import inspect
from typing import (
TYPE_CHECKING,
AbstractSet,
Any,
Callable,
Iterator,
Mapping,
Optional,
Sequence,
Tuple,
TypeVar,
Union,
cast,
)
from typing_extensions import TypeAlias, get_args, get_origin
import dagster._check as check
from dagster._annotations import deprecated, deprecated_param, public
from dagster._config.config_schema import UserConfigSchema
from dagster._core.definitions.asset_check_result import AssetCheckResult
from dagster._core.definitions.dependency import NodeHandle, NodeInputHandle
from dagster._core.definitions.node_definition import NodeDefinition
from dagster._core.definitions.op_invocation import direct_invocation_result
from dagster._core.definitions.policy import RetryPolicy
from dagster._core.definitions.resource_requirement import (
InputManagerRequirement,
OpDefinitionResourceRequirement,
OutputManagerRequirement,
ResourceRequirement,
)
from dagster._core.definitions.utils import DEFAULT_IO_MANAGER_KEY
from dagster._core.errors import (
DagsterInvalidDefinitionError,
DagsterInvalidInvocationError,
DagsterInvariantViolationError,
)
from dagster._core.types.dagster_type import DagsterType, DagsterTypeKind
from dagster._utils import IHasInternalInit
from dagster._utils.warnings import normalize_renamed_param
from .definition_config_schema import (
IDefinitionConfigSchema,
convert_user_facing_definition_config_schema,
)
from .hook_definition import HookDefinition
from .inference import infer_output_props
from .input import In, InputDefinition
from .output import Out, OutputDefinition
from .result import MaterializeResult, ObserveResult
if TYPE_CHECKING:
from dagster._core.definitions.asset_layer import AssetLayer
from .composition import PendingNodeInvocation
from .decorators.op_decorator import DecoratedOpFunction
OpComputeFunction: TypeAlias = Callable[..., Any]
[docs]@deprecated_param(
param="version", breaking_version="2.0", additional_warn_text="Use `code_version` instead."
)
class OpDefinition(NodeDefinition, IHasInternalInit):
"""Defines an op, the functional unit of user-defined computation.
For more details on what a op is, refer to the
`Ops Overview <../../concepts/ops-jobs-graphs/ops>`_ .
End users should prefer the :func:`@op <op>` decorator. OpDefinition is generally intended to be
used by framework authors or for programatically generated ops.
Args:
name (str): Name of the op. Must be unique within any :py:class:`GraphDefinition` or
:py:class:`JobDefinition` that contains the op.
input_defs (List[InputDefinition]): Inputs of the op.
compute_fn (Callable): The core of the op, the function that performs the actual
computation. The signature of this function is determined by ``input_defs``, and
optionally, an injected first argument, ``context``, a collection of information
provided by the system.
This function will be coerced into a generator or an async generator, which must yield
one :py:class:`Output` for each of the op's ``output_defs``, and additionally may
yield other types of Dagster events, including :py:class:`AssetMaterialization` and
:py:class:`ExpectationResult`.
output_defs (List[OutputDefinition]): Outputs of the op.
config_schema (Optional[ConfigSchema): The schema for the config. If set, Dagster will check
that the config provided for the op matches this schema and will fail if it does not. If
not set, Dagster will accept any config provided for the op.
description (Optional[str]): Human-readable description of the op.
tags (Optional[Dict[str, Any]]): Arbitrary metadata for the op. Frameworks may
expect and require certain metadata to be attached to a op. Users should generally
not set metadata directly. Values that are not strings will be json encoded and must meet
the criteria that `json.loads(json.dumps(value)) == value`.
required_resource_keys (Optional[Set[str]]): Set of resources handles required by this op.
code_version (Optional[str]): (Experimental) Version of the code encapsulated by the op. If set,
this is used as a default code version for all outputs.
retry_policy (Optional[RetryPolicy]): The retry policy for this op.
