引入
回顾requests模块实现数据爬取的流程
1、指定url
2、发起请求
3、获取响应数据
4、持久化存储
其实,在上述流程中还需要较为重要的一步,就是在持久化存储之前需要进行指定数据解析。因为大多数情况下的需求,我们都会指定去使用聚焦爬虫,也就是爬取页面中指定部分的数据值,而不是整个页面的数据。因此,本次课程中会给大家详细介绍讲解三种聚焦爬虫中的数据解析方式。至此,我们的数据爬取的流程可以修改为:
1、指定url
2、发起请求
3、获取响应数据
4、数据解析
5、持久化存储
python如何实现数据解析
1、正则表达式
常用正则表达式回顾
正则练习
import re #提取出python key="javapythonc++php" re.findall('python',key)[0] ##################################################################### #提取出hello world key=" hello world " re.findall(' (.*) ',key)[0] ##################################################################### #提取170 string = '我喜欢身高为170的女孩' re.findall('d+',string) ##################################################################### #提取出http://和https:// key='http://www.baidu.com and https://boob.com' re.findall('https?://',key) ##################################################################### #提取出hello key='lalalahellohahah' #输出hello re.findall('<[Hh][Tt][mM][lL]>(.*)',key) ##################################################################### #提取出hit. key='bobo@hit.edu.com'#想要匹配到hit. re.findall('h.*?.',key) ##################################################################### #匹配sas和saas key='saas and sas and saaas' re.findall('sa{1,2}s',key)
案例:爬取图片
#!/usr/bin/env python # -*- coding:utf-8 -*- import requests import re import os if __name__ == "__main__": url = 'https://www.qiushibaike.com/pic/%s/' headers={ 'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Safari/537.36', } #指定起始也结束页码 page_start = int(input('enter start page:')) page_end = int(input('enter end page:')) #创建文件夹 if not os.path.exists('images'): os.mkdir('images') #循环解析且下载指定页码中的图片数据 for page in range(page_start,page_end+1): print('正在下载第%d页图片'%page) new_url = format(url % page) response = requests.get(url=new_url,headers=headers) #解析response中的图片链接 e = ' .*?.*?' pa = re.compile(e,re.S) image_urls = pa.findall(response.text) #循环下载该页码下所有的图片数据 for image_url in image_urls: image_url = 'https:' + image_url image_name = image_url.split('/')[-1] image_path = 'images/'+image_name image_data = requests.get(url=image_url,headers=headers).content with open(image_path,'wb') as fp: fp.write(image_data)
2、bs4解析
环境安装
- 需要将pip源设置为国内源,阿里源、豆瓣源、网易源等 - windows (1)打开文件资源管理器(文件夹地址栏中) (2)地址栏上面输入 %appdata% (3)在这里面新建一个文件夹 pip (4)在pip文件夹里面新建一个文件叫做 pip.ini ,内容写如下即可 [global] timeout = 6000 index-url = https://mirrors.aliyun.com/pypi/simple/ trusted-host = mirrors.aliyun.com - linux (1)cd ~ (2)mkdir ~/.pip (3)vi ~/.pip/pip.conf (4)编辑内容,和windows一模一样 - 需要安装:pip install bs4 bs4在使用时候需要一个第三方库,把这个库也安装一下 pip install lxml
基础使用
