iLoveDevTool
Back to Tools

Mock Data Generator

My tools
No tools yet

Loading...

Generate mock test data online (JSON/CSV/SQL)

  1. 1

    Define fields

    Fill field names and pick types row by row (numeric types take ranges), or press Example to load the preset six-field user-table schema; rows can be added or removed.

  2. 2

    Pick rows and format

    Rows 1-1000; switch output between JSON/CSV/SQL; pick the dataset toggle for your target system — generation updates live.

  3. 3

    Copy and use

    One-click copy: JSON goes straight into fixtures, CSV into spreadsheets, SQL includes table creation and inserts ready to run; for prettier SQL use the SQL Formatter tool.

Features Overview

Mock data generator online: define fields (11 types — name, username, email, phone, city, company, integer, decimal, date, UUID, boolean), batch-generate test data, and export as JSON, CSV, or SQL (CREATE TABLE + INSERTs). Chinese and English datasets switchable; fully in-browser with no signup and no limits.

Tool Description

Free online mock data generator. 11 field types (name, email, phone, city, company, numbers, dates, UUID and more), 1-1000 rows batch generation, output as JSON, CSV, or SQL create-and-insert statements; Chinese dataset included, fully in-browser with no signup.

Last updated

Capabilities Checklist

  • 11 safe field types: names (Chinese/English), usernames (in-pool pinyin/English), email, phone (shape-real only), city, company, integers and decimals (configurable ranges), dates (valid dates in 2020-2025), UUID, and boolean — covering mainstream test-data needs, with no sensitive ID types.
  • Three output formats: JSON (array of objects, Chinese preserved), CSV (RFC 4180 escaping, numbers unquoted to keep types), and SQL (CREATE TABLE plus INSERTs batched per 100 rows, with column-name sanitization and reserved-word avoidance) — no second hop needed.
  • Chinese dataset: real-looking Chinese names composed from surname and given-name pools, cities and company names from common pools, and domestic email domains — Chinese systems no longer get awkward English faker output.
  • Live schema editing: add/remove field rows, switch types, expand numeric ranges; a preset six-field user-table schema loads in one click; 1-1000 rows.
  • Uniform field-name rules: empty names become field_N and duplicates get suffixes (consistent across all three formats); everything generates locally in the browser, nothing uploaded.

User feedback

Your feedback will be reviewed (and redacted if needed) before it appears publicly. Please do not include emails, passwords, or other personal information.

Your feedback helps us do better ❤️

Related Tools

JSON Formatter Online

Format, validate, and explore JSON in your browser with a tree view, exact error lines, and JSONPath, with nothing uploaded

UUID Generator

Generate UUID v4/v7 online in bulk

SQL Formatter

Format, beautify, and minify SQL online across 20 dialects

Related Guides

How to Make Fake Test Data: Field Types and JSON/CSV/SQL in Practice

Where does test-environment data come from? This guide covers the common fake-data field types (name, email, phone, city, company, numbers, dates) and the shape each must preserve — why ID numbers should never be generated, why phones are only shape-real, and how to avoid February 30th — plus escaping and table-creation practice for JSON, CSV, and SQL output, ending with an online generator that turns a schema into usable test data.

Frequently Asked Questions

Which field types and output formats are supported?
11 field types: name, username, email, phone, city, company, integer, decimal, date, UUID, and boolean (integers/decimals take ranges; dates are valid dates within 2020-2025). Three output formats: JSON arrays, CSV (RFC 4180 escaping), and SQL (CREATE TABLE + batched INSERTs with column types derived from field types). Rows: 1-1000.
Do generated emails and phone numbers belong to real people?
They are randomly composed placeholders: emails combine word pools with common domains and may coincidentally match real addresses — for placeholder display only, do not send mail to them; phone numbers are random 11-digit strings in the 1[3-9] shape and may hit numbers in active use — for test placeholders only, do not call or text them.
Why no ID-card or bank-card numbers?
A safety boundary: ID numbers carry checksum algorithms — fake numbers generated to pass the checksum could survive real-name verification flows, a substantive abuse risk, so they are not offered. Fields like phone numbers have no verification scenario and are only shape-real, keeping the placeholder risk manageable within the safe set.
Can generated data be reproduced?
No seed support: the random source is the platform's crypto-grade randomness and every run differs — intentional, to avoid generated data becoming a trackable fingerprint. For reproducible fixtures (regression tests), save the output to a file or use a library with seed support (e.g. faker.js's seed API).
What is the rule for switching between Chinese and English data?
The "Chinese data" toggle controls the dataset: on, names/cities/companies draw from Chinese pools (usernames use in-pool pinyin); off, from English pools — the dataset is independent of the interface language, so an English UI can still generate Chinese data (and vice versa), switchable at any time.