爬虫3 request3高级 代理操作、模拟登录、单线程+多任务异步协程
- HttpConnectinPool:
- 原因:
- 1.短时间内发起了高频的请求导致ip被禁
- 2.http连接池中的连接资源被耗尽
- 解决:
- 1.代理
- 2.headers中加入Conection:“close”
- 代理:代理服务器,可以接受请求然后将其转发。
- 匿名度
- 高匿:啥也不知道
- 匿名:知道你使用了代理,但是不知道你的真实ip
- 透明:知道你使用了代理并且知道你的真实ip
- 类型:
- http
- https
- 免费代理:
- www.goubanjia.com
- 快代理
- 西祠代理
- http://http.zhiliandaili.cn/ 智联HTTP的代理精灵
- cookie的处理
代理的写法示例:
import requests headers = { 'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/76.0.3809.132 Safari/537.36' } url = 'https://www.baidu.com/s?wd=ip' page_text1 = requests.get(url,headers=headers,proxies={'https':'183.166.171.51:8888'}).text with open('ip.html','w',encoding='utf-8') as fp: fp.write(page_text1)
一个代理很容易被封,这时候我们要构造一个代理池
#代理池:列表 import random proxy_list = [ {'https':'121.231.94.44:8888'}, {'https':'131.231.94.44:8888'}, {'https':'141.231.94.44:8888'} ] url = 'https://www.baidu.com/s?wd=ip' page_text = requests.get(url,headers=headers,proxies=random.choice(proxy_list)).text with open('ip.html','w',encoding='utf-8') as fp: fp.write(page_text)
如何构造代理池呢?其中一个方法如下
from lxml import etree ip_url = 'http://t.11jsq.com/index.php/api/entry?method=proxyServer.generate_api_url&packid=1&fa=0&fetch_key=&groupid=0&qty=4&time=1&pro=&city=&port=1&format=html&ss=5&css=&dt=1&specialTxt=3&specialJson=&usertype=2' page_text = requests.get(ip_url,headers=headers).text tree = etree.HTML(page_text) ip_list = tree.xpath('//body//text()') print(ip_list)
然后
import random headers = { 'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/76.0.3809.132 Safari/537.36', 'Connection':"close" } url = 'https://www.xicidaili.com/nn/%d' proxy_list_http = [] proxy_list_https = [] for page in range(1,20): new_url = format(url%page) ip_port = random.choice(ip_list) page_text = requests.get(new_url,headers=headers,proxies={'https':ip_port}).text tree = etree.HTML(page_text) #tbody不可以出现在xpath表达式中 tr_list = tree.xpath('//*[@id="ip_list"]//tr')[1:] for tr in tr_list: ip = tr.xpath('./td[2]/text()')[0] port = tr.xpath('./td[3]/text()')[0] t_type = tr.xpath('./td[6]/text()')[0] ips = ip+':'+port if t_type == 'HTTP': dic = { t_type: ips } proxy_list_http.append(dic) else: dic = { t_type:ips } proxy_list_https.append(dic) print(len(proxy_list_http),len(proxy_list_https))
#检测 for ip in proxy_list_http: response = requests.get('https://www/sogou.com',headers=headers,proxies={'https':ip}) if response.status_code == '200': print('检测到了可用ip')
模拟登录!!!
