from google.cloud import pubsub_v1 from google.api_core.exceptions import DeadlineExceeded from selenium import webdriver from selenium.webdriver.chrome.options import Options from selenium.webdriver.common.by import By import json import time import os from concurrent.futures import TimeoutError import pandas as pd # os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = "C:/Users/Karen/AppData/Roaming/gcloud/application_default_credentials.json" # Create an empty dataframe to store the prices df = pd.DataFrame(columns=['Zip Code', 'Price']) print(f"Created a dataframe: {df}") # Configure the Selenium driver options options = Options() options.add_argument('--disable-gpu') driver = webdriver.Chrome(options=options) # Set up Pub/Sub subscription project_id = "fp-20230425" subscription_name = "fp-subscription-NFL-zipcodes" subscriber = pubsub_v1.SubscriberClient() subscription_path = subscriber.subscription_path(project_id, subscription_name) # Load the product page and extract the __NEXT_DATA__ data url = 'https://www.sayweee.com/en/product/Want-Want-Senbei-Rice-Crackers/17119' driver.delete_all_cookies driver.get(url) # time.sleep(5) time.sleep(5) # Close the cookies setting window try: cookie_button = driver.find_element(By.XPATH, "/html/body/div[3]/div[2]/div/div/div[3]/button") print(f"cookie_button = {cookie_button}") cookie_button.click() except: pass time.sleep(5) # Clear any modal dialog try: modal_close = driver.find_element(By.XPATH, "//*[contains(concat(' ', normalize-space(@class), ' '), ' icon-icon-close-s ')]") print(f"modal_close = {modal_close}") modal_close.click() time.sleep(2) except: pass # Define callback function for Pub/Sub messages def callback(message): # Extract zip code from message data zip_code = message.message.data.decode('utf-8') print(f"Received zip code: {zip_code}") try: try: time.sleep(5) # Clear any modal dialog modal_close = driver.find_element(By.XPATH, "//*[contains(concat(' ', normalize-space(@class), ' '), ' icon-icon-close-s ')]") # print(f"modal_close = {modal_close}") modal_close.click() time.sleep(2) except: pass try: time.sleep(1) # Clear any notice notice_close = driver.find_element(By.XPATH, "/html/body/div[3]/div/div[2]/div/div[2]/div/div/div[1]/div[2]/button") # print(f"notice_close = {notice_close}") notice_close.click() time.sleep(1) except: pass # Enter zip code in the input field and click the "Check" button zip_button = driver.find_element(By.XPATH, '/html/body/div[2]/div/header/div[3]/div/div/div[2]/div[1]') # print(f"{zip_button:}") zip_button.click() time.sleep(1) zip_input = driver.find_element(By.XPATH, '//div/label/following-sibling::input') # print(f"{zip_input:}") zip_input.click() # print("clearing zip_input") zip_input.send_keys("\b\b\b\b\b") # print("sending zip code") zip_input.send_keys(zip_code) check_button = driver.find_element(By.XPATH, "//button[text() = 'Confirm']") # print(f"{check_button:}") check_button.click() time.sleep(2) try: time.sleep(5) # Clear any notice notice_close = driver.find_element(By.XPATH, "/html/body/div[3]/div/div[2]/div/div[2]/div/div/div[1]/div[2]/button") print(f"notice_close = {notice_close}") notice_close.click() time.sleep(2) except: pass # Extract the price of rice crackers for this zip code ''' price_element = driver.find_element(By.XPATH, '/html/body/div[2]/div/div[1]/div[1]/div/div/div[2]/div[2]/div[1]/div[4]/div[1]/div') price = price_element.text.strip()[1:] print(price) ''' next_data_element = driver.find_element(By.XPATH, "//script[@type='application/ld+json']") next_data_text = next_data_element.get_attribute('innerHTML') next_data_start_index = next_data_text.index('{') next_data_end_index = next_data_text.rindex('}') + 1 next_data_json = next_data_text[next_data_start_index:next_data_end_index] next_data = json.loads(next_data_json) price = next_data['offers']['price'] # Print the price print(price) subscriber.acknowledge(ack_ids = [message.ack_id], subscription = subscription_path) # acknowledge receipt of message # Append the zip code and price to the dataframe #append_to_df(zip_code, price) new_row = pd.DataFrame({'Zip Code': [zip_code], 'Price': [price]}) return pd.concat([df, new_row], ignore_index=True) #print(f"Recorded the zip code and price: {zip_code},{price}") #print(f"Updated a dataframe: {df}") except Exception as e: print(f"Caught exception: {e}") print(f"Price not found for zip code {zip_code}") price = 'Price not found' return df import time # Start receiving messages (sync version) while True: pull_request = pubsub_v1.types.PullRequest(max_messages = 1, subscription = subscription_path) messages = [] try: pull_response = subscriber.pull(request = pull_request) messages = pull_response.received_messages except DeadlineExceeded: pass if messages: df = callback(messages[0]) print(f"Updated a dataframe: {df}") else: print(f"no messages, trying again at {time.asctime()}") # Start receiving messages (async version) #flow_control = pubsub_v1.types.FlowControl(max_messages = 1) #streaming_pull_future = subscriber.subscribe(subscription_path, # callback=callback, # flow_control=flow_control) # print(f'Started receiving messages from subscription {subscription_path}...') # Wrap subscriber in a 'with' block to automatically call close() when done. # with subscriber: # try: # When `timeout` is not set, result() will block indefinitely, # unless an exception is encountered first. # streaming_pull_future.result(timeout=timeout) # except TimeoutError: # streaming_pull_future.cancel() # Trigger the shutdown. # streaming_pull_future.result() # Block until the shutdown is complete. # Keep the script running indefinitely to continue receiving messages #while True: # time.sleep(10)