1 # -*- coding:utf-8 -*-
  2 import requests
  3 from bs4 import BeautifulSoup
  4 import os
  5 import time
  6 import re
  7 import random
  8 
  9 se = requests.session()
 10 
 11 
 12 class Pixiv():
 13 
 14     def __init__(self):
 15         self.base_url = 'https://accounts.pixiv.net/login?lang=zh&source=pc&view_type=page&ref=wwwtop_accounts_index'
 16         self.login_url = 'https://accounts.pixiv.net/api/login?lang=zh'
 17         self.target_url = 'http://www.pixiv.net/search.php?' \
 18                           'word=%E5%9B%9B%E6%9C%88%E3%81%AF%E5%90%9B%E3%81%AE%E5%98%98&order=date_d&p='
 19         self.main_url = 'http://www.pixiv.net'
 20         # headers只要这两个就可以了,之前加了太多其他的反而爬不上
 21         self.headers = {
 22             'Referer': 'https://accounts.pixiv.net/login?lang=zh&source=pc&view_type=page&ref=wwwtop_accounts_index',
 23             'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) '
 24                           'AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
 25         }
 26         self.pixiv_id = 'userid'
 27         self.password = '*****'
 28         self.post_key = []
 29         self.return_to = 'http://www.pixiv.net/'
 30         self.load_path = 'D:\psdcode\Python\pixiv_pic'
 31         self.ip_list = []
 32 
 33     def login(self):
 34         post_key_html = se.get(self.base_url, headers=self.headers).text
 35         post_key_soup = BeautifulSoup(post_key_html, 'lxml')
 36         self.post_key = post_key_soup.find('input')['value']
 37         # 上面是去捕获postkey
 38         data = {
 39             'pixiv_id': self.pixiv_id,
 40             'password': self.password,
 41             'return_to': self.return_to,
 42             'post_key': self.post_key
 43         }
 44         se.post(self.login_url, data=data, headers=self.headers)
 45 
 46     def get_proxy(self):
 47         html = requests.get('http://haoip.cc/tiqu.htm')
 48         ip_list_temp = re.findall(r'r/>(.*?)<b', html.text, re.S)
 49         for ip in ip_list_temp:
 50             i = re.sub('\n', '', ip)
 51             self.ip_list.append(i.strip())
 52             print(i.strip())
 53 
 54     ''' 会被反爬,改成使用代理
 55         def get_tml(self, url):
 56             response = se.get(url, headers=self.headers)
 57             return response
 58     '''
 59     def get_html(self, url, timeout, proxy=None, num_entries=5):
 60         if proxy is None:
 61             try:
 62                 return se.get(url, headers=self.headers, timeout=timeout)
 63             except:
 64                 if num_entries > 0:
 65                     print('获取网页出错,5秒后将会重新获取倒数第', num_entries, '次')
 66                     time.sleep(5)
 67                     return self.get_html(url, timeout, num_entries = num_entries - 1)
 68                 else:
 69                     print('开始使用代理')
 70                     time.sleep(5)
 71                     ip = ''.join(str(random.choice(self.ip_list))).strip()
 72                     now_proxy = {'http': ip}
 73                     return self.get_html(url, timeout, proxy = now_proxy)
 74         else:
 75             try:
 76                 return se.get(url, headers=self.headers, proxies=proxy, timeout=timeout)
 77             except:
 78                 if num_entries > 0:
 79                     print('正在更换代理,5秒后将会重新获取第', num_entries, '次')
 80                     time.sleep(5)
 81                     ip = ''.join(str(random.choice(self.ip_list))).strip()
 82                     now_proxy = {'http': ip}
 83                     return self.get_html(url, timeout, proxy = now_proxy, num_entries = num_entries - 1)
 84                 else:
 85                     print('使用代理失败,取消使用代理')
 86                     return self.get_html(url, timeout)
 87 
 88     def get_img(self, html, page_num):
 89         li_soup = BeautifulSoup(html, 'lxml')  # 传入第page_num页的html
 90         li_list = li_soup.find_all('li', attrs={'class', 'image-item'})   # 找到li所在位置
 91         # print('get_list succeed')
 92         # print(li_list)
 93         for li in li_list:
 94             href = li.find('a')['href']  # 直接提取第一个href
 95             # print('get_href succeed')
 96             # print(href)
 97             jump_to_url = self.main_url + href  # 跳转到目标的url
 98             # print('get_jump_to_url succeed')
 99             jump_to_html = self.get_html(jump_to_url, 3).text  # 获取图片的html
100             # print('get_jump_to_html succeed')
101 
102             img_soup = BeautifulSoup(jump_to_html, 'lxml')
103             img_info = img_soup.find('div', attrs={'class', 'works_display'})\
104                 .find('div', attrs={'class', '_layout-thumbnail ui-modal-trigger'})
105             # 找到目标位置的信息
106             if img_info is None:  # 有些找不到url,如果不continue会报错
107                 continue
108             self.download_img(img_info, jump_to_url, page_num)  # 去下载这个图片
109 
110     def download_img(self, img_info, href, page_num):
111         title = img_info.find('img')['alt']  # 提取标题
112         src = img_info.find('img')['src']  # 提取图片位置
113         src_headers = self.headers
114         src_headers['Referer'] = href  # 增加一个referer,否则会403,referer就像上面登陆一样找
115         try:
116             html = requests.get(src, headers=src_headers)
117             img = html.content
118         except:  # 有时候会发生错误导致不能获取图片.直接跳过这张图吧
119             print('获取该图片失败')
120             return False
121 
122         title = title.replace('?', '_').replace('/', '_').replace('\\', '_').replace('*', '_').replace('|', '_')\
123             .replace('>', '_').replace('<', '_').replace(':', '_').replace('"', '_').strip()
124         # 去掉那些不能在文件名里面的.记得加上strip()去掉换行
125 
126         if os.path.exists(os.path.join(self.load_path, str(page_num), title + '.jpg')):
127             for i in range(1, 100):
128                 if not os.path.exists(os.path.join(self.load_path, str(page_num), title + str(i) + '.jpg')):
129                     title = title + str(i)
130                     break
131         # 如果重名了,就加上一个数字
132         print('正在保存名字为: ' + title + ' 的图片')
133         with open(title + '.jpg', 'ab') as f:  # 图片要用b
134             f.write(img)
135         print('保存该图片完毕')
136 
137     def mkdir(self, path):
138         path = path.strip()
139         is_exist = os.path.exists(os.path.join(self.load_path, path))
140         if not is_exist:
141             print('创建一个名字为 ' + path + ' 的文件夹')
142             os.makedirs(os.path.join(self.load_path, path))
143             os.chdir(os.path.join(self.load_path, path))
144             return True
145         else:
146             print('名字为 ' + path + ' 的文件夹已经存在')
147             os.chdir(os.path.join(self.load_path, path))
148             return False
149 
150     def work(self):
151         self.login()
152         for page_num in range(1, 51):  # 太多页了,只跑50页
153             path = str(page_num)  # 每一页就开一个文件夹
154             self.mkdir(path)  # 创建文件夹
155             # print(self.target_url + str(page_num))
156             now_html = self.get_html(self.target_url + str(page_num), 3)  # 获取页码
157             self.get_img(now_html.text, page_num)  # 获取图片
158             print('第 {page} 页保存完毕'.format(page=page_num))
159             time.sleep(2)  # 防止太快被反
160 
161 
162 pixiv = Pixiv()
163 pixiv.work()

P站phython爬虫带注释（搬运）

总共已有 0 条评论