cookie的处理
- 手动处理:将cookie封装到headers中
- 自动处理:session对象。可以创建一个session对象,改对象可以像requests一样进行请求发送。不同之处在于如果在使用session进行请求发送的过程中产生了cookie,则cookie会被自动存储在session对象中。
手动加上cookie:
#对雪球网中的新闻数据进行爬取https://xueqiu.com/ headers = { 'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/76.0.3809.132 Safari/537.36', # 'Cookie':'aliyungf_tc=AQAAAAl2aA+kKgkAtxdwe3JmsY226Y+n; acw_tc=2760822915681668126047128e605abf3a5518432dc7f074b2c9cb26d0aa94; xq_a_token=75661393f1556aa7f900df4dc91059df49b83145; xq_r_token=29fe5e93ec0b24974bdd382ffb61d026d8350d7d; u=121568166816578; device_id=24700f9f1986800ab4fcc880530dd0ed' } url = 'https://xueqiu.com/v4/statuses/public_timeline_by_category.json?since_id=-1&max_id=20349203&count=15&category=-1' page_text = requests.get(url=url,headers=headers).json() page_text
自动添加cookie:
#创建session对象 session = requests.Session() session.get('https://xueqiu.com',headers=headers) url = 'https://xueqiu.com/v4/statuses/public_timeline_by_category.json?since_id=-1&max_id=20349203&count=15&category=-1' page_text = session.get(url=url,headers=headers).json() page_text
- 验证码的识别
- 超级鹰:
- 注册:(用户中心身份)
- 登陆:
- 创建一个软件:899370
- 下载示例代码
- 打码兔
- 云打码
超级鹰示例
import requests from hashlib import md5 class Chaojiying_Client(object): def __init__(self, username, password, soft_id): self.username = username password = password.encode('utf8') self.password = md5(password).hexdigest() self.soft_id = soft_id self.base_params = { 'user': self.username, 'pass2': self.password, 'softid': self.soft_id, } self.headers = { 'Connection': 'Keep-Alive', 'User-Agent': 'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0)', } def PostPic(self, im, codetype): """ im: 图片字节 codetype: 题目类型 参考 http://www.chaojiying.com/price.html """ params = { 'codetype': codetype, } params.update(self.base_params) files = {'userfile': ('ccc.jpg', im)} r = requests.post('http://upload.chaojiying.net/Upload/Processing.php', data=params, files=files, headers=self.headers) return r.json() def ReportError(self, im_id): """ im_id:报错题目的图片ID """ params = { 'id': im_id, } params.update(self.base_params) r = requests.post('http://upload.chaojiying.net/Upload/ReportError.php', data=params, headers=self.headers) return r.json() #识别古诗文网中的验证码 def tranformImgData(imgPath,t_type): chaojiying = Chaojiying_Client('bobo328410948', 'bobo328410948', '899370') im = open(imgPath, 'rb').read() return chaojiying.PostPic(im, t_type)['pic_str'] url = 'https://so.gushiwen.org/user/login.aspx?from=http://so.gushiwen.org/user/collect.aspx' page_text = requests.get(url,headers=headers).text tree = etree.HTML(page_text) img_src = 'https://so.gushiwen.org'+tree.xpath('//*[@id="imgCode"]/@src')[0] img_data = requests.get(img_src,headers=headers).content with open('./code.jpg','wb') as fp: fp.write(img_data) tranformImgData('./code.jpg',1004)
然后就可以轻松登录古诗文 网站啦!(注意验证码的刷新的机制和动态变化的请求参数)
- 动态变化的请求参数
- 通常情况下动态变化的请求参数都会被隐藏在前台页面源码中
(这里直接在页面搜__VIEWSTATE值,然后抓下来用它)
(用session 发送请求,保持验证码的一致性!)
s = requests.Session() url = 'https://so.gushiwen.org/user/login.aspx?from=http://so.gushiwen.org/user/collect.aspx' page_text = s.get(url,headers=headers).text tree = etree.HTML(page_text) img_src = 'https://so.gushiwen.org'+tree.xpath('//*[@id="imgCode"]/@src')[0] img_data = s.get(img_src,headers=headers).content with open('./code.jpg','wb') as fp: fp.write(img_data) #动态获取变化的请求参数 __VIEWSTATE = tree.xpath('//*[@id="__VIEWSTATE"]/@value')[0] __VIEWSTATEGENERATOR = tree.xpath('//*[@id="__VIEWSTATEGENERATOR"]/@value')[0] code_text = tranformImgData('./code.jpg',1004) print(code_text) login_url = 'https://so.gushiwen.org/user/login.aspx?from=http%3a%2f%2fso.gushiwen.org%2fuser%2fcollect.aspx' data = { '__VIEWSTATE': __VIEWSTATE, '__VIEWSTATEGENERATOR': __VIEWSTATEGENERATOR, 'from':'http://so.gushiwen.org/user/collect.aspx', 'email': 'www.zhangbowudi@qq.com', 'pwd': 'bobo328410948', 'code': code_text, 'denglu': '登录', } page_text = s.post(url=login_url,headers=headers,data=data).text with open('login.html','w',encoding='utf-8') as fp: fp.write(page_text)
