微博图片爬取

准备

从浏览器中获取自己账号的Cookie、User-Agent、uid，在自己B站主页上按F12，点进network。

在自己py文件同文件夹中创建node.json，将{“weibojiedian”:{“3897000000”: “M_HmUe44444”}}，因为代码会打开此文件，没有的话会报错。
此json文件是为了记录此次下载到哪个文章，以防重复下载。
代码将图片链接保存到桌面TXT文件中，可以用下载软件如：IDM等自行下载

本代码有3处需自行输入：

# #python3.7 windows

import requests
from bs4 import BeautifulSoup as bs
import re
import time
from tkinter import *
import winreg,os,time
import json
def get_desktop():
    key = winreg.OpenKey(winreg.HKEY_CURRENT_USER,r'Software\Microsoft\Windows\CurrentVersion\Explorer\Shell Folders',)
    return winreg.QueryValueEx(key, "Desktop")[0]

class Wb():
    def __init__(self,b):
        self.chucu = []
        self.chuc = []
        self.uidlist = []
        self.uidset = set()
        self.path1 = r'node.json'
        self.path2 = b

    def fw(self, url):
        headers = {
            'User-Agent': '需自己输入',
            'Cookie': '需自己输入'}
        try:
            r = requests.get(url, headers=headers, timeout=5)
            time.sleep(2)
            if r.status_code == 200:
                return r.content.decode('utf-8')
            else:
                self.chuc.append(url)
                return ('chuc')
        except:
            self.chucu.append(url)
            return ('chuc')

    def uidget(self):
        i = 1
        while True:
            url = 'https://weibo.cn/需自己输入/follow?page=' + str(i)
            res = self.fw(url)
            i += 1
            if res == 'chuc' or '取消关注' not in res:
                #                print('break')
                break
            else:
                uide = re.findall('remark.uid=\d+', res)
                for line in uide:
                    uid = re.findall('\d\d\d+', line)[0]
                    if uid not in self.uidset:
                        self.uidlist.append(uid)
                        self.uidset.add(uid)

    def chuli_a(self, uid, bi):
        i = 1
        a = 1
        global bj
        bj = bi
        while True:
            page_list = uid + '?page=' + str(i)
            i += 1
            if '1' in self.stop:
                break
            else:
                pass
            r = self.fw(page_list)
            if r == 'chuc':
                print('出错：' + page_list)
            if '转发' not in r:
                break
            else:
                page_div_list = r.split('<div class="c" id="')
                for pic, pic_page in enumerate(page_div_list):
                    find = re.findall('M_.........', pic_page)
                    if pic == 0:
                        pass
                    elif bi in find:
                        self.stop.append('1')
                        break
                    else:
                        if a == 1:
                            bj = find[0]
                        self.chuli_er(pic_page, uid)
                        a += 1

    def chuli_er(self, pic_page, uid):
        if '组图共' in pic_page:
            find = re.findall('<a href="https://weibo.cn/mblog/picAll/\S+.rl=1">组图共\d张</a>', pic_page)
            pic_page_list = re.findall("https://weibo.cn/mblog/picAll/\S+rl=1", find[0])
            req = self.fw(pic_page_list[0])
            soup_a = bs(req, 'html.parser')
            pic_url = soup_a.find_all('img', alt="图片加载中...")
            pic_url_list = [img.attrs['src'].replace('thumb180', 'large') for img in pic_url]
            self.xr(pic_url_list, uid)
        elif '原图' in pic_page:
            find = re.findall(
                '<img src="http://..\d.sinaimg.cn/wap180/\S+(?:.jpg|.gif|.bmp|.png|.jpeg)" alt="图片" class="ib"',
                pic_page)
            find_a = find[0]
            pic_url_list_a = re.findall('http://..\d.sinaimg.cn/wap180/\S+(?:.jpg|.gif|.bmp|.png|.jpeg)', find_a)
            pic_url_list = [pic_url_list_a[0].replace('wap180', 'large')]
            self.xr(pic_url_list, uid)
        elif 'sinaimg.cn' not in pic_page or 'sinaimg.cn/upload' in pic_page:
            pass
        else:
            #            print(pic_page)
            print(pic_page)

    def xr(self, list, uid):
        fi = re.findall('\d+', uid)[0]
        with open(self.path2, 'a') as f:
            for line in list:
                f.write(line + '\n')

    def run(self):
        global bj
        self.uidget()
        with open(self.path1, 'r') as f:
            load_dict = json.load(f)
            dic = load_dict['weibojiedian']
            for uid in self.uidlist:
                print(uid)
                self.stop = []
                bi = dic.get(uid, "NA")
                print(bi)
                uid_li = 'https://weibo.cn/u/' + str(uid)
                self.chuli_a(uid_li, bi)
                dic[uid] = bj

        dic_w = str(dic)
        with open(self.path1, 'w+') as f:
            json.dump(load_dict, f)
            print('page finish')
        print(self.chuc)
        print(self.chucu)

if __name__ == '__main__':
    dspath = get_desktop()
    localtime = time.localtime()
    time_str = time.strftime("%m-%d", localtime)
    name = '微博图片爬取' + time_str + '.txt'
    path = os.path.join(dspath, name)
    Wb(path).run()