How It Works Under the Hood
Every generation run seeds a fast 32-bit `Mulberry32` pseudo-random generator with your numeric seed (`seed`), ensuring the exact same schema and seed always output identical rows across developers or CI runs before serializing to formatted JSON, RFC 4180 CSV, or SQL `INSERT` batches.
Step-by-Step Instructions
Choose your row count (`1–500`) and keep or randomize the numeric PRNG seed.
Customize column names, select field types (`UUID v4`, `Full Name`, `Email`, `Status`, etc.), or add/remove columns.
Switch output mode between `JSON`, `CSV`, and `SQL` (with custom table name).
Copy or download the exported dataset immediately.
Key Features & Performance
- Deterministic Mulberry32 seeded PRNG yields identical snapshots across re-runs for reliable unit tests
- Supports 9 field types: UUID v4, full name, email, city, ISO date, USD amount, IPv4, boolean, and status
- Instant single-click output switching between JSON array, RFC-escaped CSV, and batched SQL `INSERT INTO` statements
- Supports up to 500 rows directly inside the browser with sub-millisecond rendering
- 100% synthetic data: zero chance of exposing real user records during demos or screenshots
Methodology, Formulas & Technical Standards
Synthetic UUIDs format deterministic random 128-bit nibbles with standard version `4` and variant `8/9/a/b` bits. CSV serializer quotes fields containing commas, quotes, or line breaks, while SQL serializer safely escapes single quotes (`''`).
Common Real-World Use Cases
- Seeding local PostgreSQL/SQLite tables during backend schema migrations
- Populating UI pagination, sorting, and chart demos without touching production databases
- Generating identical CSV import fixtures across QA team members using a shared seed
- Creating quick JSON mocks for React/Vue prototypes
Worked Examples & Edge Cases
Seeded Reproducibility
Rows: 12 | Seed: 2026 | Columns: id (uuid), role (enum: admin,editor,viewer)
Produces identical 12 JSON/CSV/SQL records every time seed 2026 is used
Share the seed number in PR comments instead of huge JSON fixture pastes.
SQL Batch Export
Format: SQL | Table: users
INSERT INTO users (id, full_name, email) VALUES (...);
Strings, booleans, and numbers are typed cleanly for standard SQL engines.
Limitations, Scope & Accuracy Notes
- • Browser UI caps row generation at 500 rows to keep instant textarea editing responsive.
- • Generated emails and IPs use fictional domains (`@example.net`) and test-net ranges so accidental outbound notifications cannot deliver.
Local Browser Processing & Privacy
Traditional web converters upload your documents, snippets, or media files to remote servers. Seeded Mock Dataset Generator (JSON / CSV / SQL) runs inside your browser tab:
Your input text and files are processed locally without uploading payloads to a processing server.
Microdeck does not log or store the files or inputs you paste into tool workspaces (normal static hosting/CDN requests still load page assets).
Working outputs stay in your browser tab and are released when you reset the tool or close the tab.
Frequently Asked Questions
Why does changing the seed regenerate every field deterministically?
Using a seeded Mulberry32 PRNG instead of `Math.random()` lets you reproduce exact datasets reliably in automated tests or teammate demos.
How do Enum columns work?
Select `Enum List` and enter comma-separated choices like `admin,editor,viewer` — each row samples one value pseudo-randomly.
Does SQL output escape single quotes inside names or text?
Yes. Standard SQL single-quote doubling (`''`) and unquoted numbers/booleans are applied automatically.
Is any generated dataset sent to an analytics server?
No. Generation and export are 100% local.
Explore Related Micro-Tools
Instant CSV to JSON & Table Converter
Fast reactive CSV to JSON parser with zero ads and zero reloads
SQL Formatter & Query Beautifier
Pretty-print or minify SQL queries across PostgreSQL, MySQL, SQLite, T-SQL & BigQuery dialects
JSON Schema Draft Validator
Validate JSON payloads against JSON Schema rules (`required`, `type`, `enum`, `format`, `additionalProperties`, numeric/string bounds)