mappersmith

What is mappersmith?

Mappersmith is a lightweight HTTP client library for JavaScript that simplifies the process of making HTTP requests. It provides a declarative way to define API endpoints and offers features like middleware support, request/response transformation, and retries.

What are mappersmith's main functionalities?

Declarative API Definition

Mappersmith allows you to define your API endpoints in a declarative manner. This example shows how to define a client with endpoints for fetching all users and fetching a user by ID.

const { forge } = require('mappersmith');

const api = forge({
  clientId: 'my-api',
  host: 'https://api.example.com',
  resources: {
    User: {
      all: { path: '/users' },
      byId: { path: '/users/{id}' }
    }
  }
});

api.User.all().then(response => console.log(response.data));

Middleware Support

Mappersmith supports middleware, allowing you to intercept and modify requests and responses. This example demonstrates how to log requests and responses using middleware.

const { forge, configs } = require('mappersmith');

configs.middleware = [
  (request) => {
    console.log('Request:', request);
    return request;
  },
  (response) => {
    console.log('Response:', response);
    return response;
  }
];

const api = forge({
  clientId: 'my-api',
  host: 'https://api.example.com',
  resources: {
    User: {
      all: { path: '/users' }
    }
  }
});

api.User.all().then(response => console.log(response.data));

Request/Response Transformation

Mappersmith allows you to transform requests and responses. This example shows how to transform the response to extract user names from the list of users.

const { forge } = require('mappersmith');

const api = forge({
  clientId: 'my-api',
  host: 'https://api.example.com',
  resources: {
    User: {
      all: { path: '/users', transform: (response) => response.data.map(user => user.name) }
    }
  }
});

api.User.all().then(userNames => console.log(userNames));

Retries

Mappersmith provides built-in support for retrying failed requests. This example configures the client to retry failed requests up to 3 times with exponential backoff.

const { forge, configs } = require('mappersmith');

configs.retry = { retries: 3, factor: 2, minTimeout: 1000 };

const api = forge({
  clientId: 'my-api',
  host: 'https://api.example.com',
  resources: {
    User: {
      all: { path: '/users' }
    }
  }
});

api.User.all().then(response => console.log(response.data));

Other packages similar to mappersmith

axios

Axios is a popular promise-based HTTP client for JavaScript. It offers a simple API for making HTTP requests and supports features like interceptors, request/response transformation, and automatic JSON parsing. Compared to Mappersmith, Axios is more widely used and has a larger community, but it does not provide a declarative way to define API endpoints.

fetch

Fetch is a built-in web API for making HTTP requests in modern browsers. It provides a low-level API for making requests and handling responses. While Fetch is more lightweight and has no external dependencies, it lacks the higher-level abstractions and features like middleware and retries that Mappersmith offers.

superagent

Superagent is a small, progressive HTTP request library for Node.js and browsers. It provides a flexible API for making HTTP requests and supports features like plugins, request/response transformation, and retries. Compared to Mappersmith, Superagent is more flexible but requires more manual setup for defining API endpoints.

README

Mappersmith

Mappersmith is a lightweight, isomorphic, dependency-free, rest client mapper for javascript. It helps you map your API to use at the client and/or server, giving you all the flexibility you want to customize requests or write your own gateways.

https://www.npmjs.com/package/mappersmith

Install

NPM

npm install mappersmith

Browser

Download the tag/latest version from the build folder.

Build from the source

Install the dependencies

npm install

Build

npm run build
npm run release # for minified version

Requiring in Node.js

var Mappersmith = require('mappersmith/node');

Usage

To create a client for your API, you will need to provide a simple manifest, which must have host and resources keys. Each resource has a name and a list of methods with its definitions, like:

var manifest = {
  host: 'http://my.api.com',
  resources: {
    Book: {
      all:  {path: '/v1/books.json'},
      byId: {path: '/v1/books/{id}.json'}
    },
    Photo: {
      byCategory: {path: '/v1/photos/{category}/all.json'}
    }
  }
}

You can specify an HTTP method for every API call, but if you don't, GET will be used. For instance, let's say you can save a photo:

...
Photo: {
  save: {method: 'POST', path: '/v1/photos/{category}/save'}
}
...

