python+selenium+PhantomJS抓取網(wǎng)頁動(dòng)態(tài)加載內(nèi)容

from selenium import webdriver
 
 
driver = webdriver.PhantomJS()
driver.get('http://www.baidu.com')  #加載網(wǎng)頁
data = driver.page_source  #獲取網(wǎng)頁文本
driver.save_screenshot('1.png')  #截圖保存
print(data)
driver.quit()

from selenium import webdriver
from selenium.webdriver.common.desired_capabilities import DesiredCapabilities
 
 
dcap = dict(DesiredCapabilities.PHANTOMJS) #設(shè)置useragent
dcap['phantomjs.page.settings.userAgent'] = ('Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:25.0) Gecko/20100101 Firefox/25.0 ') #根據(jù)需要設(shè)置具體的瀏覽器信息
driver = webdriver.PhantomJS(desired_capabilities=dcap) #封裝瀏覽器信息
driver.get('http://www.baidu.com')  #加載網(wǎng)頁
data = driver.page_source  #獲取網(wǎng)頁文本
driver.save_screenshot('1.png')  #截圖保存
print(data)
driver.quit()

from selenium import webdriver
 
driver = webdriver.PhantomJS()
driver.set_page_load_timeout(5) #設(shè)置超時(shí)時(shí)間
driver.get('http://www.baidu.com')
print(driver.title)
driver.quit()

from selenium import webdriver
 
driver = webdriver.PhantomJS()
driver.set_page_load_timeout(5)
driver.get('http://www.baidu.com')
try:
  driver.get('http://www.baidu.com')
  driver.find_element_by_id('kw') # 通過ID定位
  driver.find_element_by_class_name('s_ipt') # 通過class屬性定位
  driver.find_element_by_name('wd') # 通過標(biāo)簽name屬性定位
  driver.find_element_by_tag_name('input') # 通過標(biāo)簽屬性定位
  driver.find_element_by_css_selector('#kw') # 通過css方式定位
  driver.find_element_by_xpath("http://input[@id='kw']") # 通過xpath方式定位
  driver.find_element_by_link_text("貼吧") # 通過xpath方式定位
  print(driver.find_element_by_id('kw').tag_name ) # 獲取標(biāo)簽的類型
except Exception as e:
  print(e)
driver.quit()

from selenium import webdriver
driver = webdriver.PhantomJS()
try:
  driver.get('http://www.baidu.com')  #訪問百度首頁
  driver.save_screenshot('1.png')
  driver.get('http://www.sina.com.cn') #訪問新浪首頁
  driver.save_screenshot('2.png')
  driver.back()              #回退到百度首頁
  driver.save_screenshot('3.png')
  driver.forward()            #前進(jìn)到新浪首頁
  driver.save_screenshot('4.png')
except Exception as e:
  print(e)
driver.quit()