python爬虫之爬取微博评论（4）

news2025/1/12 16:10:14

一、获取单页评论

随机选取一个微博，例如下面这个

【#出操死亡女生家属... - @冷暖视频的微博 - 微博 (weibo.com)

1、fn+f12，然后点击网络，搜索评论内容，然后预览，就可以查看到网页内容里面还有评论内容

2、编写代码，获取网页信息，url是点击网络，然后点击标头，就会出现一个请求url

# requets是一个爬虫的第三方库，需要单独安装
import requests

# url是一访问网站的地址
url = 'https://weibo.com/ajax/statuses/buildComments?flow=0&is_reload=1&id=5024086457849590&is_show_bulletin=2&is_mix=0&count=10&uid=6532102014&fetch_level=0&locale=zh-CN'
# 1.发送请求
response = requests.get(url=url)
response.encoding = 'gbk'   #防止中文乱码
# 2.打印网页数据
print(response.text)

3、但是我们发现这个并不是像我们想的一样，将网页的文本用文字的方式呈现，那么接下来我们要用到一个方法，让我们能够用文字的方式输出网页数据----->定义请求头。为什么要定义请求头，因为从网站的设置初衷，它是不想服务于我们的爬虫程序的，网站正常只想服务于客户的访问服务，那么我们要将我们的爬虫程序伪装成为正常的服务。通常我们只需要设置 cookie 、referee、user-agent就够了(如果有些特殊的网站可能需要我们有其他的参数)

点击网络，点击标头，标头界面向下划动，找到 cookie 、referer 、user-agent 。

import requests

# 请求头
headers = {
    # 用户身份信息
    'cookie': 'XSRF-TOKEN=ZcbKzr5C4_40k_yYwTHIEj7k; PC_TOKEN=4496aa7595; login_sid_t=e6e7e18ba091dcafc2013982ea8fa895; cross_origin_proto=SSL; WBStorage=267ec170|undefined; _s_tentry=cn.bing.com; UOR=cn.bing.com,weibo.com,cn.bing.com; Apache=1040794478428.4973.1713353134174; SINAGLOBAL=1040794478428.4973.1713353134174; ULV=1713353134177:1:1:1:1040794478428.4973.1713353134174:; wb_view_log=1287*8051.9891666173934937; SUB=_2A25LG8JKDeRhGeFJ71MX9S3Lzj2IHXVoWVuCrDV8PUNbmtANLVnQkW9Nf_9NjBaZTVOL8AH-RMG38C00YruaYRtp; SUBP=0033WrSXqPxfM725Ws9jqgMF55529P9D9W5kReTqg1GgdNOFZ.hiJO0G5NHD95QNS0BpSo-0S0-pWs4DqcjMi--NiK.Xi-2Ri--ciKnRi-zNS0MXeKqfe0MfeBtt; ALF=02_1715945242; WBPSESS=FH255CAr_cfIbZ29-Y520e5NsKBGpFZni0Bmim3vDfjXHIPxgSSGqvAfC_UQmc3W2RLUzLHkkX4YI-_Pn1KHeHJhkeHw5kFxeJYgMYDr9t5bvBCMRkcG_IvV3Y2XiVRlu9ZS91UwfD5AH5MY7jhkfw==',
    # 防盗链
    'referer': 'https://weibo.com/6532102014/Oa6B7wW2i',
    # 浏览器基本信息
    'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36 Edg/123.0.0.0'
}

url = 'https://weibo.com/ajax/statuses/buildComments?flow=0&is_reload=1&id=5024086457849590&is_show_bulletin=2&is_mix=0&count=10&uid=6532102014&fetch_level=0&locale=zh-CN'
# 1.发送请求
response = requests.get(url=url, headers=headers)

# 2.打印网页数据

print(response.json()['data'][0]['text_raw'])

4、获取更多字节内容，然后存储到excel表格中

import requests
import csv

f = open('评论.csv', mode='a', encoding='utf-8-sig', newline='')
csv_write = csv.writer((f))
csv_write.writerow(['id', 'screen_name', 'text_raw', 'like_counts', 'total_number', 'created_at'])

# 请求头
headers = {
    # 用户身份信息
    'cookie': 'XSRF-TOKEN=ZcbKzr5C4_40k_yYwTHIEj7k; PC_TOKEN=4496aa7595; login_sid_t=e6e7e18ba091dcafc2013982ea8fa895; cross_origin_proto=SSL; WBStorage=267ec170|undefined; _s_tentry=cn.bing.com; UOR=cn.bing.com,weibo.com,cn.bing.com; Apache=1040794478428.4973.1713353134174; SINAGLOBAL=1040794478428.4973.1713353134174; ULV=1713353134177:1:1:1:1040794478428.4973.1713353134174:; wb_view_log=1287*8051.9891666173934937; SUB=_2A25LG8JKDeRhGeFJ71MX9S3Lzj2IHXVoWVuCrDV8PUNbmtANLVnQkW9Nf_9NjBaZTVOL8AH-RMG38C00YruaYRtp; SUBP=0033WrSXqPxfM725Ws9jqgMF55529P9D9W5kReTqg1GgdNOFZ.hiJO0G5NHD95QNS0BpSo-0S0-pWs4DqcjMi--NiK.Xi-2Ri--ciKnRi-zNS0MXeKqfe0MfeBtt; ALF=02_1715945242; WBPSESS=FH255CAr_cfIbZ29-Y520e5NsKBGpFZni0Bmim3vDfjXHIPxgSSGqvAfC_UQmc3W2RLUzLHkkX4YI-_Pn1KHeHJhkeHw5kFxeJYgMYDr9t5bvBCMRkcG_IvV3Y2XiVRlu9ZS91UwfD5AH5MY7jhkfw==',
    # 防盗链
    'referer': 'https://weibo.com/6532102014/Oa6B7wW2i',
    # 浏览器基本信息
    'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36 Edg/123.0.0.0'
}

