网易云音乐热歌榜爬取(用Jsonpath解析Json数据 + 面向对象写法)

要注意的一点是Jsonpath是从0开始数的，Xpath是从1开始数的

一般写法（函数式）

import requests
from requests.exceptions import RequestException
import re
import json
import jsonpath
import csv
import time

headers = {
    "User-Agent": 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.110 Safari/537.36',
}

fp = open('D:/网易云音乐Top200.csv','wt',newline='',encoding='utf8')
writer = csv.writer(fp)
writer.writerow(('歌名','歌手','图片链接','上次排名'))

def get_html(url):
    try:
        content = requests.get(url,headers=headers)
        if content.status_code == requests.codes.OK:
            return content.text
        else:
            print('debug1')
            return None
    except RequestException:
        print('debug2')
        return None

def get_json_data(html):
    json_content = re.findall('<textarea id="song-list-pre-data" style="display:none;">(.*?)</textarea>',html,re.S)
    # print(json_content) 此时的json_content是真正的json格式
    result = json.loads(json_content[0])
    #注意此时的result已经不能拿去www.json.cn解析了，因为已经转化为python对象是字典
    #json只有Array和object组成，如果是Array要记得加一个[x]
    '''测试提取结果
    print(result[0])#result[0]是指JavaScript里的第一个对象
    print(result[0]["name"])#title
    print(result[0]["artists"][0]["name"])#artist
    print(result[0]["album"]["picUrl"])
    print(result[0]["lastRank"])#上一次的排名
    '''#测试成功，对应换成jsonpath
    for section in result:
        title = jsonpath.jsonpath(section,expr='$.name')[0]
        artist = jsonpath.jsonpath(section,expr='$.artists..name')[0]
        picture_link = jsonpath.jsonpath(section,expr='$.album.picUrl')[0]
        lastRank = jsonpath.jsonpath(section,expr='$.lastRank')
        if not lastRank:
            lastRank = '等于当前排名'
        else:
            lastRank = lastRank[0]
        writer.writerow((title,artist,picture_link,lastRank))

if __name__ == '__main__':
    '''值得注意的是这里url要把源网页的
    'https://music.163.com/#/discover/toplist?id=3778678'中的/#删除才可以得到
    因为可以观察到网页的url和解析网页时的url是这个地方有差别，故尝试
    '''
    url = 'https://music.163.com/discover/toplist?id=3778678'
    html = get_html(url)
    get_json_data(html)
    

尝试了一下面向对象写法，封装一只小蜘蛛

import requests
from requests.exceptions import RequestException
import re
import json
import jsonpath
import csv
import time

class CloudMusicSpider:
    def __init__(self):
        self.headers = {
            "User-Agent": 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.110 Safari/537.36',
        }
        self.url = 'https://music.163.com/discover/toplist?id=3778678'

    def parse_url(self,url):
        try:
            response = requests.get(url,headers=self.headers)
            if response.status_code == requests.codes.OK:
                return response.text
            else:
                return None
        except RequestException:
            return None

    def get_json_data(self,html):
        fp = open('D:/网易云音乐Top200.csv', 'wt', newline='', encoding='utf8')
        writer = csv.writer(fp)
        writer.writerow(('歌名', '歌手', '图片链接', '上次排名'))
        json_content = re.findall('<textarea id="song-list-pre-data" style="display:none;">(.*?)</textarea>', html,re.S)
        result = json.loads(json_content[0])
        for section in result:
            title = jsonpath.jsonpath(section, expr='$.name')[0]
            artist = jsonpath.jsonpath(section, expr='$.artists..name')[0]
            picture_link = jsonpath.jsonpath(section, expr='$.album.picUrl')[0]
            lastRank = jsonpath.jsonpath(section, expr='$.lastRank')
            if not lastRank:
                lastRank = '等于当前排名'
            else:
                lastRank = lastRank[0]
            writer.writerow((title, artist, picture_link, lastRank))

    def runspider(self):
        html = self.parse_url(self.url)
        self.get_json_data(html)

if __name__ == '__main__':
    cloud_music_spider = CloudMusicSpider()
    cloud_music_spider.runspider()
    

刚刚爬取好的数据在文件里面会遇到乱码的现象

解决方法是：把这个刚刚爬好的文件以记事本的形式打开，并另存为另外的一个csv文件。


然后再打开就可以看到结果了