# 普通单线程 和线程池的速度对比
from time import sleep
import time
from multiprocessing.dummy import Pool
start = time.time()
urls = [
'http://www.baidu.com',
'http://www.sougou.com',
'http://www.qq.com',
'https://www.iqiyi.com/'
]
def get_request(url):
print('正在下载',url)
time.sleep(2)
print('OK了',url)
# pool = Pool(3)
# pool.map(get_request,urls)
for url in urls:
get_request(url)
print('总耗时:',time.time()-start)
单线程+多任务异步协程
- 协程
- 在函数(特殊的函数)定义的时候,如果使用了async修饰的话,则改函数调用后会返回一个协程对象,并且函数内部的实现语句不会被立即执行
- 任务对象
- 任务对象就是对协程对象的进一步封装。任务对象==高级的协程对象==特殊的函数
- 任务对象时必须要注册到事件循环对象中
- 给任务对象绑定回调:爬虫的数据解析中
- 事件循环
- 当做是一个容器,容器中必须存放任务对象。
- 当启动事件循环对象后,则事件循环对象会对其内部存储任务对象进行异步的执行。
- aiohttp:支持异步网络请求的模块
模板如下 import asyncio def callback(task): #作为任务对象的回调函数 print('i am callback and ',task.result()) # task.result()就是异步函数的返回值 async def test(): print('i am test()') return 'bobo' c = test() #封装了一个任务对象 task = asyncio.ensure_future(c) task.add_done_callback(callback) # 绑定回调 #创建一个事件循环的对象 loop = asyncio.get_event_loop() loop.run_until_complete(task)
异步 I/O
asyncio 是用来编写 并发 代码的库,使用 async/await 语法。
asyncio 被用作多个提供高性能 Python 异步框架的基础,包括网络和网站服务,数据库连接库,分布式任务队列等等。
asyncio 往往是构建 IO 密集型和高层级 结构化 网络代码的最佳选择。
import asyncio import time start = time.time() #在特殊函数内部的实现中不可以出现不支持异步的模块代码 async def get_request(url): await asyncio.sleep(2) print('下载成功:',url) urls = [ 'www.1.com', 'www.2.com' ] tasks = [] for url in urls: c = get_request(url) task = asyncio.ensure_future(c) tasks.append(task) loop = asyncio.get_event_loop() #注意:挂起操作需要手动处理 loop.run_until_complete(asyncio.wait(tasks)) print(time.time()-start)
爬虫应用:
import requests import aiohttp import time import asyncio s = time.time() urls = [ 'http://127.0.0.1:5000/bobo', 'http://127.0.0.1:5000/jay' ] # async def get_request(url): # page_text = requests.get(url).text # return page_text async def get_request(url): async with aiohttp.ClientSession() as s: #这边不能用不支持异步的requests async with await s.get(url=url) as response: page_text = await response.text() print(page_text) return page_text tasks = [] for url in urls: c = get_request(url) task = asyncio.ensure_future(c) tasks.append(task) loop = asyncio.get_event_loop() loop.run_until_complete(asyncio.wait(tasks)) print(time.time()-s)
multiprocessing包是Python中的多进程管理包。
1、示例:
爬虫脚本:
import aiohttp
import asyncio
import time
from lxml import etree
start = time.time()
urls = [
'http://127.0.0.1:5000/bobo',
'http://127.0.0.1:5000/jay',
'http://127.0.0.1:5000/tom',
'http://127.0.0.1:5000/bobo',
'http://127.0.0.1:5000/jay',
'http://127.0.0.1:5000/tom',
'http://127.0.0.1:5000/bobo',
'http://127.0.0.1:5000/jay',
'http://127.0.0.1:5000/tom',
'http://127.0.0.1:5000/bobo',
'http://127.0.0.1:5000/jay',
'http://127.0.0.1:5000/tom'
]
#特殊的函数:请求发送和响应数据的捕获
#细节:在每一个with前加上async,在每一个阻塞操作的前边加上await
async def get_request(url):
async with aiohttp.ClientSession() as s:
#s.get(url,headers,proxy="http://ip:port",params)
async with await s.get(url) as response:
page_text = await response.text()#read()返回的是byte类型的数据
return page_text
#回调函数
def parse(task):
page_text = task.result()
tree = etree.HTML(page_text)
parse_data = tree.xpath('//li/text()')
print(parse_data)
tasks = []
for url in urls:
c = get_request(url)
task = asyncio.ensure_future(c)
task.add_done_callback(parse)
tasks.append(task)
loop = asyncio.get_event_loop()
loop.run_until_complete(asyncio.wait(tasks))
print(time.time()-start)
示例服务器:
from flask import Flask from time import sleep app = Flask(__name__) @app.route('/index') def index(): sleep(2) return 'hello' @app.route('/index1') def index1(): sleep(2) return 'hello1' if __name__ == '__main__': app.run()