contributing.md

Contributing to backend

General notes

Please make sure to run eslint and prettier before commiting. This ensures code style consistency, and removes potential bugs. You can do this by running npm run predeploy in the functions folder, or individually running the scripts in package.json.

Conversion models

This is a guide for writing functions for the number format converter API.

The way the API converts between all the formats, is to first convert the input into decimal, and then convert the decimal into the output format.

Therefore each number format needs a function to encode from decimal, and decode to decimal.

Each number format has a conversion model that follows this standard:

interface IModel {
    // function to convert a decimal number into the specified format
    encode: (decimal: number) => string,
    // function to convert a number in the specified format to a decimal number
    decode: (representation: string) => number
}

This interface can be found in src/models/IModel.ts

The structure of each model is already done for you, you just need to fill in the logic to make it functional.

Encoding and decoding

For extra clarity sake:

The encode function must take a Number as an argument, and return the value of that number encoded in the format of the model.

The decode function must take a string representation of a number in the format of the model and return the decimal value as a Number

Test cases

Please write test cases if you are writing a new model. The structure of the test suite can be seen in TestBase64.

For each model (octal, binary, hex etc) there should be a test suite that checks that the conversions are correct. This is done in two ways:

1. known outcomes
2. reverse conversion

For known outcomes, you can find some correct conversions online. For example, you could search 300 in roman numerals to find it should be CCC, and then write a test case that checks if the roman numeral model encoding of 300 is equal to CCC. You should do at least one of these for the encode function and the decode function.

Another way of testing is to generate a random number, encode it using the model, and then decode the output to test if it gives the original random number. This can be written once, and then looped through a thousand times.

Please make sure your test cases are passing before you pull request, as they are not checked by continuous integration actions.