【爬虫练习之glidedsky】爬虫-基础2

题目
链接

爬虫往往不能在一个页面里面获取全部想要的数据，需要访问大量的网页才能够完成任务。
这里有一个网站，还是求所有数字的和，只是这次分了1000页。

思路

• 找到调用接口

可以看到后面有个参数page来控制页码

代码实现

import requests
import re

url = 'http://www.glidedsky.com/level/web/crawler-basic-2'

headers = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.4389.90 Safari/537.36 Edg/89.0.774.54',
    'Cookie': ''
}

res = 0
for i in range(1, 1001):
    temp_url = url + '?page=' + str(i)
    response = requests.get(temp_url, headers=headers)
    html = response.text
    pattern = re.compile('<div class="col-md-1">.*?(\d+).*?</div>', re.S)
    n_list = re.findall(pattern, html)
    for n in n_list:
        res += int(n)

print(f'Result: {res}')

使用多线程实现更快爬取：

import requests
import re
import threading

url = 'http://www.glidedsky.com/level/web/crawler-basic-2'
total_threads = 10  # 设置线程数量
lock = threading.Lock()  # 创建一个锁，用于线程间的数据同步
res = 0


def worker(thread_id):
    global res
    for i in range(thread_id, 1001, total_threads):
        temp_url = url + '?page=' + str(i)
        response = requests.get(temp_url, headers=headers)
        html = response.text
        pattern = re.compile('<div class="col-md-1">.*?(\d+).*?</div>', re.S)
        n_list = re.findall(pattern, html)
        with lock:
            for n in n_list:
                res += int(n)

threads = []
headers = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.4389.90 Safari/537.36 Edg/89.0.774.54',
    'Cookie': ''
}

# 创建并启动线程
for i in range(total_threads):
    thread = threading.Thread(target=worker, args=(i,))
    thread.start()
    threads.append(thread)

# 等待所有线程执行完成
for thread in threads:
    thread.join()

print(f'Result: {res}')

使用异步函数

import aiohttp
import asyncio
import re

url = 'http://www.glidedsky.com/level/web/crawler-basic-2'
total_requests = 1000  # 总共地请求次数
concurrent_requests = 10  # 同时并发的请求数量
res = 0

headers = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.4389.90 Safari/537.36 Edg/89.0.774.54',
    'Cookie': ''
}

async def fetch_url(session, temp_url):
    async with session.get(temp_url, headers=headers) as response:
        html = await response.text()
        pattern = re.compile('<div class="col-md-1">.*?(\d+).*?</div>', re.S)
        n_list = re.findall(pattern, html)
        return [int(n) for n in n_list]

async def main():
    async with aiohttp.ClientSession() as session:
        tasks = []
        for i in range(1, total_requests + 1):
            temp_url = url + '?page=' + str(i)
            tasks.append(fetch_url(session, temp_url))
            if len(tasks) >= concurrent_requests or i == total_requests:
                results = await asyncio.gather(*tasks)
                for n_list in results:
                    for n in n_list:
                        global res
                        res += n
                tasks = []


loop = asyncio.get_event_loop()
loop.run_until_complete(main())

print(f'Result: {res}')

时间统计：同步的方式大概80s，多线程和异步时间差不多都是20s左右