README.md

pypdown

A Pydantic model-based approach to data pipelining with file I/O linting.

Features

• Pydantic model-based approach to data pipelining
• File I/O linting for robust pipeline execution
• Easy-to-use API for defining and running pipeline steps
• Support for callback functions and keyword argument-based file paths

Installation

pip install pypdown

Quick Start

from pypdown import run_step
from pypdown.models import Step
from pydantic import BaseModel
from pathlib import Path


class StepParams(BaseModel):
    input_file: Path = "input.txt"
    output_file: Path = "output.txt"
    final_file: Path = "final.txt"


def process_input(input_file: Path, output_file: Path, config: StepParams):
    """Process input file and create output file."""
    output_file.write_text(input_file.read_text().upper())


def finalize_output(output_file: Path, final_file: Path, config: StepParams):
    """Process output file and create final file."""
    final_file.write_text(f"Processed: {output_file.read_text()}")


config = StepParams()

# Define your pipeline tasks by reference to config field names
task_refs = [
    {
        "src": ["input_file"],
        "dst": ["output_file"],
        "fn": process_input,
    },
    {
        "src": ["output_file"],
        "dst": ["final_file"],
        "fn": finalize_output,
    },
]

# Create a Step
step = Step(name="Example Pipeline Step", task_refs=task_refs, config=config)

# Run the step
run_step(step)

Documentation

For full documentation, please visit pypdown.vercel.app.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.