Tiny 500b fetch "barely-polyfill"

• Tiny: about 500 bytes of ES3 gzipped
• Minimal: just fetch() with headers and text/json responses
• Familiar: a subset of the full API
• Supported: supports IE8+ (assuming Promise is polyfilled of course!)
• Standalone: one function, no dependencies
• Modern: written in ES2015, transpiled to 500b of old-school JS

🤔 What's Missing?

• Uses simple Arrays instead of Iterables, since Arrays are iterables
• No streaming, just Promisifies existing XMLHttpRequest response bodies
• Use in Node.JS is handled by isomorphic-unfetch

• Unfetch
• Installation
• Usage: As a Polyfill
• Usage: As a Ponyfill
• Examples & Demos
• API
• Caveats
• Contribute
• License

For use with node and npm:

npm i unfetch

Otherwise, grab it from unpkg.com/unfetch.

This automatically "installs" unfetch as window.fetch() if it detects Fetch isn't supported:

import 'unfetch/polyfill'

// fetch is now available globally!
fetch('/foo.json')
  .then( r => r.json() )
  .then( data => console.log(data) )

This polyfill version is particularly useful for hotlinking from unpkg:

<script src="https://unpkg.com/unfetch/polyfill"></script>
<script>
  // now our page can use fetch!
  fetch('/foo')
</script>

With a module bundler like rollup or webpack, you can import unfetch to use in your code without modifying any globals:

// using JS Modules:
import fetch from 'unfetch'

// or using CommonJS:
const fetch = require('unfetch')

// usage:
fetch('/foo.json')
  .then( r => r.json() )
  .then( data => console.log(data) )

The above will always return unfetch(). (even if window.fetch exists!)

There's also a UMD bundle available as unfetch/dist/unfetch.umd.js, which doesn't automatically install itself as window.fetch.

Real Example on JSFiddle ➡️

// simple GET request:
fetch('/foo')
  .then( r => r.text() )
  .then( txt => console.log(txt) )


// complex POST request with JSON, headers:
fetch('/bear', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({ hungry: true })
}).then( r => {
  open(r.headers.get('location'));
  return r.json();
})

While one of Unfetch's goals is to provide a familiar interface, its API may differ from other fetch polyfills/ponyfills. One of the key differences is that Unfetch focuses on implementing the fetch() API, while offering minimal (yet functional) support to the other sections of the Fetch spec, like the Headers class or the Response class. Unfetch's API is organized as follows:

This function is the heart of Unfetch. It will fetch resources from url according to the given options, returning a Promise that will eventually resolve to the response.

Unfetch will account for the following properties in options:

• method: Indicates the request method to be performed on the target resource (The most common ones being GET, POST, PUT, PATCH, HEAD, OPTIONS or DELETE).
• headers: An Object containing additional information to be sent with the request, e.g. { 'Content-Type': 'application/json' } to indicate a JSON-typed request body.
• credentials: ⚠ Accepts a "include" string, which will allow both CORS and same origin requests to work with cookies. As pointed in the 'Caveats' section, Unfetch won't send or receive cookies otherwise. The "same-origin" value is not supported. ⚠
• body: The content to be transmitted in request's body. Common content types include FormData, JSON, Blob, ArrayBuffer or plain text.

These methods are used to handle the response accordingly in your Promise chain. Instead of implementing full spec-compliant Response Class functionality, Unfetch provides the following methods and attributes:

Returns true if the request received a status in the OK range (200-299).

Contains the status code of the response, e.g. 404 for a not found resource, 200 for a success.

A message related to the status attribute, e.g. OK for a status 200.

Will return another Object with the same shape and content as response.

Will return the response content as plain text, JSON and Blob, respectively.

Again, Unfetch doesn't implement a full spec-compliant Headers Class, emulating some of the Map-like functionality through its own functions:

• headers.keys: Returns an Array containing the key for every header in the response.
• headers.entries: Returns an Array containing the [key, value] pairs for every Header in the response.
• headers.get(key): Returns the value associated with the given key.
• headers.has(key): Returns a boolean asserting the existence of a value for the given key among the response headers.

Adapted from the GitHub fetch polyfill readme.

The fetch specification differs from jQuery.ajax() in mainly two ways that bear keeping in mind:

• By default, fetch won't send or receive any cookies from the server, resulting in unauthenticated requests if the site relies on maintaining a user session.

fetch('/users', {
  credentials: 'include'
});

• The Promise returned from fetch() won't reject on HTTP error status even if the response is an HTTP 404 or 500. Instead, it will resolve normally, and it will only reject on network failure or if anything prevented the request from completing.

  To have fetch Promise reject on HTTP error statuses, i.e. on any non-2xx status, define a custom response handler:

fetch('/users')
  .then(response => {
    if (response.ok) {
      return response;
    }
    // convert non-2xx HTTP responses into errors:
    const error = new Error(response.statusText);
    error.response = response;
    return Promise.reject(error);
  })
  .then(response => response.json())
  .then(data => {
    console.log(data);
  });

First off, thanks for taking the time to contribute! Now, take a moment to be sure your contributions make sense to everyone else.

Found a problem? Want a new feature? First of all see if your issue or idea has already been reported. If it hasn't, just open a new clear and descriptive issue.

Pull requests are the greatest contributions, so be sure they are focused in scope, and do avoid unrelated commits.

💁 Remember: size is the #1 priority.

Every byte counts! PR's can't be merged if they increase the output size much.

• Fork it!
• Clone your fork: git clone https://github.com/<your-username>/unfetch
• Navigate to the newly cloned directory: cd unfetch
• Create a new branch for the new feature: git checkout -b my-new-feature
• Install the tools necessary for development: npm install
• Make your changes.
• npm run build to verify your change doesn't increase output size.
• npm test to make sure your change doesn't break anything.
• Commit your changes: git commit -am 'Add some feature'
• Push to the branch: git push origin my-new-feature
• Submit a pull request with full remarks documenting your changes.

MIT License © Jason Miller