Loading

线程的创建

【一】threading模块介绍

  • 多线程创建和多进程创建很像
  • 我的理解是threading模块的作者遵循了鸭子类型
  • 所以和multiprocessing模块的使用方法那么像

【二】开启线程的两种方式

方式一

  • 直接调用 Thread 方法
from threading import Thread
import time


def task(name):
    print(f'{name}任务开始')
    time.sleep(2)
    print(f'{name}任务结束')


if __name__ == '__main__':
    t = Thread(target=task, args=('学习',))
    t.start()
    print('主线程')

方式二

  • 继承Thread类
from threading import Thread
import time


class MyThread(Thread):
    def __init__(self, name):
        super().__init__()
        self.name = name

    def run(self) -> None:
        print(f'{self.name}任务开始')
        time.sleep(2)
        print(f'{self.name}任务结束')


if __name__ == '__main__':
    t = MyThread(name='打篮球')
    t.start()
    print('主线程')
  • 用法基本和multiprocessing模块,也就是创建进程的方式一模一样

【三】查看ID

  • 用threading模块里面的current_thread方法
from threading import Thread,current_thread
import time


def task(name):
    print(f'{name}任务开始')
    print(current_thread().native_id)
    time.sleep(2)
    print(f'{name}任务结束')


if __name__ == '__main__':
    t = Thread(target=task, args=('学习',))
    t.start()
    print('主线程')
  • 用os模块里面的getpid()方法
from threading import Thread
import time
from os import getpid

def task(name):
    print(f'{name}任务开始')
    print(getpid())
    time.sleep(2)
    print(f'{name}任务结束')


if __name__ == '__main__':
    t = Thread(target=task, args=('学习',))
    t.start()
    print('主线程')

【四】多线程并发的socket服务端

  • 服务端
from threading import Thread
import socket
from socket import SOL_SOCKET, SO_REUSEADDR

server = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
server.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1)
server.bind(('127.0.0.1', 8080))
server.listen(5)


def talk(conn):
    while True:
        try:
            msg = conn.recv(1024).decode('utf-8')
            print(f'客户端:>>>>{msg}')

            conn.send(msg.upper().encode('utf-8'))
        except Exception as error:
            print(error)
            break
    conn.close()


def threading_version(conn):
    t = Thread(target=talk, args=(conn,))
    t.start()


def main_threading():
    while True:
        conn, addr = server.accept()
        threading_version(conn=conn)


if __name__ == '__main__':
    main_threading()

  • 客户端(不变)
from socket import *

# 不写参数:默认是TCP协议
# (1)创建客户端对象
client = socket()

# (2)绑定 IP PORT
# (2)建立链接桥梁 --(呼应客户端的 ip 和 port)
IP = '127.0.0.1'
PORT = 8080
client.connect((IP, PORT))

# (4)链接循环
while True:
    # (4.1)向服务端发数据
    msg_to_server = input('请输入消息:>>>>').strip()

    if not msg_to_server:
        continue
    client.send(msg_to_server.encode('utf-8'))

    # 接受服务器返回的数据
    data_from_server = client.recv(1024)
    print(data_from_server.decode('utf-8'))
    if msg_to_server == 'q':
        break
client.close()

【五】单进程,多进程,多线程速度比较

  • 通过一个爬虫案例
import requests
import os
import time
from lxml import etree
from multiprocessing import Process
from threading import Thread


def timer(func):
    def inner(*args, **kwargs):
        start_time = time.time()
        func(*args, **kwargs)
        print(f'总耗时{time.time() - start_time}秒')

    return inner


def create_url_list():
    url_list = []
    for i in range(1, 6):
        if i == 1:
            url = 'https://pic.netbian.com/4kmeinv/'
            url_list.append(url)
        else:
            url = f'https://pic.netbian.com/4kmeinv/index_{i}.html'
            url_list.append(url)
    return url_list


def spider_data(url_list):
    data_dic = []
    for url in url_list:
        response = requests.get(url=url)
        response.encoding = 'gbk'
        response_text = response.text

