Hyparquet

Hyparquet is a pure JavaScript library for parsing Apache Parquet files directly in the browser (or Node.js), enabling zero-server data access via efficient HTTP range requests.

This is Hyparquet: your high-performance, browser-native Parquet parser. It’s a pure JavaScript library, zero dependencies, designed to read massive Apache Parquet files directly from cloud storage (like S3) without a backend server. We leverage HTTP range requests and metadata analysis to fetch only the necessary data, dramatically reducing network load and latency. At a core engine size of only ~9.7 KB (gzipped), it ensures near-instant initialization. Hyparquet supports the full Parquet specification, including all 22 data types and various compression codecs, making it the most compliant parser available for in-browser data science and analytics.

https://github.com/hyparam/hyparquet