Before generating QR codes you'll need to install the dependencies.

%pip install "qrcode[pil]"

Once installed you can generate a QR code by running:

import qrcode
img = qrcode.make('qr codes from python')

You can save these images to disk by running:

img.save("filename.png")

To read QR codes we'll need another dependency.

%pip install opencv-python==4.4.0.44

Given this dependency you can retreive the code from a QR code via;

import cv2
import matplotlib.pylab as plt 

# read the QRCODE image
img = cv2.imread("qrcode.png")

detector = cv2.QRCodeDetector()
data, bbox, straight_qrcode = detector.detectAndDecode(img)

You can explore many settings when writing a QR code.

qr = qrcode.QRCode(
    version=1,
    error_correction=qrcode.constants.ERROR_CORRECT_H, # L -> M -> Q -> H
    box_size=5,
    border=4,
)
qr.add_data('https//calmcode.io')
qr.make(fit=True) # Setting fit=True ensures the minimum size.

img = qr.make_image(fill_color="black", back_color="white")
img

If you're interested in running the same error-correcting experiment, you'll appreciate the following helper function.

import matplotlib.pylab as plt
from PIL import Image
import numpy as np

def convert_msg(img_path):
    img_broken = Image.open(img_path)
    np_arr = np.array(img_broken.convert('RGB'))
    data, bbox, straight_qrcode = detector.detectAndDecode(np_arr)
    print(f"found msg: {data}")
    return img_broken

convert_msg('high-quality.png')

There are many ways to serve QR codes but we've implemented one with FastAPI as an example.

import io
import qrcode

from fastapi import FastAPI
from starlette.responses import StreamingResponse

app = FastAPI()

@app.get("/generate/{message}")
def generate(message: str):
    img = qrcode.make(message)
    buf = io.BytesIO()
    img.save(buf)
    buf.seek(0) # important here!
    return StreamingResponse(buf, media_type="image/jpeg")

You can run this locally by running this command from the terminal:

uvicorn fastqrapi:app --reload

These commands do assume that both fastapi and uvicorn are installed. If you've never worked with FastAPI before you might appreciate our series of videos here.