_atransform() — langchain Function Reference

Architecture documentation for the _atransform() function in base.py from the langchain codebase.

Function python LangChainCore Runnables calls 1 called by 2

Entity Profile

LangChainCore→ Runnables→ _atransform() — langchain Function Reference

Dependency Diagram

graph TD
  6a8e40f9_b621_2878_43d7_e38d2dca02c5["_atransform()"]
  c4bf8d59_69f9_f8e5_efba_837c20df2360["RunnableSequence"]
  6a8e40f9_b621_2878_43d7_e38d2dca02c5 -->|defined in| c4bf8d59_69f9_f8e5_efba_837c20df2360
  df93336b_2161_a3ab_4adf_2f7000c37a15["_atransform()"]
  df93336b_2161_a3ab_4adf_2f7000c37a15 -->|calls| 6a8e40f9_b621_2878_43d7_e38d2dca02c5
  178252a4_c169_63ca_3a82_a5f7a1dcd26b["_atransform()"]
  178252a4_c169_63ca_3a82_a5f7a1dcd26b -->|calls| 6a8e40f9_b621_2878_43d7_e38d2dca02c5
  df93336b_2161_a3ab_4adf_2f7000c37a15["_atransform()"]
  6a8e40f9_b621_2878_43d7_e38d2dca02c5 -->|calls| df93336b_2161_a3ab_4adf_2f7000c37a15
  style 6a8e40f9_b621_2878_43d7_e38d2dca02c5 fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/core/langchain_core/runnables/base.py lines 3488–3511

    async def _atransform(
        self,
        inputs: AsyncIterator[Input],
        run_manager: AsyncCallbackManagerForChainRun,
        config: RunnableConfig,
        **kwargs: Any,
    ) -> AsyncIterator[Output]:
        steps = [self.first, *self.middle, self.last]
        # stream the last steps
        # transform the input stream of each step with the next
        # steps that don't natively support transforming an input stream will
        # buffer input in memory until all available, and then start emitting output
        final_pipeline = cast("AsyncIterator[Output]", inputs)
        for idx, step in enumerate(steps):
            config = patch_config(
                config,
                callbacks=run_manager.get_child(f"seq:step:{idx + 1}"),
            )
            if idx == 0:
                final_pipeline = step.atransform(final_pipeline, config, **kwargs)
            else:
                final_pipeline = step.atransform(final_pipeline, config)
        async for output in final_pipeline:
            yield output