How to open a Parquet file

A .parquet file will not open in Excel, Google Sheets, or a text editor. You need a reader that understands the format. ParquetView is one: it runs in Chrome, Safari, Firefox, and Edge.

Steps

1. Open the ParquetView viewer.
2. Drag a .parquet file onto the page, or click the upload area. CSV, TSV, and JSON work too.
3. Wait for the grid. For large files a preview loads first; use SQL if you need the full set.
4. Open the Info panel for column types and metadata. Use filters on a header, or switch to the SQL tab.

Nothing to install. Local file bytes stay in the browser (see privacy).

If the file is huge

Do not start by exporting everything. Select a few columns, run SELECT col, count(*) FROM file GROUP BY 1 LIMIT 100, or filter in the grid. Parquet is designed so those operations read a slice of the file, not a giant CSV equivalent in memory.

If you only have Python

You can still use pandas.read_parquet() or duckdb.sql("SELECT * FROM 'file.parquet' LIMIT 20"). That is the right tool when the next step is a notebook. It is the wrong tool when you just received a file in Slack and want to see the columns.

Remote files

Local drop is the default. If Azure Blob is configured, you can sign in and open a blob from your own storage account — that download is between your browser and Azure, not a ParquetView upload API. There is no “paste any S3 URL and we fetch it on our servers” path.

Related

• What is a Parquet file?
• Convert Parquet to CSV
• Run SQL on Parquet