序

• 作为网易云音乐的忠实用户，当然需要搞搞它～
• 在爬取的过程中是有点难度的，需要花费一些精力的。
• 像是歌曲ID的url拼接、或是关键参数的加密破解获取等等～

正

样式图

• 获取这种歌单的歌曲和歌曲下面的所有评论～

1.下载歌曲

• 下载歌曲唯一需要注意的点就是对歌曲id进行真正的url拼接

  
  
  样式图
  
  

  获取此id就可以进行拼接了

• 原本的歌曲url
  https://music.163.com/#/song?id=1407358755
• 其实真实的url是这个（将id获取下来）
  http://music.163.com/song/media/outer/url?id=1407358755.mp3
  由此我进行拼接

#拼接url
musicUrl = 'http://music.163.com/song/media/outer/url?id=' + u.split('=')[1] + '.mp3'

• 注意️：携带请求头请求
• 完整下载歌曲的代码

# -*- coding: utf-8 -*-
import requests,ssl
from lxml import etree
from copyheaders import headers_raw_to_dict
import urllib.request
ssl._create_default_https_context = ssl._create_unverified_context

headers = '''
authority: music.163.com
method: GET
path: /
scheme: https
accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9
accept-encoding: gzip, deflate, br
accept-language: zh-CN,zh;q=0.9,en;q=0.8,zh-HK;q=0.7,zh-TW;q=0.6
cache-control: max-age=0
cookie: _iuqxldmzr_=32; _ntes_nnid=df5a2087ca2abb70b4bd3d6170b71dd8,1576462996114; _ntes_nuid=df5a2087ca2abb70b4bd3d6170b71dd8; WM_TID=FAq2Izmu0OhBQVVRVEdtrgGglTVpU2XD; WM_NI=Ixn8HGuhi6m6UnVf3aZbSOHiXTaPPkbaWn0cillgVZAecQsakDyUwcw%2B2U1SU2FcAMPShnFqcQcZ247MWWbO96%2B4DOE37wV%2BhJnSdG7mHVrvW4dQZSkLaSdnzu%2Blo8umOWg%3D; WM_NIKE=9ca17ae2e6ffcda170e2e6eed8d85fa59996d8b56682928eb3c85e929b9abbb825b1a9ad83b2538cedbdadf92af0fea7c3b92a8ba8a2d9cb4db7b8a7d7f03e97b29a87b26b9abe9db0cb639c8cafb1d361a2bf998fb369abb4a488d55283bbadb2f26ba99cb7b6f669b5f5feb6b45997ee8a84c447948da88fc57ba18fadbbb140f59399b8e145fbeae5d5c74587edf998d521ac96a2b4e74df5b388b7f17e95ecacb2d959a994afb4e766baabe58dee34b7b381b8e637e2a3; JSESSIONID-WYYY=Dtz2qrqubvoAeyIW1F1NCF7uF2yk3wSP%2B8Qo3jyBoIk6%2BPjP2JYn970Cc2Sf1%2Bp3%5CfFEvaoueYJmHAtnQobcm8k4hWagBqBQ4ThVZQwdennHsWMke%2Fk6vU28EqQk5uvjUvaAPCcg%2Bi7n3N4CI33dtB7cNKJ%2FkbhW9uCneafzS6i7qOfd%3A1579005850713
referer: https://music.163.com/
sec-fetch-mode: navigate
sec-fetch-site: same-origin
sec-fetch-user: ?1
upgrade-insecure-requests: 1
user-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/79.0.3945.117 Safari/537.36'''
h = bytes(headers, encoding="utf-8")
headers = headers_raw_to_dict(h)

musicUrl_list = []
def download_parse():
    url='https://music.163.com/playlist?id=2008272804'
    datas = requests.get(url,headers=headers)
    sku_html = etree.HTML(datas.text)
    # print(datas.text)
    # 歌曲链接
    result_data = sku_html.xpath('//ul[@class="f-hide"]/li/a/@href')
    # 歌曲名称
    result_data_name = sku_html.xpath('//ul[@class="f-hide"]/li/a/text()')
    # print(result_data)
    print(len(result_data))
    for index,u in enumerate(result_data):
        #拼接url
        musicUrl = 'http://music.163.com/song/media/outer/url?id=' + u.split('=')[1] + '.mp3'
        # print(musicUrl)
        lists = [result_data_name[index],musicUrl]
        musicUrl_list.append(lists)
    # 下载列表中的全部歌曲，并以歌曲名命名下载后的文件，文件位置为当前文件夹
    for i in musicUrl_list:
        url = i[1]
        name = i[0]
        try:
            print('正在下载', name)
            urllib.request.urlretrieve(url, './song/%s.mp3' % name)
            print('下载成功')
        except Exception as e:
            print(e)
            print('下载失败')
    return result_data

