爬取目标

1.本次代码是在python2上运行通过的，python3的最需改2行代码，用到其它python模块

• selenium 2.53.6 +firefox 44
• BeautifulSoup
• requests

2.爬取目标网站，我的博客：https://home.cnblogs.com/u/yoyoketang
爬取内容：爬我的博客的所有粉丝的名称，并保存到txt

3.由于博客园的登录是需要人机验证的，所以是无法直接用账号密码登录，需借助selenium登录

selenium获取cookies

1.大前提：先手工操作浏览器，登录我的博客，并记住密码
（保证关掉浏览器后，下次打开浏览器访问我的博客时候是登录状态）
2.selenium默认启动浏览器是一个空的配置，默认不加载配置缓存文件，这里先得找到对应浏览器的配置文件地址，以火狐浏览器为例
3.使用driver.get_cookies()方法获取浏览器的cookies

# coding:utf-8
import requests
from selenium import webdriver
from bs4 import BeautifulSoup
import re
import time
# firefox浏览器配置文件地址
profile_directory = r‘C:\Users\admin\AppData\Roaming\Mozilla\Firefox\Profiles\yn80ouvt.default‘
# 加载配置
profile = webdriver.FirefoxProfile(profile_directory)
# 启动浏览器配置
driver = webdriver.Firefox(profile)

driver.get("https://home.cnblogs.com/u/yoyoketang/followers/")

time.sleep(3)
cookies = driver.get_cookies()  # 获取浏览器cookies
print(cookies)
driver.quit()

（注：要是这里脚本启动浏览器后，打开的博客页面是未登录的，后面内容都不用看了，先检查配置文件是不是写错了）

requests添加登录的cookies

1.浏览器的cookies获取到后，接下来用requests去建一个session，在session里添加登录成功后的cookies

s = requests.session()  # 新建session

# 添加cookies到CookieJar
c = requests.cookies.RequestsCookieJar()
for i in cookies:
    c.set(i["name"], i[‘value‘])

s.cookies.update(c)  # 更新session里cookies

计算粉丝数和分页总数

1.由于我的粉丝的数据是分页展示的，这里一次只能请求到45个，所以先获取粉丝总数，然后计算出总的页数

# 发请求
r1 = s.get("https://home.cnblogs.com/u/yoyoketang/relation/followers")

soup = BeautifulSoup(r1.content, "html.parser")

# 抓取我的粉丝数
fensinub = soup.find_all(class_="current_nav")
print fensinub[0].string
num = re.findall(u"我的粉丝\((.+?)\)", fensinub[0].string)
print u"我的粉丝数量：%s"%str(num[0])

# 计算有多少页，每页45条
ye = int(int(num[0])/45)+1
print u"总共分页数：%s"%str(ye)

保存粉丝名到txt

# 抓取第一页的数据
fensi = soup.find_all(class_="avatar_name")
for i in fensi:
    name = i.string.replace("\n", "").replace(" ","")
    print name
    with open("name.txt", "a") as f:  # 追加写入
        f.write(name.encode("utf-8")+"\n")

# 抓第二页后的数据
for i in range(2, ye+1):
    r2 = s.get("https://home.cnblogs.com/u/yoyoketang/relation/followers?page=%s"%str(i))
    soup = BeautifulSoup(r1.content, "html.parser")
    # 抓取我的粉丝数
    fensi = soup.find_all(class_="avatar_name")

    for i in fensi:
        name = i.string.replace("\n", "").replace(" ","")
        print name
        with open("name.txt", "a") as f:  # 追加写入
            f.write(name.encode("utf-8")+"\n")

参考代码：

# coding:utf-8
import requests
from selenium import webdriver
from bs4 import BeautifulSoup
import re
import time

# firefox浏览器配置文件地址
profile_directory = r‘C:\Users\admin\AppData\Roaming\Mozilla\Firefox\Profiles\yn80ouvt.default‘

s = requests.session()  # 新建session
url = "https://home.cnblogs.com/u/yoyoketang"

def get_cookies(url):
    ‘‘‘启动selenium获取登录的cookies‘‘‘
    try:
        # 加载配置
        profile = webdriver.FirefoxProfile(profile_directory)
        # 启动浏览器配置
        driver = webdriver.Firefox(profile)
        driver.get(url+"/followers")

        time.sleep(3)
        cookies = driver.get_cookies()  # 获取浏览器cookies
        print(cookies)
        driver.quit()
        return cookies
    except Exception as msg:
        print(u"启动浏览器报错了：%s" %str(msg))
def add_cookies(cookies):
    ‘‘‘往session添加cookies‘‘‘
    try:
        # 添加cookies到CookieJar
        c = requests.cookies.RequestsCookieJar()
        for i in cookies:
            c.set(i["name"], i[‘value‘])

        s.cookies.update(c)  # 更新session里cookies
    except Exception as msg:
        print(u"添加cookies的时候报错了：%s" % str(msg))

def get_ye_nub(url):
    ‘‘‘获取粉丝的页面数量‘‘‘
    try:
        # 发请求
        r1 = s.get(url+"/relation/followers")
        soup = BeautifulSoup(r1.content, "html.parser")
        # 抓取我的粉丝数
        fensinub = soup.find_all(class_="current_nav")
        print(fensinub[0].string)
        num = re.findall(u"我的粉丝\((.+?)\)", fensinub[0].string)
        print(u"我的粉丝数量：%s"%str(num[0]))

        # 计算有多少页，每页45条
        ye = int(int(num[0])/45)+1
        print(u"总共分页数：%s"%str(ye))
        return ye
    except Exception as msg:
        print(u"获取粉丝页数报错了，默认返回数量1 ：%s"%str(msg))
        return 1

def save_name(nub):
    ‘‘‘抓取页面的粉丝名称‘‘‘
    try:
        # 抓取第一页的数据
        if nub <= 1:
            url_page = url+"/relation/followers"
        else:
            url_page = url+"/relation/followers?page=%s" % str(nub)
        print(u"正在抓取的页面：%s" %url_page)
        r2 = s.get(url_page, verify=False)
        soup = BeautifulSoup(r2.content, "html.parser")
        fensi = soup.find_all(class_="avatar_name")
        for i in fensi:
            name = i.string.replace("\n", "").replace(" ","")
            print(name)
            with open("name.txt", "a") as f:  # 追加写入
                f.write(name.encode("utf-8")+"\n")

            # python3的改成下面这两行
            # with open("name.txt", "a", encoding="utf-8") as f:  # 追加写入
            #     f.write(name+"\n")

    except Exception as msg:
        print(u"抓取粉丝名称过程中报错了 ：%s"%str(msg))

if __name__ == "__main__":
    cookies = get_cookies(url)
    add_cookies(cookies)
    n = get_ye_nub(url)
    for i in list(range(1, n+1)):
        save_name(i)

原文地址：https://www.cnblogs.com/jason89/p/8998209.html