【WEB UI自动化】Selenium爬取某用户的博客园首页的全部博客

忙了一上午学这个Selenium
考虑到我有备份博客园博客的需求，所以实战是仿着写了下面的代码，功能是Selenium爬取某用户博客园首页的全部博客
第一版只是把文章url和标题爬取了下来,博客网页另存为mhtml格式到本地
有时间会继续更新

from selenium import webdriver
from selenium.webdriver.common.action_chains import ActionChains
import time
import requests
from selenium.webdriver.common.keys import Keys
import win32api
import win32con

def load_photo(url, name):
    '''给定图片链接，将图片以某个名称下载到本地'''
    # url = 'http://img14.360buyimg.com/n1/s450x450_jfs/t1/148801/37/12770/118749/5f9d71e4E39f1e893/533675187c108953.jpg'
    reponse = requests.get(url)
    # name = 'd:/photo.jpg'
    with open(name, 'wb') as ft:
        ft.write(reponse.content)

def drop_scroll(browser):
    '''将滑条从头滚动到底,以便让浏览器充分加载'''
    for x in range(1, 11, 2):
        # time.sleep(0.5)
        j = x/10
        js = 'document.documentElement.scrollTop = document.documentElement.scrollHeight * %f' % j
        browser.execute_script(js)

def switch_window(browser):
    '''将browser的指令移到新打开的小窗口处'''
    # time.sleep(0.5) # 如果移转失败，请增大这个时间
    windows = browser.window_handles
    browser.switch_to.window(windows[-1])

def switch_window_back(browser):
    '''将browser的指令移回旧的小窗口'''
    windows = browser.window_handles
    browser.switch_to.window(windows[0])


def save_mhtml(DuringTime):
    win32api.keybd_event(17, 0, 0, 0)  # 按下ctrl
    win32api.keybd_event(65, 0, 0, 0)  # 按下a
    win32api.keybd_event(65, 0, win32con.KEYEVENTF_KEYUP, 0)  # 释放a
    win32api.keybd_event(83, 0, 0, 0)  # 按下s
    win32api.keybd_event(83, 0, win32con.KEYEVENTF_KEYUP, 0)  # 释放s
    win32api.keybd_event(17, 0, win32con.KEYEVENTF_KEYUP, 0)  # 释放ctrl
    time.sleep(1)  #缓冲1s
    win32api.keybd_event(13, 0, 0, 0)  # 按下enter
    win32api.keybd_event(13, 0, win32con.KEYEVENTF_KEYUP, 0)  # 释放enter
    win32api.keybd_event(13, 0, 0, 0)  # 按下enter
    win32api.keybd_event(13, 0, win32con.KEYEVENTF_KEYUP, 0)  # 释放enter
    win32api.keybd_event(27, 0, 0, 0)  # 按下Esc
    time.sleep(0.5)  # 缓冲1s
    win32api.keybd_event(27, 0, win32con.KEYEVENTF_KEYUP, 0)  # 释放Esc
    time.sleep(DuringTime)


# 构造网址
u_id= input('请输入您想查询的博客园用户用户名')
url = f'https://cnblogs.com/{u_id}'
page = int(input("要遍历的页数(从第一页开始)"))

# 打开博客园

options = webdriver.ChromeOptions()
options.add_argument('--save-page-as-mhtml')
browser = webdriver.Chrome(options=options)# 打开另存为mhtml功能
# browser.set_window_size(900, 500) # 设置窗口大小
# browser.set_window_position(300, 200) # 设置浏览器的位置
browser.get(url) #到达主页
#time.sleep(10)

# 遍历页面中每篇博客
count = 1 # 博客的编号
blog_urls=[]
for page_id in range(1,page+1):
    # 翻页
    print("Page:%d" %page_id)
    url = f'https://www.cnblogs.com/{u_id}/default.html?page={page_id}'
    browser.get(url)
    switch_window_back(browser)  # 将browser的指令移回到新标签页
    drop_scroll(browser)   # 拖到底，让页面加载完全
    blogs = browser.find_elements_by_xpath("//div[@class='postTitle']")  # 获取所有博客
    for blog in blogs:
        btn = blog.find_element_by_xpath(".//a[@class='postTitle2 vertical-middle']")
        blog_urls.append(btn.get_attribute('href'))
        print(btn.get_attribute('href'),btn.find_element_by_xpath("./span").text,sep=' ')
        print(count)
        count = count + 1



# 如果不需要下面的部分可以删了
for blog_url in blog_urls:
    browser.get(blog_url)
    save_mhtml(DuringTime=2)