爬虫——requests.get爬虫模块参数

地址和请求头参数--url和header

res = requests.get(url,headers=headers)　　向网站发起请求，并获取响应对象

参数

• url ：需要抓取的URL地址
• headers : 请求头
• timeout : 超时时间，超过时间会抛出异常

响应对象(res)属性

• encoding ：响应字符编码 res.encoding = 'utf-8'
• text ：字符串 网站源码
• content ：字节流 字符串网站源码
• status_code ：HTTP响应码
• url ：实际数据的URL地址

import requests

url = 'http://www.baidu.com/'    # 爬取百度网页
headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.1 \
    (KHTML, like Gecko) Chrome/14.0.835.163 Safari/535.1'}

res = requests.get(url, headers=headers)
print(res.encoding)     # 查看网站的编码格式 ISO-8859-1

# text属性获取响应内容（字符串）网站源码
res = requests.get(url,headers=headers)
res.encoding = 'utf-8'
html = res.text

# content属性获取响应内容（字节串 bytes）网站源码
res = requests.get(url,headers=headers)
html = res.content.decode('utf-8')

print(res.status_code)      # 查看响应码 200
print(res.url)              # 查看访问的URL地址 https://www.baidu.com/

非结构化数据的保存方式

像压缩文件zip、图片文件等都可以使用非结构化数据的保存方式

with open('xxx.jpg','wb') as f:
    f.write(res.content)

示例：保存赵丽颖图片到本地

import requests

url = 'http://dingyue.nosdn.127.net/lL1JH2YdpAWrzEhfp8BrJ8lTHa1602AEX9E7qpTpH5NzW1535203788506compressflag.jpg'
headers = {'User-Agent': 'Mozilla/5.0'}

html = requests.get(url, headers=headers).content

# 把图片保存到本地
with open('赵丽颖.jpg', 'wb') as f:
    f.write(html)

百度贴吧图片抓取

目标：抓取指定贴吧所有图片

思路

1. 获取贴吧主页URL，下一页，找到不同页的URL规律
2. 获取1页中所有帖子URL地址: [帖子链接1,帖子链接2,...]
3. 对每个帖子链接发请求，获取图片URL
4. 向图片的URL发请求，以wb方式写入本地文件

贴吧URL规律：http://tieba.baidu.com/f?kw=??&pn=50

xpath表达式

1、帖子链接xpath，这里为什么属性选择class，因为相同的元素他们要相同的样式

//div[@class="t_con cleafix"]/div/div/div/a/@href

2、图片链接xpath

//div[@class="d_post_content j_d_post_content  clearfix"]/img[@class="BDE_Image"]/@src

3、视频链接xpath

//div[@class="video_src_wrapper"]/embed/@data-video

# 注意: 此处视频链接前端对响应内容做了处理，需要查看网页源代码来查看，复制HTML代码在线格式化

百度贴吧视频抓取反爬机制（对响应内容做处理）

网页源代码是：

   

F12调试定位到的代码是：

<div class="video_src_wrapper">
    <div class="video_src_wrap_main">
        <video src="http://tb-video.bdstatic.com/tie-cae/f2358e8_0_cae.mp4" ">video>
    div>
div>

如果通过F12定位的位置，写xpath，会爬取不到，因为我们requsets爬取的是网页代码，最后还是要以网页源代码为主。

import requests
from lxml import etree
import random
import time
from urllib import parse


class BaiduImageSpider(object):
    def __init__(self):
        self.url = 'http://tieba.baidu.com/f?kw={}&pn={}'
        self.ua_list = [
            'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/14.0.835.163 Safari/535.1',
            'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0) Gecko/20100101 Firefox/6.0',
            'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; WOW64; Trident/4.0; SLCC2; .NET CLR 2.0.50727; .NET \
            CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; .NET4.0C; InfoPath.3)', ]

    # 获取html网页代码
    def get_html(self, url):
        headers = {'User-Agent': random.choice(self.ua_list)}
        html = requests.get(url=url, headers=headers).content.decode('utf-8', 'ignore')
        return html