使用流程: - 导包:from bs4 import BeautifulSoup - 使用方式:可以将一个html文档,转化为BeautifulSoup对象,然后通过对象的方法或者属性去查找指定的节点内容 (1)转化本地文件: - soup = BeautifulSoup(open('本地文件'), 'lxml') (2)转化网络文件: - soup = BeautifulSoup('字符串类型或者字节类型', 'lxml') (3)打印soup对象显示内容为html文件中的内容 基础巩固: (1)根据标签名查找 - soup.a 只能找到第一个符合要求的标签 (2)获取属性 - soup.a.attrs 获取a所有的属性和属性值,返回一个字典 - soup.a.attrs['href'] 获取href属性 - soup.a['href'] 也可简写为这种形式 (3)获取内容 - soup.a.string - soup.a.text - soup.a.get_text() 【注意】如果标签还有标签,那么string获取到的结果为None,而其它两个,可以获取文本内容 (4)find:找到第一个符合要求的标签 - soup.find('a') 找到第一个符合要求的 - soup.find('a', title="xxx") - soup.find('a', alt="xxx") - soup.find('a', class_="xxx") - soup.find('a', id="xxx") (5)find_all:找到所有符合要求的标签 - soup.find_all('a') - soup.find_all(['a','b']) 找到所有的a和b标签 - soup.find_all('a', limit=2) 限制前两个 (6)根据选择器选择指定的内容 select:soup.select('#feng') - 常见的选择器:标签选择器(a)、类选择器(.)、id选择器(#)、层级选择器 - 层级选择器: div .dudu #lala .meme .xixi 下面好多级 div > p > a > .lala 只能是下面一级 【注意】select选择器返回永远是列表,需要通过下标提取指定的对象
案例:爬取三国演义
import requests from bs4 import BeautifulSoup url = 'http://www.shicimingju.com/book/sanguoyanyi.html' header = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.149 Safari/537.36 OPR/67.0.3575.115 (Edition B2)' } req_txt = requests.get(url=url,headers=header).text soup = BeautifulSoup(req_txt,'lxml') title_list = soup.select('.book-mulu > ul > li > a') fp = open(r'C:UsersAdministratorDesktopsanguo.txt', 'w', encoding='utf-8') for i in title_list: title = i.string data_url = 'http://www.shicimingju.com' + i['href'] data_txt = requests.get(url = data_url,headers=header).text data_suop = BeautifulSoup(data_txt,'lxml') data_tag =data_suop.find('div',class_='chapter_content') data_info = data_tag.text fp.write(title + data_info + ' ') print(title + '爬取成功!!!')
3、xpath解析
引入
xpath解析是我们在爬虫中最常用也是最通用的一种数据解析方式,由于其高效且简介的解析方式受到了广大程序员的喜爱。在后期学习scrapy框架期间,也会再次使用到xpath解析。
环境安装
pip install lxml
解析原理
1、使用通用爬虫爬取网页数据
2、实例化etree对象,且将页面数据加载到该对象中
3、使用xpath函数结合xpath表达式进行标签定位和指定数据提取
常用xpath表达式
层级定位: from lxml import etree tree = etree.HTML(req_txt) tree_list = tree.xpath('/html/body/div') 属性定位: #找到class属性值为song的div标签 tree.xpath('//div[@class="song"] ') //div[@class="song"] 层级&索引定位: #找到class属性值为tang的div的直系子标签ul下的第二个子标签li下的直系子标签a //div[@class="tang"]/ul/li[2]/a 取文本: # /表示获取某个标签下的文本内容 # //表示获取某个标签下的文本内容和所有子标签下的文本内容 //div[@class="song"]/p[1]/text() //div[@class="tang"]//text() 取属性: //div[@class="tang"]//li[2]/a/@href 逻辑运算: #找到href属性值为空且class属性值为du的a标签 //a[@href="" and @class="du"] 模糊匹配: //div[contains(@class, "ng")] //div[starts-with(@class, "ta")]
etree对象实例化
- 本地文件:tree = etree.parse(文件名) tree.xpath("xpath表达式") - 网络数据:tree = etree.HTML(网页内容字符串) tree.xpath("xpath表达式")
案例一:爬取58同城二手房标题信息
import requests from lxml import etree url = 'https://hz.58.com/ershoufang/' header = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.149 Safari/537.36 OPR/67.0.3575.115 (Edition B2)' } req_txt = requests.get(url=url,headers=header).text tree = etree.HTML(req_txt) # li_list = tree.xpath('//ul[@class="house-list-wrap"]/li/div[2]/h2[@class="title"]/a/text()') li_list = tree.xpath('//ul[@class="house-list-wrap"]/li') fp = open(r'C:UsersAdministratorDesktop58title.txt', 'w', encoding='utf-8') for i in li_list: title = i.xpath('./div[2]/h2/a/text()')[0] fp.write(title + ' ') print(title)
案例二:爬取4k图片
