random_algorithm

A very simple yet original random algo made while overthinking about randomness

random_algorithm is a Python library that generates random numbers using a novel approach based on English words and the current epoch time. It utilizes ASCII values of shuffled words, reduces them to single digits, and combines these digits to generate a random number of a specified size. This implementation attempts to get as close as possible to true randomness by using time-based factors and word shuffling.

Features

• Generates random numbers based on shuffled English words.
• ASCII values of each word are reduced to single digits.
• Supports dynamic generation of random numbers of any specified size.
• Includes special logic to generate lower numbers (like 0 or 1), which can sometimes be difficult to produce with traditional randomization methods.
• Provides functions to extract seeds based on time and to reduce any number to a single digit.

Usage

Install the package in your repo

pip install random_algorithm

If you use a requirements file, add this line to it :

random_algorithm==1.1.0

You can import and use the library in your Python code like this :

from random_algorithm import gen_random

Example :

Generate a random number with a desired size of 5 digits

random_number = gen_random(desired_size=5)
print(f"Generated random number : {random_number}")

You can also use the following internal functions :

• ascii_reduce(word, index) :
  • Takes a word (ex "github") and an index (can be any positive number) as input.
  • Reduces the word to a single digit (returns an int).
• reduce_to_single_digit(n) :
  • Reduces any integer n to a single digit by summing its digits repeatedly until one digit remains.
  • For example, reduce_to_single_digit(9875) would return 2.
• get_time_seed() :
  • Generates a seed based on the current time and ASCII reduction of character permutations.

Handling errors

gen_random() will raise the following exceptions if invalid inputs are provided :

• TypeError : Raised when desired_size is not an integer.
• ValueError : Raised when desired_size is less than 1.

How it works

1. Word shuffling : The words from wordlist.txt are shuffled on each call to ensure randomness.
2. Random index : The current epoch time is used to generate a random index into the shuffled word list.
3. ASCII reduction : Each word’s ASCII values are reduced to single digits and then summed to produce a random number.
4. Low numbers handling : A special condition ensures that numbers like 0 and 1 can be generated using the current epoch time modulo operations.
5. Time Seed: get_time_seed() creates a seed value based on permutations of characters from the current time.

Side notes

The time it takes to generate numbers is pretty linear, I’d evaluate it to $O(n)$
Here’s a quick overview :

On my potato PC, an extrapolation of some results give the following formula (keep in mind that performance will change depending on your hardware) :

$$\text{Time (seconds)} \approx 0.1623 \times \text{Desired size} + 0.1468$$

As a final info, here’s the word length distribution of the dictionnary we use :

(more)

Contributing

Feel free to open an issue or a pull request if you want to contribute to this project

How to build ?

py -m pip install --upgrade pip build twine setuptools wheel
py -m build
py -m twine check dist/*
# Optional : publish to test.pypi.org
py -m twine upload --repository testpypi dist/*
# Or to pypi.org
py -m twine upload dist/*

License

This project is licensed under the MIT License - see the LICENSE file for details.
The wordlist comes from from https://github.com/dwyl/english-words.

Authors

• EDM115 - Initial work