爬虫学习（四）

1.任务

模拟登录丁香园，并抓取论坛页面所有的人员基本信息与回复帖子内容
http://www.dxy.cn/bbs/thread/626626#626626

2.实现

思路：

1.selenium模拟浏览器登陆
2.xpath匹配获取内容存储数据

代码：

import time
from selenium import webdriver
from selenium.common.exceptions import TimeoutException, NoSuchElementException

def login(browser, username, password):
    try:
        # 点击“登录”
        browser.find_element_by_link_text('登录').click()
        # 点击“电脑登录”
        browser.find_element_by_class_name('ico_pc').click()
        # 输入账号
        element = browser.find_element_by_name('username')
        element.clear()
        element.send_keys(username)
        # 输入密码
        element = browser.find_element_by_name('password')
        element.clear()
        element.send_keys(password)
        # 点击“登录按钮”
        browser.find_element_by_class_name('button').click()
    except TimeoutException:
        print('Time out')
    except NoSuchElementException:
        print('No Element')

def get_content(browser):
    print("success login")
    time.sleep(5)
    auth = browser.find_elements_by_class_name('auth')  # 昵称
    level = browser.find_elements_by_class_name('info') # 等级
    user_atten = browser.find_elements_by_class_name('user_atten')  # 用户属性
    content = browser.find_elements_by_class_name('postbody')   # 内容
    sfile = open('data.txt','a',encoding='utf-8')
    for i in range(len(content)):
        num = user_atten[i].find_elements_by_tag_name('a')
        data = str({'num':i+1,'name':auth[i].text,'level':level[i].text,'score':num[0].text,'vote':num[2].text,'dingdang':num[4].text,'content':content[i].text.replace(" ", "").replace("\n", "")}) +"\n" 
        sfile.writelines(data)
    print("success write")
    sfile.close()

if __name__ == '__main__':
    browser = webdriver.Firefox()
    browser.get('http://www.dxy.cn/bbs/thread/626626#626626')
    login(browser, 'username', 'password')
    get_content(browser)
    browser.quit()

3.结果

## 4.参考资料
爬虫学习day4
爬虫day4