scrapy+Django爬取酷我音乐

Spider/spider/kuwo.py爬虫代码：

# -*- coding: utf-8 -*-
import scrapy
import demjson
import re
import os
from ..items import MusicItem, SingerItem
from bloomfilter import Bloomfilter    #布隆过滤
 
 
class KuwoSpider(scrapy.Spider):
    name = 'kuwo'
    allowed_domains = ['kuwo.cn']
    start_urls = [
        'http://artistlistinfo.kuwo.cn/mb.slist?stype=artistlist&category=0&order=dict&pn=0&rn=100&encoding=utf8&prefix='
    ]
    
    def __init__(self, name=None, **kwargs):
        super(KuwoSpider, self).__init__(name=name, kwargs=kwargs)
        if not os.path.exists("singer.state"):
            self.bloom = Bloomfilter(10000000)
        else:
            # 存储状态文件后缀随便写
            self.bloom = Bloomfilter("singer.state")
 
    def start_requests(self):
        for x in [chr(code) for code in range(97, 123)]:
            url = self.start_urls[0] + x
            yield scrapy.Request(
                url=url,
                callback=self.parse,
                dont_filter=True,
                meta={'prefix': x}
            )
 
    def parse(self, response):
        meta = response.meta
        json_obj = demjson.decode(response.text)
        total = json_obj.get("total", "0")
        total = int(total) if total.isdigit() else 0
 
        rn = json_obj.get("rn", "100")
        rn = int(rn) if rn.isdigit() else 100
        total_page = total//rn if total % rn == 0 else total//rn+1
        # 处理数据并存储
        artistlist = json_obj.get('artistlist', [])
        for artist in artistlist:
            pic = artist.get('pic')
            if not self.bloom.test(pic):
                item = SingerItem()
                item['singer_id'] = artist.get("id")
                url = "http://search.kuwo.cn/r.s?stype=artist2music&artistid={}&pn=0&rn=100&sortby=0&show_copyright_off=1&alflac=1&pcmp4=1&encoding=utf8&vipver=MUSIC_8.7.7.0_PQ&plat=pc&devid=51016591&thost=search.kuwo.cn".format(item['singer_id'])
                yield scrapy.Request(
                    url=url,
                    callback=self.parse_music,
                    dont_filter=True,
                )
                item['singer_name'] = artist.get("name")
                item['singer_music_num'] = artist.get("music_num")
                item['singer_listen'] = artist.get("listen")
                item['singer_like'] = artist.get("like")
                item['singer_pic'] = pic
                # pic_list = pic.split("/")[:-1]
                # pic_path = "../imgs/" + "/".join(pic_list)
                # # if not os.path.exists(pic_path):
                # #     os.makedirs(pic_path)
                # os.makedirs(pic_path, exist_ok=True)
                item['singer_aartist'] = artist.get("AARTIST")
                item['singer_isstar'] = artist.get("isstar")
                item['singer_prefix'] = response.meta.get("prefix")
                yield item
                self.bloom.add(pic)
                # 数据的持久化
                self.bloom.save("singer.state")
            else:
                print("数据已经存在")
   
        pattern = re.compile(r"pn=(\d+)")
        pn = pattern.findall(response.url)
        pn = pn[0] if pn else 0
        pn = int(pn)
        pn += 1
        pattern = re.compile(r"pn=\d+")
        url = pattern.sub("pn={}".format(pn), response.url)
        print("---------------", url)
        if pn < total_page:
            yield scrapy.Request(
                url=url,
                callback=self.parse,
                dont_filter=True,
                meta=meta
            )
 
    def parse_music(self, response):
        json_obj = demjson.decode(response.text)
        for music in json_obj.get('musiclist', []):
            item = MusicItem()
            item['music_musicrid'] = music.get("musicrid")
            item['music_name'] = music.get("name")
            item['music_artist'] = music.get("artist")
            item['music_releasedate'] = music.get("releasedate")
            item['music_artistid'] = music.get("artistid")
            item['music_alnumind'] = music.get("albumid")
            item['music_album'] = music.get("album")
            yield item

items.py 代码

# -*- coding: utf-8 -*-
 
# Define here the models for your scraped items
#
# See documentation in:
# https://doc.scrapy.org/en/latest/topics/items.html
 
import scrapy
from scrapy_djangoitem import DjangoItem
from api.models import Singer, Music
 
