price-checker.py

import requests #3rd party module
from bs4 import BeautifulSoup
from datetime import datetime
from http.cookiejar import MozillaCookieJar
from amazoncaptcha import AmazonCaptcha
import traceback
import time
import os
import json

session = requests.Session()
session.cookies = MozillaCookieJar('cookies.txt')
userAgent = "Mozilla/5.0 (Macintosh; Intel Mac OS X 11_1_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.88 Safari/537.36"

OUTBOX_DIR = "../../outbox"
SCRIPT_TITLE = "amazon-price-checker"

# takes (title, price, url, max) tuple as all strings
def notify(resp):
    if not resp:
        return

    # create a file with our message contents
    output = {}
    output["source"] = SCRIPT_TITLE

    if type(resp) == str:
        # just pass the string through
        output["content"] = resp
        print(resp)
    elif type(resp) == tuple and len(resp) == 4:
        name, price, url, maxPrice = resp
        price = float(price)

        if price > maxPrice:
            # too high... skip
            return

        output["content"] = f"Amazon price alert! ${price} {name} - {url}"

    tstamp = time.strftime("%Y-%m-%d-%H-%M-%S")
    with open(f"{OUTBOX_DIR}{os.sep}{SCRIPT_TITLE}-{tstamp}.json", "w") as file:
        json.dump(output, file, indent=4)
    # or other notification method

def checkPrice(productURL, maxPrice = 9999):
    headers = { "User-Agent": userAgent }
    htmlPage = session.get(productURL, headers = headers)
    soup1 = BeautifulSoup(htmlPage.content,'html.parser')
    try:
        captcha = soup1.find('div', class_='a-row a-text-center').find('img')['src']
        if captcha:
            # amazon doesn't want us to be a robot, so we will now prove our humanity
            submitURL = f"https://www.amazon.com{soup1.find('form')['action']}"
            codeParams = [('field-keywords', AmazonCaptcha.fromlink(captcha).solve())]
            amznParams = list(map(lambda x: (x['name'], x['value']), soup1.find_all('input', type='hidden')))
            payload = dict(codeParams + amznParams)
            htmlPage = session.get(submitURL, params=payload, headers=headers)     
            soup1 = BeautifulSoup(htmlPage.content,'html.parser')
    except:
        # no captcha? captcha detection failed?
        # traceback.print_exc()
        print("Skipping captcha solving")

    # second soup because amazon uses javascript to create the price
    soup2 = BeautifulSoup(soup1.prettify(),'html.parser')
    # get product title
    productTitle = soup2.find('span', id='productTitle').getText()
    print(productTitle.strip())
    # get product price
    price = soup2.find('span', id='newBuyBoxPrice')
    price = price or soup2.find('span', id='price_inside_buybox')
    price = price or soup2.find('span', id='priceblock_ourprice')
    price = price or soup2.find('span', id='priceblock_saleprice')
    price = price or soup2.find('span', id='priceBlockStrikePriceString')
    session.cookies.save()
    try:
        productPriceStr = price.getText()
        print(productPriceStr.strip())
        return (productTitle.strip(), productPriceStr.strip().lstrip("$"), productURL, maxPrice)
    except:
        # no price?
        print("No price found / unavailable")
        pass
    return None

while True:
    try:
        notify(checkPrice("https://www.amazon.com/gp/product/B08166SLDF", 311.99))
        notify(checkPrice("https://www.amazon.com/gp/product/B07M6VMTRQ", 501.99))
    except Exception as e:
        notify(f"Hit an issue with one of the amazon price pages: {e}")
        traceback.print_exc()
    time.sleep(30)