Regex Compiler

A powerful regex builder and compiler that transforms human-friendly DSL into optimized Regular Expressions.

Part of the Open Tech Foundation ecosystem.

Regex Workbench →

---

A human-readable DSL for building, debugging, and compiling complex Regular Expressions into standard RegExp objects.

• 🎨 Intuitive DSL: Designed for clarity, simplicity, and ease of use.
• ⚡ Optimized Output: Generates clean, high-performance regular expressions.
• 🛡️ Smart Wrapping: Automatically handles non-capturing groups for safety.
• 🌐 Cross-Language Support: Designed as a target-agnostic DSL for generating regex across multiple languages.
• 📦 Zero Dependencies: Lightweight and focused on core functionality.

📦 Installation

Install the compiler using your preferred package manager:

# Bun
bun add @opentf/regex-compiler

# pnpm
pnpm add @opentf/regex-compiler

# npm
npm install @opentf/regex-compiler

🛠 Usage

Explore the library's capabilities with full TypeScript support:

Compiling a Named Capture Group

import { compileToJS, type RegexDSL } from "@opentf/regex-compiler";

const dsl: RegexDSL = [
  { capture: { name: "user", pattern: [{ repeat: { type: "word" }, oneOrMore: true }] } }
];

const result = compileToJS(dsl); 

if ('error' in result) {
  console.error(result.error);
} else {
  console.log(result.pattern); //=> "(?<user>\\w+)"
}

🛡️ Validation & Error Handling

The compiler performs deep logical validation. If the DSL is invalid, it returns an object containing an error string and a detailed issues array.

Handling Errors

const invalidDSL = [
  { repeat: "a", oneOrMore: true, count: 5 } // ❌ Conflicting quantifiers
];

const result = compileToJS(invalidDSL);

if ('error' in result) {
  console.log(result.error); 
  //=> "root.0.repeat: Conflicting quantifiers. Use only one of: count, min/max..."
  
  console.log(result.issues);
  /*=> [
    { 
      path: "root.0.repeat", 
      message: "Conflicting quantifiers. Use only one of: count, min/max..." 
    }
  ] */
}

🚀 DSL Features

Our DSL is optimized for developer productivity with a "one-form, no-confusion" standard:

1. Implicit String Literals

Any plain string is automatically treated as an escaped literal.

• ["api.v1"] ⮕ api\.v1

2. The Dual-System

• Strings = Literals (e.g., "digit" matches the word "digit").
• { type: "..." } = Keywords (e.g., { type: "digit" } matches \d).

3. Inline Flags

Declare your flags anywhere in the array.

["hello", { "flags": { "ignoreCase": true } }]

Note

The unicodeSets (v) and unicode (u) flags are mutually exclusive — using both will produce a validation error.

📝 Email Validator Example

The ultimate "Zero-Wrapper" experience:

[
  { "$": "start" },
  {
    "capture": {
      "name": "user",
      "pattern": [
        {
          "repeat": { "charSet": { "chars": "a-z0-9._%+-", "exclude": false } },
          "oneOrMore": true
        }
      ]
    }
  },
  "@",
  {
    "capture": {
      "name": "domain",
      "pattern": [
        {
          "repeat": { "charSet": { "chars": "a-z0-9-", "exclude": false } },
          "oneOrMore": true
        }
      ]
    }
  },
  ".",
  {
    "capture": {
      "name": "tld",
      "pattern": [{ "repeat": { "charSet": { "chars": "a-z", "exclude": false } }, "min": 2 }]
    }
  },
  { "$": "end" },
  { "flags": { "ignoreCase": true } }
]

📄 License

This project is licensed under the MIT License.