以前抓取动态网页是用PhantomJS + Selenium + ChromeDriver,但是新版的Selenium不支持PhantomJS了,程序跑的时候总会跳出一些warnings.
现在的操作是放弃PhantomJS,直接用headless ChromeDriver。可直接在Google主页下载个ChromeDriver,都是支持headless的。
下面的程序就是启动driver,抓取数据,关闭driver的例子。记得要关掉driver,不然会占内存。
# -*- coding: UTF-8 -*-
'''
@version: Python 3.6
@introduction:
@author:
@date: 2018-3
'''
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
# 启动driver
def init_web_driver():
global driver
chrome_options = Options()
chrome_options.add_argument('--headless')
chrome_options.add_argument('--disable-gpu')
driver_path = 'E:\chromedriver.exe' #这里放的就是下载的driver本地路径
driver = webdriver.Chrome(chrome_options=chrome_options, executable_path = driver_path)
# 关掉driver
def close_web_driver():
driver.quit()
def get_data():
driver.get('https://www.baidu.com')
driver.implicitly_wait(10) # wait up to 10 seconds for the elements to become available
# ====== 网页中静态部分抓取,采用BeautifulSoup去解析
html = driver.page_source # 获取网页html
html_soup = BeautifulSoup(html.text,"lxml")
time.sleep(0.1)
coin_list = html_soup .find(name='table', attrs={"class": "table maintable"})
# 页面元素的提取请查看 BeautifulSoup的用法
# ====== 网页中动态部分抓取,采用driver自带的方法
# 下面展示的从调用百度搜索,在搜索框中输入"headless chrome",然后获取结果。具体自行百度driver的用法
text = driver.find_element_by_css_selector('#kw')
search = driver.find_element_by_css_selector('#su')
text.send_keys('headless chrome')
# search
search.click()
driver.get_screenshot_as_file('search-result.png')
results = driver.find_elements_by_xpath('//div[@class="result c-container "]')
for result in results:
res = result.find_element_by_css_selector('a')
title = res.text
link = res.get_attribute('href')
print ('Title: %s \nLink: %s\n' % (title, link) )
if __name__ == '__main__':
init_web_driver()
get_data()
close_web_driver()