Parquet Viewer Online

Open a Parquet file, browse it in a fast table, and search, sort, filter and export it. The file is read by your browser; it is not uploaded.

Loading the Parquet Viewer…

What this tool does

Parquet is a typed, columnar, self-describing binary format: unlike CSV, a Parquet file already knows its own column names and types, so there is no delimiter or header to guess. This viewer opens the file, shows its real schema and rows in a fast table, and lets you search, sort, filter and export — the same controls as the CSV Viewer, built on the same underlying engine.

File structure: information CSV cannot have

“Show file structure” reads facts straight from the Parquet file's own footer: the format version, how many row groups it has, and, per physical column, which compression codec and encoding were used and how many bytes it takes compressed versus uncompressed. These numbers come from the file itself, not an estimate.

Nested columns are shown honestly, not flattened

A Parquet file can have list and struct (nested-object) columns. This viewer never silently flattens one into several new columns or drops it: a nested value is shown as its own readable text (for example ["red", "blue"] or {city: NYC, zip: 10001}), and the file-structure panel shows exactly how Parquet actually stores it — a struct or list is physically written as more than one column on disk, and that panel lists each one rather than merging them back into one row.

Browser-local processing

The file is handed to DuckDB running as WebAssembly inside your browser, in a background Web Worker. Nothing about opening or querying it involves a server: the engine, its Parquet component, and this page's own code are all served from this site, and the browser is told to refuse requests to any other site while the tool runs. See the privacy page for exactly what that does and does not guarantee.

Exporting

“Export CSV” downloads the current search, filters and sort order applied to the whole file as a CSV file, not just the current page. A nested column is exported as the same readable text shown in the table. Any text value that would be read as a formula by a spreadsheet program (starting with =, +, -, @, a tab or a carriage return) is prefixed with a quote character so it opens as plain text instead.

Large files

Files are read by your device, so performance depends on your browser and its available memory. This tool warns you above roughly 100 MB, warns more strongly above roughly 500 MB, and declines to open a file above roughly 2 GB rather than risk crashing the tab. Parquet's compression means a large file on disk can still decode to a lot of memory; these are practical guidelines, not a guarantee.