"""A generally useful event scheduler class.

​

Each instance of this class manages its own queue.

No multi-threading is implied; you are supposed to hack that

yourself, or use a single instance per application.

​

Each instance is parametrized with two functions, one that is

supposed to return the current time, one that is supposed to

implement a delay.  You can implement real-time scheduling by

substituting time and sleep from built-in module time, or you can

implement simulated time by writing your own functions.  This can

also be used to integrate scheduling with STDWIN events; the delay

function is allowed to modify the queue.  Time can be expressed as

integers or floating point numbers, as long as it is consistent.

​

Events are specified by tuples (time, priority, action, argument, kwargs).

As in UNIX, lower priority numbers mean higher priority; in this

way the queue can be maintained as a priority queue.  Execution of the

event means calling the action function, passing it the argument

sequence in "argument" (remember that in Python, multiple function

arguments are be packed in a sequence) and keyword parameters in "kwargs".

The action function may be an instance method so it

has another way to reference private data (besides global variables).

"""

​

# XXX The timefunc and delayfunc should have been defined as methods

# XXX so you can define new kinds of schedulers using subclassing

# XXX instead of having to define a module or class just to hold

# XXX the global state of your particular time and delay functions.

​

import time

import heapq

from collections import namedtuple

try:

    import threading

except ImportError:

    import dummy_threading as threading

try:

    from time import monotonic as _time

except ImportError:

    from time import time as _time

​

__all__ = ["scheduler"]