multitimer

A pure-python auto-repeating timer that can be stopped and restarted multiple times.

multitimer.MultiTimer is similar to threading.Timer, but allows the timer to repeat multiple times. Additionally, MultiTimer can be started and stopped multiple times (unlike threading.Timer).

Overview

multitimer.MultiTimer(interval, function, args=None, kwargs=None, count=-1, runonstart=True)

Creates a timer that will run function with arguments args and keyword arguments kwargs, after interval seconds have passed, a total of count times.

If runonstart==True, then function will be called immediately when .start() is called.

If args is None (the default) then an empty list will be used. If kwargs is None (the default) then an empty dict will be used.

If count == -1 (the default), the timer will repeat indefinitely, or until .stop() is called.

Start this timer by calling .start(). Once started, calling .stop() will terminate the timer's loop and not produce any further calls to function. Note that if function is currently in the middle of running, it will finish the current iteration and not be interrupted.

ontimeout and params were deprecated in 0.2 and replaced by function, args and kwargs to match the threading.Timer API. ontimeout and params have been removed in 0.3.

Since the underlying mechanism is purely based on python threads & events, the overall processor load & memory usage are minimal. Note that the timing accuracy is typically to within about 10 ms, depending on the platform.

Installation & usage

$ pip install multitimer

import multitimer
import time

def job():
	print("I'm working...")

# This timer will run job() five times, one second apart
timer = multitimer.MultiTimer(interval=1, function=job, count=5)

# Pauses for one interval before starting job() five times
timer = multitimer.MultiTimer(interval=1, function=job, count=5, runonstart=False)


# You can specify input parameters for the _function_ function
def job2(foo):
	print(foo)

timer = multitimer.MultiTimer(interval=1, function=job2, kwargs={'foo':"I'm still working..."})

# Also, this timer would run indefinitely...
timer.start()

# ...unless it gets stopped
time.sleep(5)
timer.stop()

# and potentially waited for (in case an iteration was in progress)
timer.join()


# If a mutable object is used to specify input parameters, it can be changed after starting the timer
output = {'foo':"Doin' my job again."}
timer = multitimer.MultiTimer(interval=1, function=job2, kwargs=output, count=5)
timer.start()

time.sleep(3.5)
output['foo'] = "I'd like to be done now."

# Note: While this feature can be useful, be aware that changing arguments while the timer is running may result in some
# race conditions. multitimer is multithreaded but does not currently have any sort of locking mechanisms in place to
# ensure that operations are atomic. 


# And a MultiTimer can be re-started by just calling start() again
time.sleep(2)
output['foo'] = 'Please just let me be...'
timer.start()
time.sleep(4.5)
timer.stop()

Releases

0.3, 2020-11-27

• Add a .join() method to wait for a timer that has been stopped to complete its final iteration. (Thanks, @pakal!)
• Remove ontimeout and params arguments (deprecated in 0.2)
• Properly pass args to RepeatingTimer
• Fix error if .stop() called before .start()

0.2, 2019-01-17

• Replace time.clock() calls with time.perf_counter(), as time.clock is deprecated since python 3.3 and doesn't provide consistent behavior across different platforms.
• Replace ontimeout with function, and params with args and kwargs, to match the threading.Timer API. ontimeout and params are deprecated and will be removed in v0.3.
• Add lots of code comments to better explain how the module works.

0.1, 2018-02-15

• Initial release

Meta

Josh Burnett - [email protected]

Distributed under the MIT license. See LICENSE.txt for more information.

https://github.com/joshburnett/multitimer

Hope you find this useful!