mocker-data-generator

A simplified way to generate massive mock data based on a schema, using the awesome fake/random data generators like (Faker-JS, ChanceJs, CasualJs and RandExpJs), all in one tool to generate your fake data for testing.

Now the library has been migrated 100% to typescript typing are included.

You can test online here: https://danibram.github.io/mocker-data-generator/

Install the module and the awesome generator you want: npm install mocker-data-generator @faker-js/faker

Import it

var mocker = require('mocker-data-generator').default // (vanilla way)
var { faker } = require('@faker-js/faker')


// or

import mocker from 'mocker-data-generator' // (ES6 or Typescript way)
import { faker } from '@faker-js/faker'

Then use it:

var user = {
    firstName: {
        faker: 'person.firstName()'
    },
    lastName: {
        faker: 'person.lastName()'
    },
    country: {
        faker: 'location.country()'
    },
    createdAt: {
        faker: 'date.past()'
    },
    username: {
        function: function () {
            return (
                this.object.lastName.substring(0, 5) +
                this.object.firstName.substring(0, 3) +
                Math.floor(Math.random() * 10)
            )
        }
    }
}
var group = {
    description: {
        faker: 'lorem.paragraph()'
    },
    users: [
        {
            function: function () {
                return this.generators.faker.helpers.arrayElement(this.db.user)
                    .username
            },
            length: 10,
            fixedLength: false
        }
    ]
}
var conditionalField = {
    type: {
        values: ['HOUSE', 'CAR', 'MOTORBIKE']
    },
    'object.type=="HOUSE",location': {
        faker: 'location.city()'
    },
    'object.type=="CAR"||object.type=="MOTORBIKE",speed': {
        faker: 'number.int()'
    }
}

// Using traditional callback Style

mocker()
    .addGenerator('faker', faker)
    .schema('user', user, 2)
    .schema('group', group, 2)
    .schema('conditionalField', conditionalField, 2)
    .build(function (error, data) {
        if (error) {
            throw error
        }
        console.log(util.inspect(data, { depth: 10 }))

        // This returns an object
        // {
        //      user:[array of users],
        //      group: [array of groups],
        //      conditionalField: [array of conditionalFields]
        // }
    })

// Using promises

mocker()
    .addGenerator('faker', faker)
    .schema('user', user, 2)
    .schema('group', group, 2)
    .schema('conditionalField', conditionalField, 2)
    .build()
    .then(
        data => {
            console.log(util.inspect(data, { depth: 10 }))
            // This returns an object
            // {
            //      user:[array of users],
            //      group: [array of groups],
            //      conditionalField: [array of conditionalFields]
            // }
        },
        err => console.error(err)
    )

// Synchronously

// This returns an object
// {
//      user:[array of users],
//      group: [array of groups],
//      conditionalField: [array of conditionalFields]
// }
var data = mocker()
    .addGenerator('faker', faker)
    .schema('user', user, 2)
    .schema('group', group, 2)
    .schema('conditionalField', conditionalField, 2)
    .buildSync()

console.log(util.inspect(data, { depth: 10 }))

NOTE: For the demo above you will also need to import util i.e. var util = require('util') or import util from 'util'

Data generation goes with model based composed by generators, the generators can have access to the data generated and to the entity generated. Generators run synchronously, take care of the related entities!!

• addGenerator(name, generator, runFn): Add a new generator, now generators are not included in the package, so you have to add whathever generator you want.

  • name (String): key of the generator this should match with the key in the schema.
  • generator (library): Library if the generator
  • runFN (function) (Optional) (generator: any, input: any) => any : It compose a exec function for the selected generator, only for custom generators, see randexp in the examples folder

• schema(name, schema, generationType): Add a new schema, you must specify this params:

  • name (String): Name of the schema.
  • schema (JSON): The schema you define
  • generationType (integer or JSON): In this field you specify how you will generate this schema. 3 ways:
    • Integer to specify how many objects of this schema you want.
    • JSON with this object {max: '<maximunValues>'} you can also optionally pass min {max: '<maximunValues>', min: '<minimumValues>', this will generate a range of objects of this schema, between (0 and max) or (min and max) randomly.
    • JSON with this object {uniqueField: '<yourUniqueField>'} this means that this field () is an array and you want to generate entities with this unique values

• reset(): Clean the internal DB.

• restart(): Clean the internal DB and all the schemas inside.

• build(callback): This methods start to produce the data and wrap it to the callback function, the callback funtion have 2 parameters, error and data generated.

