Csse layout 536 input_data

docs/changelog.rst

@@ -36,6 +36,16 @@ New Features
 
 Enhancements
 ++++++++++++
+- (536b) ``v1.AtomicResult.convert_v`` learned a ``external_input_data`` option to inject that field (if known) rather than using incomplete reconstruction from the v1 Result. may not be the final sol'n.
+- (536b) ``v2.FailedOperation`` gained schema_name and schema_version=2.
+- (536b) ``v2.AtomicResult`` no longer inherits from ``v2.AtomicInput``. It gained a ``input_data`` field for the corresponding ``AtomicInput`` and independent ``id`` and ``molecule`` fields (the latter being equivalvent to ``v1.AtomicResult.molecule`` with the frame of the results; ``v2.AtomicResult.input_data.molecule`` is new, preserving the input frame). Gained independent ``extras``
+- (536b) Both v1/v2 ``AtomicResult.convert_v()`` learned to handle the new ``input_data`` layout.
+- (:pr:`357`, :issue:`536`) ``v2.AtomicResult``, ``v2.OptimizationResult``, and ``v2.TorsionDriveResult`` have the ``success`` field enforced to ``True``. Previously it could be set T/F. Now validation errors if not T. Likewise ``v2.FailedOperation.success`` is enforced to ``False``.
+- (:pr:`357`, :issue:`536`) ``v2.AtomicResult``, ``v2.OptimizationResult``, and ``v2.TorsionDriveResult`` have the ``error`` field removed. This isn't used now that ``success=True`` and failure should be routed to ``FailedOperation``.
+- (:pr:`357`) ``v1.Molecule`` had its schema_version changed to a Literal[2] (remember Mol is one-ahead of general numbering scheme) so new instances will be 2 even if another value is passed in. Ditto ``v2.BasisSet.schema_version=2``. Ditto ``v1.BasisSet.schema_version=1`` Ditto ``v1.QCInputSpecification.schema_version=1`` and ``v1.OptimizationSpecification.schema_version=1``.
+- (:pr:`357`) ``v2.AtomicResultProperties``, ``v2.QCInputSpecification``, ``v2.OptimizationSpecification`` lost its schema_version until we see if its really needed.
+- (:pr:`357`) ``v2.OptimizationSpecification`` gained extras field
+- (:pr:`357`) ``v1.FailedOperation.extras`` and ``v2.FailedOperation.extras`` default changed from None to {}
 * Fix a lot of warnings originating in this project.
 * `Molecule.extras` now defaults to `{}` rather than None in both v1 and v2. Input None converts to {} upon instantiation.
 * ``v2.FailedOperation`` field `id` is becoming `Optional[str]` instead of plain `str` so that the default validates.

qcelemental/models/types.py

@@ -10,6 +10,6 @@
     DeprecationWarning,
 )
 
-# Array = qcelemental.models.v1.Array
+Array = qcelemental.models.v1.Array
 # ArrayMeta = qcelemental.models.v1.ArrayMeta
 # TypedArray = qcelemental.models.v1.TypedArray

qcelemental/models/v1/__init__.py

@@ -1,4 +1,4 @@
-from . import types
+from . import types  # ever used?
 from .align import AlignmentMill
 from .basemodels import AutodocBaseSettings  # remove when QCFractal merges `next`
 from .basemodels import ProtoModel
@@ -8,6 +8,7 @@
 from .procedures import Optimization  # scheduled for removal
 from .procedures import (
     OptimizationInput,
+    OptimizationProtocols,
     OptimizationResult,
     OptimizationSpecification,
     QCInputSpecification,
@@ -18,7 +19,8 @@
 from .results import Result  # scheduled for removal
 from .results import ResultInput  # scheduled for removal
 from .results import ResultProperties  # scheduled for removal
-from .results import AtomicInput, AtomicResult, AtomicResultProperties, AtomicResultProtocols
+from .results import AtomicInput, AtomicResult, AtomicResultProperties, AtomicResultProtocols, WavefunctionProperties
+from .types import Array
 
 
 def qcschema_models():

qcelemental/models/v1/basis.py

@@ -1,6 +1,12 @@
 from enum import Enum
 from typing import Dict, List, Optional
 
+try:
+    from typing import Literal
+except ImportError:
+    # remove when minimum py38
+    from typing_extensions import Literal
+
 from pydantic.v1 import ConstrainedInt, Field, constr, validator
 
 from ...exceptions import ValidationError
@@ -155,7 +161,7 @@ class BasisSet(ProtoModel):
         "qcschema_basis",
         description=(f"The QCSchema specification to which this model conforms. Explicitly fixed as qcschema_basis."),
     )
-    schema_version: int = Field(  # type: ignore
+    schema_version: Literal[1] = Field(  # type: ignore
         1,
         description="The version number of :attr:`~qcelemental.models.BasisSet.schema_name` to which this model conforms.",
     )
@@ -175,6 +181,10 @@ class Config(ProtoModel.Config):
         def schema_extra(schema, model):
             schema["$schema"] = qcschema_draft
 