效果图

2.获取歌曲下的评论

• 难度就在评论的采集获取，在获取时需要携带两个参数（params和encSecKey）才能对获得评论信息。
  参数
  
  
  post请求网页
• 通过携带两个参数对网页进行post请求且返回json字符，再对此进行解码

sku_data = requests.post(url=self.url,data=dataf,headers=self.headers)
json_dict = json.loads(sku_data.text)

• 不过我们暂时只能获得到第一页的热门评论与最新评论，如此我们就需要对这两个参数进行解析与破解，看到这此的js加密我很懵逼
• 由此，我在知乎找到一位大佬做了很详细的分析=》链接

简单来说，就是找到关键js文件（js加密的‘发起者’），在其中寻找参数相关的代码，利用Fiddler进行调试，并且分别对四种参数进行分析比对，两次aes加密，从而得出params的结果，encSecKey是常量，直接使用即可。

js参数位置

• 相关代码

    first_param = '{rid:"", offset:"0", total:"true", limit:"20", csrf_token:""}'
    second_param = "010001"
    third_param = "00e0b509f6259df8642dbc35662901477df22677ec152b5ff68ace615bb7b725152b3ab17a876aea8a5aa76d2e417629ec4ee341f56135fccf695280104e0312ecbda92557c93870114af6c9d05c4f7f0c3685b7a46bee255932575cce10b424d813cfe4875d3e82047b97ddef52741d546b8e289dc6935b3ece0462db0a22b8e7"
    forth_param = "0CoJUm6Qyw8W8jud"
    def get_params(self,i):
        if i == 0:
            first_param = '{rid:"", offset:"0", total:"true", limit:"20", csrf_token:""}'
        else:
            offset = str(i * 20)
            first_param = '{rid:"", offset:"%s", total:"%s", limit:"20", csrf_token:""}' % (offset, 'flase')
        iv = "0102030405060708"
        first_key = self.forth_param
        second_key = 16 * 'F'
        h_encText = self.AES_encrypt(first_param, first_key, iv).decode('utf-8')
        h_encText = self.AES_encrypt(h_encText, second_key, iv)
        return h_encText

    def get_encSecKey(self):
        encSecKey = "257348aecb5e556c066de214e531faadd1c55d814f9be95fd06d6bff9f4c7a41f831f6394d5a3fd2e3881736d94a02ca919d952872e7d0a50ebfa1769a7a62d512f5f1ca21aec60bc3819a9c3ffca5eca9a0dba6d6f7249b06f5965ecfff3695b54e1c28f3f624750ed39e7de08fc8493242e26dbc4484a01c76f739e135637c"
        return encSecKey

    def AES_encrypt(self,text, key, iv):
        pad = 16 - len(text) % 16
        text = text + pad * chr(pad)
        encryptor = AES.new(key, AES.MODE_CBC, iv)
        encrypt_text = encryptor.encrypt(text)
        encrypt_text = base64.b64encode(encrypt_text)
        return encrypt_text

• 获取评论数以及页数

    def get_page(self,params, encSecKey):
        json_text = self.get_json(params, encSecKey)
        json_dict = json.loads(json_text.text)
        total_comment = json_dict['total']
        page = (total_comment / 20) + 1
        print('***查询到评论共计%d条,%d页***' % (total_comment, page))
        return page

• 完善代码，循环获取每页评论
  如果想要热门评论的话，把comments改成hotcomments就好了～

# -*- coding: utf-8 -*-
import requests,json
from copyheaders import headers_raw_to_dict
from Crypto.Cipher import AES
import base64