• buildSync(): Synchronous version of build(callback). Returns generated data or throws an error.

• Normal string key: indicates the key.
• Commaseparated string key: indicates that there is a conditional, before the comma you must specify a conditional (you have all level fields generated in this moment), then you must specify the field if the conditional is true see the example.

• static: For fixed fields

  {
      static: 'hello im fixed field'
  }

• self: get himself object, and evaluate the string, so you can get calculated fields.

  • eval (Optional): Also now you can pass, eval to true, to use like in versions < 2.6.0

  {
      self: 'id'
  }   //will get the id of the generated entity
  
  {
      self: 'id',
      eval: true
  }   // will get the first user id

• db: get the db, and evaluate the string, so you can access to this entities.

  • eval (Optional): Also now you can pass, fast to true, eval to true, to use like in versions < 2.6.0

  {
      db: 'user[0].id'
  }   // will get the first user id
  
  {
      db: 'user[0].id',
      eval: true
  }   // will get the first user id

• eval: evaluate the current string, remember that all context are injected (generators, db the actual object). With this eval field, you must pass an exactly JSON syntax:

  {
      eval: 'object.id'
  }
  
  // OR
  
  {
      eval: 'db.user[0]'
  }
  
  // OR
  
  {
      eval: 'faker.lorem.words()'
  }

• hasOne: You can pass 2 parameters:

  • hasOne: the name of the related entity, get one random.

  • get (Optional): String that will be evaluated over the random related entity.

  • eval (Optional): Only affects if get is passed, the get param only support dotted paths, with eval=true you can use an eval string, this impacts on the performance

        {
            hasOne: 'user'
        }   // this populate the field with one random user
    
        // OR:
    
        {
            hasOne: 'user',
            get: 'id'
        }   // this populate the field with one id of a random user
    
    
        // OR:
    
        {
            hasOne: 'user',
            get: 'id',
            eval: true
        }   // this populate the field with one id of a random user with eval string

• hasMany: You can pass 4 parameters:

  • hasMany: the name of the related entity, get one random.

  • amount (Optional): Fixed number of related entities to get.

  • min (Optional): Minimum entities to get, buy default is 1, if you want the chance to have empty arrays please specify min to 0.

  • max (Optional): Maximum entities to get.

  • get (Optional): String that will be evaluated over the random related entity.

  • eval (Optional): Get will only support dotted paths, with eval= true you can get from an evaluable string

  • unique (Optional): hasMany will get unique values from the entity (Make sure that you have many unique data in the source)

        // In this case we will get 1 user (hasMany)
        {
            hasMany: 'user'
        }   // this populate the field with one random user
    
        // OR:
    
    
        {
            hasMany: 'user',
            amount: 1, //optional
        }   // In this case we will get 1 (amount) user (hasMany)
    
        // OR:
    
        {
            hasMany: 'user',
            max: 3 // optional
        }   // In this case we will get as max 3 (max) users (hasMany)
    
        // OR:
    
    
        {
            hasMany: 'user',
            min: 1 //optional
            max: 3 //optional
        }   // In this case we will get bettween min 1 (min) and max 3 (max) users (hasMany)
    
        // OR:
    
        {
            hasMany: 'user',
            get: 'id'
        }   // In this case we will get the id (get) from 1 random user (hasMany)

• incrementalId: For incremental numeric ids, pass the start number to increment. If you put incrementalId = true it takes from 0 the ids.

  {
      incrementalId: 0
  }

• function: No params are passed, only context (this), in this you have {db, object, faker, chance}, and you can use faker or chance functions, object (the specified model), db (actual data generated)

        { function: function(){
  
            // this.db
            // this.object
            // this.generators
  
            return yourValue
        } }
  
        // OR:
  
        { function(){
  
            // this.db
            // this.object
            // this.generators
  
            return yourValue
        } }

• [Array]: you can pass an array that indicates an array of data you can create, passing in the first field the generator (function, faker, or array(not Tested)), and in the second field pass a config object (length, fixedLentgh)

  • length: to know how many values

  • fixedLength (Optional): true to create always same amount of values in the array, false to generate a random number between 0 and 'length' value. False by default.

  • concat (Optional): An stringuified array ex: '[object.id, db.users.id]'. This should be an evaluable string to concat with the array that are generating. Also takes in mind that if you have a fixedLength, should not increase the length.

  • strictConcat (Optional): true to remove duplicates in the concatenated string array, when it is calculated. False by default.

