使用Windows 7按计划触发python脚本的最简单方法是什么？

Question

When I try to run it using the Task Scheduler, it never works. It gives me weird errors like 0x1 and the script is never run. I have googled this problem and can't really find a sufficient, easy solution. (Someone suggested writing a batch script in Windows but I'm not exactly sure how to do that so I'd rather try different avenues first).

On the Task Scheduler GUI in Windows 7, I go to start a program/script and enter the python.exe directory. The argument that I add to the scheduler GUI is the location of the python script. I run it with the highest privileges (check box) to no avail (it does not work).

If there is an alternative way of running a python script on a schedule, I would be happy to hear about it. I'd rather not use a seemingly ad-hoc method like apscheduler which may require my script to always be running (?). Perhaps there is a way to daemonize this process? I did indeed try using celery but it did not work.

For example:

from celery.task import periodic_task
from celery.schedules import crontab

@periodic_task(run_every=timedelta(seconds=30))
def every_30_seconds():
    print("Running periodic task!")

doesn't work either because apparently it is a deprecated decorator. Any help would be appreciated. The print statement is never printed.

Thanks.

Answer 1

I would seriously reconsider Task Scheduler.

I've just set up a quick Python script to run as a scheduled task on my machine, and it works. (I am however using Windows 8.1 instead of Windows 7, but I can't imagine that would make much difference.)

If you can't get a Python script working on its own, you could write a short batch script to start your Python script: something like the following should work:

@echo off
python YourScript.py > output.log 2> errors.log

Feel free to use the full path to the Python executable and/or the various files mentioned in this line.

Task Scheduler works fine when it does work, but when it can't run a task it can be difficult to find what the problem is, as Task Scheduler will typically show only the exit code. Writing output/errors to file allows you to see any output or traceback that Python may have generated and help you understand what the problem could be.

Answer 2

The following two classes demonstrate how easy it can be to schedule a task in Python and separate the scheduler process from the activity process. By dividing the two tasks, you prevent any problems in the activity from propagating into the scheduler (such as memory leaks et cetera).

class Scheduler(sched.scheduler):

    "Scheduler(path, extension) -> Scheduler instance"

    SECOND_1 = 1
    MINUTE_1 = SECOND_1 * 60
    HOUR_1 = MINUTE_1 * 60

    def __init__(self, path, extension):
        "Initialize the scheduler and schedule the first run after a minute."
        super().__init__()
        self.__path = path
        self.__extension = extension
        self.__processed_files = set(self.matched_files)
        self.enter(self.MINUTE_1, 0, self.process_new_files)

    @property
    def matched_files(self):
        "Read-only property of files matching path and extension."
        return glob.iglob('{}\\*.{}'.format(self.__path, self.__extension))

    @property
    def unprocessed_files(self):
        "Read-only property of files that have not been processed."
        return set(self.matched_files) - self.__processed_files

    def process_new_files(self):
        "Schedule to run later and process any unprocessed files."
        self.enter(self.HOUR_1, 0, self.process_new_files)
        print('Checking for unprocessed files @', time.strftime('%c'))
        for file in self.unprocessed_files:
            print('Processing:', repr(file), '...', end='')
            processor = Processor(file)
            processor.start()
            processor.join()
            self.__processed_files.add(file)
            print(' DONE')
        print()

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

class Processor(multiprocessing.Process):

    "Processor(file) -> Processor instance"

    def __init__(self, file):
        "Initialize processor with the file that needs to be processed."
        super().__init__()
        self.__file = file

    def run(self):
        "In a new process, run the original program while logging errors."
        log_errors(process_files, (self.__file, 'Items.txt'))

Answer 3

i'd revisit the task scheduler, assuming always running while PC is on, try the following:

1) open task scheduler

2) in the menubar click "Action" and then "Create Task"

3) under the "General" tab, give it a name and description, and configure your security options

4) under the trigger tab, click "New..."

5) in the "Begin the task" dropbox select "At Startup"

6) check the "repeat task every" box and then set the repeat interval

7) click "Ok" then open "Actions" Tab

8) add the path to python in the "programs/script" box

9a) either add the complete path to your script in the "Add arguments" box or:

9b) add the directory of your script into the "start in" box and then just the script name to the arguments box

10) click ok

this should work no problems HOWEVER if there are spaces in any of your path names, you will need to enclose them in quotes, else they will be interpreted as separate arguments and python will be unable to do anything with it.

James