 
class SingerItem(DjangoItem):
    django_model = Singer
 
 
class MusicItem(DjangoItem):
    django_model = Music

Spider/spider/pipelines.py代码（保存爬取的数据）

# -*- coding: utf-8 -*-
 
# Define your item pipelines here
#
# Don't forget to add your pipeline to the ITEM_PIPELINES setting
# See: https://doc.scrapy.org/en/latest/topics/item-pipeline.html
 
 
class SpiderPipeline(object):
    def process_item(self, item, spider):
        return item
 
 
from scrapy.pipelines.images import ImagesPipeline, FilesPipeline
from scrapy.http import Request
from urllib.parse import urljoin
from .items import SingerItem, MusicItem
 
 
class MyImagesPipeline(FilesPipeline):
    def get_media_requests(self, item, info):
        if 'singer_pic' in dict(item):
            pic = item['singer_pic']
            pic = urljoin("http://img1.sycdn.kuwo.cn/star/starheads/", pic)
            return [Request(pic, meta={
                'path': item['singer_pic']
            })]
        else:
            music_id = item['music_musicrid']
            music_src ="http://antiserver.kuwo.cn/anti.s?rid=MUSIC_{}&format=aac|mp3&type=convert_url&response=url".format(music_id)
            return [
                Request(music_src, meta={
                    'path': item['music_musicrid'] + '.acc'
                })
            ]
 
    def file_path(self, request, response=None, info=None):
        path = request.meta.get('path')
        if '/' in path:
            return "../imgs/" + path
        else:
            return "../musics/" + path
 
    def item_completed(self, results, item, info):
 
        print('=====', results)
 
        if 'singer_pic' in dict(item):
            pass
        else:
            status, value = results[0] if results else (0, {'path': '没有路径'})
            item['music_src'] = value.get('path').replace("../", "")
            item['music_lrc_src'] = ""
        item.save()
        return item

settings.py代码：

ITEM_PIPELINES = {
   'Spider.pipelines.SpiderPipeline': 300,
   'Spider.pipelines.MyImagesPipeline': 300,
}
 
 
FILES_URLS_FIELD = "singer_pic"
FILES_STORE = "../imgs/"

common/orm2json.py 代码：

对数据进行json序列化

from django.db.models.query import QuerySet
import datetime
 
 
def object_to_json(model, ignore=None):
    if ignore is None:
        ignore = []
    if type(model) in [QuerySet, list]:
        json = []
        for element in model:
            json.append(_django_single_object_to_json(element, ignore))
        return json
    else:
        return _django_single_object_to_json(model, ignore)
 
 
def _django_single_object_to_json(element, ignore=None):
    return dict([(attr, getattr(element, attr)) for attr in [f.name for f in element._meta.fields]])

api/models.py代码：

from django.db import models
 
 
class Singer(models.Model):
    singer_id = models.IntegerField()
    singer_name = models.CharField(max_length=200)
    singer_music_num = models.IntegerField()
    singer_listen = models.IntegerField()
    singer_like = models.IntegerField()
    singer_pic = models.CharField(max_length=200)
    singer_aartist = models.CharField(max_length=200)
    singer_isstar = models.IntegerField()
    singer_prefix = models.CharField(max_length=200, default='')
    singer_ishot = models.BooleanField(default=False)
 
 
class Music(models.Model):
    music_musicrid = models.IntegerField()
    music_name = models.CharField(max_length=200)
    music_artist = models.CharField(max_length=200)
    music_releasedate = models.CharField(max_length=200)
    music_artistid = models.IntegerField()
    music_album = models.CharField(max_length=200)
    music_alnumind = models.IntegerField()
    # 本地地址，不是远程地址
    music_src = models.CharField(max_length=200)
    music_lrc_src = models.CharField(max_length=200)

在KuWO写Django的配置文件settings.py 中需要做一下配置：

DATABASES = {
    'default': {
        'ENGINE': 'django.db.backends.mysql',
        # 数据库名称
        'NAME': 'kuwodb',
        'USER': 'root',
        'PASSWORD': '123456',
        'POST': '12.0.0.1',
        'PORT': 3306
    }
}

# redis数据库配置
CACHES = {
    'default': {
        'BACKEND': 'django_redis.cache.RedisCache',
        'LOCATION': 'redis://127.0.0.1:6379',
        "OPTIONS": {
            "CLIENT_CLASS": "django_redis.client.DefaultClient",
 
