ai_graph.pipeline

AI-Graph Pipeline Module.

This module provides the core pipeline management functionality for AI-Graph, allowing you to chain together multiple processing steps in a Chain of Responsibility pattern.

The pipeline module contains:

  • Pipeline: Main class for managing and executing pipeline steps

  • Utilities for pipeline construction and management

Example

>>> from ai_graph.pipeline import Pipeline
>>> from ai_graph.step import AddKeyStep, DelKeyStep
>>>
>>> # Create a pipeline with multiple steps
>>> pipeline = Pipeline("DataProcessor")
>>> pipeline.add_step(AddKeyStep("status", "processing"))
>>> pipeline.add_step(AddKeyStep("timestamp", "2025-01-01"))
>>> pipeline.add_step(DelKeyStep("temp_data"))
>>>
>>> # Execute the pipeline
>>> result = pipeline.process({"input": "data", "temp_data": "remove_me"})
>>> # result = {"input": "data", "status": "processing", "timestamp": "2025-01-01"}