Component step

A pipeline step that executes a component on the input state and returns the output.

Author

Dimitrij Ray (dimitrij.ray@gdplabs.id) Kadek Denaya (kadek.d.r.diana@gdplabs.id)

References

NONE

ComponentStep(name, component, input_state_map=None, output_state=None, runtime_config_map=None, fixed_args=None, input_map=None, retry_config=None, error_handler=None, cache_store=None, cache_config=None)

Bases: BasePipelineStep, HasInputsMixin

A pipeline step that executes a specific component.

This step wraps a component, manages its inputs and outputs, and integrates it into the pipeline.

Attributes:

Name	Type	Description
name	str	A unique identifier for this pipeline step.
component	Component	The component to be executed in this step.
input_map	dict[str, str | Val] | None	Unified input map.
output_state	str | list[str] | None	Key(s) to extract from the component result and add to the pipeline state. If None, the component is executed but no state updates are performed.
retry_policy	RetryPolicy | None	Configuration for retry behavior using LangGraph's RetryPolicy.
error_handler	BaseStepErrorHandler | None	Strategy to handle errors during execution.

Initializes a new ComponentStep.

Parameters:

Name	Type	Description	Default
name	str	A unique identifier for this pipeline step.	required
component	Component	The component to be executed in this step.	required
input_state_map	dict[str, str]	Mapping of component input arguments to pipeline state keys. Keys are input arguments expected by the component, values are corresponding state keys.	None
output_state	str | list[str]) | None	Key(s) to extract from the component result and add to the pipeline state. If None, the component is executed but no state updates are performed. Defaults to None.	None
runtime_config_map	dict[str, str] | None	Mapping of component input arguments to runtime configuration keys. Keys are input arguments expected by the component, values are runtime configuration keys. Defaults to None.	None
fixed_args	dict[str, Any] | None	Fixed arguments to be passed to the component. Defaults to None.	None
input_map	InputMapSpec | None	Unified input map. Can be a dict (arg -> str|Val) or a list with elements: 1. str for identity mapping 2. dict[str, str] for state/config mapping 3. dict[str, Val] for fixed args. Defaults to None.	None
retry_config	RetryConfig | None	Configuration for retry behavior using GLLM Core's RetryConfig. Defaults to None, in which case no retry config is applied.	None
error_handler	BaseStepErrorHandler | None	Strategy to handle errors during execution. Defaults to None, in which case the RaiseStepErrorHandler is used.	None
cache_store	'BaseCache' | None	Cache store to be used for caching. Defaults to None, in which case no cache store is used.	None
cache_config	dict[str, Any] | None	Cache configuration to be used for caching. Defaults to None, in which case no cache configuration is used.	None

execute(state, runtime) async

Executes the component and processes its output.

This method validates inputs, prepares data, executes the component, and formats the output for integration into the pipeline state.

Parameters:

Name	Type	Description	Default
state	PipelineState	The current state of the pipeline, containing all data.	required
runtime	Runtime[dict[str, Any] | BaseModel]	Runtime information for this step's execution.	required

Returns:

Type	Description
dict[str, Any] | None	dict[str, Any] | None: The update to the pipeline state after this step's operation, or None if output_state is None. When not None, this includes new or modified data produced by the component, not the entire state.

Raises:

Type	Description
RuntimeError	If an error occurs during component execution.
TimeoutError	If the component execution times out.
CancelledError	If the component execution is cancelled.