pipeline.port

Port primitives for the pipeline framework.

Contains the core port-related classes:

PortReference: typed handle to a specific module port (for IDE support)
ChainedConnection: batches multiple port connections with the | operator
_PortDataCheck: internal Pydantic model that validates port data at runtime
Port: a single data slot on a PipeModule (input or output)

Classes

class pipeline.port.PortReference

A reference to a specific port of a module for IDE-friendly port access.

Inherits from:: <ast.Subscript object at 0x0000028E217659F0>

Methods:

__init__(module: PipeModule, port_name: str, port_direction: Literal[input, output], expected_data_type: Type[PortDataType] | str | None = None)

Initialize PortReference.

Parameters

modulePipeModule: The module that owns this port
port_namestr: The name of the port port_direction : Literal[“input”, “output”] The direction of port (‘input’ or ‘output’)
expected_data_typeType[PortDataType] | str | None: The expected data type for this port (for IDE hints)

Properties:

data: Get the actual data from the port.

class pipeline.port.ChainedConnection

Represents a chain of port connections for batch linking.

Methods:

__init__(connections: list)

to_links() → list[Link]: Convert chained connections to Link objects.

class pipeline.port.Port

Methods:

__init__(ptype: PortType | list[PortType] = PortType.NumberArray, data: Any = None, pdoc: str | None = None) → None

Initialize the Port object.

Parameters

ptypePortType | list[PortType], default: PortType.Field: The data type of in or out of the port which can be one type or several types.
dataAny, default: None: The data in or out of the port.
pdocstr, default: None: The description of the port.

Properties:

data