Schema diff — compare two files’ columns
Drop two files (Parquet, Avro, CSV, TSV, JSON or JSONL) to see how their schemas differ — which columns were added or removed and which changed type. Perfect for catching schema drift between two versions of a table. It all runs in your browser; neither file is uploaded.
Drop a file or click to browse
First file (before)
Drop a file or click to browse
Second file (after)
Frequently asked questions
- What does the diff show?
- Column-level changes between two files' schemas: columns added, columns removed, and columns whose type changed. It's a schema diff, not a row-by-row data diff.
- What is schema drift?
- When a dataset's shape changes between versions — a column renamed, dropped, added or retyped — often silently breaking a downstream pipeline. Diffing two snapshots catches it before it ships.
- Which formats can I compare?
- Any mix of Parquet, Avro, CSV, TSV, JSON, JSONL and Excel — the two files don't have to be the same format.
- Are the files uploaded?
- No. Both files are read in your browser with DuckDB-WASM; only their schemas are compared, and neither file leaves your device.