Parquet Explorer
Open a Parquet file and explore its rows and columns. Filter or group rows, preview the results and download a data report.
How do I open and explore a Parquet file?
Upload an unencrypted Parquet file and inspect its columns. Preview rows, choose fields, apply filters or group values into a summary. Download a ZIP with CSV data and a JSON report. The separate processing service removes temporary files when the job ends.
Your result
A little clarity, right here.
Choose a file and run the tool. Your result will appear here, ready to download.
Three simple steps
How to use Parquet Explorer
Upload your file and inspect its schema to find exact column names, data types, row counts and compression details.
Choose a row preview or aggregation, then set the relevant column names, filters, grouping and range options.
Run the service and download the selected result with its schema, typed-value encodings and processing details.
Common questions
Good to know
Understand the result.
Keep the original.
Is this tool free, and are my files uploaded?
This tool is free with no account required. When you run the tool, your input goes to a separate processing service. Temporary files are deleted when the job ends. You can download the result to your device.
What are the limits and details?
One file up to 8 MB. Schema inspection supports 50 top-level columns, 200 leaves, 256 row groups, six nested levels and five million metadata rows, with a 1 MB footer limit. Preview and aggregation support at most 100,000 source rows, two million source cells and 128 MiB declared/decoded Arrow data. Export up to 5,000 preview rows or 1,000 aggregate groups; the page shows 100 rows and 200 characters per cell. ZIP downloads are capped at 6 MiB and JSON reports at 8 MB; cells at 64,000 characters or 48,000 binary bytes. Nested values have up to 2,000 items and are preview-only. Filters combine with AND and support scalar text, number, boolean and ISO date/time values; binary and duration fields are preview-only for filters. JSON preserves nulls, exact decimal/large-integer text and tagged binary/nonfinite values. CSV prefixes formula-like or numeric text for spreadsheets; blank CSV cells cannot distinguish null from empty text. Floating aggregates use floating arithmetic; integer/decimal means use up to 100 significant decimal digits. Encrypted files, arbitrary SQL, remote datasets and custom Arrow extensions are unsupported. Processing stops after two minutes.