下面是我的代码的很大一部分，基本上，如果您向下滚动到 execute_subscripts() 函数，您可以看到我有两个通过 execfile 运行的脚本工作精美，它们显示 prints，它们将 traceback 错误保存到错误文件中。

我正在尝试将第二个脚本变成一个不等待自己完成就可以继续下一个脚本的脚本。

如您所见，我尝试将 subprocess 与 Popen 一起使用来启动一个静默的隐藏窗口...但是它似乎没有运行而且我有不知道如何正确使用 p.communicate() 函数来检索 tracebacks 和/或 prints。

我也... 需要帮助创建某种超时/终止开关，所以如果通过 Popen 或 execfile 路径下标如果它在 5 分钟内没有完成，则跳过该循环或重试并在它立即再次失败时跳过。

我知道我可能不应该暂时使用 strftime……但是那部分对我来说工作正常，所以我认为没有必要改变它。

from datetime import date, timedelta
from sched import scheduler
from time import time, sleep, strftime
import random
import traceback
import subprocess

s = scheduler(time, sleep)
random.seed()

def periodically(runtime, intsmall, intlarge, function):

     ## Get current time
    currenttime = strftime('%H:%M:%S')

    ## If currenttime is anywhere between 23:40 and 23:50 then...
    if currenttime > '23:40:00' and currenttime < '23:50:00':

        ## Open the error logging file as the variable "errors"
        errors = open('MISC/ERROR(S).txt', 'a')

        ## Try to...
        try:
            ## Call the clear subscript.
            execfile("SUBSCRIPTS/CLEAR.py", {})
        ## On exception (fail)...
        except Exception:
            ## Write the entire traceback error to file...
            errors.write(traceback.format_exc() + '\n')
            errors.write("\n\n")

        ## Close and exit the error logging file. 
        errors.close()

        ## Update time
        currenttime = strftime('%H:%M:%S')

    ## Idle time
    while currenttime >= '23:40:00' and currenttime <= '23:59:59' or currenttime >= '00:00:00' and currenttime <= '11:30:00':

        ## Update time
        currenttime = strftime('%H:%M:%S')
        print currenttime, "Idling..."
        sleep(10)

        ## Update time
        currenttime = strftime('%H:%M:%S')

    ## Initiate the scheduler.
    runtime += random.randrange(intsmall, intlarge)
    s.enter(runtime, 1, function, ())
    s.run()

def execute_subscripts():

    st = time()
    print "Running..."
    errors = open('MISC/ERROR(S).txt', 'a')

    try: 
        execfile("SUBSCRIPTS/TESTSCRIPT.py", {})
    except Exception:
        errors.write(traceback.format_exc() + '\n')
        errors.write("\n\n")

    try: 
        execfile("SUBSCRIPTS/TEST.py", {})
    except Exception:
        errors.write(traceback.format_exc() + '\n')
        errors.write("\n\n")
##    subprocess.Popen(["pythonw", "SUBSCRIPTS/TEST.py", "0"], shell=True)

    try: 
        execfile("SUBSCRIPTS/TESTSCRIPTTest.py", {})
    except Exception:
        errors.write(traceback.format_exc() + '\n')
        errors.write("\n\n")

    try: 
        execfile("SUBSCRIPTS/TESTTESTTEST.py", {})
    except Exception:
        errors.write(traceback.format_exc() + '\n')
        errors.write("\n\n")

    errors.close()
    print """The whole routine took %.3f seconds""" % (time() - st)

while True:
    periodically(50, -25, +90, execute_subscripts)

任何想法将不胜感激

添加了赏金，希望有人知道如何实现这一目标。

提前致谢
海福莱克斯

我希望脚本能够执行的操作的示例...

1. 下标 1 - 在后台运行，将打印和错误从 subscript1.py 发送到 main.py，不要等待它完成，转到下标 2，10 秒后超时(或者尽可能接近 10 秒，或者在调用所有下标后超时。)

2. 下标 2 - 在后台运行，将打印和错误从 subscript2.py 发送到 main.py，在进入下标 3 之前等待它完成，10 秒后超时(或尽可能接近 10 秒，或者在调用所有下标后超时。)

3. 下标 3 - 在后台运行，将打印和错误从 subscript3.py 发送到 main.py，在进入下标 4 之前等待它完成，10 秒后超时(或尽可能接近 10 秒，或者在调用所有下标后超时。)