import requests import os from lxml import etree url = 'http://pic.netbian.com/4kmeinv/' header = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.149 Safari/537.36 OPR/67.0.3575.115 (Edition B2)' } # 手动设置相应编码 # req_data = requests.get(url=url, headers=header) # req_data.encoding = 'utf-8' req_txt = requests.get(url=url, headers=header).text tree = etree.HTML(req_txt) li_list = tree.xpath('//ul[@class="clearfix"]/li') if not os.path.exists(r'C:UsersAdministratorDesktoppic'): os.makedirs(r'C:UsersAdministratorDesktoppic') for i in li_list: pic_url = 'http://pic.netbian.com' + i.xpath('./a/img/@src')[0] print(pic_url) pic_alt = i.xpath('./a/img/@alt')[0] + '.jpg' # 通用处理中文乱码问题的解决方案 pic_alt = pic_alt.encode('iso-8859-1').decode('gbk') pic_info = requests.get(url=pic_url, headers=header).content pic_name = pic_alt + '.jpg' pic = r'C:UsersAdministratorDesktoppic/' + pic_name with open(pic,'wb') as f: f.write(pic_info) print(pic_name + '下载完成!!!') print('over')
案例三:爬取全国城市信息
import requests from lxml import etree url = 'https://www.aqistudy.cn/historydata' header = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.149 Safari/537.36 OPR/67.0.3575.115 (Edition B2)' } req_txt = requests.get(url=url, headers=header).text tree = etree.HTML(req_txt) # hot_li_list = tree.xpath('//div[@class="hot"]/div[@class="bottom"]/ul/li') # all_li_list = tree.xpath('//div[@class="all"]/div[@class="bottom"]/ul/div[2]/li') # hot_city_list = [] # all_city_list = [] # for i in hot_li_list: # hot_city = i.xpath('./a/text()')[0] # hot_city_list.append(hot_city) # # for i in all_li_list: # all_city = i.xpath('./a/text()')[0] # all_city_list.append(all_city) # # print(hot_city_list) # print(all_city_list) # print(len(all_city_list)) # 使用一个 xpth 获取热门城市和全部城市 all_list = [] all = tree.xpath('//div[@class="hot"]/div[@class="bottom"]/ul/li/a | //div[@class="all"]/div[@class="bottom"]/ul/div[2]/li/a') for i in all: city = i.xpath('./text()')[0] all_list.append(city) print(all_list) print(len(all_list))
案例四:爬取站长素材中的免费模板
import requests import os from lxml import etree header = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.149 Safari/537.36 OPR/67.0.3575.115 (Edition B2)' } url = 'http://sc.chinaz.com/jianli/free_%d.html' for u in range(1, 3): if u == 1: new_url = 'http://sc.chinaz.com/jianli/free.html' else: new_url = format(url % u) req_txt = requests.get(url=new_url, headers=header).text tree = etree.HTML(req_txt) tree_list = tree.xpath('//div[@id = "container"]') if not os.path.exists(r'C:UsersAdministratorDesktopjianli'): os.makedirs(r'C:UsersAdministratorDesktopjianli') for i in tree_list: list_url = i.xpath('.//a[1]/@href') for li in list_url: jianli_txt = requests.get(url=li, headers=header).text jianli_tree = etree.HTML(jianli_txt) jianli_url = jianli_tree.xpath('//ul[@class="clearfix"]/li/a/@href')[1] jianli_txt = requests.get(url=jianli_url, headers=header).content janli_name = jianli_url.split('/')[-1] file_url = r'C:UsersAdministratorDesktopjianli/' + janli_name with open(file_url, 'wb') as f: f.write(jianli_txt) print(janli_name + ' 爬取成功!!!') print('爬取完毕')
验证码识别
what is 验证码?
