klokke 0.1.0

A small library for high level measurement of code execution time

klokke

klokke is a small library for high level measurement of code execution time. klokke keeps track of running timers, and nests them according to execution order. This allows you to keep track of which sub-sections of code contributes the most to your overall time spent. In addition, a decorator @timed is provided for timing function execution.

Examples

Timing a snippet of code:

>>> from time import sleep
>>> from klokke import Timer
>>> with Timer("Something expensive") as timer:
...     sleep(5)
... 
>>> print(timer)
Something expensive: 5.004069805145264 seconds

Nesting timers:

>>> from time import sleep
>>> from klokke import Timer
>>> with Timer("outer") as outer:
...     with Timer("inner") as inner:
...         sleep(1)
...     sleep(2)
... 
>>> print(outer)
outer: 3.0084269046783447 seconds
Of which:
  inner: 1.0031757354736328 seconds

Using a decorator:

>>> from time import sleep
>>> from klokke import Timer, timed
>>> @timed
... def foo():
...     sleep(2)
...     return 5
... 
>>> with Timer("bar") as t:
...     x = foo() + foo()
... 
>>> print(t)
bar: 4.010395050048828 seconds
Of which:
  __main__.foo: 4.010124683380127 seconds

Setting up timers to automatically print on completion:

>>> from time import sleep
>>> from klokke import Timer
>>> with Timer("Something expensive", autoprint=print) as timer:
...     sleep(5)
... 
Something expensive: 5.000231981277466 seconds

All keyword arguments to Timer can also be passed through the @timed decorator:

>>> from time import sleep
>>> from klokke import timed
>>> import logging
>>> logger = logging.getLogger("foo")
>>> @timed(name="foo_log", autoprint=logger.warning)
... def foo():
...     sleep(1)
... 
>>> foo()
foo_log: 1.0042669773101807 seconds