微博视频爬取

在前文爬取图片的基础上，修改其中的json筛选条件就可以

前文

https://blog.csdn.net/weixin_43596589/article/details/122215981

还是因为线程下载会个别出错，所以我选择生成的链接使用迅雷

get_pics_url函数如下

    def get_pics_url(self):
        i = 1
        url_list=[]
        while True:
            url = self.start_url + '&page={}'.format(i)
            headers = {'User-Agent': get_ua()}
            r = requests.get(url, headers=headers)
            _json = json.loads(r.text)
            items = _json["data"]["cards"]
            flag = _json['ok']
            if flag == 1:  # 爬取数据标志+一个手动控制标志
                for v in items:
                    picslist = v.get('mblog')
                    if picslist is not None:
                        pageInfoList = picslist.get('page_info')
                        if pageInfoList is not None:
                            urlsList = pageInfoList.get('urls')
                            if urlsList is not None:
                                img_url = urlsList.get('mp4_720p_mp4')
                                if img_url is not None:
                                    url_list.append(img_url)
                                else:
                                    img_url = urlsList.get('mp4_hd_mp4')
                                    if img_url is None:
                                        print(urlsList)
                                    url_list.append(img_url)

            else:
                #1.06页数显示出现问题
                t1.insert(END, f'***在第{i}页终止***\n')
                t1.see(END)
                t1.update()
                if r1_var.get() == 1:
                    big_dir=disk+':/WeiBo_Pics'
                    os.startfile(big_dir)
                break

            i += 1
        print("url共有个数")
        print(len(url_list))
        # return url_list

#网上找的将整个list写入txt
        file = open(user_name_selected+'mp4.txt', 'w')
        for i in range(len(url_list)):
            s = str(url_list[i]).replace('{', '').replace('}', '').replace("'", '').replace(':', ',') + '\n'
            file.write(s)
        file.close()

        return ""