爬虫入门Task03

为什么会出现IP被封

网站为了防止被爬取，会有反爬机制，对于同一个IP地址的大量同类型的访问，会封锁IP，过一段时间后，才能继续访问。

如何应对IP被封的问题

有几种套路：
修改请求头，模拟浏览器（而不是代码去直接访问）去访问
采用代理IP并轮换
设置访问时间间隔

如何获取代理IP地址

从该网站获取： https://www.xicidaili.com/
inspect -> 鼠标定位：
要获取的代理IP地址，属于class = "odd"标签的内容：代码如下，获取的代理IP保存在proxy_ip_list列表中

# 案例代码
from bs4 import BeautifulSoup
import requests
import time

def open_proxy_url(url):
    user_agent = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.119 Safari/537.36'
    headers = {'User-Agent': user_agent}
    try:
        r = requests.get(url, headers = headers, timeout = 20)
        r.raise_for_status()
        r.encoding = r.apparent_encoding
        return(r.text)
    except:
        print('无法访问网页' + url)


def get_proxy_ip(response):
    proxy_ip_list = []
    soup = BeautifulSoup(response, 'html.parser')
    proxy_ips  = soup.select('.odd')#选择标签
    for proxy_ip in proxy_ips:
        ip = proxy_ip.select('td')[1].text
        port = proxy_ip.select('td')[2].text
        protocol = proxy_ip.select('td')[5].text
        if protocol in ('HTTP','HTTPS'):
            proxy_ip_list.append(f'{protocol}://{ip}:{port}')
    return proxy_ip_list

if __name__ == '__main__':
    proxy_url = 'https://www.xicidaili.com/'
    text = open_proxy_url(proxy_url)
    proxy_ip_filename = 'proxy_ip.txt'
    with open(proxy_ip_filename, 'w') as f:
        f.write(text)
    text = open(proxy_ip_filename, 'r').read()
    proxy_ip_list = get_proxy_ip(text)
    print(proxy_ip_list)

通过bs4的find_all(‘tr’)来获取所有IP：

def get_proxy_ip(response):
    proxy_ip_list = []
    soup = BeautifulSoup(response, 'html.parser')
    proxy_ips = soup.find(id = 'ip_list').find_all('tr')
    for proxy_ip in proxy_ips:
        if len(proxy_ip.select('td')) >=8:
            ip = proxy_ip.select('td')[1].text
            port = proxy_ip.select('td')[2].text
            protocol = proxy_ip.select('td')[5].text
            if protocol in ('HTTP','HTTPS','http','https'):
                proxy_ip_list.append(f'{protocol}://{ip}:{port}')
    return proxy_ip_list

使用代理

proxies的格式是一个字典：
{‘http’: ‘http://IP:port‘,‘https’:'https://IP:port‘}
把它直接传入requests的get方法中即可
web_data = requests.get(url, headers=headers, proxies=proxies)

def open_url_using_proxy(url, proxy):
    user_agent = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.119 Safari/537.36'
    headers = {'User-Agent': user_agent}
    proxies = {}
    if proxy.startswith('HTTPS'):
        proxies['https'] = proxy
    else:
        proxies['http'] = proxy
    try:
        r = requests.get(url, headers = headers, proxies = proxies, timeout = 10)
        r.raise_for_status()
        r.encoding = r.apparent_encoding
        return (r.text, r.status_code)
    except:
        print('无法访问网页' + url)
        return False
url = 'http://www.baidu.com'
text = open_url_using_proxy(url, proxy_ip_list[0])

确认代理IP地址有效性

无论是免费还是收费的代理网站，提供的代理IP都未必有效，我们应该验证一下，有效后，再放入我们的代理IP池中，以下通过几种方式：访问网站，得到的返回码是200真正的访问某些网站，获取title等，验证title与预计的相同访问某些可以提供被访问IP的网站，类似于“查询我的IP”的网站，查看返回的IP地址是什么验证返回码。

def check_proxy_avaliability(proxy):
    url = 'http://www.baidu.com'
    result = open_url_using_proxy(url, proxy)
    VALID_PROXY = False
    if result:
        text, status_code = result
        if status_code == 200:
            print('有效代理IP: ' + proxy)
        else:
            print('无效代理IP: ' + proxy)

改进：确认网站title

def check_proxy_avaliability(proxy):
    url = 'http://www.baidu.com'
    text, status_code = open_url_using_proxy(url, proxy)
    VALID = False
    if status_code == 200:
        if r_title:
            if r_title[0] == '<title>百度一下，你就知道</title>':
                VALID = True
    if VALID:
        print('有效代理IP: ' + proxy)
    else:
        print('无效代理IP: ' + proxy)

