Files
python-bootcamp/fp2_8.py
T

56 lines
2.1 KiB
Python
Raw Normal View History

2023-05-03 19:04:01 -04:00
import json
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.common.by import By
import time
from google.cloud import pubsub_v1
import os
os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = "C:/Users/Karen/AppData/Roaming/gcloud/application_default_credentials.json"
options = Options()
options.add_argument('--headless')
options.add_argument('--disable-gpu')
driver = webdriver.Chrome(options=options)
# Set up Pub/Sub subscription
project_id = "fp-20230425"
subscription_name = "fp-subscription-zipcodes"
subscriber = pubsub_v1.SubscriberClient()
subscription_path = subscriber.subscription_path(project_id, subscription_name)
def callback(message):
try:
# Get the zip code from the message
zip_code = message.data.decode('utf-8')
print(f"Received zip code: {zip_code}")
# Load the webpage with the zip code
url = 'https://www.sayweee.com/en/product/Want-Want-Senbei-Rice-Crackers/17119'
driver.get(url)
2023-05-04 19:19:11 -04:00
driver.find_element(By.XPATH, '/html/body/div[4]/div/div[2]/div/div[2]/div[2]/div/div[1]/form/div/div/div[1]/div/div/div[2]/input').send_keys(zip_code)
driver.find_element(By.XPATH, '/html/body/div[4]/div/div[2]/div/div[2]/div[2]/div/div[1]/form/div/div/div[1]/div/div/div[3]/div').click()
2023-05-03 19:04:01 -04:00
# Extract the price from the webpage
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 result
print(f"Price of rice crackers in zip code {zip_code}: {price}")
except Exception as e:
print(f"Error processing message: {str(e)}")
finally:
message.ack()
subscriber.subscribe(subscription_path, callback=callback)
while True:
time.sleep(1)