        html_obj = etree.HTML(response_text)
        li_list = html_obj.xpath('//*[@id="main"]/div[3]/ul/li')

        for li in li_list:
            src = "https://pic.netbian.com/" + li.xpath('./a/img/@src')[0]
            title = li.xpath('./a/b/text()')[0]
            data_dic.append({'src': src, 'title': title})
    return data_dic


def save_data(file_title, file_src):
    file_name = 'image'
    file_path = os.path.join(os.path.dirname(__file__), file_name)
    os.makedirs(file_path, exist_ok=True)
    file_path = os.path.join(file_path, f'{file_title}.png')
    data = requests.get(file_src)
    with open(file_path, 'wb') as fp:
        fp.write(data.content)
    print(f'当前图片{file_title}保存完成')


@timer
def main_normal():
    url_list = create_url_list()
    data_list = spider_data(url_list=url_list)
    for data in data_list:
        image_src = data.get('src')
        image_title = data.get('title')
        save_data(image_title, image_src)


@timer
def main_process():
    url_list = create_url_list()
    data_list = spider_data(url_list=url_list)
    p_list = []
    for data in data_list:
        image_src = data.get('src')
        image_title = data.get('title')
        p = Process(target=save_data, args=(image_title, image_src))
        p.start()
        p_list.append(p)
    for p in p_list:
        p.join()


@timer
def main_thread():
    url_list = create_url_list()
    data_list = spider_data(url_list=url_list)
    p_list = []
    for data in data_list:
        image_src = data.get('src')
        image_title = data.get('title')
        p = Thread(target=save_data, args=(image_title, image_src))
        p.start()
        p_list.append(p)
    for p in p_list:
        p.join()


if __name__ == '__main__':
    ...
# main_normal() 34s
# main_process() 9.7s
# main_thread()  7s

  • 可以看到在多IO阻塞时的速度比较结果为
  • 多线程>多进程>单进程

【六】线程之间数据共享

from threading import Thread

age = 20


def task():
    global age
    age -= 2
    print(f'我是子线程的age:>>>>{age}')


if __name__ == '__main__':
    t = Thread(target=task, args=())
    t.start()
    print(f'我是主线程的age:>>>>{age}')

'''
我是子线程的age:>>>>18
我是主线程的age:>>>>18
'''
  • 可以看到我在子线程中修改了主线程的age
  • 主线程的age也修改了
  • 说明线程之间数据共享

【七】不同的线程的进程ID是相同的

from threading import Thread
import os

age = 20


def task():
    global age
    age -= 2
    print(f'我是子线程的age:>>>>{age}')
    print(f'进程ID:>>>>{os.getpid()}')


if __name__ == '__main__':
    t = Thread(target=task, args=())
    t.start()
    print(f'我是主线程的age:>>>>{age}')
    print(f'进程ID:>>>>{os.getpid()}')

'''
进程ID:>>>>15792
进程ID:>>>>15792
'''

【八】查看线程名称

  • 首先要导入threading模块下的current_thread方法
  • 通过这个方法.name就可以拿到当前线程的名称
from threading import Thread, current_thread


def task(i):
    print(f'线程程{i}名称:>>>>{current_thread().name}')


if __name__ == '__main__':
    t = [Thread(target=task, args=(i,)).start() for i in range(1, 6)]
    print(f'主线程名称:>>>>{current_thread().name}')

'''
线程程1名称:>>>>Thread-1 (task)
线程程2名称:>>>>Thread-2 (task)
线程程3名称:>>>>Thread-3 (task)
线程程4名称:>>>>Thread-4 (task)
线程程5名称:>>>>Thread-5 (task)
主线程名称:>>>>MainThread
'''

【九】查看活跃的线程数量

  • 首先要导入threading模块下的active_count方法
from threading import Thread, current_thread, active_count
import time

def task(i):
    print(f'线程程{i}名称:>>>>{current_thread().name}')
    time.sleep(2)


if __name__ == '__main__':
    t = [Thread(target=task, args=(i,)).start() for i in range(1, 6)]
    print(f'主线程名称:>>>>{current_thread().name}')
    print(f'当前活动线程的数量{active_count()}')

