抓取无限滚动页面

Question

我正在尝试使用 python 进行网络抓取，这是一个在巴西（5 安达尔）非常有名的租赁 houses/apartment 网站。

我需要输入每个元素并在其中抓取一些信息。关于如何做的任何提示？因为它是无限滚动类型的页面？

OBS：现在我已经可以输入每个元素并抓取数据了。我唯一的问题是继续滚动/抓取新数据。

这是站点的 link：https://www.quintoandar.com.br/alugar/imovel/sao-paulo-sp-brasil 和它的 img

这是我目前所拥有的。它已经在处理第一批项目

from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
import time
import numpy as np
import pandas as pd

#Initializing the webdriver
options = webdriver.ChromeOptions()

#Change the path to where chromedriver is in your home folder.
path = 'chromedriver'
driver = webdriver.Chrome(executable_path=path, options=options)
driver.set_window_size(1600, 1024)

url = 'https://www.quintoandar.com.br/alugar/imovel/sao-paulo-sp-brasil'
driver.get(url)
time.sleep(5)

num_houses = 40
houses=[]

#Fix (scrolling the page a few items and going back to initial)
aux = driver.find_elements_by_xpath("//div[@class='sc-1qwl1yl-0 igVsBW']")
driver.execute_script("arguments[0].scrollIntoView();", aux[12])
time.sleep(1)
driver.execute_script("arguments[0].scrollIntoView();", aux[0])
time.sleep(1)

house_buttons = driver.find_elements_by_xpath("//div[@class='sc-1qwl1yl-0 igVsBW']")

for house_button in house_buttons:
    if (not 'Sem tempo pra procurar' in house_button.text) and (not 'Ainda não encontrou seu lar' in house_button.text):

        house_button.click()

        #Wait for new tab
        time.sleep(2)

        #Switch to it
        driver.switch_to.window(driver.window_handles[1])

        #Wait page load its infos
        time.sleep(4)

        try:
            title = driver.find_element_by_xpath("//h1[@class='sc-1q9n36n-0 ghXeyc sc-bdVaJa hgGleC']").text
            address = driver.find_element_by_xpath("//p[@data-testid='listing-address-subtitle']").text
        except:
            title = address = np.nan
        
        #General Infos
        try:
            infos = driver.find_elements_by_xpath("//div[@class='MuiGrid-root tptht-0 fAvqys MuiGrid-item MuiGrid-grid-xs-3 MuiGrid-grid-sm-3 MuiGrid-grid-md-1']")
            size = infos[0].text
            bedroom = infos[1].text
            bathroom = infos[2].text
            garage = infos[3].text
            floor = infos[4].text
            pet = infos[5].text
            furniture = infos[6].text
            subway = infos[7].text
        except:
            size = bedroom = bathroom = garage = floor = pet = furniture = subway = np.nan

        #Price Infos
        infos = driver.find_elements_by_xpath("//li[contains(@class, 'MuiListItem-root rf1epz-0')]")
        for info in infos:
            if 'Aluguel' in info.text: rent = info.text
            elif 'Condomínio' in info.text: other = info.text
            elif 'IPTU' in info.text: taxes = info.text
            elif 'Seguro incêndio' in info.text: insurance = info.text
            elif 'Taxa de serviço' in info.text: services = info.text
            elif 'Total' in info.text: total = info.text

        houses.append({
            "Title":title,
            "Address":address,
            "Size":size,
            "Bedroom":bedroom,
            "Garage":garage,
            "Floor":floor,
            "Pet":pet,
            "Size":size,
            "Subway":subway,
            "Rent":rent,
            "Other":other,
            "Taxes":taxes,
            "Insurance":insurance,
            "Services":services,
            "Total":total
        })

        #Close Tab and go back to main
        driver.close()
        driver.switch_to.window(driver.window_handles[0])
        time.sleep(.5)

Answer 1

你需要做的是：

找到您需要的主页按钮
向下滚动到此按钮
点击按钮，切换标签，获取数据，返回主标签
转到第 1 步。

这是执行此操作的代码（提取数据除外）：

    from selenium import webdriver
    import time
    import numpy as np


    url = 'https://www.quintoandar.com.br/alugar/imovel/sao-paulo-sp-brasil'
    xpath_house_buttons = "//div[@class='sc-1qwl1yl-0 igVsBW']"
    x_path_title = "//h1[@class='sc-1q9n36n-0 ghXeyc sc-bdVaJa hgGleC']"
    x_path_address = "//p[@data-testid='listing-address-subtitle']"
    num_houses = 40
    houses = []


    def scroll_to_house_button(driver, num_btn) -> bool:
        """
        returns true if it could scroll to the house button
        """
        try:
            house_buttons = driver.find_elements_by_xpath(xpath_house_buttons)
            driver.execute_script("arguments[0].scrollIntoView();", house_buttons[num_btn])
            return True
        except:
            return False
    def switch_to_house_tab(driver) -> bool:
        """
        returns true if switching tab was successful
        """
        try:
            driver.switch_to.window(driver.window_handles[1])
            return True
        except:
            return False
    def switch_to_main_tab(driver) -> bool:
        """
        returns true if switching tab was successful
        """
        try:
            driver.switch_to.window(driver.window_handles[0])
            return True
        except:
            return False
    def get_house_button_index(house_buttons, houses_scraped_text, index):
        """
        returns house_button's index in house_buttons
        """
        # at the beginning, the house to scrape is given by its index
        if len(houses_scraped_text) < 5:
            return index
        # afterwards, we try to find by comparing the buttons' content
        else:
            for i in reversed(range(len(house_buttons))):
                if (house_buttons[i].text == houses_scraped_text[-1]) and (house_buttons[i - 1].text == houses_scraped_text[-2]):
                    return i + 1


    # Initializing the webdriver
    driver = webdriver.Chrome()
    driver.set_window_size(1600, 1024)
    driver.get(url)


    # get data
    i = 0
    houses_scraped_text = []
    while len(houses) < num_houses:
        house_buttons = driver.find_elements_by_xpath(xpath_house_buttons)

        # as house_buttons never exceeds a length of 30,
        # we need a smart way of getting the next one
        index_btn = get_house_button_index(house_buttons, houses_scraped_text, i)
        house_button = house_buttons[index_btn]
        houses_scraped_text.append(house_button.text)

        # can't scroll to house button => wait 1 sec
        while not scroll_to_house_button(driver, num_btn=index_btn):
            time.sleep(1)
        print("scroll to house -- house", i + 1)

        # filter houses to be scraped
        if not ((not 'Sem tempo pra procurar' in house_button.text) and (not 'Ainda não encontrou seu lar' in house_button.text)):
            print("house filtered -- house", i + 1, "\n")
            i += 1 # you have to increment here to not loop over the same house forever
            continue

        # new house tab not open yet => wait 1 sec
        while len(driver.window_handles) != 2: # check number of open tabs
            house_button.click()
            time.sleep(1)
        print("new house tab opened -- house", i + 1)

        # can't switch yet => wait 1 sec
        while not switch_to_house_tab(driver):
            time.sleep(1)
        print("switched to new house tab -- house", i + 1)

        ##################
        # LOAD DATA HERE #
        ##################
        print("data loaded -- house", i + 1)

        # close tab
        driver.close()
        # can't switch back to main => wait 1 sec
        while not switch_to_main_tab(driver):
            time.sleep(1)
        print("house tab closed & switched back to main -- house", i + 1)
        
        print(len(houses), "house scraped\n")
        i += 1

抓取无限滚动页面

Scraping an Infinite Scroll Page

python

selenium

web-scraping

infinite-scroll