Gluify

A type-safe pipeline library for TypeScript that glues functions from different libraries together.

Why Gluify?

JavaScript lacks native pipe operators (like R's |> or PowerShell's |). While TC39's F# pipeline proposal is in the works, Gluify provides a production-ready solution today for creating type-safe, composable data pipelines.

Unlike Lodash (which only chains its own methods) or Ramda (which requires curried functions), Gluify lets you chain ANY functions from different libraries together.

Features

✅ Lazy Evaluation - Nothing executes until .run() or .runAsync() ✅ Async Support - Mix sync/async functions freely with .pipeAsync(), single await at end ✅ Error Handling - Built-in .catch(), .recover(), .when() for robust pipelines ✅ 27+ Utility Methods - Built-in, type-safe, no separate imports ✅ Type Safety - Full TypeScript generics, conditional types ✅ Clean API - Single import, intuitive chaining ✅ Zero Dependencies - Lightweight and fast

Installation

npm install gluify

Quick Start

import { gluify } from 'gluify';

// Basic pipeline
const result = gluify(() => [1, 2, 3, 4, 5])
  .filter(n => n % 2 === 0)
  .map(n => n * 2)
  .reduce((sum, n) => sum + n, 0)
  .run();  // 12

// Async pipeline
const user = await gluify(fetchUser, userId)
  .pipe(validateUser)
  .pipe(enrichProfile)
  .catch(error => defaultUser)
  .runAsync();

// String processing
const result = gluify(() => '  hello world  ')
  .trim()
  .toUpperCase()
  .replace('WORLD', 'GLUIFY')
  .run();  // "HELLO GLUIFY"

Core Concepts

1. Lazy Evaluation

Operations are stored and only executed when you call .run() or .runAsync():

const chain = gluify(expensiveFunction)
  .pipe(operation1)
  .pipe(operation2);  // Nothing executed yet!

const result = chain.run();  // NOW it executes

2. Type-Safe Chaining

TypeScript knows which methods are available based on the current type:

gluify(() => [1, 2, 3])
  .map(n => n * 2)      // ✅ Available - T is array
  .filter(n => n > 2)   // ✅ Available - still array
  .trim()               // ❌ Error - trim only for strings

3. Async Made Easy

No more nested await or .then() chains. Use .pipeAsync() to handle Promises in the chain:

// Before (painful)
const data = await fetchData(id);
const processed = await processData(data);
const validated = await validate(processed);

// After (clean) - pipeAsync awaits Promises automatically
const result = await gluify(fetchData, id)
  .pipeAsync(data => processData(data))  // Awaits fetchData result
  .pipeAsync(processed => validate(processed))  // Awaits processData result
  .runAsync();

API Reference

Core Methods

gluify(fn, ...args)

Start a pipeline with an initial function and arguments.

gluify(() => 42)
gluify(Math.sqrt, 16)
gluify(fetchUser, userId)

.pipe(fn, ...args)

Add any function to the pipeline.

.pipe(x => x * 2)
.pipe(myFunction)
.pipe(add, 10)  // Additional args

.pipeAsync(fn, ...args)

Async pipe - awaits Promises before applying the function. Use this when the previous operation returns a Promise and you need the resolved value.

// When initial function returns Promise<User>
await gluify(fetchUser, userId)
  .pipeAsync(user => user.profile)  // Awaits Promise, receives User
  .pipe(profile => profile.name)     // Regular pipe for sync transform
  .runAsync();

.run()

Execute the pipeline synchronously and return the result.

const result = gluify(() => 5)
  .pipe(x => x * 2)
  .run();  // 10

.runAsync()

Execute the pipeline asynchronously (handles both sync and async functions).

const result = await gluify(asyncFn)
  .pipeAsync(resolvedValue => transform(resolvedValue))
  .pipe(syncFn)
  .runAsync();

.tap(fn)

Execute a side effect without changing the value (useful for logging).

gluify(() => 42)
  .tap(x => console.log('Debug:', x))
  .pipe(x => x * 2)
  .run();  // Logs: "Debug: 42", returns 84

Error Handling

.catch(handler)

Catch errors and provide recovery logic.

gluify(mightFail)
  .pipe(processData)
  .catch(error => ({ fallback: 'data' }))
  .run();

.recover(fallbackValue)

Simple fallback value on error.

gluify(parseJSON, invalidJSON)
  .recover({ error: true })
  .run();

.when(predicate, fn)

Conditionally execute a function.

gluify(() => user)
  .when(u => u.age >= 18, markAsAdult)
  .run();

Array Utilities

• map(fn) - Transform each element
• filter(predicate) - Keep elements matching predicate
• reduce(fn, initial) - Reduce to single value
• find(predicate) - Find first matching element
• some(predicate) - Check if any element matches
• every(predicate) - Check if all elements match
• take(n) - Get first n elements
• skip(n) - Skip first n elements
• sort(compareFn?) - Sort array
• reverse() - Reverse array
• flat() - Flatten one level
• unique() - Get unique elements

Object Utilities

• pick(...keys) - Select specific keys
• omit(...keys) - Exclude specific keys
• keys() - Get object keys
• values() - Get object values
• entries() - Get key-value pairs
• merge(other) - Merge with another object

String Utilities

• trim() - Remove whitespace
• split(separator) - Split into array
• join(separator?) - Join array into string
• replace(search, replace) - Replace substring
• toUpperCase() - Convert to uppercase
• toLowerCase() - Convert to lowercase

General Utilities

• defaultTo(fallback) - Provide default for null/undefined
• isNil() - Check if null or undefined
• clone() - Shallow clone

Real-World Examples

Data Processing Pipeline

interface Product {
  id: number;
  name: string;
  price: number;
  category: string;
  inStock: boolean;
}

const topProducts = gluify(() => products)
  .filter(p => p.inStock)
  .filter(p => p.category === 'Electronics')
  .map(p => ({ name: p.name, price: p.price }))
  .sort((a, b) => b.price - a.price)
  .take(10)
  .map(p => `${p.name} ($${p.price})`)
  .join(', ')
  .run();

API Error Handling

const userData = await gluify(fetchUser, userId)
  .catch(error => {
    console.error('Fetch failed:', error);
    return getCachedUser(userId);
  })
  .pipe(enrichUserData)
  .when(user => user.needsValidation, validateUser)
  .runAsync();

String Processing

const processedText = gluify(() => 'apple,banana,cherry')
  .split(',')
  .map(fruit => fruit.trim())
  .map(fruit => fruit.toUpperCase())
  .filter(fruit => fruit.length > 5)
  .join(' | ')
  .run();  // "BANANA | CHERRY"

Limitations

Piping to non-first arguments: Gluify always pipes to the first argument. For other positions, use a lambda:

// ❌ Not supported
.pipe(add, 10)  // Where does piped value go?

// ✅ Use a lambda instead
.pipe(x => add(10, x))  // Clear and explicit

This keeps the library simple, type-safe, and easy to understand.

Why Not Lodash or Ramda?

Feature	Gluify	Lodash	Ramda
Chain any function	✅	❌ (only Lodash methods)	⚠️ (requires currying)
Lazy evaluation	✅	✅	❌
Built-in async support	✅	❌	❌
Type safety	✅	⚠️ (limited)	⚠️ (complex)
Error handling	✅	❌	❌
Zero dependencies	✅	❌	❌

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT © oharu121

Links

• npm package
• GitHub repository
• Issues