Create a one-column UUID CSV for spreadsheet work, imports, QA datasets and seed fixtures. The export includes a stable header and quoted values so the file can move cleanly between common data tools.
Key takeaways
- Use a clear header such as uuid or id.
- Keep one canonical UUID per row and validate before import.
- Apply a database unique constraint even when generated values should be unique.
For implementation context, continue with UUID JSON, Bulk UUIDs, and SQL. These pages cover the closest generator, comparison, validation or storage decisions without repeating this guide.
A predictable CSV structure
A single uuid header followed by one quoted identifier per row works for most importers. Quoting is conservative and makes it easy to add columns later without changing the parsing rules.
UUIDs do not contain commas, but a real CSV parser is still safer than splitting lines manually once the file gains names, notes or localized values.
Spreadsheet and import checks
UUID text is not numeric, so spreadsheets should preserve it without scientific notation. Before a production import, verify the row count, reject blank lines, validate every identifier and decide how duplicates should be handled.
Fixtures versus production IDs
Bulk UUID files are useful for deterministic test setup and one-time migrations. For live records, generate the identifier in the application or database transaction that creates the row so ownership and uniqueness rules stay close to the write.