Codetiming

1.4.0 · active · verified Sat Apr 11

Codetiming is a flexible and customizable Python library for timing code execution. It allows users to measure the performance of different code blocks, functions, or entire scripts using a simple `Timer` class, context manager, or decorator. The current version is 1.4.0, and releases are made on an as-needed basis.

Warnings

breaking Explicit support for Python 3.6 was dropped in version 1.4.0. Users on Python 3.6 should pin their `codetiming` dependency to `<1.4.0`.
Fix: Upgrade to Python 3.7+ or pin `codetiming<1.4.0` in your dependencies.
gotcha When using `Timer` as a decorator on `async` functions, the reported time might be 0 or incorrect. This is a known open issue.
Fix: For `async` functions, prefer using `Timer` as a class or context manager explicitly measuring `await` calls, or check the GitHub issues for updates on fix.
gotcha The `text` argument for `Timer` is not an f-string. If you want to include f-string-like formatting (e.g., a variable from the current scope) within the `text` string, you must use double curly braces to escape the parts that `codetiming` will format.
Fix: Use double curly braces for `codetiming`'s placeholders if mixing with f-string formatting, e.g., `Timer(text=f'Task {task_name} finished in {{:.4f}} seconds')`.
gotcha Named timers accumulate elapsed time across multiple uses. If you reuse a `Timer` instance or a name for `Timer` and expect a fresh measurement, you need to manually reset or manage these timings (e.g., retrieve statistics and then delete the timer from `Timer.timers`).
Fix: Be aware of the accumulation behavior of `Timer.timers`. If single measurements are needed for a named timer, create a new `Timer` instance or delete the existing named timer from `Timer.timers` before reuse. Example: `del Timer.timers['my_timer_name']`.

Install

pip install codetiming Install stable version

Imports

Timer
```
from codetiming import Timer
```
While tutorials might show `from timer import Timer`, the installed package is `codetiming`, requiring import from `codetiming`.

Quickstart

Demonstrates `codetiming.Timer` used as a class, a context manager, and a decorator. It also shows how to access accumulated statistics for named timers.

from codetiming import Timer
import time

# As a class
t = Timer(name="class method")
t.start()
time.sleep(0.01)
t.stop()

# As a context manager
with Timer(name="context manager"):
    time.sleep(0.02)

# As a decorator
@Timer(name="decorator")
def my_function():
    time.sleep(0.03)

my_function()

# Accessing accumulated times for named timers
print(f"Total time for 'class method': {Timer.timers.total('class method'):.4f} seconds")
print(f"All named timers: {Timer.timers.items()}")

view raw JSON →