是一种区分用户是计算机还是人的公共全自动程序。验证码可以防止:恶意破解密码、刷票、论坛灌水,有效防止某个黑客对某一个特定注册用户用特定程序暴力破解方式进行不断的登陆尝试,实际上用验证码是现在很多网站通行的方式,我们利用比较简易的方式实现了这个功能。
验证码和爬虫的爱恨情仇
有些时候,我们想要去爬取一些基于某些用户的用户信息,比如爬取张三新浪微博的博文和好友数量……那么这些时候,我们就需要使用爬虫进行某些平台的模拟登陆,登陆成功后爬取该用户的相关用户信息。模拟登陆的过程中,往往会出现验证码的阻挠,并且我们还不可以不处理验证码,否则是不可能实现模拟登陆的。因此,接下来我们就一起来看一下,如何处理或者识别验证码。
云打码平台
处理验证码,最通用的方式就是使用三方平台帮我们对相关的验证码进行识别和处理。其中云打码平台就是一个非常通用的打码平台。
http://www.yundama.com/demo.html
该平台可以帮助我们识别的验证码有如下类型:
使用流程
云打码平台:
1、在官网中进行注册(普通用户和开发者用户)
2、登录开发者用户:
3、实例代码的下载(开发文档-》调用示例及最新的DLL-》PythonHTTP实例下载)
4、创建一个软件:我的软件-》添加新的软件
5、使用示例代码中的源码文件中的代码进行修改,让其识别验证码图片中的数据值
示例代码展示
平台提供的类
注意:使用该平台识别验证码,必须将验证码事先下载到本地,然后再将本地存储的验证码提交给平台的示例程序进行识别操作。
import http.client, mimetypes, urllib, json, time, requests ###################################################################### class YDMHttp: apiurl = 'http://api.yundama.com/api.php' username = '' password = '' appid = '' appkey = '' def __init__(self, username, password, appid, appkey): self.username = username self.password = password self.appid = str(appid) self.appkey = appkey def request(self, fields, files=[]): response = self.post_url(self.apiurl, fields, files) response = json.loads(response) return response def balance(self): data = {'method': 'balance', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey} response = self.request(data) if (response): if (response['ret'] and response['ret'] < 0): return response['ret'] else: return response['balance'] else: return -9001 def login(self): data = {'method': 'login', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey} response = self.request(data) if (response): if (response['ret'] and response['ret'] < 0): return response['ret'] else: return response['uid'] else: return -9001 def upload(self, filename, codetype, timeout): data = {'method': 'upload', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey, 'codetype': str(codetype), 'timeout': str(timeout)} file = {'file': filename} response = self.request(data, file) if (response): if (response['ret'] and response['ret'] < 0): return response['ret'] else: return response['cid'] else: return -9001 def result(self, cid): data = {'method': 'result', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey, 'cid': str(cid)} response = self.request(data) return response and response['text'] or '' def decode(self, filename, codetype, timeout): cid = self.upload(filename, codetype, timeout) if (cid > 0): for i in range(0, timeout): result = self.result(cid) if (result != ''): return cid, result else: time.sleep(1) return -3003, '' else: return cid, '' def report(self, cid): data = {'method': 'report', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey, 'cid': str(cid), 'flag': '0'} response = self.request(data) if (response): return response['ret'] else: return -9001 def post_url(self, url, fields, files=[]): for key in files: files[key] = open(files[key], 'rb'); res = requests.post(url, files=files, data=fields) return res.text
平台提供的调用程序
###################################################################### # 用户名 username = 'username' # 密码 password = 'password' # 软件ID,开发者分成必要参数。登录开发者后台【我的软件】获得! appid = 1 # 软件密钥,开发者分成必要参数。登录开发者后台【我的软件】获得! appkey = '22cc5376925e9387a23cf797cb9ba745' # 图片文件 filename = 'getimage.jpg' # 验证码类型,# 例:1004表示4位字母数字,不同类型收费不同。请准确填写,否则影响识别率。在此查询所有类型 http://www.yundama.com/price.html codetype = 1004 # 超时时间,秒 timeout = 60 # 检查 if (username == 'username'): print('请设置好相关参数再测试') else: # 初始化 yundama = YDMHttp(username, password, appid, appkey) # 登陆云打码 uid = yundama.login(); print('uid: %s' % uid) # 查询余额 balance = yundama.balance(); print('balance: %s' % balance) # 开始识别,图片路径,验证码类型ID,超时时间(秒),识别结果 cid, result = yundama.decode(filename, codetype, timeout); print('cid: %s, result: %s' % (cid, result)) ######################################################################
