SAService Architect Back to designer
SAService ArchitectDocumentation
Python DSLAPI: Streams
Python API reference pipeline.input() | pipeline.sink()

Input and Sink

Bind a stream graph to connector endpoints for inbound and outbound transport.

When to use

Use Input to admit endpoint messages and Sink to submit graph messages through an endpoint.

Behavior

  • Input may define a primary source or sources for endpoint patterns that participate in a cycle.
  • Sink may reference an error stream for failures returned by the transport adapter.
  • Endpoint and value type compatibility is validated across all uses of an endpoint.

Pipeline.input() properties

Inbound boundary.

PropertyTypeDescription
namestr

Human-readable stream name; the symbolic key is derived automatically.

Required
sourceStream | None

Optional primary upstream stream.

Default: None
x / yfloat

Designer canvas coordinates.

Default: 0 / 0
endpointEndpoint

Endpoint consumed by the stream.

Required
value_typeTypeDefinition | DataType | str | None

Output or boundary payload contract.

Default: None
sourcesSequence[Stream]

Additional sources.

Default: ()

Pipeline.sink() properties

Outbound boundary.

PropertyTypeDescription
namestr

Human-readable stream name; the symbolic key is derived automatically.

Required
sourceStream | None

Optional primary upstream stream.

Default: None
x / yfloat

Designer canvas coordinates.

Default: 0 / 0
endpointEndpoint

Endpoint called or published to.

Required
value_typeTypeDefinition | DataType | str | None

Output or boundary payload contract.

Default: None
error_streamstr | None

Symbolic failure consumer reference.

Default: None

Python example

consume = orders.input(
    "Consume Orders", endpoint=order_topic, value_type=order
)
publish = orders.sink(
    "Publish Result", endpoint=result_topic, value_type=order_result
)
consume >> publish