Restlastic

Your restful APIs with zero coding in less than 30 seconds...

I spend so many time googling for a easy way to have a free full feature REST API without finding what i want, so i decided to create restlastic to get the ideas from json-server, the modularity of sails.js, the scalability and features of elasticsearch and kibana.

See the 5min video presentation

GET /restlastic?price=free
{
    features : {
    create : "Your Restful APIs in less than 30 sec",
    code : "0 line of code",
    enjoy : "Pagination, search, filters, fulltext,...",
    generate : "Instantly swagger, postman, mocha",
    dashboard : "Realtime graph with kibana",
    scalable : "To infinity and beyond",
    customise : "based on node.js, sails.js, elasticsearch"
  },
  tagline: "You Know, for API"
}

Example

Create a products API (/data/v1/products)

curl -iX POST \
    -H 'Content-Type: application/json; charset=utf-8' \
    -d '{ "name":"banana", "status":"available", "price":12 }' \
    'http://api.restlastic.com/data/v1/products'

Now you got

GET, POST , PUT, PATCH, DELETE actions available
Pagination, fulltext search, filters operators, fields filter, subressources, hierarchical json...
Automatically managed fields : id, creation_date and modification_date, etag
Swagger documentation template with data example in swagger editor
Postman request template
Access to kibana for dashboarding your data
Modular and scalable on premise solution with restlastic, sails.js, elasticsearch, kibana

Also when doing requests, its good to know that

Your request body JSON should be object enclosed, just like the GET output. (for example {"name": "Foobar"})
Id values are not mutable. Any id value in the body that differ from id in url will send an error
A POST, PUT or PATCH request should include a Content-Type: application/json header to use the JSON in the request body. Otherwise it will result in a 200 OK but without changes being made to the data.

Install

You can install Restlastic with docker-compose, npm (if elasticsearch already installed) or use cloud demo beta

with docker

It will install restlastic, elasticsearch and kibana

$> git clone https://github.com/restlatic/restlatic.git
$> cd restlastic
$> docker-compose up -d

Then use http://localhost (api), http://localhost:9200 (elasticsearch), http://localhost:5601 (kibana)

with npm

You must have elasticsearch already installed

$> npm install restlastic
$> npm start

Then use http://localhost:1337

Use the beta online demo

/!\ All data are deleted each day

Use api.restlatic.com, elastic.restlatic.com, kibana.restlatic.com

Routes

Based on the previous POST command, here are all the default routes. You can also add your own routes using sails.js routes.

Plural routes

 GET sample/v1/products
 GET sample/v1/products/search?q=
 GET sample/v1/products/:id
 POST sample/v1/products
 POST sample/v1/products/:id
 PATCH sample/v1/products/:id
 PUT sample/v1/products/:id
 DELETE sample/v1/products/:id
 GET sample/v1/products/swagger.yaml
 GET sample/v1/products/postman.json

Partial data

Add fields= to retrieve only some fields (Use . to access deep properties)

  GET sample/v1/products/:id?fields=id,name,creation_date
  GET sample/v1/products?fields=id,name
  GET sample/v1/products?fields=adress
  GET sample/v1/products?fields=adress.locality

Filter

Use . to access deep properties

GET sample/v1/products?title=elastifull&author=clodio
GET sample/v1/products?id=1,2
GET sample/v1/products?id=1&id=2
GET sample/v1/products?author.name=clodio

Slice

Add start_index and/or count (pagination and total_results will be included in response with next and previous link)

GET sample/v1/products?start_index=20
GET sample/v1/products?start_index=20&count=3
 {
   "data":[...],
   "paging": {
     "total_results":30,
     "prev": "http://api.restlastic.com/sample/v1/products?start_index=17&count=3",
     "next": "http://api.restlastic.com/sample/v1/products?start_index=23&count=3",
   }
 }

Operators

Add _gte,_lte,_gt,_lt for getting a range

GET sample/v1/products?price_gte=10&price_lte=20
GET sample/v1/products?creation_date_gte=now-1d/d
GET sample/v1/products?creation_date_gte=now-1d
GET sample/v1/products?creation_date_gte=2014-06-18T23:59:59Z

Add _ne to exclude a value

GET sample/v1/products?price_ne=20

add _exists=true to find records with the field, _exists=false to find records without fields

GET sample/v1/products?address.locality_exists=true
GET sample/v1/products?address.locality_exists=false

