codefabrik/random-number-generator repository overview

⁠random-number-generator

This project is a http endpoint for serving random numbers within a given range. The Fortuna implementation of Jochen Voss (https://github.com/seehuhn/fortuna⁠) is used.

⁠Usage

⁠URL Parameters

• min - lower bounce for random numbers
• maxExclusive - upper bounce exclusive for random numbers
• count - count of random numbers to get
• format - optional, possible values:
  • json (default)
  • int32_le
  • int32_be

⁠Response

In case for json format:

{
    "min": 1,
    "maxExclusive": 10,
    "count": 5,
    "randomNumbers": [4, 2, 7, 8, 1]
}

In case for int32_le or int32_be format the response is binary.

⁠Cryptographic strength

The RNG Service is using an implementation of Fortuna. To ensure the cryptographical strength cycling and reseeding was implemented in addition.

⁠Cycling

The cycling causes the RNG to skip a random amount of random numbers (1 up to 9).

• A background process is executing the cycling in a random time period (every 1000 up to 10000 milliseconds).
• When requesting random numbers then cycling is interrupting the process in a random step (every 100 up to 600 random numbers).

⁠Testing

See README.md⁠ in folder test for instructions.