Skip to main content

Set of tools for throttling and controlling the execution timing of Python code.

Project description

pypi version python versions pipeline status coverage report docs

This project offers some convenient tools for throttling and controlling the execution timing of functions or iterative blocks of Python code.

Key Features

  • Simple and time-accurate loop iterations.

  • Support for synchronous and asynchronous programming.

  • Rate limiting consecutive function calls.

  • Rate measurement for loops.

Installation

$ pip install pythrottle

Getting started

Throttle

A basic use for throttling the execution of a code block is using Throttle.loop() (or Throttle.aloop() for asynchronous mode). This will allow execution of the code every 1 / rate seconds:

from throttle import Throttle

rate = 2.0     # Target rate
t = Throttle(interval=(1 / rate))

for i in t.loop():
    # Do something
    print(f"Iteration {i}")

The next example code records a 15-seconds video file from the default video source at an accurate frame rate of 24 fps using OpenCV.

import cv2
from throttle import Throttle

rate = 24.0             # Target frame rate
cap = cv2.VideoCapture(0)
out = cv2.VideoWriter('output.avi', cv2.VideoWriter_fourcc(*'XVID'),
                      rate, (640, 480))

t = Throttle(interval=(1 / rate))

for _ in t.loop(duration=15.0):
    ret, frame = cap.read()    # Frame capture
    out.write(frame)           # Save frame to output file

    # Display the resulting frame
    cv2.imshow('frame', frame)
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

If you simply sleep() for 1 / 24 seconds between frame capture, there would be a difference between the capture rate and the output video rate because of the time required for frame capture. If you also add image processing (motion detection, text overlay…), the delay could cause the output to be completely out of sync.

Throttle decorators

You can also use throttle.throttle() and throttle.athrottle() decorators to limit the number of calls to a function. In the next example, the function hello() is decorated to rate-limit the /throttled endpoint, using a Flask server. Only 2 requests will be served every 5 seconds.

from flask import Flask
from throttle import throttle

app = Flask(__name__)

@app.route("/throttled")
@throttle(limit=2, interval=5, on_fail=("Limit reached :(", 429))
def hello():
    return "Hi, Throttle!"

if __name__ == '__main__':
    app.run()

Decorators can be nested to create more complex throttling rules.

Rate Meter

RateMeter class is useful for measuring the rate of an iterative code taking into account only the last few seconds, so the measured value is kept updated.

The next code block prints the execution rate of a loop that starts looping at 10 ips (iterations per second) and decreases up to 5 ips. In each iteration, the rate is displayed and updated taking into account the iterations history of the last 2 seconds.

import time
from rate_meter import RateMeter

rate_meter = RateMeter(interval=2.0)

for i in range(100):
    rate_meter.update()
    measured_rate = rate_meter.rate()
    print(f"Rate: {rate_meter.rate()}")
    time.sleep(0.1 + i * 0.001)

License

Distributed under the terms of the MIT License.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pythrottle-0.1.1.tar.gz (6.3 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page