    [{
       // Any generator
           // Faker
       [name of the generator injected]: 'path inside the generator'
           // If faker is injected with .addGenerator('faker', faker) then you can use:
       faker: 'helpers.arrayElement(db.users).userId'
           // Function that has included index, length and self that refers at the actual array generation
       function: function (index, length, self){ return /**/ }
    
       // Array config
       length: 10,
       fixedLength: true
    
       // Concat
       concat: '[db.users[0].userId, db.users[1].userId]'
       strictConcat: true
    }]

It happens!Mocker now is independant of the generators so I hope this will give you more freedom. This is an example that you can check on the examples folder, where 2 generators are used:

    var mocker = require('../build/main').default
    var { faker } = require('@faker-js/faker')
    var Randexp = require('randexp')
    var util = require('util')

    var user = {
        firstName: {
            faker: 'person.firstName()'
        },
        notes: {
            randexp: /hello+ (world|to you)/
        }
    }

    mocker()
        .addGenerator('faker', faker)
        .addGenerator('randexp', Randexp, function (generator, input) {
            return new generator(input).gen()
        })
        .schema('user', user, 2)
        .build(function (error, data) {
            if (error) {
                throw error
            }
            console.log(util.inspect(data, { depth: 10 }))
        })

• [virtual]: Boolean, if you pass this option, this mean that this field will not appear at the output entity. But you can use during the generation.

    {
        // Any generator
            // Faker
        [name of the generator injected]: 'path inside the generator'
            // If faker is injected with .addGenerator('faker', faker) then you can use:
        faker: 'helpers.arrayElement(db.users).userId'
            // Static
        static: 'any static field'
            // Function
        function: function (){ return /**/ }

        // With the virtual option
        virtual: true

    }

Initialize mocker with the config, and then generate any entity with promises style, use generate function that accepts the name of the model and the amount of data to generate. Like the example:

mocker()
    .addGenerator('faker', faker)
    .schema('user', user, 2)
    .schema('group', group, 2)
    .schema('conditionalField', conditionalField, 2)
    .build(function(err, data) {
        console.log(util.inspect(data, { depth: 10 }))
        // This returns an object
        // {
        //      user:[array of users],
        //      group: [array of groups],
        //      conditionalField: [array of conditionalFields]
        // }
    })

You can also pass instead of the number, an object with the a config, from now {uniqueField}. If this field exists tells to the generator that instead of init a fixed length of data, generate an amount of data depending of the values of the field you will specify. You have 2 way to deal with this, check the examples See the output of this example:

//
// First way, using an 'values' embedded object
//

var cat = {
    name: {
        values: ['txuri', 'pitxi', 'kitty']
    }
}
var m = mocker()
    .addGenerator('faker', faker)
    .schema('cat', cat, 10)
    .schema('cat2', cat, { uniqueField: 'name' })
    .build(function(err, data) {
        console.log(util.inspect(data, { depth: 10 }))
    })

//
// Second way, without 'values' embedded.
//

var cat = {
    name: ['txuri', 'pitxi', 'kitty']
}
var m = mocker()
    .addGenerator('faker', faker)
    .schema('cat', cat, 10)
    .schema('cat2', cat, { uniqueField: 'name' })
    .build(function(err, data) {
        console.log(util.inspect(data, { depth: 10 }))
    })

In version >= 2.6.0, eval option was introduced to run mocker-data-generator like olders versions, so by default is running without eval: faker, chance, casual, hasMany, hasOne, db and self. This means that this methods loose habilities, when eval is not passed, but this are the speed results with eval active (old way) and without (new way)

faker eval old:  0.969ms
faker now:       0.215ms
chance eval old: 0.559ms
chance now:      0.099ms
casual eval old: 0.360ms
casual now:      0.026ms

You can visit the repo url here: https://github.com/danibram/mocker-api-tester

Or visit the api directly: https://mocker-api.herokuapp.com/

Run npm install;npm run dev to watch the project, webpack compile the code automatically. Run npm build to build the normal and minified version.

json-schema-faker is awesome and works really nice, but i need a simplified and fast way to generate mock data for my projects, so i created this.

I couldn't do this without this awesome libraries, so thanks to all:

• Faker: [https://github.com/faker-js/faker]
• Chance: [https://github.com/victorquinn/chancejs]
• Casual: [https://github.com/boo1ean/casual]
• RandExpJs: [https://github.com/fent/randexp.js]
• typescript-starter: [https://github.com/bitjson/typescript-starter]

Licensed under the MIT license. 2022