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
π€ 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
npm install --save unfetch
Otherwise, grab it from unpkg.com/unfetch.
Usage: As a Polyfill
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>
Usage: As a Ponyfill
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:
var 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 β‘οΈ
// 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 beingGET
,POST
,PUT
,PATCH
,HEAD
,OPTIONS
orDELETE
).headers
: AnObject
containing additional information to be sent with the request, e.g.{ 'Content-Type': 'application/json' }
to indicate a JSON-typed request body.credentials
: β 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. βbody
: The content to be transmited in request's body. Common content types includeFormData
,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 anArray
containing thekey
for every header in the response.headers.entries
: Returns anArray
containing the[key, value]
pairs for everyHeader
in the response.headers.get(key)
: Returns thevalue
associated with the givenkey
.headers.has(key)
: Returns aboolean
asserting the existence of avalue
for the givenkey
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( checkStatus )
.then( r => r.json() )
.then( data => {
console.log(data);
});
function checkStatus(response) {
if (response.ok) {
return response;
} else {
var error = new Error(response.statusText);
error.response = response;
return Promise.reject(error);
}
}
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.
π 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.