requests+bs4爬取豌豆荚排行榜及下载排行榜app
爬取排行榜应用信息
爬取豌豆荚排行榜app信息
- app_detail_url - 应用详情页url
- app_image_url - 应用图片url
- app_name - 应用名称
- app_install_count - 下载量
- app_size - 应用大小
- app_info - 应用简介
1.分析:
- 目标url: https://www.wandoujia.com/top/app
- 在网页中,发现了加载更多按钮,点击后,为异步请求,请求url为:
https://www.wandoujia.com/wdjweb/api/top/more?resourceType=0&page=2&ctoken=kuoxO3QZz7JKIJtuA6RXibwL
- 修改page的值,可以得到响应数据,则可以直接爬接口数据,page范围为1~41
2. 爬取数据过程
- 发送请求
- 解析数据
- 保存数据到MySQL数据库
代码
# top_app.py
import requests
from bs4 import BeautifulSoup
from wandoujia.mysql_control import MySQL
# 请求函数
def get_data(url):
response = requests.get(url)
return response.json().get('data')
# 解析数据
def parse_data(json_data):
data = json_data.get('content')
# print(data)
soup = BeautifulSoup(data, 'lxml')
# 找出所以的li标签(每个app都在一个li标签里)
li_list = soup.find_all(name='li', attrs={'class': 'card'})
# print(li_list)
for li in li_list:
# 获取app详情页url
app_detail_url = li.find(name='a').get('href')
# print('应用详情页:', app_detail_url)
# 获取app图片url
img_tag = li.find(name='img')
# print(img_tag)
# app图片url
app_image_url = img_tag.get('data-original')
# print('应用图片:', app_image_url)
# 应用名称
app_name = img_tag.get('alt')
# print('应用名称:', app_name)
# 获取应用下载量
app_install_count = li.find(name='span', attrs={'class': 'install-count'}).text
# print('应用下载量:', app_install_count)
# 获取应用大小
try:
app_size = li.find(name='span', attrs={'title': re.compile('MB')}).text
except:
app_size = ''
# print('应用大小:', app_size)
# 获取应用简介
app_info = li.find(name='div', attrs={'class': 'comment'}).text.strip()
# print('应用简介:', app_info)
yield app_detail_url, app_image_url, app_name, app_install_count, app_size, app_info
# 保存数据
def save_data(generator_data, mysql_obj):
for data in generator_data:
# print(data)
sql = 'insert into top_app(app_detail_url, app_image_url, app_name, app_install_count, app_size, app_info) ' \
'values(%s, %s, %s, %s, %s, %s)'
mysql_obj.execute(sql, data)
print(f'{data[2]} 数据已爬取成功')
print('*' * 100)
if __name__ == '__main__':
# 实例化数据库对象
mysql_obj = MySQL()
# 拼接url
for page in range(1, 42):
url = f'https://www.wandoujia.com/wdjweb/api/top/more?resourceType=0&page={page}&ctoken=kuoxO3QZz7JKIJtuA6RXibwL'
# 发送请求
json_data = get_data(url)
# 解析数据
generator_data = parse_data(json_data)
# 保存数据
save_data(generator_data, mysql_obj)
MySQL数据库
# mysql.py
import pymysql
class MySQL:
def __init__(self):
self.client = pymysql.connect(
host='127.0.0.1',
port=3306,
database='wandoujia',
user='root',
password='admin',
charset='utf8',
autocommit=True
)
self.cursor = self.client.cursor(pymysql.cursors.DictCursor)
def execute(self, sql, args):
try:
self.cursor.execute(sql, args)
except Exception as e:
print(e)
def close(self):
self.cursor.close()
self.client.close()
爬取详情页下载链接并下载
爬取豌豆荚排行榜app详情页
- 分析:
- 目标url:在top_app文件中已有爬取的函数,可直接使用得到app_detail_url
- 详情页面分析:
- <div class="download-wp">下存在a标签<a class="normal-dl-btn">,href属性为下载链接
- 爬取数据过程:
- 使用上面top_app.py中的get_data(),和parse_data()可得到每个app的详情页
- 发送请求
- 解析数据
- 多线程下载app
代码
# top_app_detail.py
import os
from concurrent.futures import ThreadPoolExecutor
import requests
from bs4 import BeautifulSoup
from wandoujia import top_app
# 获取详情页数据
def get_detail_data(data):
response = requests.get(data[0])
# print(response.text)
return response.text
# 解析数据
def parse_detail_data(response):
soup = BeautifulSoup(response, 'lxml')
app_download_url = soup.find(name='a', attrs={'class': 'normal-dl-btn'}).get('href')
# print(f'应用名称: {data[2]}, 下载链接: {app_download_url}')
return data[2], app_download_url
# 保存数据
def download_app(app_name_download_url_tuple):
# 下载app
app_name = os.path.join(save_dir, app_name_download_url_tuple[0])
print(app_name_download_url_tuple[0], '开始下载')
app_file = requests.get(app_name_download_url_tuple[1])
# print(app_name)
with open(app_name, 'wb') as f:
for line in app_file.iter_lines():
f.write(line)
print(app_name_download_url_tuple[0], '下载完成')
if __name__ == '__main__':
# app保存文件夹
save_dir = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'top_app')
# 多线程下载,下载线程数为3
pool = ThreadPoolExecutor(3)
# 拼接url
for page in range(1, 42):
url = f'https://www.wandoujia.com/wdjweb/api/top/more?resourceType=0&page={page}&ctoken=kuoxO3QZz7JKIJtuA6RXibwL'
# 获取详情页url
# 发送请求
json_data = top_app.get_data(url)
# 解析数据
generator_data = top_app.parse_data(json_data)
# 爬取详情页
for data in generator_data:
# print(data)
# 获取详情页数据
detail_response = get_detail_data(data)
# 解析详情页数据
app_name_download_url_tuple = parse_detail_data(detail_response)
# 单线程请求并下载app
# download_app(app_name_download_url_tuple)
# 使用多线程请求并下载app,下载线程数不宜过多,否则会很慢
pool.submit(download_app, app_name_download_url_tuple)