Examples:
.. code-block:: python
def _add_one(_context, inputs):
yield Output(inputs["num"] + 1)
OpDefinition(
name="add_one",
ins={"num": In(int)},
outs={"result": Out(int)},
compute_fn=_add_one,
)
"""
_compute_fn: Union[Callable[..., Any], "DecoratedOpFunction"]
_config_schema: IDefinitionConfigSchema
_required_resource_keys: AbstractSet[str]
_version: Optional[str]
_retry_policy: Optional[RetryPolicy]
def __init__(
self,
compute_fn: Union[Callable[..., Any], "DecoratedOpFunction"],
name: str,
ins: Optional[Mapping[str, In]] = None,
outs: Optional[Mapping[str, Out]] = None,
description: Optional[str] = None,
config_schema: Optional[Union[UserConfigSchema, IDefinitionConfigSchema]] = None,
required_resource_keys: Optional[AbstractSet[str]] = None,
tags: Optional[Mapping[str, Any]] = None,
version: Optional[str] = None,
retry_policy: Optional[RetryPolicy] = None,
code_version: Optional[str] = None,
):
from .decorators.op_decorator import DecoratedOpFunction, resolve_checked_op_fn_inputs
ins = check.opt_mapping_param(ins, "ins")
input_defs = [
inp.to_definition(name) for name, inp in sorted(ins.items(), key=lambda inp: inp[0])
] # sort so that input definition order is deterministic
if isinstance(compute_fn, DecoratedOpFunction):
resolved_input_defs: Sequence[InputDefinition] = resolve_checked_op_fn_inputs(
decorator_name="@op",
fn_name=name,
compute_fn=cast(DecoratedOpFunction, compute_fn),
explicit_input_defs=input_defs,
exclude_nothing=True,
)
self._compute_fn = compute_fn
_validate_context_type_hint(self._compute_fn.decorated_fn)
else:
resolved_input_defs = input_defs
self._compute_fn = check.callable_param(compute_fn, "compute_fn")
_validate_context_type_hint(self._compute_fn)
code_version = normalize_renamed_param(
code_version,
"code_version",
version,
"version",
)
self._version = code_version
check.opt_mapping_param(outs, "outs")
output_defs = _resolve_output_defs_from_outs(
compute_fn=compute_fn, outs=outs, default_code_version=code_version
)
self._config_schema = convert_user_facing_definition_config_schema(config_schema)
self._required_resource_keys = frozenset(
check.opt_set_param(required_resource_keys, "required_resource_keys", of_type=str)
)
self._retry_policy = check.opt_inst_param(retry_policy, "retry_policy", RetryPolicy)
positional_inputs = (
self._compute_fn.positional_inputs()
if isinstance(self._compute_fn, DecoratedOpFunction)
else None
)
super(OpDefinition, self).__init__(
name=name,
input_defs=check.sequence_param(resolved_input_defs, "input_defs", InputDefinition),
output_defs=check.sequence_param(output_defs, "output_defs", OutputDefinition),
description=description,
tags=check.opt_mapping_param(tags, "tags", key_type=str),
positional_inputs=positional_inputs,
)
def dagster_internal_init(
*,
compute_fn: Union[Callable[..., Any], "DecoratedOpFunction"],
name: str,
ins: Optional[Mapping[str, In]],
outs: Optional[Mapping[str, Out]],
description: Optional[str],
config_schema: Optional[Union[UserConfigSchema, IDefinitionConfigSchema]],
required_resource_keys: Optional[AbstractSet[str]],
tags: Optional[Mapping[str, Any]],
version: Optional[str],
retry_policy: Optional[RetryPolicy],
code_version: Optional[str],
) -> "OpDefinition":
return OpDefinition(
compute_fn=compute_fn,
name=name,
ins=ins,
outs=outs,
description=description,
config_schema=config_schema,
required_resource_keys=required_resource_keys,
tags=tags,
version=version,
retry_policy=retry_policy,
code_version=code_version,
)
@property
def node_type_str(self) -> str:
return "op"
@property
def is_graph_job_op_node(self) -> bool:
return True
@public
@property
def name(self) -> str:
"""str: The name of this op."""
return super(OpDefinition, self).name
@public
@property
def ins(self) -> Mapping[str, In]:
"""Mapping[str, In]: A mapping from input name to the In object that represents that input."""
return {input_def.name: In.from_definition(input_def) for input_def in self.input_defs}
@public
@property
def outs(self) -> Mapping[str, Out]:
"""Mapping[str, Out]: A mapping from output name to the Out object that represents that output."""
return {output_def.name: Out.from_definition(output_def) for output_def in self.output_defs}
@property
def compute_fn(self) -> Union[Callable[..., Any], "DecoratedOpFunction"]:
return self._compute_fn
@public
@property
def config_schema(self) -> IDefinitionConfigSchema:
"""IDefinitionConfigSchema: The config schema for this op."""
return self._config_schema
@public
@property
def required_resource_keys(self) -> AbstractSet[str]:
"""AbstractSet[str]: A set of keys for resources that must be provided to this OpDefinition."""
return frozenset(self._required_resource_keys)
@public
@deprecated(breaking_version="2.0", additional_warn_text="Use `code_version` instead.")
@property
def version(self) -> Optional[str]:
"""str: Version of the code encapsulated by the op. If set, this is used as a
default code version for all outputs.