Add _like to filter using like

GET sample/v1/products?title_like=server*

Add _prefix to filter with prefix

GET sample/v1/products?name_prefix=cl

Add _regex to filter with RegExp

GET sample/v1/products?name_regex=clo.?dio

Add _fuzzy to filter with fuzzy search

GET sample/v1/products?votes_fuzzy=2 --> votes : 1,2,3
GET sample/v1/products?name_fuzzy=cladio --> name : clodio
GET sample/v1/products?votes_fuzzy=cldio --> name : clodio

Full-text search

Add /search?q= to do a full search on all fields

GET sample/v1/products/search?q=*c

Relationships

Ressources can be linked to others with a _, in this case subRessource will have a products_id field with the value :id

GET sample/v1/products/:id/_subRessource
GET sample/v1/products/:id/_subRessource/search?q=
GET sample/v1/products/:id/_subRessource/:subRessource_id
POST sample/v1/products/:id/_subRessource
POST sample/v1/products/:id/_subRessource/:subRessource_id
PATCH sample/v1/products/:id/_subRessource/:subRessource_id
PUT sample/v1/products/:id/_subRessource/:subRessource_id
DELETE sample/v1/products/:id/_subRessource/:subRessource_id

Dates

creation_date and modification_date are automatically managed with RFC3369

{
  "creation_date": "2014-06-18T23:59:59Z",
  "modification_date": "2014-06-18T23:59:59Z",
}

You can use now, now-1d, now+1d/d,... when searching/filtering (see elastic.co)

GET sample/v1/products?creation_date_gte=now-1d

Etag

All ressources have a version, you can use it to cache data or to modify a specific version

GET sample/v1/products/1
Headers :
  etag: 123456
Body:
  {
    "id":"1",
    "name":"banana",
    "etag":"123456"
  }

--> send a etag header and a etag inside the result

you can use header If-None-Match to retrieve data only if modified, to modify data only if not modified

Automatic or managed data definitions with elasticsearch

You can create a new ressource or a new field easily by calling POST (ex : /sample/v1/products). It will create an index in elasticsearch named sample_products. If you want to manage precisely the types of your fields, you must use [elasticsearch mapping(https://www.elastic.co/guide/en/elasticsearch/reference/current/mapping.html). By default, all the fields will considered as string, interger and dates depending on the first POST. If you want to change a type or delete a field you must delete the index before changing the mapping.

Retrieve mapping

curl -XGET http://localhost:9200/sample_products/_mapping/products

create index

curl -XPUT 'http://localhost:9200/sample_products/'

create mapping

curl -XPUT 'http://localhost:9200/sample_products/_mapping/products' \
-d'{
 "products": {
   "properties":
     {
      "creation_date":{
         "type":"date","format":"strict_date_optional_time||epoch_millis"
      },
      "etag":{
         "type":"long"
      },
      "id":{
         "type":"string"
      },
      "modification_date":{
         "type":"date","format":"strict_date_optional_time||epoch_millis"
      },
      "name":{
         "type":"string"
      },
      "price":{
         "type":"long"
      },
      "status":{
         "type":"string"
      }
    }
  }
}'

Business Intelligence

Since data are stored in ELK you can make graphs with kibana, and use specific elastic request (aggregations,...). to use data you must configure an index pattern corresponding to your data (/sampple/v1/products will have sample_products index)

see kibana for more information

Swagger

You can get a swagger template of your Restull API

Open directly in swagger editor http://editor.swagger.io/#/?import=http://api.restlastic.com/sample/v1/products/swagger.yaml&no-proxy

or download the file

GET http://api.restlastic/sample/v1/products/swagger.yaml

Postman

You can get a postman template of the Restull API

GET http://api.restlastic.com/sample/v1/products/postman.json

Generate random data

You can easily create data programmatically.

  //import-data.js
  var request = require('request');

  var dns="http://localhost:1337/sample/v1/users/";
  var body={};

  // Create 1000 users
  for (var i = 0; i < 1000; i++) {
    body = { id: i, name: 'user' + i };
    request.put({
        url: dns + i,
        body: body,
        rejectUnauthorized: false,
        json: true
      }, function (err, res, body) {
        if (err) {
          console.log(err);
        }
      }
    );
  }
  // Then launch $> node import-data.js

Tip use modules like faker, casual or chance to create random semantic data.

Add routes

You can add your own routes using sails.js routes in /config/routes.js file.

License

MIT - Restlastic

Credits

json-server
Homepage based on Bootstrap and Bootswatch
Icons from Font Awesome
Web fonts from Google

clodio/restelastic