nauti

The fastest way to rid your strings of naughty words, supercharged by Rust

The What

nauti is a profanity filter/swear censor for NodeJS that's designed to be faster than other modules on offer.

Most npm package profanity filters use JS regexes to clean serially, nauti uses native Rust code & parallelism across all cores to achieve awesome performance.

The How

For the sake of keeping the readme nice and clean, let's pretend 'Nasty' and 'Ugly' are swear words

const nauti = require('nauti')
 
const dodgyStrings = ['Perfectly fine', 'A nasty phrase', 'Very ugly words']
const cleanedStrings = nauti.cleanStrings(dodgyStrings)
 
// ['Perfectly fine', 'A ***** phrase', 'Very **** words']
 
const dodgyLeaderboard = [{ score: 10, nickname: 'Adam' }, { score: 5, name: 'Nasty name' }]
const cleanedObjects = nauti.cleanObjects(dodgyLeaderboard, 'nickname')
 
// [{ score: 10, nickname: 'Adam' }, { score: 5, name: '***** name' }]

The Why

nauti was developed to filter the leaderboard for a game I developed.

Once the leaderboard grew to around fifteen-thousand records, bad-words-plus - which I was using - began to take an unreasonable amount of time to filter players' usernames. Players are bound to abandon the leaderboard viewer if it takes almost half a minute to load!

nauti filters that same board four-hundred-and-seven times faster on a 16 thread CPU

bad-words-plus: 25.295s
nauti: 62.113ms