type-level-regexp

🔤🔍 Type-Level RegExp (WIP)

TypeScript type-level RegExp parser and matcher implemented using template literals.

Examples in Playground, test

🚧 🚧 Work In Progress, PRs and issues are welcome 🚧 🚧

Quick Setup

1. Add type-level-regexp dependency to your project

# Using pnpm
pnpm i -D type-level-regexp

# Using npm
npm i -D type-level-regexp

1. Import createRegExp from type-level-regexp, create a TypedRegExp and pass it to String.match or String.replace functions.

Basic Usage

match result will be fully typed if match against a literal stirng, or shows emumerated results if match against a dynamic string.

import { createRegExp, spreadRegExpIterator } from 'type-level-regexp'

/** string.match() */
const regExp = createRegExp('foO(?<g1>b[a-g]r)(?:BAz|(?<g2>qux))', ['i'])
const matchResult = 'prefix foobarbaz suffix'.match(regExp) // matching literal string
matchResult[0] // 'foobarbaz'
matchResult[1] // 'bar'
matchResult[3] // show type error `type '3' can't be used to index type 'RegExpMatchResult<...>`
matchResult.length // 3
matchResult.index // 7
matchResult.groups // { g1: "bar"; g2: undefined; }

/** string.replace() */
const regExp2 = createRegExp('(\\d{4})[-.](?<month>\\w{3,4})[-.](\\d{1,2})')
const replaceResult = '1991-Sept-15'.replace(regExp2, '$<month> $3, $1')
replaceResult // 'Sept 15, 1991'

/** string.matchAll() */
const regExp3 = createRegExp('c[a-z]{2}', ['g'])
const matchALlIterator = 'cat car caw cay caw cay'.matchAll(regExp3)
const spreadedResult = spreadRegExpIterator(matchALlIterator)
spreadedResult[2][0] // 'caw'
spreadedResult[3].index // 12

For TypeScript library authors, you can also import individual generic types to parse and match RegExp string at type-level and combine with your library's type-level features.

import { ParseRegExp, MatchRegExp } from 'type-level-regexp'

type MatchResult = MatchRegExp<'fooBAR42', 'Fo[a-z](Bar)\\d{2}', 'i'>

type Matched = MatchResult[0] // 'fooBAR42'
type First = MatchResult[1] // 'BAR'

type RegExpAST = ParseRegExp<'foo(?<g1>bar)'>
// [{
//     type: "string";
//     value: "foo";
// }, {
//     type: "namedCapture";
//     name: "g1";
//     value: [{
//         type: "string";
//         value: "bar";
//     }];
// }]

Origin & Notice

The main purpose of this project is to test and demonstrate the possibility and limitations of writing a RegExp parser/matcher in TypeScript's type-level. Note that this may not be practically useful, but rather an interesting showcase.

The idea for this project originated while I was working on improving the type hints of string.match and replace in magic-regexp (created by the most inspiring, resourceful, and kind Daniel Roe from Nuxt, definitely check it out if you are working with RegExp and TypeScript!).

As the complexity grows, I start working on this separated repo to increase development speed and try out different iterations. Hopefully, it can be ported back to magic-regexp, and even Gabriel Vergnaud's awesome hotscript when it's stable with good performance.

❤️ Testing, feedbacks and PRs are welcome!

Features

• Export createRegExp function to create a TypedRegExp that can be pass to String match and replace functions and gets fully typed match result.
• Enhance types of RegExp related String functions (.match, matchAll, .replace...) for literal or dynamic typed string.
• Result of String functions matched exactly as runtime result.
• Support all common RegExp tokens (incl. Lookarounds, Backreferences...etc), quantifiers (incl. greedy/lazy) and (g,i) flags.
• Export helper functions spreadRegExpMatchArray and spreadRegExpIterator to get tuple type of match results and iterators.
• Provide generic type ParseRegExp to parse and RegExp string to AST.
• Provide generic type MatchRegExp to match giving string with a parsed RegExp.
• Provide generic type ResolvePermutation to permutation all possible matching string of given RegExp if possible (due to TypeScript type-level limitation)
• More details please see playground, or test in Tests or Stackblitz. (examples in index.test-d.ts)

That's it! You can now use Tailwind classes in your Nuxt app ✨

Example - type-safe args in replacing function of string.replace()

Example - spreaded string.matchAll() with union of RegExp pattern remain as tuple

RegExp Tokens & Flags

Tokens	Description	Support
.	Matches any single character.	✅
*, *?	Matches zero or more occurrences (Greedy/Lazy).	✅
+, *?	Matches one or more occurrences (Greedy/Lazy).	✅
?, ??	Matches zero or one occurrence (Greedy/Lazy).	✅
^	Matches the start of a line.	✅
$	Matches the end of a line.	✅
\s, \S	Matches any whitespace, non-whitespace character.	✅
\d, \D	Matches any digit, non-digit character.	✅
\w, \W	Matches any word, non-word character.	✅
\b, \B	Matches a word-boundary, non-word-boundary.	✅
[abc]	Matches any character in the set.	✅
[^abc]	Matches any character not in the set.	✅
()	Creates a capturing group.	✅
(?:)	Creates a non-capturing group.	✅
(?<name>)	Creates a named-capturing group.	✅
|	Matches either the expression before or after the vertical bar.	✅
{n}	Matches exactly n occurrences.	✅
{n,}	Matches at least n occurrences.	✅
{n,m}	Matches between n and m occurrences.	✅
(?=), (?!)	Positive/Negative lookahead.	✅
(?<=), (?<!)	Positive/Negative lookbehind.	✅

Flags	Description	Support
g	Global matching (matches all occurrences).	✅
i	Case-insensitive matching.	✅

💻 Development

• Clone this repository
• Enable Corepack using corepack enable (use npm i -g corepack for Node.js < 16.10)
• Install dependencies using pnpm install
• Run interactive tests using pnpm dev

License

Made with 🔥 and ❤️

Published under MIT License.

!-- Badges -->