Python爬虫爬取网易云音乐全部评论

  1. 云栖社区>
  2. 博客>
  3. 正文

Python爬虫爬取网易云音乐全部评论

AwesomeTang 2018-01-27 00:22:00 浏览931
展开阅读全文
img_7711508f8984ed2df5c8b451b2e63a9a.png
beautiful now.png
思路整理
  1. 访问网易云音乐单曲播放界面,我们可以看到当我们翻页的时候网址是没有变化的,这时候我们大致可以确定评论是通过post形式加载的;
    img_be061d6087b08b72c4ae5c2ffe5fa13f.png
    .

    2.接下来就打开控制台找我们要的评论藏在哪里就好了。
    我们在http://music.163.com/weapi/v1/resource/comments/R_SO_4_32019002?csrf_token=发现了我们要的评论,包括热门评论,我们注意看下R_SO_4_后面的数字,其实就是每首歌的id,如果我们想一次性爬取多首歌曲的评论的话,可以通过每次传入歌曲id来实现;
img_b50f9a99554b9cdae1271a8c9abb43c5.png
image.png
  1. 我们接下来看下需要post的数据,有两个值params和encSecKey,本以为就是页码之类的,看到这两个值我其实是懵逼的,很显然是加密过了的,不过我不知道他是怎么加密的,后面在知乎上找到了解决方法,各位可以去知乎看看,我就不赘述了,因为我也没看明白……;
img_9d777047194683c3b8b5ec32f6223fc1.png
image.png

代码部分

加密

前文说了,这部分参考了知乎的一位答主,各位可以去知乎看看,我这边只是稍微改了下就拿来用了,点这里跳转

first_param = '{rid:"", offset:"0", total:"true", limit:"20", csrf_token:""}'
second_param = "010001"
third_param = "00e0b509f6259df8642dbc35662901477df22677ec152b5ff68ace615bb7b725152b3ab17a876aea8a5aa76d2e417629ec4ee341f56135fccf695280104e0312ecbda92557c93870114af6c9d05c4f7f0c3685b7a46bee255932575cce10b424d813cfe4875d3e82047b97ddef52741d546b8e289dc6935b3ece0462db0a22b8e7"
forth_param = "0CoJUm6Qyw8W8jud"

def get_params(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 = forth_param
    second_key = 16 * 'F'
    h_encText = AES_encrypt(first_param, first_key, iv)
    h_encText = AES_encrypt(h_encText, second_key, iv)
    return h_encText


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

def AES_encrypt(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
获取页码以及评论

获取页码数是为了加入循环获取每页的评论,代码如下;

def get_json(url, params, encSecKey):
    data = {
         "params": params,
         "encSecKey": encSecKey
    }
    response = requests.post(url, headers=headers, data=data,proxies = proxies)
    return response.content

def get_page(url):
    params = get_params(0);
    encSecKey = get_encSecKey();
    json_text = get_json(url, params, encSecKey)
    json_dict = json.loads(json_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

最后就是把json数据按照你想要的保存下来就好了,如果只想要热门评论的话,把comments改成hotcomments就好了。
完整代码如下:

#coding = utf-8
from Crypto.Cipher import AES
import base64
import requests
import json
import time
import pandas as pd
import random

headers = {
    'Referer': 'http://music.163.com/song?id=531051217',
    'User-Agent':'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36',
    'Cookie': 'JSESSIONID-WYYY=%5CuiUi%5C%2FYs%2FcJcoQ5xd3cBhaHw0rEfHkss1s%2FCfr92IKyg2hJOrJquv3fiG2%2Fn9GZS%2FuDH8PY81zGquF4GIAVB9eYSdKJM1W6E2i1KFg9%5CuZ4xU6VdPCGwp4KOUZQQiWSlRT%2F1r07OmIBn7yYVYN%2BM2MAalUQnoYcyskaXN%5CPo1AOyVVV%3A1516866368046; _iuqxldmzr_=32; _ntes_nnid=7e2e27f69781e78f2c610fa92434946b,1516864568068; _ntes_nuid=7e2e27f69781e78f2c610fa92434946b; __utma=94650624.470888446.1516864569.1516864569.1516864569.1; __utmc=94650624; __utmz=94650624.1516864569.1.1.utmcsr=baidu|utmccn=(organic)|utmcmd=organic; __utmb=94650624.8.10.1516864569'
}
proxies = {'http':'http://221.200.107.118','https':'http://116.2.25.251'}



first_param = '{rid:"", offset:"0", total:"true", limit:"20", csrf_token:""}'
second_param = "010001"
third_param = "00e0b509f6259df8642dbc35662901477df22677ec152b5ff68ace615bb7b725152b3ab17a876aea8a5aa76d2e417629ec4ee341f56135fccf695280104e0312ecbda92557c93870114af6c9d05c4f7f0c3685b7a46bee255932575cce10b424d813cfe4875d3e82047b97ddef52741d546b8e289dc6935b3ece0462db0a22b8e7"
forth_param = "0CoJUm6Qyw8W8jud"

def get_params(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 = forth_param
    second_key = 16 * 'F'
    h_encText = AES_encrypt(first_param, first_key, iv)
    h_encText = AES_encrypt(h_encText, second_key, iv)
    return h_encText


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

def AES_encrypt(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_json(url, params, encSecKey):
    data = {
         "params": params,
         "encSecKey": encSecKey
    }
    response = requests.post(url, headers=headers, data=data,proxies = proxies)
    return response.content

def get_page(url):
    params = get_params(0);
    encSecKey = get_encSecKey();
    json_text = get_json(url, params, encSecKey)
    json_dict = json.loads(json_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



if __name__ == "__main__":
    start_time = time.time()
    url = "http://music.163.com/weapi/v1/resource/comments/R_SO_4_32019002?csrf_token="
    page = get_page(url)
    for i in range(page):
        params = get_params(i);
        encSecKey = get_encSecKey();
        json_text = get_json(url, params, encSecKey)
        json_dict = json.loads(str(json_text))['comments']
        for t in list(range(len(json_dict))):
            if t == 0:
                rdata=pd.DataFrame(pd.Series(data=json_dict[t])).T
            else:
                rdata=pd.concat([rdata,pd.DataFrame(pd.Series(data=json_dict[t])).T])
        if i == 0:
            commentdata=rdata
        else:
            commentdata=pd.concat([commentdata,rdata])           
        print('***正在保存第%d页***'%(i+1))
        time.sleep(random.uniform(0.2,0.5)) 
    commentdata.to_excel('NetEase_Music_Spider.xls',sheet_name='sheet1')
    end_time = time.time()
    print "程序耗时%f秒." % (end_time - start_time)
    print '***NetEase_Music_Spider@Awesome_Tang***'

本次爬的是最近一直循环的<beautiful now--Zedd/Jon Bellion>,评论共计37429条,1872页,程序耗时1036.046966秒,接近20分钟。


Notes

各位爬的时候一定要使用代理IP,我后面准备爬周董最近的新歌<等你下课>的评论的,爬到5000多页也就是差不多10W条的时候,被封IP了,导致我们整个公司的网络都一段时间内不能访问网易云音乐的评论,包括手机连Wi-Fi...

img_01723493df20ae3327aedad475be7b23.png
image.png


Peace~

网友评论

登录后评论
0/500
评论
AwesomeTang
+ 关注