Parquet X-ray (GitHub Repo)

Back to Devoured - October 01, 2026

DEVOURED

Parquet X-ray (GitHub Repo)

Data infrastructure GitHub

Parquet X-ray allows developers to inspect Parquet metadata and file structure in the browser without downloading massive datasets.

What: Parquet X-ray is a browser-based tool that uses the hyparquet library to visualize Parquet file internals, including row groups, column chunks, bloom filters, and min/max statistics. It reads only the file footer and indexes, allowing a 2 GB file to be parsed by downloading roughly 3 MB.

Why it matters: Understanding how data is physically laid out on disk is crucial for optimizing read performance, yet most CLI tools require full file downloads, making remote inspection difficult.

Takeaway: Try the tool on Hugging Face to inspect remote Parquet files by pasting the dataset URL.

Decoder

Row group: A horizontal partition of data in a Parquet file, typically containing a set of rows.
Column chunk: The data for a specific column within a row group, stored in a contiguous block.

Original article

See how a Parquet file is laid out on disk: row groups, column chunks, pages, indexes, bloom filters and the footer.

Paste a Hub URL, an hf:// path, any URL that allows CORS range requests, or open a local file. Only the footer and page indexes are downloaded, so a 2 GB file opens after reading about 3 MB. Parsing uses hyparquet.

Development

npm install
npm run dev       # http://localhost:5173/?url=sensors.parquet
npm run verify    # lint, type-check, tests, build

See CONTRIBUTING.md for how the code is organized.

License

Apache 2.0

Explore all 58 articles in today's digest