Loading Avro to CSV converter...

How to Convert Avro to CSV - Step by Step Guide

Step 1

Input Your Avro Data

Paste your Avro data (with schema) into the editor or upload a .json file containing both schema and data.

•
Paste directly: Copy your Avro data with schema
•
Upload file: Click the Upload button to select a file
•
Try sample: Click the Sample button to load example Avro data
Step 2

Automatic CSV Conversion

The converter automatically extracts field names from your Avro schema and converts the data records into CSV rows:

•
Schema check: Checks the schema and every record before anything is written, and names the record and field that don't match
•
CSV headers: Creates CSV header row from schema field names
•
Exact values: Long IDs, decimals like 14.50 and text like 00501 come out exactly as typed; commas, quotes and line breaks are quoted
•
Excel ready: Output works with Microsoft Excel, Google Sheets, and other spreadsheet tools
Step 3

Download or Copy CSV Output

Once converted, download the CSV file or copy it to your clipboard for use in spreadsheet applications:

•
Download: Save as .csv file
•
Copy: One-click copy to clipboard
•
Import to Excel: Open directly in Microsoft Excel or Google Sheets

FAQs About Avro to CSV Conversion

Why convert Avro to CSV?

Avro is built for pipelines and storage, but most people who need to look at the data live in Excel or Google Sheets. A CSV of the records is the easiest thing to hand to someone for a quick check, a report or an import into a BI tool.

What does the input need to look like?

One JSON object with two fields: "schema", holding your Avro record schema, and "data", holding a list of records (a single record works too). Each field in the schema becomes a CSV column, in the order the schema lists them. Avro's own JSON encoding, where union values are wrapped like {"string": "a@b.com"}, is accepted as well and the wrapper is dropped in the CSV.

What if a record does not match the schema?

Nothing is converted. The page checks every record against the schema first and tells you which record and which field is wrong, for example that "age" should be an int but holds "thirty". That way a half-wrong CSV never ends up in your spreadsheet.

How are nested records, arrays and maps written?

A nested record gets one column per field, named with a dot, such as address.city and address.zip. Map keys work the same way. Arrays stay together in one cell as compact JSON, for example ["a","b"], so a list never spills across columns.

Will long IDs and decimals stay exactly as I typed them?

Yes. A long like 8901240544102066246 or -9223372036854775808 is written with every digit, and a double typed as 14.50 stays 14.50 in the CSV. Text such as "00501" is left alone too. Keep in mind that Excel itself shows only 15 significant digits, so open long IDs as text there if you need all of them.

Does it handle commas, quotes and line breaks inside values?

Yes. Any value with a comma, a double quote, a line break or spaces at either end is wrapped in quotes, and quotes inside it are doubled, following the usual CSV rules (RFC 4180). The downloaded file starts with a UTF-8 marker so Excel shows accented letters and emoji correctly.

Can I upload a binary .avro file?

Not directly. This page reads the JSON form of Avro data. If you have a binary container file, turn it into JSON first, for example with avro-tools tojson, and paste the records here together with the schema. Your data is converted in the browser and is not uploaded or kept.