request、bs4爬虫
一 先从爬虫案例开始
爬虫和反爬虫之间的斗争,看似反爬虫占据着主动权,但最后都爬虫者获胜,只是付出代价大小的问题;本文只在技术层面探讨爬虫入门知识,爬虫本质上就是猜和博弈,经验很重要,主要就是分析请求和web接口,因此,想学爬虫,必须先了解web;只要浏览器能够获得的数据,都能够通过爬虫获取的到,就看伪装的彻底不彻底了,这里先从几个简单爬虫说起:
1 汽车之家
这应该是跟爬取百度一样简单的网站,该网站完全没有设防,无需伪装成浏览器,不需要cookie,也无需登录:
import requests from bs4 import BeautifulSoup response = requests.get("https://www.autohome.com.cn/news/") response.encoding = 'gbk' soup = BeautifulSoup(response.text,'html.parser') div = soup.find(name='div',attrs={'id':'auto-channel-lazyload-article'}) li_list = div.find_all(name='li') for li in li_list: title = li.find(name='h3') if not title: continue p = li.find(name='p') a = li.find(name='a') print(title.text) print(a.attrs.get('href')) print(p.text) img = li.find(name='img') src = img.get('src') src = "https:" + src print(src) # 再次发起请求,下载图片 file_name = src.rsplit('/',maxsplit=1)[1] ret = requests.get(src) with open(file_name,'wb') as f: f.write(ret.content)
2 抽屉新热榜
第一步爬取网页内容,需要伪装成浏览器:
import requests from bs4 import BeautifulSoup r1 = requests.get( url='https://dig.chouti.com/', headers={ 'user-agent':'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36' } ) soup = BeautifulSoup(r1.text,'html.parser') # 标签对象 content_list = soup.find(name='div',id='content-list') # print(content_list) # [标签对象,标签对象] item_list = content_list.find_all(name='div',attrs={'class':'item'}) for item in item_list: a = item.find(name='a',attrs={'class':'show-content color-chag'}) print(a.text.strip()) # print(a.text)
更进一步,给通过文章点赞,注意陷阱,点赞需要cookie,但是不是登录后响应中的cookie,而是第一次加载网页时响应中的cookie,可以看出如果爬虫直接从登录那一步开始发请求是不行的,因为正常的浏览器都是先访问一下页面才登录的,从点赞使用的cookie可以看出,这是反爬虫的一种方案,这就要求爬虫者具有一个的分析能和经验了,不然即使获取到了cookie也是一个烟雾弹,具体如下:
import requests # 1. 查看首页 r1 = requests.get( url='https://dig.chouti.com/', headers={ 'user-agent':'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36' } ) # 2. 提交用户名和密码 r2 = requests.post( url='https://dig.chouti.com/login', headers={ 'user-agent':'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36' }, data={ 'phone':'自己先注册', 'password':'自己先注册', 'oneMonth':1 }, cookies=r1.cookies.get_dict() ) # 3. 点赞 r3 = requests.post( url='https://dig.chouti.com/link/vote?linksId=20435396', headers={ 'user-agent':'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36' }, cookies=r1.cookies.get_dict() ) print(r3.text)
3 自动登录GitHub
这里登录的时候是form表单提交,需要获取csrf token值防止跨站伪造请求,登录这里的 token值跟登录名和密码一块放在了请求体中;这也是分析得到的,如果一个网站登录或提交数据需要token值的话,放在请求头中还是请求体中,或者cookie中,或者其他参数中,需要自己具体真实登录一下就能分析到了;具体如下:
# 1. GET,访问登录页面 """ - 去HTML中找隐藏的Input标签获取csrf token - 获取cookie """ # 2. POST,用户名和密码 """ - 发送数据: - csrf - 用户名 - 密码 - 携带cookie """ # 3. GET,访问https://github.com/settings/emails """ - 携带 cookie """ import requests from bs4 import BeautifulSoup # # 1. 访问登陆页面,获取 authenticity_token i1 = requests.get('https://github.com/login') soup1 = BeautifulSoup(i1.text, features='lxml') tag = soup1.find(name='input', attrs={'name': 'authenticity_token'}) authenticity_token = tag.get('value') c1 = i1.cookies.get_dict() i1.close() # 1. 携带authenticity_token和用户名密码等信息,发送用户验证 form_data = { "authenticity_token": authenticity_token, "utf8": "", "commit": "Sign in", "login": "自己注册", 'password': '自己注册' } i2 = requests.post('https://github.com/session', data=form_data, cookies=c1) c2 = i2.cookies.get_dict() c1.update(c2) i3 = requests.get('https://github.com/settings/repositories', cookies=c1) soup3 = BeautifulSoup(i3.text, features='lxml') list_group = soup3.find(name='div', class_='listgroup') from bs4.element import Tag for child in list_group.children: if isinstance(child, Tag): project_tag = child.find(name='a', class_='mr-1') size_tag = child.find(name='small') temp = "项目:%s(%s); 项目路径:%s" % (project_tag.get('href'), size_tag.string, project_tag.string, ) print(temp)
4 自动登录拉钩网