        },
    },
}

common/middleware.py 中自定义中间件的配置
MIDDLEWARE = [
    'django.middleware.security.SecurityMiddleware',
    'django.contrib.sessions.middleware.SessionMiddleware',
    'django.middleware.common.CommonMiddleware',
    'django.middleware.csrf.CsrfViewMiddleware',
    # 这是自定义的中间件，至少要放在SessionMiddleware之后
    'common.middleware.MyCustomMiddleware',
    # 用于过滤UserAgent的中间件
    'common.middleware.BadUserAgentMiddleware',
    # 用于过滤哪些ip可以访问系统的中间件
    'common.middleware.GoodIpMiddlleware',
    # # 判定cookie中是否有指定字段
    # 'common.middleware.BadCookieMiddleware',
    # 限定ip访问次数的中间件
    'common.middleware.SlowSpeedMiddleware',
    'django.contrib.auth.middleware.AuthenticationMiddleware',
    'django.contrib.messages.middleware.MessageMiddleware',
    'django.middleware.clickjacking.XFrameOptionsMiddleware',
]

common/middleware.py代码：

设置安全问题

import re
 
from django.conf import settings
from django.http import HttpResponsePermanentRedirect, HttpResponseBadRequest, HttpResponseForbidden, HttpResponseNotFound
from django.utils.deprecation import MiddlewareMixin
import time
import demjson
 
 
class MyCustomMiddleware(MiddlewareMixin):
    def process_request(self, request):
       print('有人来访问后台了')
 
    def process_response(self, request, response):
        return response
 
 
class BadUserAgentMiddleware(MyCustomMiddleware):
    def process_request(self, request):
        user_agent = request.META.get('HTTP_USER_AGENT', '')
        print(user_agent)
        if not user_agent:
            return HttpResponseBadRequest(content='你是一个爬虫吧？')
        if 'python' in user_agent or 'requests' in user_agent or 'scrapy' in user_agent:
            return HttpResponseBadRequest(content='你是一个框架写的爬虫吧')
 
 
class GoodIpMiddlleware(MiddlewareMixin):
    def process_ruquest(self, request):
        ip = request.META.get('REMOVE_ADDR')
        if ip in ['127.0.0.1', 'localhost']:
            return HttpResponseForbidden(content='你的ip禁止访问该系统!!!')
 
 
class BadCookieMiddleware(MiddlewareMixin):
    def process_request(self, request):
        cookies = request.COOKIES
        if 'my_name' not in cookies:
            return HttpResponseBadRequest(content='不是一个好cookie')
 
 
VISIT_TOTAL_TIME = 60
VISIT_PER_SECOND = 10
AllOW = {}
 
 
class SlowSpeedMiddleware(MiddlewareMixin):
 
    ip = '1.1.1.1'
 
    def process_request(self, request):
        ctime = time.time()
        ip = self.ip
        if ip not in AllOW:
            AllOW[ip] = [ctime, ]
        else:
            time_list = AllOW[ip]
            while True:
                last_time = time_list[-1] if time_list else None
                if not last_time:
                    break
                if ctime - VISIT_TOTAL_TIME > last_time:
                    time_list.pop()
                else:
                    break
            if len(AllOW[ip]) > VISIT_PER_SECOND:
                error_msg = {
                    'msg': '访问频率太快啦！限制你的{}!{}秒后再试！！！'.format(ip, self.wait())
                }
                return HttpResponseNotFound(content=demjson.encode(error_msg), content_type='application/json')
            AllOW[ip].insert(0, ctime)
 
    def wait(self):
        ip = self.ip
        ctime = time.time()
        first_in_time = AllOW[ip][-1]
        wt = VISIT_TOTAL_TIME - (ctime - first_in_time)
        return int(wt)

然后在Terminal中运行命令进行数据迁移，建立数据表，在此之前要先把数据库建好

python manage.py makemigrations

python manage.py migrate

STATICFILES_DIRS = [
    os.path.join(BASE_DIR, "static"),
    os.path.join(BASE_DIR, "imgs"),
]
 
CACHES = {
    'default': {
        'BACKEND': 'django_redis.cache.RedisCache',
        'LOCATION': 'redis://127.0.0.1:6379',
        "OPTIONS": {
            "CLIENT_CLASS": "django_redis.client.DefaultClient",
 