With the manifest in your hands, you are able to forge your client:

var Client = Mappersmith.forge(manifest)

And then, use it as defined:

// without callbacks
Client.Book.byId({id: 3})

// with all callbacks
Client.Book.byId({id: 3}, function(data, stats) {
  // success callback
}).fail(function() {
  // fail callback
}).complete(function() {
  // complete callback, it will always be called
})

Success callback arguments

The success callback will receive two arguments: the first one will be data, returned by your API; and the second one will be a stats object. The stats object hold information of the request, like the elapsed time between your call and callback execution.

Be aware that plugins hooked to Mappersmith can include more stats into this object, like CachedGateway which includes if the call got an cache hit or miss.

The default stats in the object are: url, params, timeElapsed and timeElapsedHumanized. Example:

{
  url: 'http://my.api.com/v1/books.json?language=en',
  host: 'http://my.api.com',
  path: '/v1/books.json?language=en',
  params: {language: 'en'},
  timeElapsed: 6.745000369846821,
  timeElapsedHumanized: '6.75 ms'
}

Parameters

If your method doesn't require any parameter, you can just call it without them:

Client.Book.all() // http://my.api.com/v1/books.json

Every parameter that doesn't match a pattern ({parameter-name}) in path will be sent as part of the query string:

Client.Book.all({language: 'en'}) // http://my.api.com/v1/books.json?language=en

Default parameters

It is possible to configure default parameters for your resources, just use the key params in the definition. It will replace params in the URL or include query strings, for example, imagine that our manifest has the method byYear in the resource Photo:

...
Photo: {
  byYear: {
    path: '/v1/photos/{year}.json',
    params: {year: new Date().getFullYear(), category: 'cats'}
  }
}
...

If we call it without any params and new Date().getFullYear() is 2015, it will generate the following URL:

Client.Photo.byYear();
// http://my.api.com/v1/photos/2015.json?category=cats

And, of course, we can override the defaults:

Client.Photo.byYear({category: 'dogs'});
// http://my.api.com/v1/photos/2015.json?category=dogs

Message Body

To send values in the request body (usually for POST or PUT methods) you will use the special parameter body:

Client.Photo.save({
  category: 'family',
  body: {year: 2015, tags: ['party', 'family']}
})

It will create a urlencoded version of the object (year=2015&tags[]=party&tags[]=family). If the body used is not an object it will use the original value. If body is not possible as a special parameter for your API you can configure it with another value, just pass the new name as the third argument of method forge:

var Client = Mappersmith.forge(manifest, Mappersmith.VanillaGateway, 'data')
...
Client.Photo.save({
  category: 'family',
  data: {year: 2015, tags: ['party', 'family']}
})

Processors

You can specify functions to process returned data before they are passed to success callback:

...
Book: {
  all:  {
    path: '/v1/books.json',
    processor: function(data) {
      return data.result;
    }
  }
}
...

Alternative host

There are some cases where a resource method reside in another host, in those cases you can use the host key to configure a new host or to disable the resolution.

var manifest = {
  host: 'http://new-host.com/api/v2',
  resources: {
    MyResouce: {
      all: {path: '/all.json'},
      byId: {path: '/{id}.json', host: 'http://old-host.com/api/v1'},
      other: {path: '{url}', host: false}
    }
  }
}

var Client = Mappersmith.forge(manifest);
Client.MyResource.all()
// http://new-host.com/api/v2/all.json

Client.MyResource.byId({id: 1})
// http://old-host.com/api/v1/1.json

Client.MyResource.other({url: 'http://host.com/other/'})
// http://host.com/other/

Compact Syntax

If you find tiring having to map your API methods with hashes, you can use our incredible compact syntax:

...
Book: {
  all: 'get:/v1/books.json', // The same as {method: 'GET', path: '/v1/books.json'}
  byId: '/v1/books/{id}.json' // The default is GET, as always
},
Photo: {
  // The same as {method: 'POST', path: '/v1/photos/{category}/save.json'}
  save: 'post:/v1/photos/{category}/save'
}
...

A downside is that you can't use processor functions with compact syntax.

Gateways

Mappersmith allows you to customize the transport layer. There are gateways for browser and server (Nodejs). You can use the default Mappersmith.VanillaGateway (client only), the included Mappersmith.JQueryGateway (client only), NodeVanillaGateway (server only) or write your own version. Check the list of available gateways at the bottom of the readme.

