timework 0.2.8

Cross-platform python module to set execution time limits (timer, timeout and iterative deepening) as decorators.

timework

Cross-platform python module to set execution time limits ^{timer, timeout, iterative} as decorators.

Install

pip install timework

Usage

import timework as tw

timework.timer

A decorator measuring the execution time.

timework.TimeError contains three parts:

• TimeError.message string, <decorated function name>: <time used> seconds used
• TimeError.result return values of the function being decorated
• TimeError.detail float, time used

Notice: In @timer decorator, timeout is used to raise a Error after the decorated function finishes, but fails to finish within timeout seconds. If you want to terminate the function with a timeout limit, please use @limit.

import logging

@tw.timer(logging.warning)
def timer_demo_a():
    i = 0
    while i < 2 ** 23:
        i += 1
    return i

@tw.timer(print, detail=True)
def timer_demo_b():
    i = 0
    while i < 2 ** 24:
        i += 1
    return i

@tw.timer(timeout=1)
def timer_demo_c():
    i = 0
    while i < 2 ** 25:
        i += 1
    return i

a = timer_demo_a()
b = timer_demo_b()

try:
    c = timer_demo_c()
except tw.TimeError as e:
    print('error:', e.message)
    c = e.result

print(a, b, c)

WARNING:root:timer_demo_a: 0.496672 seconds used
START:  Tue Feb 18 15:06:45 2020
FINISH: Tue Feb 18 15:06:46 2020
timer_demo_b: 0.989352 seconds used
error: timer_demo_c: 1.9817 seconds used
8388608 16777216 33554432

timework.limit

A decorator limiting the execution time.

timework.TimeError only contains:

• TimeError.message string, <decorated function name>: <timeout> seconds exceeded
• TimeError.result None, unused
• TimeError.detail None, unused

@tw.limit(3)
def limit_demo(m):
    i = 0
    while i < 2 ** m:
        i += 1
    return i

try:
    s = limit_demo(4)
except tw.TimeError as e:
    print(e)
else:
    print('result:', s)

try:
    s = limit_demo(30)
except tw.TimeError as e:
    print(e)
else:
    print('result:', s)

result: 16
limit_demo: 3 seconds exceeded

timework.iterative

A decorator used to process iterative deepening.

timework.TimeError contains three parts:

• TimeError.message string, <decorated function name>.iterative_deepening: <timeout> seconds exceeded
• TimeError.result return values at current level of the iterative deepening search
• TimeError.detail collections.deque, results of the previous levels

Notice: Please use keyword arguments when calling the function that is being decorated. Make sure the max-depth variable of the inner function is given at key^{(key='max_depth' by default)} and whose value should be the maximum search-depth^(integer).

@tw.iterative(3)
def iterative_demo_a(max_depth):
    i = 0
    while i < 2 ** max_depth:
        i += 1
    return max_depth, i

@tw.iterative(3, history=5, key='depth')
def iterative_demo_b(depth):
    i = 0
    while i < 2 ** depth:
        i += 1
    return depth

try:
    s = iterative_demo_a(max_depth=10)
except tw.TimeError as e:
    print(e.message)
    print(e.result, e.detail)
else:
    print('result:', s)

try:
    s = iterative_demo_a(max_depth=25)
except tw.TimeError as e:
    print(e.message)
    print(e.result, e.detail)
else:
    print('result:', s)

try:
    s = iterative_demo_b(depth=25)
except tw.TimeError as e:
    print(e.message)
    print(e.result, e.detail)
else:
    print('result:', s)

result: (10, 1024)
iterative_demo_a.iterative_deepening: 3 seconds exceeded
(20, 1048576) deque([(20, 1048576)], maxlen=1)
iterative_demo_b.iterative_deepening: 3 seconds exceeded
20 deque([16, 17, 18, 19, 20], maxlen=5)

You can read docstrings for more details.

License

MIT License © bugstop