HdrHistogramPyo3

A PyO3-based Python extension that wraps Rust implementations as Polars expression plugins, featuring high-performance histogram statistics for latency and performance analysis.

Features

• HdrHistogram Statistics: Compute percentiles, mean, and summaries from numeric columns using the efficient hdrhistogram Rust crate
• String Transformations: Element-wise string operations (e.g., pig_latinnify)
• Polars Integration: Use as native Polars expressions within DataFrames
• High Performance: Direct Rust compilation via PyO3 with optimized memory management

Quick Start

import polars as pl
from hdrhistogram_pyo3 import hdr_mean, hdr_percentile, hdr_stats_summary

# Create a DataFrame with latency measurements
df = pl.DataFrame({"latency_ms": [10.0, 15.0, 20.0, 25.0, 30.0]})

# Compute statistics using HdrHistogram
result = df.select(
    mean=hdr_mean("latency_ms"),
    p50=hdr_percentile("latency_ms", 50.0),
    p95=hdr_percentile("latency_ms", 95.0),
    summary=hdr_stats_summary("latency_ms"),
)

print(result)

Development

See AGENTS.md for comprehensive development guide including:

• Build and test workflows
• Architecture and design patterns
• Adding new expressions
• Common gotchas and troubleshooting

Quick Commands

make venv              # Create virtual environment
make install           # Build extension
make test              # Run tests
make run               # Run example
make pre-commit        # Full validation (fmt + lint + test)

Stack

• Rust 2021 (nightly) with hdrhistogram v7
• PyO3 0.27 for Python bindings
• Polars 0.53+ with plugin system
• Maturin for build orchestration