Hapax Quick Start Guide

This guide will help you get started with Hapax in under 5 minutes. For more in-depth information, see the Comprehensive Guide.

Installation

pip install hapax

Basic Usage

1. Define Simple Operations

Operations are the basic building blocks in Hapax. They are pure functions decorated with @ops:

from hapax import ops
from typing import List

@ops  # Uses function name as operation name
def tokenize(text: str) -> List[str]:
    return text.split()

@ops
def remove_stops(words: List[str]) -> List[str]:
    stops = {'the', 'a', 'an'}
    return [w for w in words if w not in stops]

2. Compose Operations

Operations can be composed using the >> operator:

# Use >> to chain operations
pipeline = tokenize >> remove_stops

# Use the pipeline
result = pipeline("The quick brown fox")  # ['quick', 'brown', 'fox']

3. Create a Graph

For more complex pipelines, you can use the @graph decorator:

from hapax import graph

@graph  # Uses function name as graph name
def text_pipeline(text: str) -> List[str]:
    return tokenize >> remove_stops

# Use the graph
result = text_pipeline("The quick brown fox")

Add Monitoring (Optional)

import openlit

# Initialize monitoring
openlit.init(otlp_endpoint="http://127.0.0.1:4318")

# Your operations are now automatically monitored!

Next Steps

• Read the full documentation for more features
• Learn about the Graph API for building complex pipelines
• Check out the OpenLit integration for monitoring
• Try evaluation decorators for content safety

That's it! You're ready to build type-safe data processing pipelines with Hapax.