url = 'https://weibo.com/ajax/statuses/buildComments?flow=0&is_reload=1&id=5024086457849590&is_show_bulletin=2&is_mix=0&count=10&uid=6532102014&fetch_level=0&locale=zh-CN'


# 1.发送请求
response = requests.get(url=url, headers=headers)
# 2.获取数据

# 3.提取数据
json_data = response.json()
data_list = json_data['data']

for data in data_list:
    text_raw = data['text_raw']
    id = data['id']
    created_at = data['created_at']
    like_counts = data['like_counts']
    total_number = data['total_number']
    screen_name = data['user']['screen_name']
    print(id, screen_name, text_raw, like_counts, total_number, created_at)
    # 4.保存数据
    csv_write.writerow([id, screen_name, text_raw, like_counts, total_number, created_at])

二、获取多页评论

仔细观察，多条评论应该都会包含buildComments字段

然后搜索buildComments字段，点进去看，会发现每一个路径都对应20条评论，但是我们人为输入每一条路径不显示，所以我们需要观察他们的规律

我们发现除开第一个分页格式不一样，其他分页的url完全一致，但是max_id的值不一样。我们可以多观察几组（例如第二个分页的max_id是第三个分页URL中的max_id）

因此代码如下，

import requests
import csv

f = open('评论.csv', mode='a', encoding='utf-8-sig', newline='')
csv_write = csv.writer((f))
csv_write.writerow(['id', 'screen_name', 'text_raw', 'like_counts', 'total_number', 'created_at'])

# 请求头
headers = {
    # 用户身份信息
    'cookie': 'XSRF-TOKEN=ZcbKzr5C4_40k_yYwTHIEj7k; PC_TOKEN=4496aa7595; login_sid_t=e6e7e18ba091dcafc2013982ea8fa895; cross_origin_proto=SSL; WBStorage=267ec170|undefined; _s_tentry=cn.bing.com; UOR=cn.bing.com,weibo.com,cn.bing.com; Apache=1040794478428.4973.1713353134174; SINAGLOBAL=1040794478428.4973.1713353134174; ULV=1713353134177:1:1:1:1040794478428.4973.1713353134174:; wb_view_log=1287*8051.9891666173934937; SUB=_2A25LG8JKDeRhGeFJ71MX9S3Lzj2IHXVoWVuCrDV8PUNbmtANLVnQkW9Nf_9NjBaZTVOL8AH-RMG38C00YruaYRtp; SUBP=0033WrSXqPxfM725Ws9jqgMF55529P9D9W5kReTqg1GgdNOFZ.hiJO0G5NHD95QNS0BpSo-0S0-pWs4DqcjMi--NiK.Xi-2Ri--ciKnRi-zNS0MXeKqfe0MfeBtt; ALF=02_1715945242; WBPSESS=FH255CAr_cfIbZ29-Y520e5NsKBGpFZni0Bmim3vDfjXHIPxgSSGqvAfC_UQmc3W2RLUzLHkkX4YI-_Pn1KHeHJhkeHw5kFxeJYgMYDr9t5bvBCMRkcG_IvV3Y2XiVRlu9ZS91UwfD5AH5MY7jhkfw==',
    # 防盗链
    'referer': 'https://weibo.com/6532102014/Oa6B7wW2i',
    # 浏览器基本信息
    'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36 Edg/123.0.0.0'
}


def get_next(next='count=10'):
    url = f'https://weibo.com/ajax/statuses/buildComments?flow=0&is_reload=1&id=5024086457849590&is_show_bulletin=2&is_mix=0&{next}&count=20&uid=6532102014&fetch_level=0&locale=zh-CN'
    response = requests.get(url=url, headers=headers)
    json_data = response.json()

    data_list = json_data['data']
    max_id = json_data['max_id']
    for data in data_list:
        text_raw = data['text_raw']
        id = data['id']
        created_at = data['created_at']
        like_counts = data['like_counts']
        total_number = data['total_number']
        screen_name = data['user']['screen_name']
        print(id, screen_name, text_raw, like_counts, total_number, created_at)

        csv_write.writerow([id, screen_name, text_raw, like_counts, total_number, created_at])

    if max_id != 0:
        max_str = 'max_id=' + str(max_id)
        get_next(max_str)
        
get_next()