4. 下标 4 - 在后台运行，将打印和错误从 subscript4.py 发送到 main.py，不要等待它完成，转到下标 5，10 后超时秒(或尽可能接近 10 秒，或在调用所有下标后超时。)

5. 下标 5 - 在后台运行，将打印和错误从 subscript5.py 发送到 main.py，在进入下一个下标(或在本例中为循环结束)之前等待它完成， 10 秒后超时(或尽可能接近 10 秒，或调用所有下标后超时。)

shx2 的打印和追溯

[pid=9940] main running command: C:\Python27\python.exe SUB/subscript1.py (is_bg=False)
[pid=9940] main running command: C:\Python27\python.exe SUB/subscript1.py (is_bg=True)

Traceback (most recent call last):
  File "C:\Test\main.py", line 21, in <module>
    bg_proc1 = run_subscript(cmd, is_bg = True)
  File "C:\Test\main.py", line 10, in run_subscript
    return (cmd > sys.stdout) & BG  # run in background
  File "C:\Python27\lib\site-packages\plumbum\commands\modifiers.py", line 81, in __rand__
    return Future(cmd.popen(), self.retcode)
  File "C:\Python27\lib\site-packages\plumbum\commands\base.py", line 317, in popen
    return self.cmd.popen(args, **kwargs)
  File "C:\Python27\lib\site-packages\plumbum\commands\base.py", line 233, in popen
    return self.cmd.popen(self.args + list(args), **kwargs)
  File "C:\Python27\lib\site-packages\plumbum\machines\local.py", line 104, in popen
    **kwargs)
  File "C:\Python27\lib\site-packages\plumbum\machines\local.py", line 253, in _popen
    stderr = stderr, cwd = str(cwd), env = env, **kwargs)  # bufsize = 4096
  File "C:\Python27\lib\subprocess.py", line 703, in __init__
    errread, errwrite) = self._get_handles(stdin, stdout, stderr)
  File "C:\Python27\lib\subprocess.py", line 851, in _get_handles
    c2pwrite = msvcrt.get_osfhandle(stdout.fileno())
UnsupportedOperation: fileno

编辑:

             | --> # Sub 1.py # --> Sequential with timeout --> Started: 11:30.00 --> Estimated Completion: 11:30.01 (1 Second) --> Timeout at 11:30:10 (10 Seconds) --> # Sub 2.py # --> Sequential with timeout --> Started: 11:30.02 (or after time Sub 1.py's timeout) --> Estimated Completion: 11:30.03 (1 Second) --> Timeout at 11:30:13 (10 Seconds) --> # Sub 3.py # --> Sequential with timeout --> Started: 11:30.04 (or after time Sub 2.py's timeout) --> Estimated Completion: 11:30.08 (3 Seconds) --> Timeout at 11:30:18 (10 Seconds)
             |                                                                                                                                                  ^                                                                                                                                                                             ^
             |                                                                                                                                                  |                                                                                                                                                                             |
             | --------------------------------------------------------------------------------------------------------------------------------------------------                                                                                                                                                                             |
             | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
             |
Scheduler -->|
             | --> Sub 4.py --> Nonsequential with timeout --> Started: 11:30.00 --> Estimated Completion: 11:30.05 (5 Seconds) --> Timeout at 11:30:10 (15 Seconds)
             |
             | --> Sub 5.py --> Nonsequential with timeout --> Started: 11:30.00 --> Estimated Completion: 11:30.02 (2 Seconds) --> Timeout at 11:30:10 (10 Seconds)
             |
             | --> Sub 6.py --> Nonsequential with timeout --> Started: 11:30.00 --> Estimated Completion: 11:30.10 (10 Seconds) --> Timeout at 11:30:10 (25 Seconds)

希望这有助于直观地展示我想要实现的目标

最佳答案

正如您在问题中已经指出的那样，您实际上是在问两个不同的问题(在后台运行，并强制超时)。幸运的是，两者的简短答案是一样的:

使用 Plumbum !

