BBQr JavaScript / TypeScript implementation

This is an implementation of BBQr - Better Bitcoin QR for the browser, written in TypeScript.

Installation

Install with npm (or your package manager of choice):

npm install bbqr

Usage

Low-level splitting/joining

import { joinQRs, splitQRs } from 'bbqr';

// Create a Uint8Array containing the raw bytes of a PSBT, transaction or other data
const input = new Uint8Array([
  112,
  115,
  98,
  116,
  255, // ... rest of PSBT bytes
]);

// ...

const fileType = 'P'; // 'P' is for PSBT

const splitResult = splitQRs(input, fileType, {
  // these are optional - default values are shown
  encoding: 'Z', // Zlib compressed base32 encoding
  minSplit: 1, // minimum number of parts to return
  maxSplit: 1295, // maximum number of parts to return
  minVersion: 5, // minimum QR code version
  maxVersion: 40, // maximum QR code version
});

// the QR code version chosen for best efficiency
console.log(splitResult.version);

// the actual encoding used - could be '2' (uncompressed base32) if the 'Z' option didn't provide a smaller result
console.log(splitResult.encoding);

// the QR code parts
console.log(splitResult.parts);

// now we do this in reverse and get back the bytes

const reassembled = joinQRs(splitResult.parts);

console.log(reassembled.fileType === fileType); // true
console.log(reassembled.encoding === splitResult.encoding); // true
console.log(reassembled.raw.every((byte, i) => byte === input[i])); // true

Detecting the File Type

In the previous example, we provided the raw bytes and specified the File Type to use. You can also detect the file type by calling detectFileType with a Uint8Array, a File object or a string (the contents of a text file).

This is especially useful for PSBTs and Bitcoin transactions as detectFileType can also detect these in HEX and Base64 format.

If a PSBT or transaction is not detected, the file type will be:

• J for a text file that can be successfully parsed as JSON.
• U for all other text files.
• B for all other binary files.

For other supported type codes, you should do the conersion to Uint8Array not rely on detectFileType.

import { detectFileType } from 'bbqr';

// hex representation of a Bitcoin transaction
const contents =
  '01000000010000000000000000000000000000000000000000000000000000000000000000ffffffff0704ffff001d0104ffffffff0100f2052a0100000043410496b538e853519c726a2c91e61ec11600ae1390813a627c66fb8be7947be63c52da7589379515d4e0a604f8141781e62294721166bf621e73a82cbf2342c858eeac00000000';

const detected = await detectFileType(contents);

// 'T' for transaction
console.log(detected.fileType);

// Uint8Array of the transaction bytes
console.log(detected.raw);

Generating QR Code Images

The result from splitQRs can be passed to renderQRImage to generate (animated) PNG images of the QR codes. The result is an ArrayBuffer which can easily be converted to a suitable format.

import { renderQRImage } from 'bbqr';

// get an ArrayBuffer containing the PNG image data
const imgBuffer = await renderQRImage(splitResult.parts, splitResult.version, {
  // optional settings - values here are the defaults
  frameDelay: 250,
  randomizeOrder: false,
});

// convert to data URL for display
const base64String = btoa(String.fromCharCode(...new Uint8Array(imgBuffer)));
const dataUrl = `data:image/png;base64,${base64String}`;

document.body.innerHTML += `<img src="${dataUrl}">`;

Developing

This library is built with Vite.

• The index.html and demo.ts files are for local development.
• Start the Vite dev server: npm run dev.
• Start the Vitest tests in watch mode: npm run test:watch
• Building for production: npm run build