"""
return self._version
@public
@property
def retry_policy(self) -> Optional[RetryPolicy]:
"""Optional[RetryPolicy]: The RetryPolicy for this op."""
return self._retry_policy
@public
@property
def tags(self) -> Mapping[str, str]:
"""Mapping[str, str]: The tags for this op."""
return super(OpDefinition, self).tags
[docs] @public
def alias(self, name: str) -> "PendingNodeInvocation":
"""Creates a copy of this op with the given name."""
return super(OpDefinition, self).alias(name)
[docs] @public
def tag(self, tags: Optional[Mapping[str, str]]) -> "PendingNodeInvocation":
"""Creates a copy of this op with the given tags."""
return super(OpDefinition, self).tag(tags)
[docs] @public
def with_hooks(self, hook_defs: AbstractSet[HookDefinition]) -> "PendingNodeInvocation":
"""Creates a copy of this op with the given hook definitions."""
return super(OpDefinition, self).with_hooks(hook_defs)
[docs] @public
def with_retry_policy(self, retry_policy: RetryPolicy) -> "PendingNodeInvocation":
"""Creates a copy of this op with the given retry policy."""
return super(OpDefinition, self).with_retry_policy(retry_policy)
def is_from_decorator(self) -> bool:
from .decorators.op_decorator import DecoratedOpFunction
return isinstance(self._compute_fn, DecoratedOpFunction)
def get_output_annotation(self) -> Any:
if not self.is_from_decorator():
raise DagsterInvalidInvocationError(
f"Attempted to get output annotation for {self.node_type_str} '{self.name}', "
"which was not constructed from a decorated function."
)
return cast("DecoratedOpFunction", self.compute_fn).get_output_annotation()
def all_dagster_types(self) -> Iterator[DagsterType]:
yield from self.all_input_output_types()
def iterate_node_defs(self) -> Iterator[NodeDefinition]:
yield self
def iterate_op_defs(self) -> Iterator["OpDefinition"]:
yield self
T_Handle = TypeVar("T_Handle", bound=Optional[NodeHandle])
def resolve_output_to_origin(
self, output_name: str, handle: T_Handle
) -> Tuple[OutputDefinition, T_Handle]:
return self.output_def_named(output_name), handle
def resolve_output_to_origin_op_def(self, output_name: str) -> "OpDefinition":
return self
def get_inputs_must_be_resolved_top_level(
self, asset_layer: "AssetLayer", handle: Optional[NodeHandle] = None
) -> Sequence[InputDefinition]:
handle = cast(NodeHandle, check.inst_param(handle, "handle", NodeHandle))
unresolveable_input_defs = []
for input_def in self.input_defs:
if (
not input_def.dagster_type.loader
and not input_def.dagster_type.kind == DagsterTypeKind.NOTHING
and not input_def.has_default_value
and not input_def.input_manager_key
):
input_asset_key = asset_layer.asset_key_for_input(handle, input_def.name)
# If input_asset_key is present, this input can be resolved
# by a source asset, so input does not need to be resolved
# at the top level.
if input_asset_key:
continue
unresolveable_input_defs.append(input_def)
return unresolveable_input_defs
def input_has_default(self, input_name: str) -> bool:
return self.input_def_named(input_name).has_default_value
def default_value_for_input(self, input_name: str) -> InputDefinition:
return self.input_def_named(input_name).default_value
def input_supports_dynamic_output_dep(self, input_name: str) -> bool:
return True
def with_replaced_properties(
self,
name: str,
ins: Optional[Mapping[str, In]] = None,
outs: Optional[Mapping[str, Out]] = None,
config_schema: Optional[IDefinitionConfigSchema] = None,
description: Optional[str] = None,
) -> "OpDefinition":
return OpDefinition.dagster_internal_init(
name=name,
ins=ins
or {input_def.name: In.from_definition(input_def) for input_def in self.input_defs},
outs=outs
or {
output_def.name: Out.from_definition(output_def) for output_def in self.output_defs
},
compute_fn=self.compute_fn,
config_schema=config_schema or self.config_schema,
description=description or self.description,
tags=self.tags,
required_resource_keys=self.required_resource_keys,
code_version=self._version,
retry_policy=self.retry_policy,
version=None, # code_version replaces version
)
def copy_for_configured(
self,
name: str,
description: Optional[str],
config_schema: IDefinitionConfigSchema,
) -> "OpDefinition":
return self.with_replaced_properties(
name=name,
description=description,
config_schema=config_schema,
)
def get_resource_requirements(
self,
outer_context: Optional[object] = None,
) -> Iterator[ResourceRequirement]:
