Skip to main content
Faker is a special datasource connector that generates realistic mock data for testing and development. It’s useful for:
  • Building demos without real data
  • Testing your Evidence app locally
  • Creating sample datasets for documentation

​
Installation

The Faker datasource is included with Evidence by default.

​
Configuration

Faker doesn’t require database credentials. Instead, you define data schemas using YAML configuration files.

​
Basic Setup

1

Add Faker Source

Navigate to Settings in your Evidence app (localhost:3000/settings) and add a Faker datasource.
2

Create Schema Files

Create .yaml files in your sources/faker/ directory to define your mock data schemas.
3

Run Sources

Generate the mock data:
npm run sources

​
Schema Definition

Define your mock data structure in YAML files. Faker uses the @faker-js/faker library. Example: sources/faker/users.yaml 
tables:
  users:
    rows: 100
    columns:
      - name: user_id
        type: number
        generator: datatype.number
        min: 1
        max: 10000
      
      - name: first_name
        type: string
        generator: person.firstName
      
      - name: last_name
        type: string
        generator: person.lastName
      
      - name: email
        type: string
        generator: internet.email
      
      - name: signup_date
        type: date
        generator: date.past
        years: 2
      
      - name: is_active
        type: boolean
        generator: datatype.boolean

​
Available Generators

Faker provides hundreds of data generators organized by category: Person Data:
  • person.firstName - First names
  • person.lastName - Last names
  • person.fullName - Full names
  • person.jobTitle - Job titles
Internet Data:
  • internet.email - Email addresses
  • internet.url - URLs
  • internet.userName - Usernames
  • internet.ipv4 - IP addresses
Commerce Data:
  • commerce.productName - Product names
  • commerce.price - Prices
  • commerce.department - Departments
Date/Time:
  • date.past - Past dates
  • date.future - Future dates
  • date.recent - Recent dates
  • date.between - Dates between two dates
Numbers:
  • datatype.number - Random numbers
  • datatype.float - Random floats
  • datatype.boolean - Boolean values
Location:
  • location.city - City names
  • location.country - Countries
  • location.streetAddress - Street addresses
See the Faker.js API documentation for the complete list of generators.

​
Advanced Features

​
Seeded Data

Use the same seed value to generate consistent, reproducible data: 
seed: 12345
tables:
  products:
    rows: 50
    columns:
      - name: product_name
        type: string
        generator: commerce.productName

​
Relationships Between Tables

Create related data across multiple tables: 
tables:
  orders:
    rows: 500
    columns:
      - name: order_id
        type: number
        generator: datatype.number
      
      - name: user_id
        type: number
        generator: datatype.number
        min: 1
        max: 100  # References user_id range from users table
      
      - name: order_date
        type: date
        generator: date.recent
        days: 90

​
Custom Value Ranges

Control the output of generators with parameters: 
columns:
  - name: price
    type: number
    generator: commerce.price
    min: 10
    max: 1000
    dec: 2  # Decimal places
  
  - name: quantity
    type: number
    generator: datatype.number
    min: 1
    max: 100

​
Example Schemas

​
E-commerce Dataset

seed: 42
tables:
  customers:
    rows: 200
    columns:
      - name: customer_id
        type: number
        generator: datatype.number
        min: 1000
        max: 9999
      - name: name
        type: string
        generator: person.fullName
      - name: email
        type: string
        generator: internet.email
      - name: city
        type: string
        generator: location.city
      - name: registration_date
        type: date
        generator: date.past
        years: 3
  
  orders:
    rows: 1000
    columns:
      - name: order_id
        type: number
        generator: datatype.number
        min: 10000
        max: 99999
      - name: customer_id
        type: number
        generator: datatype.number
        min: 1000
        max: 9999
      - name: order_date
        type: date
        generator: date.recent
        days: 365
      - name: total_amount
        type: number
        generator: commerce.price
        min: 20
        max: 500
      - name: status
        type: string
        generator: helpers.arrayElement
        values: ['pending', 'processing', 'shipped', 'delivered', 'cancelled']

​
Use Cases

Demo Applications

Build impressive demos without needing real data or database access

Testing

Generate large datasets to test performance and edge cases

Development

Develop locally without connecting to production databases

Documentation

Create consistent example datasets for tutorials and guides

​
Troubleshooting

Make sure you’ve run npm run sources after creating or modifying your schema files. The Faker plugin generates static Parquet files that Evidence uses.
Check the Faker.js API docs to verify the generator name. Generator names are case-sensitive and follow the format category.method.
Increase the rows value in your schema file. Note that very large datasets (>100,000 rows) may take longer to generate.

Build docs developers (and LLMs) love

Get started for freeTalk to us