python爬虫，多线程与生产者消费者模式

in dreaming2023-09-14 19:07

使用队列完成生产者消费者模式
使用类创建多线程提高爬虫速度

python 复制代码

'''
https://sc.chinaz.com/tupian/index.html
https://sc.chinaz.com/tupian/index_2.html
https://sc.chinaz.com/tupian/index_3.html
'''

from threading import Thread
from queue import Queue
import requests
from bs4 import BeautifulSoup
import os

headers = {
    'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/116.0.0.0 Safari/537.36 Edg/116.0.1938.69',
}
class Put_Thread(Thread):
    def __init__(self, url_queue, img_queue):
        super().__init__()
        self.url_queue = url_queue
        self.img_queue = img_queue

    def run(self):
        while not self.url_queue.empty():
            url = self.url_queue.get()
            self.fetch_url(url)

    def fetch_url(self, url):
        response = requests.get(url, headers=headers)
        response.encoding = 'utf-8'
        soup = BeautifulSoup(response.text, 'lxml')
        data_list = soup.find_all('img', class_='lazy')
        for i in data_list:
            title = i.get('alt')
            href = 'https:' + i.get('data-original').replace('_s', '')
            self.img_queue.put((title, href))

class Get_Thread(Thread):
    def __init__(self, img_queue):
        super().__init__()
        self.img_queue = img_queue

    def run(self):
        while True:
            try:
                img_data = self.img_queue.get(timeout=3)
            except:
                break
            else:
                title, href = img_data
                if not os.path.exists('./image'):
                    os.mkdir('./image')
                with open('./image/' + title + '.jpg', 'wb') as f:
                    resp = requests.get(href, headers=headers).content
                    f.write(resp)
                print(title, '保存成功！')

def main():
    '''存放url'''
    url_queue = Queue()
    '''存放图片的地址和名称'''
    img_queue = Queue()

    url_queue.put('https://sc.chinaz.com/tupian/index.html')
    for i in range(1,11):
        url = 'https://sc.chinaz.com/tupian/index_{}.html'.format(i)
        url_queue.put(url)

    for i in range(41):
        t1 = Put_Thread(url_queue, img_queue)
        t1.start()
        t2 = Get_Thread(img_queue)
        t2.start()

if __name__ == '__main__':
    main()
    print('\n************主线程已结束************\n')

通过队列可以让线程之间进行通信
创建继承Thread的类创建线程，run()会在线程start时执行
吃cpu性能