Amazon.com.au

gistfile1.txt

#!/usr/bin/env python3
# Script for scanning amazon.com.au for ps5
# Aleksandar Josifoski about.me/josifsk for Timothy Yang via upwork
# Date created: 2022 January

# Dependencies:
# pip3 install -U selenium 

# Create on pi, /home/pi/amznau2 directory & place geckodriver there
# Start program with python3 amznau2.py
# After ps5 showing in stock, this script will make screenshot photo.png
# Also will save html page in file html.txt
#########################################################################
import getpass
username = getpass.getuser()
comp = username

if comp == 'josifoski':
    dir_in = '/data/upwork/Timothy_Yang/'
    geckodriver_path = '/data/Scrape/geckodriver'
else:
    dir_in = '/home/pi/amznau2/'
    geckodriver_path = '/home/pi/amznau2/geckodriver'
    

brows = 'firefox'
headless = True

import codecs
import datetime

timeout = 10
#########################################################################
from selenium import webdriver
from selenium.webdriver.firefox.options import Options
from selenium.webdriver.firefox.service import Service
from selenium.webdriver.common.keys import Keys
from selenium.common.exceptions import TimeoutException
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.common.action_chains import ActionChains
import sys
import os
import time
import traceback
import random
import requests
import re
import html

time1 = time.time()

driver = None

def openurl(url):
    '''function to open url using selenium'''
    global driver
    try:
        driver.get(url)
        print('loading ' + url)
    except Exception as e:
        now = str(datetime.datetime.now())[:16]
        print(str(e))
        
def scroll_down(sbypx):
    '''function to scroll by sbypx pixels'''
    global driver
    driver.execute_script("window.scrollBy(0, %d);" % (sbypx))
    time.sleep(0.5)

def setbrowser():
    ''' function for preparing browser for automation '''
    print("Preparing browser")
    global driver

    # firefox
    if brows.lower() in ('firefox',):
        options = Options()
        profile = webdriver.FirefoxProfile()
        #options.set_preference("javascript.enabled", True)
        #options.profile = firefox_profile
        #options.set_preference('network.proxy.type', 1)
        #options.set_preference('network.proxy.socks', '127.0.0.1')
        #options.set_preference('network.proxy.socks_port', 9050)
        #options.set_preference('network.proxy.socks_remote_dns', False)
        #options.add_argument(f'user-agent=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/79.0.3945.79 Safari/537.36')
        profile.set_preference("general.useragent.override", f'user-agent=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/79.0.3945.79 Safari/537.36')
        if headless:
            options.headless = True
        service = Service(geckodriver_path)
        #driver = webdriver.Firefox(service=service, options=options)
        driver = webdriver.Firefox(firefox_profile=profile, service=service, options=options)


    driver.maximize_window()
    driver.implicitly_wait(timeout)


def calculate_time():
    '''function to calculate elapsed time'''
    time2 = time.time()
    hours = int((time2-time1)/3600)
    minutes = int((time2-time1 - hours * 3600)/60)
    sec = time2 - time1 - hours * 3600 - minutes * 60
    print("processed in %dh:%dm:%ds" % (hours, minutes, sec))

def driverquit():
    ''' quiting session '''
    global driver
    driver.quit()
    driver = None
    calculate_time()
    sys.exit()
    
def restartff():
    ''' restart firefox '''
    global driver
    driver.quit()
    driver = None
    setbrowser()
    print('Preparing browser done')
    
def login():
    ''' login '''
    global driver
    openurl('https://amazon.com.au')
    print('Login attempt')
    #some = input('press enter')
    try:
        el = driver.find_element(By.ID, "nav-link-accountList-nav-line-1").click()
        time.sleep(5)
        print('Sign in clicked')
    except Exception as e:
        print(str(e))
    try:
        el = driver.find_element(By.ID, "ap_email").send_keys('[email protected]')
        time.sleep(2)
        print('email entered')
    except Exception as e:
        print(str(e))
    try:
        el = driver.find_element(By.ID, "continue").click()
        time.sleep(3)
        print('continue clicked')
    except Exception as e:
        print(str(e))
    try:
        el = driver.find_element(By.ID, "ap_password").send_keys('Upw0rker')
        time.sleep(2)
        print('password entered')
    except Exception as e:
        print(str(e))
    try:
        el = driver.find_element(By.ID, "signInSubmit").click()
        time.sleep(15)
        print('Submit clicked, login successful!')
    except Exception as e:
        print(str(e))

def main():
    ''' main function '''
    global driver
    
    while True:
        # ps5
        openurl('https://www.amazon.com.au/gp/product/B08HHV8945/')
        time.sleep(5)
        try:
            xpath = '//span[@class="a-color-price a-text-bold" and text()="Currently unavailable."]'
            el = driver.find_element(By.XPATH, xpath)

        except Exception as e:
            if 'Tried to run command without establishing a connection' in str(e):
                restartff()
            driver.save_screenshot(dir_in + 'photo.png')
            ht = driver.page_source
            with codecs.open(dir_in + 'html.txt', 'w', 'utf8') as f:
                f.write(html.unescape(ht))
            break

        print('sleeping 120 seconds')
        time.sleep(120)
        

if __name__ == '__main__':
    setbrowser()
    print('Preparing browser done')
    try:
        main()
        driverquit()

    except Exception as e:
        print(str(e))
        print(traceback.format_exc())
        driverquit()