Multi-format 1D/2D barcode image processing library.
Ported from ZXing project (written in Java).
ZXing ("zebra crossing") is an open-source, multi-format 1D/2D barcode image processing library implemented in Java, with ports to other languages.
WIP: There's still some things we gotta do here, see Status and Roadmap for what is currently done and what's planned next.
| 1D product | 1D industrial | 2D |
|---|---|---|
| QR Code | ||
| Code 128 (no docs) | ||
| PDF 417 (in-progress) | ||
See some demo examples in browser.
The library can be used from browser with TypeScript (include anything from src/browser however you must do the packaging yourself) or with plain javascript (see below). It can also be used from node (see also below). The library is using separate builds for node and browser to allow different ES targeting.
Examples below are for QR barcode, all other supported barcodes work similary.
npm i @zxing/library --save
To use from JS you need to include what you need from build/umd folder (for example zxing.min.js).
See some demo examples for browser code examples with javascript.
All the examples are using ES6, be sure is supported in your browser or modify as needed (eg. var instead of const etc.).
The browser library is using the MediaDevices web api which is marked as experimental as of this writing. You can use external polyfills like webrtc-adapter to increase browser compatiblity.
Also, note that the library is using the TypedArray (Int32Array, Uint8ClampedArray, etc.) which are not available in older browsers (for example Android 4 default browser, etc.). You can use a polyfill library like core-js to support these browsers.
To display the input from the video camera you will need to add a video element in the html page:
<video id="video" width="300" height="200" style="border: 1px solid gray"></video>To start decoding, first obtain a list of video input devices with:
const codeReader = new ZXing.BrowserQRCodeReader();
codeReader.getVideoInputDevices()
.then((videoInputDevices) => {
videoInputDevices.forEach((element) => {
console.log(`${element.label}, ${element.deviceId}`);
});
.catch((err) => {
console.error(err);
});If there is just one input device you can use the first deviceId and the video element id (in the example below is also 'video') to decode:
const firstDeviceId = videoInputDevices[0].deviceId;
codeReader.decodeFromInputVideoDevice(firstDeviceId, 'video')
.then((result) => {
console.log(result.text);
}).catch((err) => {
console.error(err);
});If there are more input devices then you will need to chose one for codeReader.decodeFromInputVideoDevice device id parameter.
You can also provide undefined for the device id parameter in which case the library will automatically choose the camera, prefering the main (environment facing) camera if more are available:
codeReader.decodeFromInputVideoDevice(undefined, 'video')
.then((result) => {
console.log(result.text);
}).catch((err) => {
console.error(err);
});A full working example for QR Code from Video Camera is provided in the examples.
Similar as above you can use a video element in the html page:
<video id="video" width="300" height="200" style="border: 1px solid gray"></video>And to decode the video from an url:
const codeReader = new ZXing.BrowserQRCodeReader();
const videoSrc = 'your url to a video';
codeReader.decodeFromVideoSource(videoSrc, 'video')
.then((result) => {
console.log(result.text);
}).catch((err) => {
console.error(err);
});You can also decode the video url without showing it in the page, in this case no video element is needed in html.
codeReader.decodeFromVideoSource(videoSrc)
.then((result) => {
console.log(result.text);
}).catch((err) => {
console.error(err);
});A full working example for QR Code from Video File is provided in the examples.
Similar as above you can use a img element in the html page (with src attribute set):
<img id="img" src="qrcode-image.png" width="200" height="300" style="border: 1px solid gray"></img>And to decode the image:
const codeReader = new ZXing.BrowserQRCodeReader();
const img = document.getElementById('img');
codeReader.decodeFromImage(img)
.then((result) => {
console.log(result.text);
}).catch((err) => {
console.error(err);
});You can also decode the image url without showing it in the page, in this case no img element is needed in html:
const imgSrc = 'url to image';
codeReader.decodeFromImage(undefined, imgSrc)
.then((result) => {
console.log(result.text);
}).catch((err) => {
console.error(err);
});Or decode the image url directly from an url, with an img element in page (notice no src attribute is set for img element):
<img id="img" width="200" height="300" style="border: 1px solid gray"></img>const imgSrc = 'url to image';
codeReader.decodeFromImage('img', imgSrc); // here img is the image id from html, in our case 'img'
.then((result) => {
console.log(result.text);
}).catch((err) => {
console.error(err);
});A full working example for QR Code from Image is provided in the examples.
To generate a QR Code SVG image include 'zxing.qrcodewriter.min.js' from build/vanillajs. You will need to include an element where the SVG element will be appended:
<div id="result"></div>And then:
const codeWriter = new ZXing.BrowserQRCodeSvgWriter('result');
var svgElement = codeWriter.write(input, 300, 300);A full working example for QR Code write to SVG is provided in the examples.
Install the package:
npm i @zxing/library --save
And then include directly the classes files you need, for example:
import { BrowserQRCodeReader, VideoInputDevice } from '@zxing/library';The usage is identical with the above.
npm i @zxing/library --save
To use in node you will need to provide an implementation of LuminanceSource for an image. A starting point is SharpImageLuminanceSource from tests that is using sharp image processing node library.
No examples are availabe for now, however you can have a look at the extensive tests cases.
To decode a barcode, the library needs at some point to decode from bits to text. Also, to generate a barcode it needs to encode text to bits. Unfortunately, the state of encoding and decoding text in javascript/browser is somehow messy at the moment.
To have full support for all encodings in CharacterSetECI except Cp437 use text-encoding library. The library is used implicitly for node (and tests), but is an optional dependency for browser because is rather large (> 600k). You will need to include it yourself if you want/need to use it.
By default, in browser, TextDecoder/TextEncoder web api are used if available (take care as these are labeled as experimental as of this writing). Also, be aware that TextEncoder encodes only to UTF-8 as per spec. If these are not available the library falls back to a minimal implementation that only encodes and decodes to/from UTF-8 (see StringEncoding).
See TypeScript Port Info for information regarding porting approach and reasoning behind some of the approaches taken.
Done:
- Port root, common and qrcode format and make it compile
- Add unit test infrastructure, a first unit test and make it pass (common/BitArray.spec)
- Add all unit tests for everything in root, common and qrcode
- Add one "back box" test for qrcode
- Add all "back box" tests for qrcode
- Create browser integration module and demo UI for qrcode
- Document browser usage
- Implement QR barcode generation to SVG in browser
Todo:
- Move all these tasks to a "Project".
- Port pdf417 format with unit and browser tests and documentation
- Adapt documentation for JSDoc, generate documentation, cleanup source files
- Create automatic tests for all major current browsers
- Port aztec format with unit and browser tests
- Port multi parsing with unit and browser tests and documentation
- Port datamatrix format with unit and browser tests and documentation
- Port maxicode format with unit and browser tests and documentation
- Port oned format with unit and browser tests and documentation
- Port client/result parsing with unit and browser tests and documentation
- Documentation for using directly from TypeScript
