cloudflarepycli

Python CLI and utilities for retrieving network performance statistics.

Purpose

Retrieve near-term performance data about the service provided to a user by an ISP. The data includes up and download speeds, latency, and jitter. The CLI makes it possible to pipe the data to other processes for possible uploading, analysis etc. and/or to build a GUI for displaying current and past data. The cloudflare module is useful for varying the types of test that are done. I will use both in future projects to expand the functionality.

Install

pip install cloudflarepycli
Windows users must also: pip install wres

CLI usage

Type cfspeedtest in the environment where you installed the package. Note that this is a shell command, not a Python command.

Options

  --debug     log network io
  --json      write json to stdout instead of formatted results
  --bps       Show output in bits per second
  --version   show program's version number and exit

Cloudflare Documentation

The CloudflareSpeedtest Class is a Python module that facilitates testing the speed and latency of your internet connection to Cloudflare's servers. It utilizes the requests library to perform various tests and provides insights into download and upload speeds, latency, and other connection metrics.

Features

Measures download and upload speeds to Cloudflare's servers.
Calculates latency and jitter for connection stability assessment.
Provides information about your IP address, ISP, test location, and more.
Easy-to-use class interface to conduct various connection tests.

Usage

The CloudflareSpeedtest class provides methods to perform various connection tests and retrieve relevant information. Here's a brief overview of its main methods:

run_all(): Runs a full suite of tests, including latency, jitter, download, and upload tests. Returns a dictionary with test results.
get_metadata(): Retrieves information about your IP address, ISP, test location, and more.
run_test(test): Runs a test according to the given TestSpec, and returns timing results.

Refer to the cloudflare.py file for detailed documentation on each method and its parameters.

Stand-alone usage

Getting Started

Install the required dependencies:
```
pip install requests
```
Download the cloudflareclass.py file and include it in your project.
Import the cloudflare class in your Python script:
```
from cfspeedtest import CloudflareSpeedtest
```

Create an instance of the speedtest class and run tests:

suite = CloudflareSpeedtest()
results = suite.run_all()
print(results)

How it works

Tests for latency are done by requesting one byte packets from Cloudflare, measuring the elapsed time to get a response, and subtracting the server processing time taken from the header in the returned message. Jitter is computed as the mean of the absolute difference between the arrival of consecutive requests.

The cloudflare module makes Python requests to various endpoints of speed.cloudflare.com. Their API is not documented, as far as I know, and so that is a potential breaking point for this code.

Mirroring the performance of the Cloudflare webpage, the CLI does multiple uploads and downloads with different block sizes and the 90th percentile of all these tests is used for calculating up and download times. Results are similar to those obtained from the webpage. Tests can be varied using the cloudflare module.

Unlike Ookla's speedtest CLI, Cloudflare does not require downloading a licensed exe. Cloudflare uses test sites from its own network of caching and hosting centers. This is useful because much of the content users would be retrieving is actually coming from these centers. On the other hand, coverage may be thin in some parts of the world.

Privacy

No identifying information is sent to any website other than the IP address that servers can see in an HTTP request. Cloudflare can probably deduce something, mostly about your location and network architecture, from the tests it runs. No results are sent anywhere. Because this an application and not running in a browser, there are no cookies.

Full source is available in this package.

Background

Billions of federal dollars are being disbursed to improve broadband availability and quality, especially in rural areas. Tools are needed to assure that ISPs deliver the quality they promise. This software is a pro bono contribution to getting those tools written.

Disclaimers

No claims of any sort are made for this software. It has been tested on Windows 10 and 11, MacOS, and Raspberry Pi OS and should work on other Linux versions but not tested. Use and/or redistribute solely at your own risk. No commitment is made to maintain this software. As noted above, changes made by Cloudflare might break the functionality.

I have no affiliation with Cloudflare, any hosting service, or any ISP (except as a customer).

martinbrose/cloudflarepycli