@codecompose/words-to-numbers

Words To Numbers

This repository was forked from words-from-numbers because the project seems abandoned. This fork:

• Fixes the manifest and makes it a pure ESM package
• Removes old dependencies and Babel transpilation
• Uses Vitest to replace old testing tools
• Upgrades clj-fuzzy to latest version
• Add prettier formatting
• Replace NPM with PNPM
• Add basis for Typescript conversion

I might convert the code to Typescript later. PRs are welcome.

Usage

npm install words-to-numbers

If the whole string passed is a number then it will return a Number type otherwise it will return the original string with all instances of numbers replaced.

TODO: Add functionality for parsing mixed numbers and words. PRs welcome.

Basic Examples

import wordsToNumbers from "words-to-numbers";
wordsToNumbers("one hundred"); //100
wordsToNumbers("one hundred and five"); //105
wordsToNumbers("one hundred and twenty five"); //125
wordsToNumbers("four thousand and thirty"); //4030
wordsToNumbers("six million five thousand and two"); //6005002
wordsToNumbers("a thousand one hundred and eleven"); //1111
wordsToNumbers("twenty thousand five hundred and sixty nine"); //20569
wordsToNumbers("five quintillion"); //5000000000000000000
wordsToNumbers("one-hundred"); //100
wordsToNumbers("one-hundred and five"); //105
wordsToNumbers("one-hundred and twenty-five"); //125
wordsToNumbers("four-thousand and thirty"); //4030
wordsToNumbers("six-million five-thousand and two"); //6005002
wordsToNumbers("a thousand, one-hundred and eleven"); //1111
wordsToNumbers("twenty-thousand, five-hundred and sixty-nine"); //20569

Multiple numbers in a string

Returns a string with all instances replaced.

wordsToNumbers('there were twenty-thousand, five-hundred and sixty-nine X in the five quintillion Y')) // 'there were 20569 X in the 5000000000000000000 Y'

With Fuzzy Matching

Uses Jaro distance to find the best match for the number words. Don't rely on this being completely accurate...

import wordsToNumbers from "words-to-numbers";
wordsToNumbers("won huntred", { fuzzy: true }); //100
wordsToNumbers("too thousant and fiev", { fuzzy: true }); //2005
wordsToNumbers("tree millyon sefen hunderd and twinty sex", { fuzzy: true }); //3000726

Decimal Points

import wordsToNumbers from "words-to-numbers";
wordsToNumbers("ten point five"); //10.5
wordsToNumbers("three point one four one five nine two six"); //3.1415926

Ordinal Numbers

import wordsToNumbers from "words-to-numbers";
wordsToNumbers("first"); //1
wordsToNumbers("second"); //2
wordsToNumbers("third"); //3
wordsToNumbers("fourteenth"); //14
wordsToNumbers("twenty fifth"); //25
wordsToNumbers("thirty fourth"); //34
wordsToNumbers("forty seventh"); //47
wordsToNumbers("fifty third"); //53
wordsToNumbers("sixtieth"); //60
wordsToNumbers("seventy second"); //72
wordsToNumbers("eighty ninth"); //89
wordsToNumbers("ninety sixth"); //96
wordsToNumbers("one hundred and eighth"); //108
wordsToNumbers("one hundred and tenth"); //110
wordsToNumbers("one hundred and ninety ninth"); //199

Implied Hundreds

wordsToNumbers("nineteen eighty four", { impliedHundreds: true }); //1984
wordsToNumbers("one thirty", { impliedHundreds: true }); //130
wordsToNumbers("six sixty two", { impliedHundreds: true }); //662
wordsToNumbers("ten twelve", { impliedHundreds: true }); //1012
wordsToNumbers("nineteen ten", { impliedHundreds: true }); //1910
wordsToNumbers("twenty ten", { impliedHundreds: true }); //2010
wordsToNumbers("twenty seventeen", { impliedHundreds: true }); //2017
wordsToNumbers("twenty twenty", { impliedHundreds: true }); //2020
wordsToNumbers("twenty twenty one", { impliedHundreds: true }); //2021
wordsToNumbers("fifty sixty three", { impliedHundreds: true }); //5063
wordsToNumbers("fifty sixty", { impliedHundreds: true }); //5060
wordsToNumbers("fifty sixty three thousand", { impliedHundreds: true }); //5063000
wordsToNumbers("one hundred thousand", { impliedHundreds: true }); //100000