# 网易云音乐
class Huangye():
    def __init__(self):
        self.url = 'https://music.163.com/weapi/v1/resource/comments/R_SO_4_1399533630?csrf_token='
        self.h = '''
user-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/79.0.3945.117 Safari/537.36
'''
        h = bytes(self.h, encoding="utf-8")
        self.headers = headers_raw_to_dict(h)
        self.first_param = '{rid:"", offset:"0", total:"true", limit:"20", csrf_token:""}'
        self.second_param = "010001"
        self.third_param = "00e0b509f6259df8642dbc35662901477df22677ec152b5ff68ace615bb7b725152b3ab17a876aea8a5aa76d2e417629ec4ee341f56135fccf695280104e0312ecbda92557c93870114af6c9d05c4f7f0c3685b7a46bee255932575cce10b424d813cfe4875d3e82047b97ddef52741d546b8e289dc6935b3ece0462db0a22b8e7"
        self.forth_param = "0CoJUm6Qyw8W8jud"

    def get_params(self,i):
        if i == 0:
            first_param = '{rid:"", offset:"0", total:"true", limit:"20", csrf_token:""}'
        else:
            offset = str(i * 20)
            first_param = '{rid:"", offset:"%s", total:"%s", limit:"20", csrf_token:""}' % (offset, 'flase')
        iv = "0102030405060708"
        first_key = self.forth_param
        second_key = 16 * 'F'
        #两次aes加密
        h_encText = self.AES_encrypt(first_param, first_key, iv).decode('utf-8')
        h_encText = self.AES_encrypt(h_encText, second_key, iv)
        return h_encText

    def get_encSecKey(self):
        encSecKey = "257348aecb5e556c066de214e531faadd1c55d814f9be95fd06d6bff9f4c7a41f831f6394d5a3fd2e3881736d94a02ca919d952872e7d0a50ebfa1769a7a62d512f5f1ca21aec60bc3819a9c3ffca5eca9a0dba6d6f7249b06f5965ecfff3695b54e1c28f3f624750ed39e7de08fc8493242e26dbc4484a01c76f739e135637c"
        return encSecKey

    def AES_encrypt(self,text, key, iv):
        pad = 16 - len(text) % 16
        text = text + pad * chr(pad)
        encryptor = AES.new(key, AES.MODE_CBC, iv)

        encrypt_text = encryptor.encrypt(text)
        encrypt_text = base64.b64encode(encrypt_text)
        return encrypt_text

    def get_page(self,params, encSecKey):
        json_text = self.get_json(params, encSecKey)
        json_dict = json.loads(json_text.text)
        total_comment = json_dict['total']
        page = (total_comment / 20) + 1
        print('***查询到评论共计%d条,%d页***' % (total_comment, page))
        return page
    # 访问
    def get_json(self,params, encSecKey):
        dataf = {
            "params": params,
            "encSecKey": encSecKey
        }
        sku_data = requests.post(url=self.url,data=dataf,headers=self.headers)
        return sku_data
    #获取热门评论
    def hot_parse(self,json_data):
        # 热门评论
        for user in json_data['hotComments']:
            # print(user)
            # 用户名称
            user_name = user['user']['nickname']
            print('用户名称:',user_name)
            # 评论内容
            user_text = user['content'].replace('\r','\\r').replace('\n','\\n')
            print('评论内容:',user_text)
            # 该评论点赞数
            user_num = user['likedCount']
            print('该评论被点赞数:', user_num)
            print('-'*30)
        print('='*100)
    # 获取全部评论
    def parse(self,json_data):

        # 全部评论
        for user in json_data['comments']:
            # print(user)
            # 用户名称
            user_name = user['user']['nickname']
            print('用户名称:',user_name)
            # 评论内容
            user_text = user['content'].replace('\r','\\r').replace('\n','\\n')
            print('评论内容:',user_text)
            # 该评论点赞数
            user_num = user['likedCount']
            print('该评论被点赞数:', user_num)
            print('-'*30)
    # 主函数
    def main(self):
        params = self.get_params(0)
        encSecKey = self.get_encSecKey()
        page = self.get_page(params,encSecKey)
        json_text = self.get_json(params, encSecKey)
        json_dict = json.loads(json_text.text)
        print('热门评论==============================================================')
        self.hot_parse(json_dict)
        print('全部评论==============================================================')
        for i in range(int(page)):
            params = self.get_params(i)
            encSecKey = self.get_encSecKey()
            json_text = self.get_json(params, encSecKey)
            json_dict = json.loads(json_text.text)
            self.parse(json_dict)
            print('下一页。。。。。。。。。')
            print('='*100)

if __name__ == '__main__':
    jian = Huangye()
    jian.main()

效果图

末

• 最后呢，可根据自己想法进行储存评论数据～
• 注意：大量采集需要使用代理ip，要不然会被封哟～
• 所以说，下载歌曲的同时，将歌曲id传过来，也是可以同时抓去评论的
• 说明一点，url的R_SO_4_后面一段数字就是歌曲id

性不性感！