From schema to seed file
The developer tools share one idea: describe the fields once, then take the data out in whatever shape the next step needs. The same seeded batch can be a JSON fixture for a unit test, a SQL script for a local database and a CSV for a product manager's spreadsheet — identical records, three formats.
- Fixtures for unit and integration tests — the JSON mock data generator and custom schema builder build nested objects, arrays and typed fields. Commit the seed alongside the test and the same records come back — or commit the file when its timestamps must not move.
- Seeding a database — the SQL INSERT generator writes dialect-correct statements for PostgreSQL, MySQL, SQLite and SQL Server, with identifier quoting and string escaping handled, an optional
CREATE TABLE, and a transaction wrapper. See How to seed a database with fake data for ordering foreign keys and loading large batches. - Mocking an API — the REST API response simulator, pagination generator and GraphQL mock generator return realistic envelopes: headers, status codes, error bodies, and offset, cursor, RFC 8288
Linkheader or JSON:API pagination. The guide Mocking REST APIs for frontend development shows how to serve them with MSW or json-server. - Interchange formats — the CSV & TSV and XML, YAML & TOML generators cover import pipelines and config parsers.
- Identifiers and patterns — the UUID, ULID & Nano ID generator covers UUID v1, v4 and time-ordered v7, ULID, Nano ID, CUID2-style IDs and MongoDB ObjectIds; the regex data generator produces strings that match a pattern you supply, which is the quickest way to test a validator from both sides.
Why reproducibility matters here
A random fixture finds bugs a hand-written one never will, and then — if it cannot be regenerated — destroys the evidence. Every generator on this site draws from a seeded pseudo-random generator, so a failing seed can be replayed: same seed, same records, with dates computed relative to the moment you generate. Seeded randomness: how to make test fixtures reproducible explains the technique if you want to apply it in your own factories.