sliding-window-counter 0.0.1

Counter with a time based sliding window

Sliding Window Counter

sliding-window-counter provides time based sliding window counters. These are useful for tracking things like throughput metrics for long running applications.

Installation

Install via pip

pip3 install sliding-window-counter

Quick Start

## Creating Counters
## -----------------------------------------------------------------------------
import datetime as dt
from sliding_window_counter import SlidingWindowCounter

# Count across 1 minute with 5 second resolution
small_counter = SlidingWindowCounter(60, 5)

# Count across 1 hour with 1 minute resolution
long_counter = SlidingWindowCounter(dt.timedelta(hours=1), dt.timedelta(minutes=1))

# Count across 12 hours with 10 second resolution (not a good idea to be honest)
long_and_precise_counter = SlidingWindowCounter(dt.timdelta(hours=12), 10)

## Using Counters
## -----------------------------------------------------------------------------
import random
import time

mini_counter = SlidingWindowCounter(10, 1)

# fill it up and check it's state
for i in range(20):
    add = random.randint(1,20)
    mini_counter.increment(add)
    print(f"{i:>3}s Added: {add:<3} Window: {mini_counter.current_total:<6} Total: {mini_counter.grand_total:<6} Avg: {mini_counter.current_throughput:>5.2f}/s")
    time.sleep(1)

# Watch it drain
for i in range(12):
    time.sleep(1)
    print(f"Window: {mini_counter.current_total:<5} Total: {mini_counter.grand_total:<5} Avg: {mini_counter.current_throughput:>5.2f}/s")

Bugs, Feature Requests etc

TLDR: Please submit an issue on github.

In the case of bug reports, please help me help you by following best practices 1 2.

In the case of feature requests, please provide background to the problem you are trying to solve so to help find a solution that makes the most sense for the library as well as your usecase.

Development

The only development dependency is bash and docker. All actions are run within docker for ease of use. See ./dev.sh help for commands. Typical commands are format, lint, test, repl, build.

I am still working through open source licencing and contributing, so not taking PRs at this point in time. Instead raise and issue and I'll try get to it as soon a feasible.

Licence

This project is licenced under the MIT Licence - see LICENCE.

This project may include other open source licenced software - see NOTICE.

Authors

A project by Nicholas Hairs - www.nicholashairs.com.