# Outer requiree in this context is the outer-calling node handle. If not provided, then
# just use the op name.
outer_context = cast(Optional[Tuple[NodeHandle, Optional["AssetLayer"]]], outer_context)
if not outer_context:
handle = None
asset_layer = None
else:
handle, asset_layer = outer_context
node_description = f"{self.node_type_str} '{handle or self.name}'"
for resource_key in sorted(list(self.required_resource_keys)):
yield OpDefinitionResourceRequirement(
key=resource_key, node_description=node_description
)
for input_def in self.input_defs:
if input_def.input_manager_key:
yield InputManagerRequirement(
key=input_def.input_manager_key,
node_description=node_description,
input_name=input_def.name,
root_input=False,
)
elif asset_layer and handle:
input_asset_key = asset_layer.asset_key_for_input(handle, input_def.name)
if input_asset_key:
io_manager_key = (
asset_layer.get(input_asset_key).io_manager_key
if asset_layer.has(input_asset_key)
else DEFAULT_IO_MANAGER_KEY
)
yield InputManagerRequirement(
key=io_manager_key,
node_description=node_description,
input_name=input_def.name,
root_input=False,
)
for output_def in self.output_defs:
yield OutputManagerRequirement(
key=output_def.io_manager_key,
node_description=node_description,
output_name=output_def.name,
)
def resolve_input_to_destinations(
self, input_handle: NodeInputHandle
) -> Sequence[NodeInputHandle]:
return [input_handle]
def __call__(self, *args, **kwargs) -> Any:
from .composition import is_in_composition
if is_in_composition():
return super(OpDefinition, self).__call__(*args, **kwargs)
return direct_invocation_result(self, *args, **kwargs)
def _resolve_output_defs_from_outs(
compute_fn: Union[Callable[..., Any], "DecoratedOpFunction"],
outs: Optional[Mapping[str, Out]],
default_code_version: Optional[str],
) -> Sequence[OutputDefinition]:
from .decorators.op_decorator import DecoratedOpFunction
if isinstance(compute_fn, DecoratedOpFunction):
inferred_output_props = infer_output_props(compute_fn.decorated_fn)
annotation = inferred_output_props.annotation
description = inferred_output_props.description
else:
inferred_output_props = None
annotation = inspect.Parameter.empty
description = None
if outs is None:
return [OutputDefinition.create_from_inferred(inferred_output_props, default_code_version)]
# If only a single entry has been provided to the out dict, then slurp the
# annotation into the entry.
if len(outs) == 1:
name = next(iter(outs.keys()))
only_out = outs[name]
return [only_out.to_definition(annotation, name, description, default_code_version)]
# If multiple outputs...
# Note: we don't provide description when using multiple outputs. Introspection
# is challenging when faced with multiple outputs.
# ... and no annotation, use empty for each output annotation
if annotation == inspect.Parameter.empty:
return [
out.to_definition(
annotation_type=inspect.Parameter.empty,
name=name,
description=None,
code_version=default_code_version,
)
for (name, out) in outs.items()
]
# ... or if a single result object type, use None for each output annotation
if _is_result_object_type(annotation):
# this can happen for example when there are outputs for checks
# that get reported via a singular MaterializeResult
return [
out.to_definition(
annotation_type=type(None),
name=name,
description=None,
code_version=default_code_version,
)
for (name, out) in outs.items()
]
# ... otherwise we expect to have a tuple with entries...
if get_origin(annotation) != tuple:
raise DagsterInvariantViolationError(
"Expected Tuple annotation for multiple outputs, but received non-tuple annotation."
)
subtypes = get_args(annotation)
# ... if they are all result object entries use None
if len(subtypes) > 0 and all(_is_result_object_type(t) for t in subtypes):
# the counts of subtypes and outputs may not align due to checks results
# being passed via MaterializeResult similar to above.
return [
out.to_definition(
annotation_type=type(None),
name=name,
description=None,
code_version=default_code_version,
)
for (name, out) in outs.items()
]
# ... otherwise they should align with outputs
if len(subtypes) != len(outs):
raise DagsterInvariantViolationError(
"Expected Tuple annotation to have number of entries matching the "
f"number of outputs for more than one output. Expected {len(outs)} "
f"outputs but annotation has {len(subtypes)}."
)
return [
cur_out.to_definition(
annotation_type=subtypes[idx],
name=name,
description=None,
code_version=default_code_version,
)
for idx, (name, cur_out) in enumerate(outs.items())
]
def _validate_context_type_hint(fn):
from inspect import _empty as EmptyAnnotation
from dagster._core.decorator_utils import get_function_params
from dagster._core.definitions.decorators.op_decorator import is_context_provided
from dagster._core.execution.context.compute import (
AssetCheckExecutionContext,
AssetExecutionContext,
OpExecutionContext,
)
params = get_function_params(fn)
if is_context_provided(params):
if params[0].annotation not in [
AssetExecutionContext,
OpExecutionContext,
EmptyAnnotation,
AssetCheckExecutionContext,
]:
raise DagsterInvalidDefinitionError(
f"Cannot annotate `context` parameter with type {params[0].annotation}. `context`"
" must be annotated with AssetExecutionContext, AssetCheckExecutionContext, OpExecutionContext, or left blank."
)
def _is_result_object_type(ttype):
# Is this type special result object type
return ttype in (MaterializeResult, ObserveResult, AssetCheckResult)