Parquet to CSV Converter: Export .parquet as .csv
Convert a Parquet file to a downloadable CSV with proper quoting for commas, quotes, and newlines.
Upload a .parquet file and the page reads every row, infers the column order from the first record, and writes the result as RFC 4180-compatible CSV. Fields with commas, double quotes, or line breaks are quoted and escaped correctly. Nested values are serialized as JSON. The output is offered as a download and previewed inline so you can sanity-check the result before saving.
Parquet is great for analytical engines but terrible to hand off to a spreadsheet, BI tool, or non-engineer. CSV is the lowest common denominator: Excel, Google Sheets, Tableau, Numbers, every database loader, and every scripting language can read it. The convert step is mechanical but easy to get wrong, especially when values contain commas or quotes.
The converter reads the Parquet file locally with hyparquet, gathers all rows into plain JavaScript objects, then walks every field and applies RFC 4180 quoting rules: a value is wrapped in double quotes when it contains a comma, a double quote, or a newline, and any embedded double quote is doubled up. BigInt values are written as their decimal form. Date values are serialized as ISO-8601. Nested objects and arrays become inline JSON, which is a lossy but compact representation that round-trips back to JSON if needed.
The result is a single CSV string. The preview shows the first 50 lines so you can spot a column-order surprise before downloading. Files of a few million cells convert in a couple of seconds; very large files may take longer because both the parsed objects and the resulting string live in memory.
- 1
Open the file
Pick a .parquet file. It is parsed locally with hyparquet and converted into an array of row objects.
- 2
Preview the CSV
The first 50 output lines render inline. Verify the header row and the quoting on tricky values.
- 3
Download .csv
Click Download .csv to save the full output. The file inherits the source filename with a new extension.
Hand a dataset to a spreadsheet user
Convert a Parquet extract so a colleague who lives in Excel or Google Sheets can open it without installing anything.
Load into a BI tool that can't read Parquet
Older BI tools and quick chart builders speak CSV. The converter bridges the format gap without spinning up a notebook.
Inspect data with grep and awk
CSV plays well with command-line tools. Once converted, the file is searchable with standard shell utilities.
Seed a database table
Most relational databases include a COPY or LOAD DATA INFILE statement that consumes CSV directly.
Is my data sent to a server?
No. The whole conversion happens in your browser tab. The Parquet file is parsed locally and the CSV is built in memory.
How are nested columns handled?
Structs, lists, and maps are serialized as inline JSON. The result is still valid CSV; downstream consumers can re-parse the JSON if they need the structure.
What about timestamps and BigInt?
Dates are written as ISO-8601 strings. BigInt values are written as decimal text so spreadsheets accept them without precision loss.
Are commas and quotes escaped correctly?
Yes. The writer follows RFC 4180: any field containing a comma, double quote, or newline is wrapped in double quotes, and embedded double quotes are doubled up.
How large a file can I convert?
Up to a few hundred megabytes works in practice. Both the parsed rows and the CSV string have to fit in browser memory, so multi-gigabyte files may run out of space.