Skip to content
Go to file

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

LibPixel Uploader

A client-side JavaScript library that makes uploading images to S3 easier than any alternative.

LibPixel will handle all the complexity of S3 request signing and server-side redirects. You won't even need to add a blank page to your website just to support older browsers. All that has been done for you. In fact you won't need to touch your server-side code at all.

It supports pretty much every browser starting with IE 8, it weighs only 2.7KB (minified + gzipped), and has no dependencies.

Getting started

If you use NPM in your project you can install and require the library:

$ npm install libpixel-uploader
var LibPixelUploader = require("libpixel-uploader");

Alternatively, you can download a release and include it in your page:

<script src="/path/to/libpixel-uploader.min.js"></script>

In your LibPixel account, all you need is a source configured with your S3 bucket, and uploads enabled.

Your S3 bucket needs to be configured for CORS. You can find the CORS configuration here, and instructions on how to configure it from the S3 User Guide.

Assuming your source's path is /eu-west-1/uploads and your LibPixel domain is, you'd configure the uploader as:

var uploader = LibPixelUploader({
  host: "",
  source: "eu-west-1/uploads",
  element: "#file-input"

See the Options section for details on all the possible uploader options.


On your page will be a form which will be used to upload the images. Once the user has selected a file, there are two ways of triggering the file upload: automatically and manually.


The automatic way binds event listeners to the file input's form and uploads the image when the form submission is triggered. You can use this option by calling addEventListeners() on the uploader object:


So as soon as the form is submitted, the upload to S3 will automatically start.

Behind the scenes, this is what the LibPixel Uploader does for you:

  • When the user triggers the form submission:
    • Image upload starts.
    • The actual form submission is cancelled.
    • The submit button is disabled.
  • When the upload finishes:
    • A hidden input is appended to the form with the same name attribute as the original file input. The value of this input is the S3 key of the uploaded object.
    • The original file input's name attribute is removed.
    • The submit button is re-enabled.
    • The form submission is triggered.

The uploader does not automatically create a progress bar, or show errors, but instead gives options to listen to progress events, errors, and other events, so that you can hook your own UI elements that match your design. See the Events section for details on all available events.

It's also possible to automatically start the image uploads as soon as the user chooses a file, by using the autoStart option:

uploader.addEventListeners({ autoStart: true });

If the user chooses another file before uploading, it guarantees that you'll receive the key of the final image that was chosen. A previous upload in progress will be cancelled. If the user submits the form before the upload has finished, the actual form submission will be delayed until the upload finishes, as without the autoStart option.


Sometimes you'll want to trigger the file upload yourself, for example when not using a form, or when the flow of events described above doesn't suit your needs.

To manually trigger an upload, just call upload() on the uploader object:


For this to work, the file element must already have a file selected, otherwise either nothing will happen or the error callback will be called depending on the required option setting.

## Options


The file input where the user selects the image to upload. Can be defined in one of the following ways:

  1. String (CSS selector that uniquely identifies the file input).
  2. An HTMLInputElement object, for example, returned by document.querySelector.
  3. A jQuery object (e.g. $("#file-input")).


The LibPixel source path, without the leading / (e.g. "us-east-1/userimages" or "eu-west-1/uploads").


The LibPixel host, without a protocol (e.g. "").


Whether choosing a file is required when submitting the form when using addEventListeners() or when calling upload(). Default is true (file is required).

If no file is set and the form is submitted with required set to true, then the error callback will be called.

If required is set to false, then the form can be submitted without a file set, and any other fields in the form will be submitted to your server without any interference by the LibPixel Uploader.


By default the LibPixel Uploader will upload files via XHR, except when the FormData interface is not supported (e.g. IE 8/9). In these cases, it will automatically fall back to an iframe transport mode, where progress events are not supported.

You usually don't want to modify this option, except when testing the iframe transport mode during development, and don't want to switch to an older and unsupported browser.

Possible options are:

  • "auto" (default)
  • "xhr"
  • "iframe"
## Events

Event listeners are defined by calling functions on the uploader object which take callback functions:

uploader.progress(function (event) {
  // handle progress

Each event listener definition function returns the upload object, so you can chain multiple event callback definitions if you want:

uploader.start(function (event) {
  // handle start
}).progress(function (event) {
  // handle progress


Called every time an upload starts, once per selected file.

The event object contains the following keys:

  • id: string that uniquely identifies this upload. The same string is passed to all events for the same file.
  • progressSupported: whether progress events will be sent. Some older browsers may not support progress events, in which case this will be false.


Called whenever a progress update for the upload is available.

The event object contains the following keys:

  • id: string that uniquely identifies this upload.
  • loaded: number of bytes transferred.
  • total: number of bytes to be transferred in total (i.e. file size).
  • percentage: percentage of upload completed (between 0.0 and 1.0).


Called every time an upload finishes successfully, once per selected file.

The event object contains the following keys:

  • id: string that uniquely identifies this upload.
  • key: S3 object key of the uploaded file.


Called whenever an error occurs.

The event object contains the following keys:

  • id: string that uniquely identifies this upload.
  • message: message describing the error, in English. See I18n to learn how to customize these messages.
## I18n

If you wish to customize the error messages, you can modify the LibPixelUploader.messages object:

LibPixelUploader.messages.fileNotSelected = "You must select a file to upload";

Important: you must always modify the messages object, and never reassign it.

The available message keys and their default values are:

  fileNotSelected: "You must select a file to upload",
  invalidExtension: "File must have one of the following extensions:"
## AWS S3 CORS configuration
<?xml version="1.0" encoding="UTF-8"?>
<CORSConfiguration xmlns="">


If you wish to contribute, start by getting the project up and running. You'll need a recent version of Node.js. After cloning the GitHub repo you'll need to install all the development dependencies with NPM:

$ npm install

This project uses gulp.js for automating tasks. The defined tasks for development are:

  • gulp server: starts a development server on port 9090 which automatically rebuilds and serves the latest version at http://localhost:9090/dist/uploader.js.
  • gulp watch: watches for changes runs the build task on every change.
  • gulp build: creates a development (non-minified) version in dist/uploader.js.
  • gulp dist: creates a production (minified) version in dist/uploader.min.js.


There are both unit tests and end-to-end tests that run in real browser environments (via Sauce Labs with Selenium).

Unit tests can be run locally with Karma:

$ gulp test:unit

End-to-end tests require a Sauce Labs account and Sauce Connect.

Set the SAUCE_USERNAME and SAUCE_ACCESS_KEY environment variables to contain your credentials and export them.

You must start Sauce Connect with the following options:

$ sc -u $SAUCE_USERNAME -k $SAUCE_ACCESS_KEY -N --direct-domains "*"

Once Sauce Connect is ready, the tests can be run with:

$ gulp test:browser

The browser tests are setup to run on the following browsers:

  • Firefox
  • Chrome
  • IE 8, 9, 10, and 11.

Unfortunately the Safari Selenium WebDriver doesn't support file uploads via a file input, so it must be manually tested.


The MIT License.


A client-side JavaScript library that makes uploading images to S3 easier than any alternative.



No packages published
You can’t perform that action at this time.