs/psi4.py

@@ -208,7 +208,7 @@ def compute(self, input_model: "AtomicInput", config: "TaskConfig") -> "AtomicRe
                     error_type = "execution_error"
 
                 # Reset the schema if required
-                output_data["schema_name"] = "qcschema_atomic_output"
+                output_data["schema_name"] = "qcschema_atomic_result"
                 output_data.pop("memory", None)
                 output_data.pop("nthreads", None)
                 output_data["stdout"] = output_data.pop("raw_output", None)

qcengine/programs/qcore.py

@@ -242,7 +242,7 @@ def parse_output(self, output: Dict[str, Any], input_model: "AtomicInput") -> "A
 
         output_data["properties"] = properties
 
-        output_data["schema_name"] = "qcschema_atomic_output"
+        output_data["schema_name"] = "qcschema_atomic_result"
         output_data["success"] = True
 
         return AtomicResult(**output_data)

qcengine/programs/qcvar_identities_resources.py

@@ -3,7 +3,7 @@
 from typing import Any, Dict, List
 
 import numpy as np
-from qcelemental.models.v2 import AtomicResultProperties
+from qcelemental.models.v2 import AtomicProperties
 
 from .util import PreservingDict
 
@@ -427,7 +427,7 @@ def build_out(rawvars: Dict[str, Any], verbose: int = 1) -> None:
 }
 
 
-def build_atomicproperties(qcvars: "PreservingDict") -> AtomicResultProperties:
+def build_atomicproperties(qcvars: "PreservingDict") -> AtomicProperties:
     """For results extracted from QC output in QCDB terminology, translate to QCSchema terminology.
 
     Parameters
@@ -437,13 +437,13 @@ def build_atomicproperties(qcvars: "PreservingDict") -> AtomicResultProperties:
 
     Returns
     -------
-    atprop : AtomicResultProperties
-        Object of calculation information in QCSchema AtomicResultProperties terminology.
+    atprop : AtomicProperties
+        Object of calculation information in QCSchema AtomicProperties terminology.
 
     """
     atprop = {}
     for pv, dpv in qcvars.items():
         if pv in qcvars_to_atomicproperties:
             atprop[qcvars_to_atomicproperties[pv]] = dpv
 
-    return AtomicResultProperties(**atprop)
+    return AtomicProperties(**atprop)

qcengine/programs/rdkit.py

@@ -126,13 +126,18 @@ def compute(self, input_data: "AtomicInput", config: "TaskConfig") -> "AtomicRes
 
         ff.Initialize()
 
-        ret_data["properties"] = {"return_energy": ff.CalcEnergy() * ureg.conversion_factor("kJ / mol", "hartree")}
+        ret_data["properties"] = {
+            "return_energy": ff.CalcEnergy() * ureg.conversion_factor("kJ / mol", "hartree"),
+            "calcinfo_natom": len(jmol.symbols),
+        }
+        if input_data.specification.driver == "gradient":
+            coef = ureg.conversion_factor("kJ / mol", "hartree") * ureg.conversion_factor("angstrom", "bohr")
+            ret_data["properties"]["return_gradient"] = [x * coef for x in ff.CalcGrad()]
 
         if input_data.specification.driver == "energy":
             ret_data["return_result"] = ret_data["properties"]["return_energy"]
         elif input_data.specification.driver == "gradient":
-            coef = ureg.conversion_factor("kJ / mol", "hartree") * ureg.conversion_factor("angstrom", "bohr")
-            ret_data["return_result"] = [x * coef for x in ff.CalcGrad()]
+            ret_data["return_result"] = ret_data["properties"]["return_gradient"]
         else:
             raise InputError(f"Driver {input_data.specification.driver} not implemented for RDKit.")
 

qcengine/programs/terachem.py

@@ -182,7 +182,7 @@ def parse_output(self, outfiles: Dict[str, str], input_model: "AtomicInput") ->
 
         output_data["properties"] = properties
 
-        output_data["schema_name"] = "qcschema_atomic_output"
+        output_data["schema_name"] = "qcschema_atomic_result"
         output_data["stdout"] = outfiles["tc.out"]
         # TODO Should only return True if TeraChem calculation terminated properly
         output_data["success"] = True