    # 解析html
    def xpath_func(self, html, xpath_bds):
        parse_html = etree.HTML(html)
        r_list = parse_html.xpath(xpath_bds)
        return r_list

    # 图片抓取
    def parse_html(self, one_url):
        html = self.get_html(one_url)
        xpath_bds = '//div[@class="t_con cleafix"]/div/div/div/a/@href'
        r_list = self.xpath_func(html, xpath_bds)  # 提取帖子链接:xpath_list ['/p/32323','','']
        for r in r_list:
            t_url = 'http://tieba.baidu.com' + r  # 拼接帖子的URL地址
            self.get_image(t_url)  # 把帖子中所有图片保存到本地
            time.sleep(random.uniform(0, 2))  # 爬完1个帖子中所有图片,休眠0-2秒钟

    # 给定1个帖子URL,把帖子中所有图片保存到本地
    def get_image(self, t_url):
        html = self.get_html(t_url)

        # 使用xpath表达式的或| : 图片链接 + 视频链接
        xpath_bds = '//div[@class="d_post_content j_d_post_content  clearfix"]/img[@class="BDE_Image"]/@src | //div[@class="video_src_wrapper"]/embed/@data-video'
        img_list = self.xpath_func(html, xpath_bds)  # ['http://xxx.jpg','']
        print(img_list)
        for img in img_list:
            html_bytes = requests.get(url=img, headers={'User-Agent': random.choice(self.ua_list)}).content
            self.save_img(html_bytes, img)

    # 保存图片函数
    def save_img(self, html_bytes, img):
        filename = img[-10:]
        with open(filename, 'wb') as f:
            f.write(html_bytes)
            print('%s下载成功' % filename)

    # 主函数
    def main(self):
        name = input('请输入贴吧名:')
        begin = int(input('请输入起始页:'))
        end = int(input('请输入终止页:'))
        # 对贴吧名进行编码
        kw = parse.quote(name)
        for page in range(begin, end + 1):
            pn = (page - 1) * 50
            url = self.url.format(kw, pn)
            # 调用主线函数
            self.parse_html(url)


if __name__ == '__main__':
    spider = BaiduImageSpider()
    spider.main()

查询参数-params

res = requests.get(url,params=params,headers=headers)

url为基准的url地址，不包含查询参数，该方法会自动对params字典编码，然后和url拼接

参数类型：字典,字典中键值对作为查询参数

import requests

baseurl = 'http://tieba.baidu.com/f?'
params = {
    'kw': '赵丽颖吧',
    'pn': '50'}
headers = {'User-Agent': 'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; WOW64; Trident/4.0; SLCC2\
                                                ; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR \
                                                3.0.30729; Media Center PC 6.0; .NET4.0C; InfoPath.3)'}
# 自动对params进行编码,然后自动和url进行拼接,去发请求
res = requests.get(baseurl, headers=headers, params=params)
res.encoding = 'utf-8'
print(res.text)

Web客户端验证参数-auth

res = requests.get(url, headers=headers, auth=('username','password'))

针对于需要web客户端用户名密码认证的网站，auth = ('username','password')

达内课程笔记

import requests
from lxml import etree
import random
import os


class CodeSpider(object):
    def __init__(self):
        self.url = 'http://code.tarena.com.cn/AIDCode/aid1904/14-redis/'
        self.auth = ('tarenacode', 'code_2013')
        self.ua_list = [
            'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/14.0.835.163 Safari/535.1',
            'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0) Gecko/20100101 Firefox/6.0',
            'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; WOW64; Trident/4.0; SLCC2; .NET CLR 2.0.50727; .\
            NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; .NET4.0C; InfoPath.3)', ]

    def parse_html(self):
        # 获取响应内容
        html = requests.get(url=self.url, headers={'User-Agent': random.choice(self.ua_list)}, auth=self.auth)
        html = html.content.decode('utf-8', 'ignore')

        parse_html = etree.HTML(html)   # 解析
        r_list = parse_html.xpath('//a/@href')      # # r_list : ['../','day01','day02','redis_day01.zip']
        for r in r_list:
            if r.endswith('.zip') or r.endswith('.rar'):
                self.save_files(r)

    def save_files(self, r):
        directory = '/home/tarena/AID/redis/'
        if not os.path.exists(directory):
            os.makedirs(directory)

