进程和线程的区别, 面相对象补充, 进程, 数据共享, 锁, 进程池, 爬虫模块(requests, bs4(beautifulsoup))
一. 进程和线程的区别?
第一:
进程是cpu资源分配的最小单元。
线程是cpu计算的最小单元。
第二:
一个进程中可以有多个线程。
第三:
对于Python来说他的进程和线程和其他语言有差异,是有GIL锁。
GIL锁保证一个进程中同一时刻只有一个线程被cpu调度。
IO密集型操作可以使用多线程;计算密集型可以使用多进程;
二. 面向对象补充:
class Foo(object): def __init__(self): object.__setattr__(self, 'info', {}) # 在继承的对象中设置值的本质 def __setattr__(self, key, value): # 会拦截所有属性的的赋值语句 self.info[key] = value def __getattr__(self, item): # 拦截点号运算。当对未定义的属性名称和实例进行点号 # 运算时,就会用属性名作为字符串调用这个方法。如果继承树可以找到该属性,则不调用此方法 print(item) # name return self.info[item] obj = Foo() obj.name = 'nacho' print(obj.name) # nacho print(obj.info) # {'name': 'nacho'}
三. 进程
- 进程间数据不共享
data_list = [] def task(arg): data_list.append(arg) print(data_list) def run(): for i in range(10): p = multiprocessing.Process(target=task,args=(i,)) # p = threading.Thread(target=task,args=(i,)) p.start() if __name__ == '__main__': # win10需要用这个, linux不需要 run()
- 常用功能:
- join
- deamon
- name
- multiprocessing.current_process()
- multiprocessing.current_process().ident/pid
- 类继承方式创建进程
class MyProcess(multiprocessing.Process): def run(self): print('当前进程',multiprocessing.current_process()) def run(): p1 = MyProcess() p1.start() p2 = MyProcess() p2.start() if __name__ == '__main__': run()
四. 进程间数据共享
Queue: linux: q = multiprocessing.Queue() def task(arg,q): q.put(arg) def run(): for i in range(10): p = multiprocessing.Process(target=task, args=(i, q,)) p.start() while True: v = q.get() print(v) run() windows: def task(arg,q): q.put(arg) if __name__ == '__main__': q = multiprocessing.Queue() for i in range(10): p = multiprocessing.Process(target=task,args=(i,q,)) p.start() while True: v = q.get() print(v) Manager:(*) Linux: m = multiprocessing.Manager() dic = m.dict() def task(arg): dic[arg] = 100 def run(): for i in range(10): p = multiprocessing.Process(target=task, args=(i,)) p.start() input('>>>') print(dic.values()) if __name__ == '__main__': run() windows: def task(arg,dic): time.sleep(2) dic[arg] = 100 if __name__ == '__main__': m = multiprocessing.Manager() dic = m.dict() process_list = [] for i in range(10): p = multiprocessing.Process(target=task, args=(i,dic,)) p.start() process_list.append(p) while True: count = 0 for p in process_list: if not p.is_alive(): count += 1 if count == len(process_list): break print(dic)
五. 进程锁
import time import threading import multiprocessing lock = multiprocessing.RLock() def task(arg): print('鬼子来了') lock.acquire() time.sleep(2) print(arg) lock.release() if __name__ == '__main__': p1 = multiprocessing.Process(target=task,args=(1,)) p1.start() p2 = multiprocessing.Process(target=task, args=(2,)) p2.start()
六. 进程池
import time from concurrent.futures import ThreadPoolExecutor,ProcessPoolExecutor def task(arg): time.sleep(2) print(arg) if __name__ == '__main__': pool = ProcessPoolExecutor(6) # 取决于CPU的核心数 for i in range(10): pool.submit(task,i)
七. 爬虫:
示例:
import requests from bs4 import BeautifulSoup from concurrent.futures import ThreadPoolExecutor,ProcessPoolExecutor # 模拟浏览器发送请求 # 内部创建 sk = socket.socket() # 和抽屉进行socket连接 sk.connect(...) # sk.sendall('...') # sk.recv(...) def task(url): print(url) r1 = requests.get( url=url, headers={ 'User-Agent':'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.92 Safari/537.36' } ) # 查看下载下来的文本信息 soup = BeautifulSoup(r1.text,'html.parser') print(soup.text) # content_list = soup.find('div',attrs={'id':'content-list'}) # for item in content_list.find_all('div',attrs={'class':'item'}): # title = item.find('a').text.strip() # target_url = item.find('a').get('href') # print(title,target_url) def run(): pool = ThreadPoolExecutor(5) for i in range(1,50): pool.submit(task,'https://dig.chouti.com/all/hot/recent/%s' %i) if __name__ == '__main__': run()