python爬虫实践——爬取豌豆荚“休闲益智”游戏app

'''
主页:
    图标地址、下载次数、大小、详情页地址

详情页:
    游戏名、好评率、评论数、小编点评、下载地址、简介、网友评论、1-5张截图链接地址、
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=1&ctoken=FRsWKgWBqMBZLdxLaK4iem9B

https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=2&ctoken=FRsWKgWBqMBZLdxLaK4iem9B

https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=3&ctoken=FRsWKgWBqMBZLdxLaK4iem9B

32
'''
import requests
from bs4 import BeautifulSoup
from pymongo import MongoClient
import re

#连接mongoDB数据库
client=MongoClient('localhost',27017)
#主页信息
index_col=client['wandoujia']['index']
#详情页信息
detail_col=client['wandoujia']['detail']

# 1、发送请求
def get_page(url):
    response = requests.get(url)
    return response

# 2、开始解析
# 解析详情页
def parse_detail(text):
    soup = BeautifulSoup(text, 'lxml')
    # print(soup)

    # app名称
    try:
        name = soup.find(name="span", attrs={"class": "title"}).text
    except Exception:
        name=None
    # print(name)

    # 好评率
    try:
        love = soup.find(name='span', attrs={"class": "love"}).text
    except Exception:
        love = None
    # print(love)

    # 评论数
    try:
        commit_num = soup.find(name='a', attrs={"class": "comment-open"}).text
    except Exception:
        commit_num = None
    # print(commit_num)

    # 小编点评
    try:
        commit_content = soup.find(name='div', attrs={"class": "con"}).text
    except Exception:
        commit_content = None
    # print(commit_content)

    # app下载链接
    try:
        download_url = soup.find(name='a', attrs={"class": "normal-dl-btn"}).attrs['href']
    except Exception:
        download_url = None
    # print(download_url)

    print('''
        ============= tank ==============
        app名称:{name}
        好评率: {love}
        评论数: {commit_num}
        小编点评: {commit_content}
        app下载链接: {download_url}
        ============= end ==============
        '''.format(name='name',love='love',commit_num='commit_num',commit_content='commit_content',download_url='download_url')
         )

    #判断所有数据都存在，正常赋值
    if name and love and commit_num and commit_content and download_url:
        detail_data={
            'name':name,
            'love':love,
            'commit_num':commit_num,
            'commit_content':commit_content,
            'download_url':download_url,
        }

    #若love没有值，则设置为  没人点赞，很惨
    if not love:
        detail_data = {
            'name': name,
            'love': "没人点赞，很惨",
            'commit_num':commit_num,
            'commit_content':commit_content,
            'download_url':download_url
        }

    # 若download_url没有值，则设置为  没有安装包
    if not love:
        detail_data = {
            'name':name,
            'love':love,
            'commit_num': commit_num,
            'commit_content': commit_content,
            'download_url': "没有安装包",
        }

    #插入详情页数据
    detail_col.insert(detail_data)
    print('{name}app数据插入成功！')




# 解析主页
def parse_index(data):
    soup = BeautifulSoup(data, 'lxml')

    # 获取所有app的li标签
    app_list = soup.find_all(name='li', attrs={"class": "card"})
    for app in app_list:
        # print(app)
        # print('tank' * 1000)
        # print('tank *' * 1000)
        # print(app)
        # 图标地址
        # 获取第一个img标签中的data-original属性
        img = app.find(name='img').attrs['data-original']
        print(img)

        # 下载次数
        # 获取class为install-count的span标签中的文本
        down_num = app.find(name='span', attrs={"class": "install-count"}).text
        print(down_num)


        # 大小
        # 根据文本正则获取到文本中包含 数字 + MB（d+代表数字）的span标签中的文本
        size = soup.find(name='span', text=re.compile("d+MB")).text
        print(size)

        # 详情页地址
        # 获取class为detail-check-btn的a标签中的href属性
        # detail_url = soup.find(name='a', attrs={"class": "name"}).attrs['href']
        # print(detail_url)

        # 详情页地址
        detail_url = app.find(name='a').attrs['href']
        print(detail_url)

        # 拼接数据
        index_data = {
            'img': img,
            'down_num': down_num,
            'size': size,
            'detail_url': detail_url,
        }

        # 插入数据
        index_col.insert(index_data)
        print('主页数据插入成功！')

        # 3、往app详情页发送请求
        response = get_page(detail_url)

        # 4、解析app详情页
        parse_detail(response.text)


def main():
    for line in range(1, 33):
        url = "https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page={line}&ctoken=FRsWKgWBqMBZLdxLaK4iem9B"

        # 1、往app接口发送请求
        response = get_page(url)
        # print(response.text)
        print('*' * 1000)
        # 反序列化为字典
        data = response.json()

        # 获取接口中app标签数据
        app_li = data['data']['content']
        # print(app_li)
        # 2、解析app标签数据
        parse_index(app_li)

        #执行完所有函数关闭mongoDB客户端
        client.close()

if __name__ == '__main__':
    main()