'''
线程程1名称:>>>>Thread-1 (task)
线程程2名称:>>>>Thread-2 (task)
线程程3名称:>>>>Thread-3 (task)
线程程4名称:>>>>Thread-4 (task)
线程程5名称:>>>>Thread-5 (task)
主线程名称:>>>>MainThread
当前活动线程的数量6
'''

【九】守护线程

  • 一般情况下主线程会等待子线程运行结束才结束
  • 因为当主线程结束时,就意味着这个进程就结束了
  • 可如果其他子线程还在运行,就没办法在拿到这个进程的资源进行正常工作了
  • 我们也可以通过跟添加守护进程一样的方式为线程添加守护进程
  • 如果给子线程添加守护线程,主线程 就不会等待子线程结束了才结束
  • 而是根据除守护线程的其他线程是否结束而结束
from threading import Thread
import time


def task_1():
    print('任务一开始')
    time.sleep(1)
    print('任务一结束')


def task_2():
    print('任务二开始')
    time.sleep(2)
    print('任务二结束')


if __name__ == '__main__':
    t1 = Thread(target=task_1, daemon=True)
    t2 = Thread(target=task_2)
    t1.start()
    t2.start()
    print('我是主线程')
    
'''
任务一开始
任务二开始
我是主线程
任务一结束
任务二结束
'''
可以看到,作为守护线程t1,没有为主线程陪葬
因为主线程正在等t2结束
但是如果t1比t2运行时间长,一旦t2运行结束,主线程就会结束运行,然后t1就会跟着主线程一起陪葬
from threading import Thread
import time


def task_1():
    print('任务一开始')
    time.sleep(3)
    print('任务一结束')


def task_2():
    print('任务二开始')
    time.sleep(2)
    print('任务二结束')


if __name__ == '__main__':
    t1 = Thread(target=task_1, daemon=True)
    t2 = Thread(target=task_2)
    t1.start()
    t2.start()
    print('我是主线程')

'''
任务一开始
任务二开始
我是主线程
任务二结束
'''

【十】线程间互斥锁

抢票例子模板

import json
from threading import Thread, Lock
import os
import time
import random


def init_data():
    file_path = os.path.join(os.path.dirname(__file__), 'ticket')
    os.makedirs(file_path, exist_ok=True)
    file_name = os.path.join(file_path, 'tickets.json')
    if not os.path.exists(file_name):
        with open(file_name, 'w', encoding='utf-8') as fp:
            json.dump({'tickets': 3}, fp, ensure_ascii=False)
    return file_name


def search_tickets():
    file_name = init_data()
    with open(file_name, 'r', encoding='utf-8') as fp:
        data = json.load(fp)
        tickets_num = data.get('tickets')
    return tickets_num, data, file_name


def buy_tickets(i):
    tickets_num, data, file_name = search_tickets()
    if tickets_num > 0:
        tickets_num -= 1
        print(f'用户{i}购票成功')
        data['tickets'] = tickets_num
        with open(file_name, 'w', encoding='utf-8') as fp:
            json.dump(data, fp, ensure_ascii=False)
    else:
        print(f'用户{i}购买失败:>>>>余票不足')


def thread_main(mutex, i):
    mutex.acquire()
    time.sleep(random.randint(1, 2))
    buy_tickets(i)
    mutex.release()


if __name__ == '__main__':
    mutex = Lock()
    t_list = [Thread(target=thread_main, args=(mutex, i)) for i in range(1,6)]
    for t in t_list:
        t.start()
    for t in t_list:
        t.join()
'''
用户1购票成功
用户2购票成功
用户3购票成功
用户4购买失败:>>>>余票不足
用户5购买失败:>>>>余票不足
'''
posted @ 2024-01-18 20:01  HuangQiaoqi  阅读(4)  评论(0编辑  收藏  举报