完整代码

from bs4 import BeautifulSoup
import requests
import re
import json


def open_proxy_url(url):
    user_agent = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.119 Safari/537.36'
    headers = {'User-Agent': user_agent}
    try:
        r = requests.get(url, headers = headers, timeout = 10)
        r.raise_for_status()
        r.encoding = r.apparent_encoding
        return r.text
    except:
        print('无法访问网页' + url)


def get_proxy_ip(response):
    proxy_ip_list = []
    soup = BeautifulSoup(response, 'html.parser')
    proxy_ips = soup.find(id = 'ip_list').find_all('tr')
    for proxy_ip in proxy_ips:
        if len(proxy_ip.select('td')) >=8:
            ip = proxy_ip.select('td')[1].text
            port = proxy_ip.select('td')[2].text
            protocol = proxy_ip.select('td')[5].text
            if protocol in ('HTTP','HTTPS','http','https'):
                proxy_ip_list.append(f'{protocol}://{ip}:{port}')
    return proxy_ip_list


def open_url_using_proxy(url, proxy):
    user_agent = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.119 Safari/537.36'
    headers = {'User-Agent': user_agent}
    proxies = {}
    if proxy.startswith(('HTTPS','https')):
        proxies['https'] = proxy
    else:
        proxies['http'] = proxy

    try:
        r = requests.get(url, headers = headers, proxies = proxies, timeout = 10)
        r.raise_for_status()
        r.encoding = r.apparent_encoding
        return (r.text, r.status_code)
    except:
        print('无法访问网页' + url)
        print('无效代理IP: ' + proxy)
        return False


def check_proxy_avaliability(proxy):
    url = 'http://www.baidu.com'
    result = open_url_using_proxy(url, proxy)
    VALID_PROXY = False
    if result:
        text, status_code = result
        if status_code == 200:
            r_title = re.findall('<title>.*</title>', text)
            if r_title:
                if r_title[0] == '<title>百度一下，你就知道</title>':
                    VALID_PROXY = True
        if VALID_PROXY:
            check_ip_url = 'https://jsonip.com/'
            try:
                text, status_code = open_url_using_proxy(check_ip_url, proxy)
            except:
                return

            print('有效代理IP: ' + proxy)
            with open('valid_proxy_ip.txt','a') as f:
                f.writelines(proxy)
            try:
                source_ip = json.loads(text).get('ip')
                print(f'源IP地址为：{source_ip}')
                print('='*40)
            except:
                print('返回的非json,无法解析')
                print(text)
    else:
        print('无效代理IP: ' + proxy)


if __name__ == '__main__':
    proxy_url = 'https://www.xicidaili.com/'
    proxy_ip_filename = 'proxy_ip.txt'
    text = open(proxy_ip_filename, 'r').read()
    proxy_ip_list = get_proxy_ip(text)
    for proxy in proxy_ip_list:
        check_proxy_avaliability(proxy)

实战项目

挑战项目：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。
丁香园论坛：http://www.dxy.cn/bbs/thread/626626#626626 。

import requests, json, re, random,time
from bs4 import BeautifulSoup
from selenium import webdriver
from lxml import etree

class getUrl(object):
	"""docstring for getUrl"""
	def __init__(self):
		self.headers={
            "Connection": "keep-alive",
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 "  
                          "(KHTML, like Gecko) Chrome/51.0.2704.63 Safari/537.36",
            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",
            "Accept-Encoding": "gzip, deflate, sdch",
            "Accept-Language": "zh-CN,zh;q=0.8"
        };

	def run(self):
		browser = webdriver.Chrome('/home/ach/桌面/爬虫/学习任务/chromedriver')
		browser.get('https://auth.dxy.cn/accounts/login?service=http://www.dxy.cn/bbs/index.html')
		time.sleep(1)
		#切换账号密码登录表单
		js1 = 'document.querySelector("#j_loginTab1").style.display="none";'
		browser.execute_script(js1)
		time.sleep(1)
		js2 = 'document.querySelector("#j_loginTab2").style.display="block";'
		browser.execute_script(js2)
		#输入账号密码
		input_name = browser.find_element_by_name('username')
		input_name.clear()
		input_name.send_keys('*')# 这里为自己账号和密码
		input_pass = browser.find_element_by_name('password')
		input_pass.clear()
		input_pass.send_keys('*')
		browser.find_element_by_xpath('//*[@class="form__button"]/button').click()
		#此步骤应该有验证码，先跳过
		time.sleep(10)
		cookie = browser.get_cookies()
		cookie_dict = {i['name']:i['value'] for i in cookie}
		#转到抓取页面
		browser.get("http://www.dxy.cn/bbs/thread/626626#626626");
		html = browser.page_source
		tree = etree.HTML(html)
		user = tree.xpath('//div[@id="postcontainer"]//div[@class="auth"]/a/text()')
		content = tree.xpath('//td[@class="postbody"]')
		for i in range(0,len(user)):
			result = user[i].strip()+":"+content[i].xpath('string(.)').strip()
			#写入文件
			dir_file = open("DXY_records.txt",'a', encoding="utf-8")
			dir_file.write(result+"\n")
			dir_file.write('*' * 80+"\n")
			dir_file.close()
		print('*' * 5 +"抓取结束"+'*' * 5)


if __name__ == '__main__':
	geturl = getUrl()
	geturl.run()