Process more basic content

1. Create a multi-process in two ways

1.1 a mode (function)

# 方式一
from multiprocessing import Process
import time

def test(name):
    print(f"{name}的子进程开始...")
    time.sleep(3)
    print("子进程结束...")
if __name__ == '__main__':
    p = Process(target=test,args=("xc",)) # 要用元组哦
    p.start()
    # time.sleep(3)
    print("主进程")

1.2 Second way (s)

# 方式二
from multiprocessing import Process
import time

class My(Process):
    def __init__(self,name):
        super().__init__()
        self.name = name
    def run(self):
        print(f"{self.name}的子进程开始...")
        time.sleep(3)
        print("子进程结束...")
if __name__ == '__main__':
    p = My('xc')
    p.start()
    print("xc的主进程")

2. orphaned and zombie process

• Orphan processes: in fact, the main successor to the Implementation finished, the child process has not been performed, because it will become a zombie process, and occupancy pid number, so it will be recycled init method
• Zombie process: in fact, the child process after running over, he is not a true run to completion, but will retain some information, where pid is the most important information, zombie process, will be automatically recovered after the main program execution is over

3.pid, a process ID

from multiprocessing import Process
import time
import os

class My(Process):
    def __init__(self,name):
        super().__init__()
        self.name = name
    def run(self):
        print(f"{self.name}的子进程开始...")
        print(self.pid)
        print(os.getpid())
        print(os.getppid())
        time.sleep(3)
        print("子进程结束...")
if __name__ == '__main__':
    p = My('xc')
    print(os.getpid())
    p.start()
    print("xc的主进程")

"""
打印结果:
    11200
    xc的主进程
    xc的子进程开始...
    6664
    6664
    11200
    子进程结束...
"""

• Through the above data and our understanding of the getpid getppid method to get the process of using the pid number os module, which is to obtain getpid pid the current process, and ppid is to obtain the parent process pid
• Of course, this view is more than a way pid

4.Process join method

• Wait for the child run to completion, code execution in the main process, blocking the main program running

from multiprocessing import Process
import time
def foo():
    print("程序开始...")
    time.sleep(3)
    print("程序结束...")

if __name__ == '__main__':
    p = Process(target=foo)
    p1 = Process(target=foo)
    p2 = Process(target=foo)
    start = time.time()
    p.start()
    p1.start()
    p2.start()
    p.join()     # 会阻塞主程序的运行
    p1.join()
    p2.join()
    end = time.time()
    print("主程序结束...")
    print(start-end)
# 打印结果:
    # 程序开始...
    # 程序结束...
    # 程序开始...
    # 程序开始...
    # 程序结束...
    # 程序结束...
    # 主程序结束...
    # -3.150564193725586
# 如果表示形式为以上的AAABBB形式,那么所需的时间为3s以上,说明这种写法是主程序等待所有都在运行的子程序执行完

###################################################################################################

from multiprocessing import Process
import time
def foo():
    print("程序开始...")
    time.sleep(3)
    print("程序结束...")

if __name__ == '__main__':
    p = Process(target=foo)
    p1 = Process(target=foo)
    p2 = Process(target=foo)
    start = time.time()
    p.start()
    p.join()  # 会阻塞主程序的运行
    p1.start()
    p1.join()
    p2.start()
    p2.join()
    print("主程序结束...")
    end = time.time()
    print(start-end)
# 打印结果：
    # 程序开始...
    # 程序结束...
    # 程序开始...
    # 程序结束...
    # 程序开始...
    # 程序结束...
    # 主程序结束...
    # -9.349010705947876
# 根据上面的内容来看，如果按照ABABAB执行，就会出现子程序等其他子程序执行完再执行，就会叠加时间，就是一个串行的进程了

5.Process the name and is_alive method

• name: Check the child's name (like this name, Process-1)
• is_alive: The return value is a Boolean type, determine whether the child process ends, the end will take some time, so use a short time modules, etc., or use the join method of waiting for the end, if not end it returns True, the end returns False

from multiprocessing import Process
import time

def foo():
    print("进程开始...")
    time.sleep(3)
    print("进程结束...")

if __name__ == '__main__':
    p = Process(target=foo)
    # print(p.is_alive())
    p.start()
    print(p.name)
    print(p.is_alive())
    time.sleep(5)
    print("主程序结束...")
    print(p.is_alive())

# 自己试验下,情况就是name这个属性对应着创建线程的一个代号,它是往下迭代的例如:Process-1,Process-2这样迭代

# 关于is_alive则是表示子程序有没有结束,主要是看子程序有没有运行完,可以使用time模块测试

6.Process的terminal

• Force the end of the child process (similar to the start, is a signal to the operating system, not directly end)

from multiprocessing import Process
import time

def foo():
    print("进程开始...")
    time.sleep(3)
    print("进程结束...")

if __name__ == '__main__':
    p = Process(target=foo)
    p.start()
    print(p.is_alive())
    p.terminate()  # 强制结束子进程
    p.join() # 我们必须在这里使用join,等待进程结束,那就有人问了,你等他结束了is_alive肯定是Flase呀,但是我们的注意点是子程序没有打印数据
    print(p.is_alive())
    print("主进程结束...")

# 对于Process中的terminate的使用也和start一样,他只是去给操作系统发送了关闭子进程的信号,并不是直接结束

7. daemon

• Loyal process by object name .daemon = True way to define a daemon.

from multiprocessing import Process
import time
"""
守护进程
守护 --> 伴随
本质上也是一个子进程
主进程的代码执行完毕守护进程直接结束,但是此时主进程可能没有结束
"""
def foo():
    print("守护进程开始...")
    time.sleep(3)
    print("守护进程结束...")

if __name__ == '__main__':
    p = Process(target=foo)
    p.daemon = True # 这个就是把创建的子进程定义为守护进程,默认值应该是Flase
    p.start()
    # p.join()
    time.sleep(2)
    print("主进程结束...") # 也就是说主进程结束了,子进程就自动释放了,就是不会运行了

# 总结就是daemon这个方法把创建的子进程定义为守护进程,所谓的守护进程就是:主进程结束了,子进程也就是守护进程结束了
# 如果主进程没结束,子进程也就是守护进程便会像普通进程一样从有到无,从出现到离开
# 普通的子进程是不会管主进程的死活的,大不了变成孤儿进程继续运行,反正会有init方法回收