+    @validator("schema_version", pre=True)
+    def _version_stamp(cls, v):
+        return 1
+
     @validator("atom_map")
     def _check_atom_map(cls, v, values):
         sv = set(v)

qcelemental/models/v1/common_models.py

@@ -120,7 +120,7 @@ class FailedOperation(ProtoModel):
         ":class:`ComputeError` for more details.",
     )
     extras: Optional[Dict[str, Any]] = Field(  # type: ignore
-        None,
+        {},
         description="Additional information to bundle with the failed operation. Details which pertain specifically "
         "to a thrown error should be contained in the `error` field. See :class:`ComputeError` for details.",
     )
@@ -139,6 +139,7 @@ def convert_v(
 
         dself = self.dict()
         if version == 2:
+            # TODO if FailedOp gets a schema_version, add a validator
             self_vN = qcel.models.v2.FailedOperation(**dself)
 
         return self_vN

qcelemental/models/v1/molecule.py

@@ -9,6 +9,12 @@
 from pathlib import Path
 from typing import TYPE_CHECKING, Any, Dict, Iterable, List, Optional, Tuple, Union, cast
 
+try:
+    from typing import Literal
+except ImportError:
+    # remove when minimum py38
+    from typing_extensions import Literal
+
 import numpy as np
 from pydantic.v1 import ConstrainedFloat, ConstrainedInt, Field, constr, validator
 
@@ -119,7 +125,7 @@ class Molecule(ProtoModel):
             f"The QCSchema specification to which this model conforms. Explicitly fixed as {qcschema_molecule_default}."
         ),
     )
-    schema_version: int = Field(  # type: ignore
+    schema_version: Literal[2] = Field(  # type: ignore
         2,
         description="The version number of :attr:`~qcelemental.models.Molecule.schema_name` to which this model conforms.",
     )
@@ -370,6 +376,12 @@ def __init__(self, orient: bool = False, validate: Optional[bool] = None, **kwar
         elif validate or geometry_prep:
             values["geometry"] = float_prep(values["geometry"], geometry_noise)
 
+    @validator("schema_version", pre=True)
+    def _version_stamp(cls, v):
+        # seemingly unneeded, this lets conver_v re-label the model w/o discarding model and
+        #   submodel version fields first.
+        return 2
+
     @validator("geometry")
     def _must_be_3n(cls, v, values, **kwargs):
         n = len(values["symbols"])

qcelemental/models/v1/procedures.py

@@ -60,7 +60,7 @@ class QCInputSpecification(ProtoModel):
     """
 
     schema_name: constr(strip_whitespace=True, regex=qcschema_input_default) = qcschema_input_default  # type: ignore
-    schema_version: int = 1  # TODO
+    schema_version: Literal[1] = 1
 
     driver: DriverEnum = Field(DriverEnum.gradient, description=str(DriverEnum.__doc__))
     model: Model = Field(..., description=str(Model.__doc__))
@@ -71,6 +71,10 @@ class QCInputSpecification(ProtoModel):
         description="Additional information to bundle with the computation. Use for schema development and scratch space.",
     )
 
+    @validator("schema_version", pre=True)
+    def _version_stamp(cls, v):
+        return 1
+
 
 class OptimizationInput(ProtoModel):
     id: Optional[str] = None
@@ -110,6 +114,7 @@ def convert_v(
 
         dself = self.dict()
         if version == 2:
+            dself["input_specification"].pop("schema_version", None)
             self_vN = qcel.models.v2.OptimizationInput(**dself)
 
         return self_vN
@@ -171,8 +176,15 @@ def convert_v(
         if check_convertible_version(version, error="OptimizationResult") == "self":
             return self
 
+        trajectory_class = self.trajectory[0].__class__
         dself = self.dict()
         if version == 2:
+            # remove harmless empty error field that v2 won't accept. if populated, pydantic will catch it.
+            dself.pop("error", None)
+
+            dself["trajectory"] = [trajectory_class(**atres).convert_v(version) for atres in dself["trajectory"]]
+            dself["input_specification"].pop("schema_version", None)
+
             self_vN = qcel.models.v2.OptimizationResult(**dself)
 
         return self_vN
@@ -189,12 +201,16 @@ class OptimizationSpecification(ProtoModel):
     """
 
     schema_name: constr(strip_whitespace=True, regex="qcschema_optimization_specification") = "qcschema_optimization_specification"  # type: ignore
-    schema_version: int = 1  # TODO
+    schema_version: Literal[1] = 1
 
     procedure: str = Field(..., description="Optimization procedure to run the optimization with.")
     keywords: Dict[str, Any] = Field({}, description="The optimization specific keywords to be used.")
     protocols: OptimizationProtocols = Field(OptimizationProtocols(), description=str(OptimizationProtocols.__doc__))
 
+    @validator("schema_version", pre=True)
+    def _version_stamp(cls, v):
+        return 1
+
     @validator("procedure")
     def _check_procedure(cls, v):
         return v.lower()
@@ -282,6 +298,9 @@ def convert_v(
 
         dself = self.dict()
         if version == 2:
+            dself["input_specification"].pop("schema_version", None)
+            dself["optimization_spec"].pop("schema_version", None)
+
             self_vN = qcel.models.v2.TorsionDriveInput(**dself)
 
         return self_vN
@@ -332,8 +351,19 @@ def convert_v(
         if check_convertible_version(version, error="TorsionDriveResult") == "self":
             return self
 
