Seedman

Seedman is a seeder allows you create data for any models or collections in any database engine like MongoDB, MySQL, Postgres, etc. (Current version: Only Supports MongoDB).

Install

Using npm:

npm install --save seedman

Note: You can install it globally

Features

• Generate data from any model including relations (one-to-one, one-to-many).
• Connection for any database engine (SQL or NoSQL): MongoDB, Postgres, Mysql. (Current Version: Only Supports MongoDB)

Usage

IMPORTANT: This seeder will erase all database data before start. For avoid this, set reset field in the configuration file to false.

For usage it's neccesary define a JSON files defining properties such:

• name: Model name in the DB for populate
• properties: Model properties or fields.
• count: Amout data for populate in the model

Example:

user.json

{
  "name": "User",
  "properties" : {
      "firstName": "string",
      "lastName": "string",
      "phone": "string",
      "address": "string"
  },
  "count": 20
}

post.json

{
  "name": "Post",
  "properties" : {
      "title": "string",
      "tags": "array",
      "visits": "number"
  },
  "count": 20
}

For default, theses files must be inside a folder named seeders

Also it's important to define a configuration file in otherwise will be use a default config for the driver choosen (mongodb). It should be in the same path that the command is execute (It shouldn't be inside a folder).

seeder.config.json

{
    "user": "",
    "password": "",
    "host": "localhost",
    "port": "27017",
    "dbname": "dbname"
}

Seed props

Props	Description
name	Model name what is definied in the database
properties	Model properties
count	Amount data for populate in the model

Properties

It is possible define some type for each property.

• string
• array
• number

Seedman use faker.js for generate fake data and identify the values according to the name of the property.

For the User example:

user.json

{
  "name": "User",
  "properties" : {
      "firstName": "string",
      "lastName": "string",
      "phone": "string",
      "address": "string"
  },
  "count": 20
}

firstName, lastName, phone and address. Theses properties will use values according to their context:

{
    "_id" : 0,
    "firstName" : "Shawna",
    "lastName" : "Lind",
    "phone" : "861.700.7122",
    "address" : "46153 Earline Meadow"
}

But, there are some properties name, it won't identify according to the name and generate a random string or it depending of the type defined in the seeder can be a number or string array (length = 5).

Relations

It's possible define a relation between models:

• One-to-One
• One-to-Many

For define a new relation it's important define a new field referencing a seed, this field must be the seed name with the next properties:

• type: model
• relation: hasOne or hasMany.
• count: Records amount to create
• fkId (Optional): Custom foreign key, for default it's modelId

Example:

user.json

{
   "name": "User",
   "properties" : {
     "lastName": "string",
     "phone": "string",
     "address": "string",
     "Post": {
         "type": "model",
         "relation": "hasMany"
     }
   },
   "count": 1
}

post.json

{
    "name": "Post",
    "properties" : {
        "description": "string"
    },
    "count": 1
}

Note: If the seed to reference (Post) has count field also will generate independent records, in order to avoid this is important remove that field.

CLI

$ seedman -h
 
  Usage: seedman 
 
  A CLI interface for Seedman
 
  Options:
 
    -h, --help                    output usage information

Tests

npm run test