scraper_lazada.py

from selenium import webdriver
from selenium.webdriver.chrome.options import Options
from selenium.common.exceptions import NoSuchElementException
from selenium.webdriver.common.by import By
from selenium.webdriver.chrome.service import Service
import time

# Global driver to use throughout the script
DRIVER = None

# Wrapper to close driver if its created
def close_driver():
    global DRIVER
    if DRIVER is not None:
        DRIVER.close()
    DRIVER = None

# Function to (re)start driver
def start_driver(force_restart=False):
    global DRIVER

    if force_restart:
        close_driver()

    # Setting up the driver
    service = Service()
    options = Options()
    options.add_argument('-headless') # we don't want a chrome browser opens, so it will run in the background
    options.add_argument('-no-sandbox')
    options.add_argument('-disable-dev-shm-usage')

    DRIVER = webdriver.Chrome(service=service, options=options)

### Function to extract product info from the necessary html and json tags
def get_lazada_product_info_single(product_element, extra_info):
    """
    Extract info from a single product element from the driver.
    Args:
        product_item: (WebDriverElement) the product whose info needs to be
                        extracted.
    Returns:
        info: (dict) a dictionary of info of the product. Every product
                should at least have four pieces of information: name, price,
                link to the product page, and link to the product image.
    """
    info = {'source': 'lazada',
            'name':'',
            'price':-1,
            'product_url':'',
            'image':''}
    
    # print(product_element.get_attribute('outerHTML'))
    try:
        # Find the <a> element within the <div class="RfADt">
        product_title_element = product_element.find_element(By.XPATH, "//div[@class='RfADt']/a")
        
        # Get the text content of the <a> element
        info['name'] = product_title_element.text

    except NoSuchElementException:
        info['name'] = ""


    # price
    try:
        # Find the <span> element with class "ooOxS" within the <div class="aBrP0">
        price_element = product_element.find_element(By.XPATH, "//div[@class='aBrP0']/span[@class='ooOxS']")
    
        # Get the text content of the <span> element
        price_text = price_element.text

        # Extract the price value
        info['price'] = int(price_text.split(" ")[0].replace('.', ''))

    except (NoSuchElementException, ValueError):
        pass

    # link
    try:
        # Find the <a> element within the <div class="RfADt">
        product_link_element = product_element.find_element(By.XPATH, "//div[@class='RfADt']/a")
        
        # Get the href attribute of the <a> element
        product_link = product_link_element.get_attribute("href")
       
        # Extract the URL from the href attribute
        info['product_url'] = product_link.split("//")[1]

    except NoSuchElementException:
        pass

    # thumbnail
    try:
        # Find the <img> element within the <div class="_95X4G">
        image_element = product_element.find_element(By.XPATH, "//div[@class='_95X4G']/a/div/img")
        
        # Get the src attribute of the <img> element
        info['image'] = image_element.get_attribute("src")

    except NoSuchElementException:
        pass

    # If we decide to get extra information
    if extra_info:
        # sales
        try:
            # Find the <span> element within the <div class="_6uN7R">
            sold_element = product_element.find_element(By.XPATH, "//div[@class='_6uN7R']/span[@class='_1cEkb']/span[1]")
            
            # Get the text content of the <span> element
            info['sales'] = sold_element.text

        except (NoSuchElementException, ValueError):
            info['sales'] = 0

        try:
            # Find the <span> element within the <div class="WNoq3">
            discount_element = product_element.find_element(By.XPATH, "//div[@class='WNoq3']/span[@class='IcOsH']")
            
            # Get the text content of the <span> element
            info['discount'] = discount_element.text

        except (NoSuchElementException, ValueError):
            info['discount'] = '0'

    return info

### Function to scrape all products from a page
def get_lazada_product_info_from_page(page_url, extra_info=False):
    """
    Extract info from all products of a specfic page_url on Tiki website
    Args:
        page_url: (string) url of the page to scrape
    Returns:
        data: (list) a list of dictionary of products info. If no products
                found, return empty list.
    """
    global DRIVER
    
    data = []
    DRIVER.get(page_url) # Use the driver to get info from the product page
    time.sleep(3)

    # FIND ALL PRODUCT ITEMS
    products = DRIVER.find_elements(By.CLASS_NAME, 'Bm3ON')
    print(products)
    print(f'Found {len(products)} products')

    if len(products)>0:
        for i in products:
            product_dict = get_lazada_product_info_single(i, extra_info)
            data.append(product_dict)
    return data

### Function to get product info from a main category
def get_lazada_product_info_from_category(search_product, max_page=0, extra_info=False):
    '''
    Scrape for multiple pages of products of a category.
    Uses get_product_info_from_page().

    Args:
        cat_url: (string) a url string of a category
        max_page: (int) an integer denoting the maximum number of pages to scrape.
                  Default value is 0 to scrape all pages.
    Returns:
        products: a list in which every element is a dictionary of one product's information
    '''
    products = []
    page_n = 1
    cat_url = 'https://www.lazada.vn/catalog/?q=' + search_product

    product_list = get_lazada_product_info_from_page(cat_url, extra_info=extra_info)

    while len(product_list)>0:
        products.extend(product_list)
        page_n += 1

        # stop_flag = False if max_page <= 0 else (page_n > max_page)
        stop_flag = max_page>0 and page_n>max_page # For stopping the scrape according to max_page
        if stop_flag:
            break
        cat_url = 'https://www.lazada.vn/catalog/?page=' + page_n + '&q=' + search_product
        product_list = get_lazada_product_info_from_page(cat_url, extra_info=extra_info)

    return products

def scrap_lazada(search_product, num_max_page, extra_info):

    start_driver(force_restart=True)

    prod_data = [] # STORE YOUR PRODUCT INFO DICTIONARIES IN HERE

    prod_per_cat = get_lazada_product_info_from_category(search_product, num_max_page, extra_info=extra_info)
    prod_data.extend(prod_per_cat)

    close_driver() # Close driver when we're done

    return prod_data