        # 拼接地址,把zip文件保存到指定目录
        url = self.url + r
        # filename: /home/tarena/AID/redis/xxx.zip
        filename = directory + r
        html = requests.get(url=url, headers={'User-Agent': random.choice(self.ua_list)}, auth=self.auth).content

        with open(filename, 'wb') as f:
            f.write(html)
            print('%s下载成功' % r)


if __name__ == '__main__':
    spider = CodeSpider()
    spider.parse_html()

SSL证书认证参数-verify

response = requests.get(url=url,params=params,headers=headers,verify=False)

　　SSL证书认证参适用于没有经过 证书认证机构认证的https类型网站，一般这种网站会抛出 SSLError 异常则考虑使用此参数

verify：True(默认)检查证书认证；False（常用）忽略证书认证

代理参数-proxies

定义：代替你原来的IP地址去对接网络的IP地址。隐藏自身真实IP，避免被封。

普通代理

获取代理IP网站：西刺代理、快代理、全网代理、代理精灵、... ...

语法结构

proxies = {'协议':'协议://IP:端口号'}

# http和https是相同的
proxies = {
  'http':'http://59.172.27.6:38380',
  'https':'https://59.172.27.6:38380'
}

使用免费普通代理IP访问测试网站: http://httpbin.org/get

import requests

url = 'http://httpbin.org/get'
headers = {'User-Agent': 'Mozilla/5.0'}
# 定义代理,在代理IP网站中查找免费代理IP
proxies = {
    'http': 'http://309435365:[email protected]:16818',
    'https': 'https://309435365:[email protected]:16818'}
html = requests.get(url, proxies=proxies, headers=headers, timeout=5).text
print(html)

IP池

从西刺代理上面爬取IP，迭代测试能否使用，建立一个自己的代理IP池，随时更新用来抓取网站数据

import requests
from lxml import etree
import time
import random
from fake_useragent import UserAgent


class GetProxyIP(object):
    def __init__(self):
        self.url = 'https://www.xicidaili.com/nn/'
        self.proxies = {
            'http': 'http://163.204.247.219:9999',
            'https': 'http://163.204.247.219:9999'}

    # 随机生成User-Agent
    def get_random_ua(self):
        ua = UserAgent()        # 创建User-Agent对象
        useragent = ua.random
        return useragent

    # 从西刺代理网站上获取随机的代理IP
    def get_ip_file(self, url):
        headers = {'User-Agent': self.get_random_ua()}
        # 访问西刺代理网站国内高匿代理，找到所有的tr节点对象
        html = requests.get(url=url, proxies=self.proxies, headers=headers, timeout=5).content.decode('utf-8', 'ignore')
        parse_html = etree.HTML(html)
        # 基准xpath，匹配每个代理IP的节点对象列表
        tr_list = parse_html.xpath('//tr')
        for tr in tr_list[1:]:
            ip = tr.xpath('./td[2]/text()')[0]
            port = tr.xpath('./td[3]/text()')[0]
            # 测试ip:port是否可用
            self.test_proxy_ip(ip, port)

    # 测试抓取的代理IP是否可用
    def test_proxy_ip(self, ip, port):
        proxies = {
            'http': 'http://{}:{}'.format(ip, port),
            'https': 'https://{}:{}'.format(ip, port), }
        test_url = 'http://www.baidu.com/'
        try:
            res = requests.get(url=test_url, proxies=proxies, timeout=8)
            if res.status_code == 200:
                print(ip, ":", port, 'Success')
                with open('proxies.txt', 'a') as f:
                    f.write(ip + ':' + port + '\n')
        except Exception as e:
            print(ip, port, 'Failed')

    # 主函数
    def main(self):
        for i in range(1, 1001):
            url = self.url.format(i)
            self.get_ip_file(url)
            time.sleep(random.randint(5, 10))


if __name__ == '__main__':
    spider = GetProxyIP()
    spider.main()

