From 7baaac206f3152b417eb43a234cd572576860d7b Mon Sep 17 00:00:00 2001 From: Karen Zeng Date: Mon, 1 May 2023 12:52:50 -0400 Subject: [PATCH] fp2_4 update --- fp2_4.py | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) diff --git a/fp2_4.py b/fp2_4.py index c3a2678..f98f13a 100644 --- a/fp2_4.py +++ b/fp2_4.py @@ -2,36 +2,42 @@ from selenium import webdriver from selenium.webdriver.chrome.options import Options from selenium.webdriver.common.by import By import json +import time # Configure the Selenium driver options options = Options() -options.add_argument('--headless') +#options.add_argument('--headless') options.add_argument('--disable-gpu') driver = webdriver.Chrome(options=options) # Load the page and extract the __NEXT_DATA__ data url = 'https://www.sayweee.com/en/product/Want-Want-Senbei-Rice-Crackers/17119' driver.get(url) +time.sleep(5) next_data_element = None -try: - next_data_element = driver.find_element(By.XPATH, '//script[contains(text(), "__NEXT_DATA__")]') -except: - print('__NEXT_DATA__ element not found') +#try: +next_data_element = driver.find_element(By.XPATH, "//script[@type='application/ld+json']") +#except: + # print('__NEXT_DATA__ element not found') +#time.sleep(1000) if next_data_element is not None: next_data_text = next_data_element.get_attribute('innerHTML') + print(next_data_text) 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) # Extract the price from the __NEXT_DATA__ data - price = next_data['props']['pageProps']['product']['price'] - + # price = next_data['props']['pageProps']['product']['price'] + print(next_data) + price = next_data['offers']['price'] # Print the price print(price) else: print('__NEXT_DATA__ element not found') +time.sleep(1000) # Close the Selenium driver -driver.quit() +#driver.quit()