+        opthist_class = next(iter(self.optimization_history.values()))[0].__class__
         dself = self.dict()
         if version == 2:
+            # remove harmless empty error field that v2 won't accept. if populated, pydantic will catch it.
+            dself.pop("error", None)
+
+            dself["input_specification"].pop("schema_version", None)
+            dself["optimization_spec"].pop("schema_version", None)
+            dself["optimization_history"] = {
+                k: [opthist_class(**res).convert_v(version) for res in lst]
+                for k, lst in dself["optimization_history"].items()
+            }
+
             self_vN = qcel.models.v2.TorsionDriveResult(**dself)
 
         return self_vN

qcelemental/models/v1/results.py

@@ -797,16 +797,59 @@ def _native_file_protocol(cls, value, values):
         return ret
 
     def convert_v(
-        self, version: int
+        self,
+        version: int,
+        *,
+        external_input_data: Optional[Any] = None,
     ) -> Union["qcelemental.models.v1.AtomicResult", "qcelemental.models.v2.AtomicResult"]:
-        """Convert to instance of particular QCSchema version."""
+        """Convert to instance of particular QCSchema version.
+
+        Parameters
+        ----------
+        version
+            The version to convert to.
+        external_input_data
+            Since self contains data merged from input, this allows passing in the original input, particularly for `molecule` and `extras` fields.
+            Can be model or dictionary and should be *already* converted to the desired version.
+            Replaces ``input_data`` field entirely (not merges with extracts from self) and w/o consistency checking.
+
+        Returns
+        -------
+        AtomicResult
+            Returns self (not a copy) if ``version`` already satisfied.
+            Returns a new AtomicResult of ``version`` otherwise.
+
+        """
         import qcelemental as qcel
 
         if check_convertible_version(version, error="AtomicResult") == "self":
             return self
 
         dself = self.dict()
         if version == 2:
+            # remove harmless empty error field that v2 won't accept. if populated, pydantic will catch it.
+            dself.pop("error", None)
+
+            input_data = {
+                k: dself.pop(k) for k in list(dself.keys()) if k in ["driver", "keywords", "model", "protocols"]
+            }
+            input_data["molecule"] = dself["molecule"]  # duplicate since input mol has been overwritten
+            # any input provenance has been overwritten
+            input_data["extras"] = {
+                k: dself["extras"].pop(k) for k in list(dself["extras"].keys()) if k in []
+            }  # sep any merged extras
+            if external_input_data:
+                # Note: overwriting with external, not updating. reconsider?
+                dself["input_data"] = external_input_data
+                in_extras = (
+                    external_input_data.get("extras", {})
+                    if isinstance(external_input_data, dict)
+                    else external_input_data.extras
+                )
+                dself["extras"] = {k: v for k, v in dself["extras"].items() if (k, v) not in in_extras.items()}
+            else:
+                dself["input_data"] = input_data
+
             self_vN = qcel.models.v2.AtomicResult(**dself)
 
         return self_vN

qcelemental/models/v2/__init__.py

@@ -4,8 +4,17 @@
 from .basis import BasisSet
 from .common_models import ComputeError, DriverEnum, FailedOperation, Model, Provenance
 from .molecule import Molecule
-from .procedures import OptimizationInput, OptimizationResult, TorsionDriveInput, TorsionDriveResult
-from .results import AtomicInput, AtomicResult, AtomicResultProperties, AtomicResultProtocols
+from .procedures import (
+    OptimizationInput,
+    OptimizationProtocols,
+    OptimizationResult,
+    OptimizationSpecification,
+    QCInputSpecification,
+    TDKeywords,
+    TorsionDriveInput,
+    TorsionDriveResult,
+)
+from .results import AtomicInput, AtomicResult, AtomicResultProperties, AtomicResultProtocols, WavefunctionProperties
 
 
 def qcschema_models():

qcelemental/models/v2/basis.py

@@ -1,6 +1,12 @@
 from enum import Enum
 from typing import Dict, List, Optional
 
+try:
+    from typing import Literal
+except ImportError:
+    # remove when minimum py38
+    from typing_extensions import Literal
+
 from pydantic import Field, constr, field_validator
 from typing_extensions import Annotated
 
@@ -171,7 +177,7 @@ class BasisSet(ProtoModel):
         "qcschema_basis",
         description=f"The QCSchema specification to which this model conforms. Explicitly fixed as qcschema_basis.",
     )
-    schema_version: int = Field(  # type: ignore
+    schema_version: Literal[2] = Field(  # type: ignore
         2,
         description="The version number of :attr:`~qcelemental.models.BasisSet.schema_name` "
         "to which this model conforms.",
@@ -245,3 +251,7 @@ def _calculate_nbf(cls, atom_map, center_data) -> int:
             ret += center_count[center]
 
         return ret
+
+    @field_validator("schema_version", mode="before")
+    def _version_stamp(cls, v):
+        return 2