这里需要在请求页面的响应数据中获取两个隐藏在网页中的参数,然后把参数放在请求头中才能登录成功,这里注意,无论解析网页用的是bs4还是xpath,都不能忘记正则表达式,它能解决其他方式无法解决的一些问题,具体如下:
import re import requests r1 = requests.get( url='https://passport.lagou.com/login/login.html', headers={ 'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36', } ) X_Anti_Forge_Token = re.findall("X_Anti_Forge_Token = '(.*?)'", r1.text, re.S)[0] X_Anti_Forge_Code = re.findall("X_Anti_Forge_Code = '(.*?)'", r1.text, re.S)[0] # print(X_Anti_Forge_Token, X_Anti_Forge_Code) # print(r1.text) # r2 = requests.post( url='https://passport.lagou.com/login/login.json', headers={ 'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36', 'X-Anit-Forge-Code':X_Anti_Forge_Code, 'X-Anit-Forge-Token':X_Anti_Forge_Token, 'Referer': 'https://passport.lagou.com/login/login.html', # 上一次请求地址是什么? }, data={ "isValidate": True, 'username': '自己注册', 'password': 'ab18d270d7126ea65915c50288c22c0d', 'request_form_verifyCode': '', 'submit': '' }, cookies=r1.cookies.get_dict() ) print(r2.text)
二 requests模块
1 请求方法
requests.get
requests.post
requests.put
requests.delete
等等等等,网络中常见的请求方法都有,具体还想了解更多的,可以参考requests原码查看
另一种写法:requests.request(method='POST')
2 请求参数
2.1 url
2.2 headers
2.3 cookies
2.4 params
2.5 data
传请求体 requests.post(
...,
data={'user':'liuneng','pwd':'123'}
)
GET /index http1.1\r\nhost:c1.com\r\n\r\nuser=liuneng&pwd=123
2.6 json,传请求体
requests.post(
...,
json={'user':'liuneng','pwd':'123'}
)
GET /index http1.1\r\nhost:c1.com\r\nContent-Type:application/json\r\n\r\n{"user":"liuneng","pwd":123}
2.7 代理 proxies
# 无验证
proxie_dict = {
"http": "61.172.249.96:80",
"https": "http://61.185.219.126:3128",
}
ret = requests.get("https://www.proxy360.cn/Proxy", proxies=proxie_dict)
# 验证代理
from requests.auth import HTTPProxyAuth
proxyDict = {
'http': '77.75.105.165',
'https': '77.75.106.165'
}
auth = HTTPProxyAuth('用户名', '密码')
r = requests.get("http://www.google.com",data={'xxx':'ffff'} proxies=proxyDict, auth=auth)
print(r.text)
-----------------------------------------------------------------------------------------上面的几项必须掌握
2.8 文件上传 files
# 发送文件
file_dict = {
'f1': open('xxxx.log', 'rb')
}
requests.request(
method='POST',
url='http://127.0.0.1:8000/test/',
files=file_dict
)
2.9 认证 auth
内部:
用户名和密码,用户和密码加密,放在请求头中传给后台。
- "用户:密码"
- base64("用户:密码")
- "Basic base64("用户|密码")"
- 请求头:
Authorization: "basic base64("用户|密码")"
from requests.auth import HTTPBasicAuth, HTTPDigestAuth
ret = requests.get('https://api.github.com/user', auth=HTTPBasicAuth('wupeiqi', 'sdfasdfasdf'))
print(ret.text)
2.10 超时 timeout
# ret = requests.get('http://google.com/', timeout=1)
# print(ret)
# ret = requests.get('http://google.com/', timeout=(5, 1))
# print(ret)
2.11 允许重定向 allow_redirects
ret = requests.get('http://127.0.0.1:8000/test/', allow_redirects=False)
print(ret.text)
2.12 大文件下载 stream
from contextlib import closing
with closing(requests.get('http://httpbin.org/get', stream=True)) as r1:
# 在此处理响应。
for i in r1.iter_content():
print(i)
2.13 证书 cert
- 百度、腾讯 => 不用携带证书(系统帮你做了)
- 自定义证书
requests.get('http://127.0.0.1:8000/test/', cert="xxxx/xxx/xxx.pem")
requests.get('http://127.0.0.1:8000/test/', cert=("xxxx/xxx/xxx.pem","xxx.xxx.xx.key"))
2.14 确认 verify =False
requests.get('http://127.0.0.1:8000/test/', cert="xxxx/xxx/xxx.pem")
出处:https://www.cnblogs.com/liujiajia_me/
本文版权归作者和博客园共有,不得转载,未经作者同意参考时必须保留此段声明,且在文章页面明显位置给出原文连接。