API Stub

Simple, no database, temporary API for your prototype.

What is it ?

API Stub is a node.js based API server for prototyping. Idea was simple, sometimes I wanted to work on frontend app before I finish my backend API, or I often needed dummy data when testing out new data visualization designs. I've been using large JSON document as data stub, but I wanted something more quick to modify both schema and size of dataset.

Getting started

You will need node.js installed on your machine.

npm install api-stub --save-dev

// api.js
var API = require('api-stub');
var config = [{
  path:'/status',
  data: {status: true}
}]
var server = new API(config);
server.start(3000);
//then run the script `node api.js`

After you run the script - node api.js, http://localhost:3000/status will return following.

{ "status": true }

You can put as many endpoints as you like in config array.

Repeated dataset

If you need repeated dataset, you can specifying template field, and use placeholder text in data to set repeated amount.

var config = [{
  path:'/random',
  data: [/// info * 3 ///] // [templatename] * [how many to repeat]
  templates:{
    info:{
      message:'Hi there!'
    }
  }
}]

[{ "message": "Hi there!" },{ "message": "Hi there!" },{ "message": "Hi there!" }]

Dynamically generate data

Built in templates

There are few built in templates to generate data dynamically. (Check Templates section for more)

var config = [{
  path:'/id',
  data: [/// users * 3 ///]
  templates:{
    users:{
      id:{
        stub_type:'number',
        starting_num:1000,
        increments:1,
        type:'incremental'
      },
    }
  }
}]

[{ "id": 1000 },{ "id": 1001 },{ "id": 1002 }]

Custom functions

You can also assign your custom function to any key. When initializing API server, API Stub will creates deep copy of data & invoke any function assigned.

var config = [{
  path:'/random',
  data: {value: function(){Math.random();}}
}]

{ "value": 0.05823205946944654 } // random value everytime you start the serer

API

###server.start(port, [callback]) Start API server on specified port number ###server.stop([callback]) Stop API server ###server.profile() Returns status of the API server

• status: True if API server is running
• port: port number that API sever is using,
• running_since: timestamp of when the API server started,
• path : list of paths this API server is serving

{
  status: true,
  port: 3000,
  running_since: Fri May 15 2015 08:29:21 GMT-0400 (EDT),
  path: [ '/data', '/status' ]
}

Templates

There are 4 stub_type .

select

Select values from given choice.

property	type	value
type	string	'liner', 'random', or 'random_unique'
choice	array	array of choices
'liner' will select values from top the the array. 'random' will pick values randomly. 'random_unique' will pick values randomly but will not　draw same value twice (must provide enough choices if you are creating repeated dataset).

data:{
  stub_type:'select',
  type:'liner',
  choice:[100,200,300,400]
}

number

Numbers in set increments or randomly within specified range

property	type	value
type	string	'incremental' or 'random'
starting_num	number	starting value for 'incremental' type
increments	number	increment step for 'incremental' type
min_num	number	minimum value for 'random' type
max_num	number	maximum value for 'random' type
'incremental' will inclement values specified in increments. 'random' will pick random values between min_num and max_num.

id:{
  stub_type:'number',
  type:'incremental',
  starting_num:1000,
  increments:1
}

value:{
  stub_type:'number',
  type:'random',
  min_num:1000,
  max_num:5000
}

lipsum

Lorem Ipsum text

property	type	value
min_char_length	number	minimum character count
max_char_length	number	maximum character count

text:{
  stub_type:'lipsum',
  min_char_length:200,
  max_char_length:500,
}

datetime

parse & create datetime related text

property	type	value
type	string	'incremental' or 'random'
starting_datetime	string	starting value for 'incremental' type. Pass any string that Date.parse(str) can process
increments	string	increments for 'incremental' type in milliseconds (86400000 would be one day)
min_datetime	string	minimum date for 'random'. Pass any string that Date.parse(str) can process
max_datetime	string	maximum date for 'random'. Pass any string that Date.parse(str) can process
output_format	string	parsing method for Date object (see output_format table below)

timestamp:{
  stub_type:'datetime',
  type:'random',
  min_datetime:'2014-01-01',
  max_datetime:'2014-01-31',
  output_format:'toYMDString',
}

timestamp:{
  stub_type:'date',
  type:'incremental',
  starting_datetime:'2014-01-01',
  increments:86400000,
  output_format:'toTimeString',
}

output_format	return
toString	Fri May 15 2015 09:30:16 GMT-0400 (EDT)
toDateString	Fri May 15 2015'
toISOString	2015-05-15T13:30:16.411Z
toJSON	2015-05-15T13:30:16.411Z
toGMTString	Fri, 15 May 2015 13:30:16 GMT
toLocaleDateString	5/15/2015
toLocaleString	5/15/2015, 9:30:16 AM
toLocaleTimeString	9:30:16 AM'
toTimeString	09:30:16 GMT-0400 (EDT)'
toUTCString	Fri, 15 May 2015 13:30:16 GMT
toYMDString	2015-05-15

FAQ

Yet another API mock up tool huh?

Yep. I needed something I can generate unique data easily, quickly modify schema as I go, and avoid set up of database.

Why there is no <some feature > ?

I have been using this script as part of my dataviz prototyping. While this serves my purpose in what I do, I have not been using it for other applications (i.e consumer app). If you find some key features are missing, I'd love to hear about it.