        },
    },
}

api/views.py 代码：

从数据库查询数据，若数据库中没有则从网页上请求，然后缓存到redis中（程序运行时要开启redis服务）

from django.shortcuts import render
from django.http import JsonResponse
from common.decorate import api_json
from .models import Singer, Music
from common.orm2json import object_to_json
from django.db import connection
from django.core.cache import cache
PAGE_SIZE = 10
 
 
def index(request):
    dict1 = {
        'msg': 'hello world'
    }
    return JsonResponse(dict1)
 
 
def get_singers(request):
    msg = '查询成功'
    status = 5
    # 网页需要给的参数：分页page 每页大小size 名字首字母cname
    page = request.GET.get('page', '1')
    page = int(page) if page.isdigit() and '-' not in page and page != '0' else 1
    size = request.GET.get('size', str(PAGE_SIZE))
    size = int(size) if size.isdigit() and '-' not in size and size != '0' else PAGE_SIZE
    cname = request.GET.get('cname', 'hot')
    cname = cname.lower()
    if cname != 'hot':
        cname = cname[0] if cname and cname in get_cnames(request) else 'a'
        key = "singer_{}".format(cname)
        if cache.has_key(key):
            print('从缓存中读取', key)
            singers = cache.get(key)
        else:
            singers = Singer.objects.filter(singer_prefix=cname)
            cache.set(key, singers, 60)
    else:
        # 返回hot的数据
        key = "singer_{}".format(cname)
        singers = get_hot(request)
        cache.set(key, singers, 60)
    singer_total = len(singers)
    if size > PAGE_SIZE:
        status = False
        msg = '页码超过指定范围'
        size = PAGE_SIZE
    singers = singers[(page-1)*size: page*size]
    current_page = page
    total_page = singer_total // size if singer_total % size == 0 else singer_total // size + 1
    page_size = size
    # orm -> dict
    pages = []
    if page > total_page:
        msg = '超过总页数'
        status = False
    singers = object_to_json(singers)
 
    return_dict = {
        'msg': msg,
        'status': status,
        'singer_total': singer_total,
        'current_page': current_page,
        'total_page': total_page,
        'page_size': page_size,
        'cname': cname,
        'pages': pages,
        'singers': singers,
    }
    return return_dict
 
    # 后台需要返回的：total current_page total_page size cname singers
 
 
def get_hot(request):
    rows = Singer.objects.order_by('singer_listen').reverse()
    return rows
 
 
def get_cnames(request):
    cursor = connection.cursor()
    cursor.execute("select singer_prefix from api_singer group by singer_prefix")
    rows = cursor.fetchall()
    return rows
 
 
def get_music_by_singer_id(request):
    msg = '查询成功'
    status = 5
    singer_id = request.GET.get('singer_id')
    singer_id = int(singer_id) if singer_id and singer_id.isdigit() and '-' not in singer_id else 0
    page = request.GET.get('page', '1')
    page = int(page) if page.isdigit() and '-' not in page and page != '0' else 1
    size = request.GET.get('size', str(PAGE_SIZE))
    size = int(size) if size.isdigit() and '-' not in size and size != '0' else PAGE_SIZE
    key = "music_{}".format(singer_id)
    if cache.has_key(key):
        print('从缓存中读取', key)
        musics = cache.get(key)
    else:
        musics = Music.objects.filter(music_artistid=singer_id)
        cache.set(key, musics, 60)
    musics = object_to_json(musics)
    music_total = len(musics)
    if size > PAGE_SIZE:
        status = False
        msg = '页码超过指定范围'
        size = PAGE_SIZE
    musics = musics[(page-1)*size: page*size]
    current_page = page
    total_page = music_total // size if music_total % size == 0 else music_total // size + 1
    page_size = size
    # orm -> dict
    pages = []
    if page > total_page:
        msg = '超过总页数'
        status = False
    return_dict = {
        'msg': msg,
        'status': status,
        'music_total': music_total,
        'current_page': current_page,
        'total_page': total_page,
        'page_size': page_size,
        'pages': pages,
        'musics': musics,
    }
    return return_dict
 
 
def get_music_src_by_music_id(request):
    pass
 
 
def get_lrc_src_by_music_id(request):
    pass
 
 
@api_json
def singers(request):
    return get_singers(request)
 
 
@api_json
def musics(request):
    return get_music_by_singer_id(request)

common/decorate.py代码：

from django.http import JsonResponse
from functools import wraps
 
 
def api_json(func):
    @wraps(func)
    def _func(*args, **kwargs):
        json_obj = func(*args, **kwargs)
        return JsonResponse(json_obj)
        # return json_obj
    return _func
 
 
# 测试代码运行，在被导入其他文件中时下面代码不会起作用
if __name__ == '__main__':
    @api_json
    def hello():
        return {'name': 'zhangsan'}
    print(hello())

在路由中配置路径：（KuWo/urls.py）

from django.contrib import admin
from django.urls import path
from api import views as api_views
 
urlpatterns = [
    path('admin/', admin.site.urls),
    path('', api_views.index),
    path('singers/', api_views.singers),
    path('musics/', api_views.musics),
]