Parquet to CSV

Apache Parquet is a compact, column-oriented binary format from the data and analytics world (Spark, pandas, DuckDB, data lakes). A .parquet file stores its schema inline, packs each column compressed (usually Snappy) and is therefore very space-efficient - but unreadable in a text editor. This tool reads the Parquet content and outputs a flat CSV table you can open in Excel, LibreOffice, Google Sheets or any text editor.

Everything runs entirely locally in your browser (hyparquet, pure JavaScript, no upload, even offline). The tool evaluates the Parquet footer and column pages and supports the common page codecs Snappy, Gzip, Zstd, Brotli and LZ4. So no data is distorted, we emit 64-bit integers beyond the safe JavaScript range as a string, timestamps as ISO 8601 text and raw binary values as base64. Nested fields (lists, structs) appear in a CSV cell as compact JSON.

This is handy when a data pipeline, an export or an analytics job hands you a .parquet file and you just want to peek inside, check it in a spreadsheet or pass it to a tool that speaks CSV. CSV is universal and immediately human-readable. Note: the conversion is one-way (Parquet to CSV); it does not write back to Parquet, and Parquet column compression is lost in flat CSV.

Specifications

In 3 steps

1. Drop or tap your Parquet file (.parquet).
2. The tool reads the schema and columns locally.
3. Download the CSV (several as a ZIP).

Limitations: Parquet files with the codecs Snappy, Gzip, Zstd, Brotli and LZ4 are supported. Encrypted Parquet files are not read. Nested columns (lists/structs) are shown in the cell as JSON. Large 64-bit integers appear as a string (precision), timestamps as ISO 8601, binary values as base64. The conversion is one-way (no CSV back to Parquet). Very large files need memory accordingly.

FAQ

Related tools

Parquet to JSON · Arrow to CSV · CSV to JSON · TOML to XML · Images to CBZ · Arrow to JSON