How to write one?

var MyGateway = Mappersmith.createGateway({
  get: function() {
    // you will have:
    // - this.url
    // - this.host
    // - this.path
    // - this.params
    // - this.body
    // - this.opts
  },

  post: function() {
  }

  // and other HTTP methods
})

How to change the default?

Just provide an object created with Mappersmith.createGateway as the second argument of the method forge:

var Client = Mappersmith.forge(manifest, Mappersmith.JQueryGateway)

Specifics of each gateway

You can pass options for the gateway implementation that you are using. For example, if we are using the Mappersmith.JQueryGateway and want one of our methods to use jsonp, we can call it like:

Client.Book.byId({id: 2}, function(data) {}, {jsonp: true})

The third argument is passed to the gateway as this.opts and, of course, the accepted options vary by each implementation. The default gateway, Mappersmith.VanillaGateway, accepts a configure callback:

Client.Book.byId({id: 2}, function(data) {}, {
  configure: function(request) {
    // do whatever you want
  }
})

Global configurations and URL matching

Imagine that you are using Mappersmith.JQueryGateway and all of your methods must be called with jsonp or use a special header, it will be incredibly boring add those configurations every time. Global configurations allow you to configure gateway options and a processor that will be used for every method. Keep in mind that the processor configured in the resource will be prioritized instead to global, for example:

var manifest = {
  host: 'http://my.api.com',
  rules: [
    { // This is our global configuration
      values: {
        gateway: {jsonp: true},
        processor: function(data) { return data.result }
      }
    }
  ],
  resources: {
    Book: {
      all:  {path: '/v1/books.json'},
      byId: {path: '/v1/books/{id}.json'}
    },
    Photo: {
      byCategory: {path: '/v1/photos/{category}/all.json'}
    }
  }
}

It is possible to add some configurations based on matches in the URLs, let's include a header for every book URL:

...
rules: [
  { // This is our global configuration
    values: {
      gateway: {jsonp: true},
      processor: function(data) { return data.result }
    }
  },
  { // This will only be applied when the URL matches the regexp
    match: /\/v1\/books/,
    values: {headers: {'X-MY-HEADER': 'value'}}
  }
]
...

Just keep in mind that the configurations and processors will be prioritized by their order, and the global configurations does not have a match key.

Gateway Implementations

The gateways listed here are available through the Mappersmith namespace.

VanillaGateway

Client Only. The default gateway - it uses plain XMLHttpRequest. Accepts a configure callback that allows you to change the request object before it is used.

Available methods:

• :ok: GET
• :ok: POST
• :ok: PUT
• :ok: DELETE
• :ok: PATCH

Available options:

• emulateHTTP: sends request as POST with _method in the body and X-HTTP-Method-Override header, both with requested method as value. (default false)

JQueryGateway

Client Only. It uses $.ajax and accepts an object that will be merged with defaults. It doesn't include jquery, so you will need to include that in your page.

Available methods:

• :ok: GET
• :ok: POST
• :ok: PUT
• :ok: DELETE
• :ok: PATCH

Available options:

• emulateHTTP: sends request as POST with _method in the body and X-HTTP-Method-Override header, both with request method as value. (default false)

NodeVanillaGateway

Server Only. It uses the module http and accepts an object that will be merged with defaults.

How to access this gateway?

var Mappersmith = require('mappersmith/node');
Mappersmith.node.NodeVanillaGateway;

Available methods:

• :ok: GET
• :ok: POST
• :ok: PUT
• :ok: DELETE
• :ok: PATCH

Available options:

• emulateHTTP: sends request as POST with _method in the body and X-HTTP-Method-Override header, both with request method as value. (default false)

Extras

For gateways with transparent cache functionalities and different cache stores, take a look at:

https://github.com/tulios/mappersmith-cached-gateway

Tests

Client

1. Build the source (npm run build-test)
2. Open test.html

Server

1. npm test

Compile and release

• Compile: npm run build
• Release (minified version): npm run release

Contributors

Check it out!

https://github.com/tulios/mappersmith/graphs/contributors

Licence

See LICENCE for more details.

Changelog

0.7.0

• Included host and path into stats object