从IP池中取IP

从文件中随机获取代理IP写爬虫

import random
import requests


class BaiduSpider(object):
    def __init__(self):
        self.url = 'http://www.baidu.com/'
        self.headers = {'User-Agent': 'Mozilla/5.0'}
        self.blag = 1

    def get_proxies(self):
        with open('proxies.txt', 'r') as f:
            result = f.readlines()  # 读取所有行并返回列表
        proxy_ip = random.choice(result)[:-1]       # 获取了所有代理IP
        L = proxy_ip.split(':')
        proxy_ip = {
            'http': 'http://{}:{}'.format(L[0], L[1]),
            'https': 'https://{}:{}'.format(L[0], L[1])
        }
        return proxy_ip

    def get_html(self):
        proxies = self.get_proxies()
        if self.blag <= 3:
            try:
                html = requests.get(url=self.url, proxies=proxies, headers=self.headers, timeout=5).text
                print(html)
            except Exception as e:
                print('Retry')
                self.blag += 1
                self.get_html()


if __name__ == '__main__':
    spider = BaiduSpider()
    spider.get_html()

收费代理API

写一个获取收费开放API代理的接口

# 获取开放代理的接口
import requests
from fake_useragent import UserAgent

ua = UserAgent()  # 创建User-Agent对象
useragent = ua.random
headers = {'User-Agent': useragent}


def ip_test(ip):
    url = 'http://www.baidu.com/'
    ip_port = ip.split(':')
    proxies = {
        'http': 'http://{}:{}'.format(ip_port[0], ip_port[1]),
        'https': 'https://{}:{}'.format(ip_port[0], ip_port[1]),
    }
    res = requests.get(url=url, headers=headers, proxies=proxies, timeout=5)
    if res.status_code == 200:
        return True
    else:
        return False


# 提取代理IP
def get_ip_list():
    # 快代理：https://www.kuaidaili.com/doc/product/dps/
    api_url = 'http://dev.kdlapi.com/api/getproxy/?orderid=946562662041898&num=100&protocol=1&method=2&an_an=1&an_ha=1&sep=2'
    html = requests.get(api_url).content.decode('utf-8', 'ignore')
    ip_port_list = html.split('\n')

    for ip in ip_port_list:
        with open('proxy_ip.txt', 'a') as f:
            if ip_test(ip):
                f.write(ip + '\n')


if __name__ == '__main__':
    get_ip_list()

私密代理

1、语法结构

proxies = {
'协议':'协议://用户名:密码@IP:端口号'
}
proxies = {
    'http':'http://用户名:密码@IP:端口号',
    'https':'https://用户名:密码@IP:端口号'
}
proxies = {
    'http': 'http://309435365:[email protected]:16816',
    'https':'https://309435365:[email protected]:16816',
} 

用户名和密码会在给你API_URL的时候给你。不是你的账号和账号密码。

# 获取开放代理的接口
import requests
from fake_useragent import UserAgent

ua = UserAgent()  # 创建User-Agent对象
useragent = ua.random
headers = {'User-Agent': useragent}


def ip_test(ip):
    url = 'https://blog.csdn.net/qq_34218078/article/details/90901602/'
    ip_port = ip.split(':')
    proxies = {
        'http': 'http://1786088386:b95djiha@{}:{}'.format(ip_port[0], ip_port[1]),
        'https': 'http://1786088386:b95djiha@{}:{}'.format(ip_port[0], ip_port[1]),
    }

    res = requests.get(url=url, headers=headers, proxies=proxies, timeout=5)
    if res.status_code == 200:
        print("OK")
        return True
    else:
        print(res.status_code)
        print("错误")
        return False


# 提取代理IP
def get_ip_list():
    # 快代理：https://www.kuaidaili.com/doc/product/dps/
    api_url = 'http://dps.kdlapi.com/api/getdps/?orderid=986603271748760&num=1000&signature=z4a5b2rpt062iejd6h7wvox16si0f7ct&pt=1&sep=2'
    html = requests.get(api_url).content.decode('utf-8', 'ignore')
    ip_port_list = html.split('\n')

    for ip in ip_port_list:
        with open('proxy_ip.txt', 'a') as f:
            if ip_test(ip):
                f.write(ip + '\n')


if __name__ == '__main__':
    get_ip_list()