bijective-link-shortener
Bijective encode and decode functions to convert between Base 10 and Base 58 (default). I'm using this for a link shortener.
For more on bijective functions and link shortening, please see:
Installation
npm install bijective-link-shortener -S
Demo site
Default encoding alphabet
123456789abcdefghijkmnopqrstuvwxyzABCDEFGHJKLMNPQRSTUVWXYZ
58 characters (default). No number '0', no upper-case letter 'O', no upper-case letter 'I', no uppercase letter 'l'.
The default alphabet can be set manually and/or randomized. See API below.
Basic Usage with default encoding alphabet
const baseConvert = require const databaseId = 2347247 baseConvert.encode // 'd2KM' baseConvert.decode // 2347247
API
baseConvert.encode(num)
: Returns string
. Encodes a given seed value (such as a unique database ID) into a string based on the selected encoding alphabet.
baseConvert.decode(str)
: Returns integer
. Decodes a given string value (such as a short URL string) into a number based on the selected encoding alphabet.
baseConvert.setAlphabet(str)
: Returns string
. Assign a new encoding alpahbet and echoes it back.
baseConvert.getAlphabet()
: Returns string
. Current encoding alphabet.
baseConvert.getBase()
: Returns integer
. The length of current encoding alphabet. For example, a 58 character encoding alphabet equals a Base 58 conversion. A 33 character encoding equals a Base 33 conversion.
baseConvert.isInteger(num)
: Returns boolean
. Indicates whether a given number is an integer.
baseConvert.isPositive(num)
: Returns boolean
. Indicates whether a given number is positive.
baseConvert.isUnique(str)
: Returns boolean
. Indicates whether given string has no repeating characters.
baseConvert.randomizeAlphabet()
: Returns string
. Shuffles current encoding alphabet.