微博动态图片爬取

微博动态图片爬取

由于微博的网页端有反爬虫，需要登录，所以我们换个思路，曲线救国。

我们找到微博在浏览器上面用于手机端的调试的APL，如何找到呢？

我这边直接附上微博的手机端的地址： https://m.weibo.cn/

步骤1：

模拟搜索用户：

搜索一个用户获取到的api:

https://m.weibo.cn/api/container/getIndex?containerid=100103type=1&q=半半子&page_type=searchall

1.1 对api内参数进行处理
containerid=100103type=1&q=半半子 ——> containerid=100103type%3D1%26q%3D%E5%8D%8A%E5%8D%8A%E5%AD%90_
这个参数需要提前转码，否则无法获取数据
1.2 对用户名进行判断，通过后提取uid

2. 获取more参数
   GET
   api : https://m.weibo.cn/profile/info?uid=2830125342
   2.1 提取并处理more参数

   

3. 循环提取图片id
   GET
   api : https://m.weibo.cn/api/container/getIndex?containerid=2304132830125342_-_WEIBO_SECOND_PROFILE_WEIBO&page_type=03&page=1
   3.1 提取图片id——>pic_id
   3.2 获取发送图片用户
   3.3 根据动态创建时间生成用户唯一识别码

4. 下载图片

   我们从浏览器抓包中就会获取到后台服务器发给浏览器的图片链接

   https://wx2.sinaimg.cn/large/pic_id.jpg 
   

   浏览器打开这个链接就可以直接下载图片

爬取完整代码：

import os
import sys
import time
from urllib.parse import quote
 
import requests
 
headers = {
    'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.69 Safari/537.36'
}
 
 
def time_to_str(c_at):
    ti = time.strptime(c_at, '%a %b %d %H:%M:%S +0800 %Y')
    time_str = time.strftime('%Y%m%d%H%M%S', ti)
    return time_str
 
 
# 1. 搜索用户，获取uid
# 2. 通过uid获取空间动态关键参数
# 3. 获取动态内容
# 4. 提取图片参数
# 5. 下载图片
 
# 1. 搜索用户，获取uid
# ========= 用户名 =========
# 输入不同的用户名可切换下载的用户图片
# 用户名需要完全匹配
name = '半半子_'
# =========================
 
con_id = f'100103type=1&q={name}'
# 这个条件需要转码
con_id = quote(con_id, 'utf-8')
user_url = f'https://m.weibo.cn/api/container/getIndex?containerid={con_id}&page_type=searchall'
user_json = requests.get(url=user_url, headers=headers).json()
user_cards = user_json['data']['cards']
for card_num in range(len(user_cards)):
    if 'mblog' in user_cards[card_num]:
        if user_cards[card_num]['mblog']['user']['screen_name'] == name:
            print(f'正在获取{name}的空间')
            # 2. 通过uid获取空间动态关键参数
            user_id = user_cards[card_num]['mblog']['user']['id']
            info_url = f'https://m.weibo.cn/profile/info?uid={user_id}'
            info_json = requests.get(url=info_url, headers=headers).json()
            more_card = info_json['data']['more'].split("/")[-1]
            break
file_name = 'weibo'
if not os.path.exists(file_name):
    os.mkdir(file_name)
 
if len(more_card) == 0:
    sys.exit()
page_type = '03'
page = 0
while True:
    # 3. 获取动态内容
    page += 1
    url = f'https://m.weibo.cn/api/container/getIndex?containerid={more_card}&page_type={page_type}&page={page}'
    param = requests.get(url=url, headers=headers).json()
    cards = param['data']['cards']
    print(f'第 {page} 页')
    for i in range(len(cards)):
        card = cards[i]
        if card['card_type'] != 9:
            continue
        mb_log = card['mblog']
        # 4. 提取图片参数
        # 获取本人的图片
        pic_ids = mb_log['pic_ids']
        user_name = mb_log['user']['screen_name']
        created_at = mb_log['created_at']
        if len(pic_ids) == 0:
            # 获取转发的图片
            if 'retweeted_status' not in mb_log:
                continue
            if 'pic_ids' not in mb_log['retweeted_status']:
                continue
            pic_ids = mb_log['retweeted_status']['pic_ids']
            user_name = mb_log['retweeted_status']['user']['screen_name']
            created_at = mb_log['retweeted_status']['created_at']
        time_name = time_to_str(created_at)
        pic_num = 1
        print(f'======== {user_name} ========')
        # 5. 下载图片
        for pic_id in pic_ids:
            pic_url = f'https://wx2.sinaimg.cn/large/{pic_id}.jpg'
            pic_data = requests.get(pic_url, headers)
            # 文件名 用户名_日期(年月日时分秒)_编号.jpg
            # 例:半半子__20220212120146_1.jpg
            with open(f'{file_name}/{user_name}_{time_name}_{pic_num}.jpg', mode='wb') as f:
                f.write(pic_data.content)
                print(f'    正在下载：{pic_id}.jpg')
            pic_num += 1
        time.sleep(2)