Plumbum 极大地简化了 python 脚本的类似 shell 脚本的元素，除其他外，为在后台运行命令和强制超时提供了干净的界面。

下面是一个使用 plumbum 的例子。

在此示例中，子进程将全部运行相同的脚本 -- subscript1.py。它进行一些打印，一些休眠，有时会随机失败。

subscript1.py

import os, sys, time, random
print '[pid=%s] STARTING %s' % (os.getpid(), sys.argv[0])
for i in range(3):
    t = random.randint(1,5)
    print '[pid=%s] sleeping for %s seconds' % (os.getpid(), t)
    time.sleep(t)
# fail randomly
if t == 5:
    raise RuntimeError('random error...')
print '[pid=%s] DONE %s' % (os.getpid(), sys.argv[0])

现在，下面的主脚本 main.py 演示了如何在前台和后台运行子进程，有或没有超时，等待后台进程完成，以及处理子进程错误和超时。

主.py

import os, sys, time
from plumbum import FG, BG, ProcessExecutionError, ProcessTimedOut
from plumbum.cmd import python

cmd = python['subscript1.py']  # create the command to run (several times)

def run_subscript(cmd, is_bg = False):
    print '[pid=%s] main running command: %s (is_bg=%s)' % (os.getpid(), cmd, is_bg)
    if is_bg:
        return (cmd > sys.stdout) & BG  # run in background
    else:
        try:
            return cmd & FG  # run in foreground
        except ProcessExecutionError, e:
            print >>sys.stderr, e

# run a process in the foreground        
run_subscript(cmd, is_bg = False)

# run two processes in the background, and one in the foreground
bg_proc1 = run_subscript(cmd, is_bg = True)
time.sleep(1)
bg_proc2 = run_subscript(cmd, is_bg = True)
time.sleep(1)
run_subscript(cmd, is_bg = False)

# wait for the background processes to finish
for bg_proc in ( bg_proc1, bg_proc2 ):
    try:
        bg_proc.wait()
    except ProcessExecutionError, e:
        print >>sys.stderr, e

# run a foreground process, which will time out
print '[pid=%s] main running command: %s (will time out)' % (os.getpid(), cmd)
try:
    cmd.run(timeout = 2)
except ProcessTimedOut, e:
    # command timed out
    print >>sys.stderr, e
except ProcessExecutionError, e:
    # command failed (but did not time out)
    print >>sys.stderr, e

输出:

% python main.py
[pid=77311] main running command: /usr/local/bin/python subscript1.py (is_bg=False)
[pid=77314] STARTING subscript1.py
[pid=77314] sleeping for 1 seconds
[pid=77314] sleeping for 5 seconds
[pid=77314] sleeping for 3 seconds
[pid=77314] DONE subscript1.py
[pid=77311] main running command: /usr/local/bin/python subscript1.py (is_bg=True)
[pid=77316] STARTING subscript1.py
[pid=77316] sleeping for 5 seconds
[pid=77311] main running command: /usr/local/bin/python subscript1.py (is_bg=True)
[pid=77317] STARTING subscript1.py
[pid=77317] sleeping for 1 seconds
[pid=77311] main running command: /usr/local/bin/python subscript1.py (is_bg=False)
[pid=77317] sleeping for 5 seconds
[pid=77318] STARTING subscript1.py
[pid=77318] sleeping for 5 seconds
[pid=77316] sleeping for 2 seconds
[pid=77316] sleeping for 4 seconds
[pid=77317] sleeping for 5 seconds
[pid=77318] sleeping for 2 seconds
[pid=77318] sleeping for 3 seconds
[pid=77316] DONE subscript1.py
[pid=77318] DONE subscript1.py
Command line: ['/usr/local/bin/python', 'subscript1.py']
Exit code: 1
Stderr:  | Traceback (most recent call last):
         |   File "subscript1.py", line 13, in <module>
         |     raise RuntimeError('random error...')
         | RuntimeError: random error...
[pid=77311] main running command: /usr/local/bin/python subscript1.py (will time out)
('Process did not terminate within 2 seconds', ['/usr/local/bin/python', 'subscript1.py'])

编辑:

我现在意识到我的示例代码没有演示在后台运行命令和对其强制执行超时。为此，只需使用 cmd.bgrun(...) 而不是 cmd.run(...)。

您收到的错误与重定向有关，并且必须与您在 Windows 上运行的事实有关。这要么是 plumbum 在 Windows 上的兼容性问题，要么我的代码可能不完美，即可能有另一种方法可以使用 plumbum 使其工作。不幸的是，我没有 Windows 机器来测试它......

希望对您有所帮助。

关于python - 自定义调度程序以具有带超时/终止开关的顺序 + 半顺序脚本？，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/19649788/