How-To Guide
How to Open a Parquet File Locally
Learn how to open a Parquet file locally in your browser without Python, Spark or uploads using DuckDB and Queryfiles.app.
TL;DR
To open a Parquet file locally, use a browser-based viewer powered by DuckDB, drop your .parquet file into the app and inspect the schema and rows without sending the file to a server.
Key Takeaways
- --No local environment setup You do not need Python, pyarrow, pandas or a desktop database client just to inspect one file.
- --Fast schema discovery DuckDB reads the Parquet metadata and exposes column names and types immediately.
- --Private local workflow The file stays on your device, which is useful for internal exports and sensitive datasets.
What you need to know
Open a Parquet file locally by dragging it into Queryfiles.app. DuckDB reads the file in your browser, infers the schema and lets you inspect rows with SQL in seconds.
Parquet is efficient for analytics, but it is not human-readable and many people still reach for Python or Spark just to inspect one file. That is overkill when the real need is simply to open the file, verify the schema and run a few filters.
Queryfiles.app removes that setup. The file is opened locally in your browser with DuckDB WebAssembly, so you can inspect columns, preview rows and run SQL immediately.
Step-by-step workflow
01.Open Queryfiles.app in your browser
Start from the Queryfiles.app home page. The app runs entirely in the browser, so there is nothing to install before you begin.
02.Drop the .parquet file into the app
Drag and drop the file or use Browse. Queryfiles.app detects the format automatically and loads it into DuckDB locally.
03.Review the inferred schema
Check column names and data types to confirm that the file matches your expectations before deeper analysis.
04.Preview rows with SQL
Run a simple SELECT query, add filters or sort by the fields you care about to validate the contents of the file.
05.Iterate on your analysis
Once the file is open, continue with aggregations, date filters and other DuckDB queries without leaving the browser.
Useful first query
A first-pass query helps you confirm the file opened correctly and shows the shape of the data.
SELECT *
FROM data
LIMIT 100;Frequently asked questions
Q.Can I open a Parquet file locally without Python?
Yes. Queryfiles.app uses DuckDB in the browser, so you can open a Parquet file locally without Python, Spark or any native installation.
Q.Does the Parquet file get uploaded?
No. The file is processed locally in your browser and does not need to be uploaded to a server.
Q.Why use SQL to inspect a Parquet file?
SQL is the fastest way to preview rows, filter large datasets and validate schema assumptions when you only need ad hoc analysis.
Q.Can I use this on Linux, macOS and Windows?
Yes. Because Queryfiles.app runs in the browser, the workflow is the same across modern desktop operating systems.