案例:古诗文网验证码识别
codeText.py
import requests from lxml import etree from YDM import YDMHttp def codeText(pic, codetype): # 用户名 username = 'ysging' # 密码 password = 'abcd_1234' # 软件ID,开发者分成必要参数。登录开发者后台【我的软件】获得! appid = 10502 # 软件密钥,开发者分成必要参数。登录开发者后台【我的软件】获得! appkey = 'bee94fd2f6ac7b0d7fbb56169b137566' # 图片文件 filename = pic # 验证码类型,# 例:1004表示4位字母数字,不同类型收费不同。请准确填写,否则影响识别率。在此查询所有类型 http://www.yundama.com/price.html codetype = codetype # 超时时间,秒 timeout = 20 result = None # 检查 if (username == 'username'): print('请设置好相关参数再测试') else: # 初始化 yundama = YDMHttp(username, password, appid, appkey) # 登陆云打码 uid = yundama.login() print('uid: %s' % uid) # 查询余额 balance = yundama.balance() print('balance: %s' % balance) # 开始识别,图片路径,验证码类型ID,超时时间(秒),识别结果 cid, result = yundama.decode(filename, codetype, timeout) print('cid: %s, result: %s' % (cid, result)) return result url = 'https://so.gushiwen.org/user/login.aspx' header = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.149 Safari/537.36 OPR/67.0.3575.115 (Edition B2)' } req_txt = requests.get(url=url,headers=header).text tree = etree.HTML(req_txt) pic_url = 'https://so.gushiwen.org/' + tree.xpath('//div[@class="mainreg2"]/div[4]/img/@src')[0] print(pic_url) req_con = requests.get(url=pic_url,headers=header).content with open('./pic/code.jpg','wb') as f: f.write(req_con) codeText = codeText('./pic/code.jpg',1004) print('codeText: ',codeText) print('识别完成')
YDM.py
# -*- coding: utf-8 -*- # @File : YDM.py # @Date : 2020-04-12 # @Author : Administrator import http.client, mimetypes, urllib, json, time, requests class YDMHttp: apiurl = 'http://api.yundama.com/api.php' username = '' password = '' appid = '' appkey = '' def __init__(self, username, password, appid, appkey): self.username = username self.password = password self.appid = str(appid) self.appkey = appkey def request(self, fields, files=[]): response = self.post_url(self.apiurl, fields, files) response = json.loads(response) return response def balance(self): data = {'method': 'balance', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey} response = self.request(data) if (response): if (response['ret'] and response['ret'] < 0): return response['ret'] else: return response['balance'] else: return -9001 def login(self): data = {'method': 'login', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey} response = self.request(data) if (response): if (response['ret'] and response['ret'] < 0): return response['ret'] else: return response['uid'] else: return -9001 def upload(self, filename, codetype, timeout): data = {'method': 'upload', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey, 'codetype': str(codetype), 'timeout': str(timeout)} file = {'file': filename} response = self.request(data, file) if (response): if (response['ret'] and response['ret'] < 0): return response['ret'] else: return response['cid'] else: return -9001 def result(self, cid): data = {'method': 'result', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey, 'cid': str(cid)} response = self.request(data) return response and response['text'] or '' def decode(self, filename, codetype, timeout): cid = self.upload(filename, codetype, timeout) if (cid > 0): for i in range(0, timeout): result = self.result(cid) if (result != ''): return cid, result else: time.sleep(1) return -3003, '' else: return cid, '' def report(self, cid): data = {'method': 'report', 'username': self.username, 'password': self.password, 'appid': self.appid, 'appkey': self.appkey, 'cid': str(cid), 'flag': '0'} response = self.request(data) if (response): return response['ret'] else: return -9001 def post_url(self, url, fields, files=[]): for key in files: files[key] = open(files[key], 'rb') res = requests.post(url, files=files, data=fields) return res.text