@axanc/ts-utils

ts-utils

About

Lightweight TypeScript utility library offering a collection of essential functions to simplify common coding tasks. Designed for a maximized type-safety, efficiency, readability, and maintainability.

Install

npm install --save @axanc/ts-utils

Examples

Components from browser/ and node/ directory only works in their related environment.

• Obj – Utility functions for Object manipulation.
• Seq – Enhanced Array
• Match – Type-safe pattern matching for Enum and string unions.
• GroupsBy – Groups an array of objects by multiple criteria and maps the results.
• Chunkify – Processes an array in batches with an asynchronous function, supporting concurrency control.
• Lazy – Caches function results to prevent redundant calls.
• Gzip – Compress and decompress files (Node.js only).
• FileSplitter – Splits a large file into smaller chunks (Node.js only).

Obj

TypeScript refuses to infer types from Object constructor functions. Obj implements the same method but infer keys and value to prevent annoying and unnecessary type casting.

// Get union string from keys
const input = {a: 1, b: 2}
Object.keys(input)        // → string[]
Obj.keys(input)           // → ('a' | 'b')[]
// Handle enum
enum Status { OK = 'Ok', ERROR = 'Error' }

Obj.values(Status)  // → Status[]

Additionally, it includes helper functions and can be used as a class constructor to enable method chaining.

const res = new Obj({ironman: 2, barman: 1, catwoman: 4, batman: 3})
  .sortManual(['catwoman', 'barman', 'ironman', 'batman'])  // Values are inferred
  .mapKeys(_ => _.toUpperCase())
  .mapValues(_ => _ + 1)
  .get() // → {BARMAN: 2, IRONMAN: 3, BATMAN: 4, CATWOMAN: 5}

Seq

Enhance poor JavaScript Array.

const a = seq([
  {name: 'Dave', age: 35},
  {name: 'Charlie', age: 25},
  {name: 'Alice', age: 25},
])
a.distinct(_ => _.age)                // → [{name: 'Dave', age: 35}, {name: 'Bob', age: 30}]
a.sortByString(_ => _.name, 'z-a')    // → [{name: 'Alice', age: 25}, {name: 'Charlie', age: 25}, {name: 'Dave', age: 35}]
a.sortByNumber(_ => _.age, '0-9')     // → [{name: 'Charlie', age: 25}, {name: 'Alice', age: 25}, {name: 'Dave', age: 35}]
a.count(_ => _.age > 26)              // → 1
a.groupBy(_ => _.age)                 // → {25: [{name: 'Charlie', age: 25}, {name: 'Alice', age: 25}], 35: [{name: 'Dave', age: 35}]}
a.groupByFirst(_ => _.age)            // → {25: {name: 'Charlie', age: 25}, 35: {name: 'Dave', age: 35}}
a.groupByFirstAndApply(_ => _.age, _ => _.length) // →  {25: 2, 35: 1}
a.sum(_ => _.age)                     // → 85
a.head()                              // → {name: 'Dave', age: 35}
a.last()                              // → {name: 'Alice', age: 25}
a.reduceObject(_ => [_.name, _.age])  // → {Dave: 35, Charlie: 25, Alice: 25}
a.percent(_ => _.age === 35)          // → 33.33..%  
[1, 2].difference([1, 2, 3])          // → [3]
[1, 2].intersect([1, 2, 3])           // → [1, 2]

Match

Simple and type-safe pattern matching. Fully infer Enum and strings union.

enum Enum {
  yes = 'yes',
  no = 'no',
  unknown = 'unknown',
}

const value = Enum.yes

const res: number = match(value)
  .cases({
    [Enum.yes]: 0,
  })
  .default(() => -1) // → number

match(value)
  .cases({
    // Cases can only contains keys of related type
    [Enum.yes]: () => 1,
    [Enum.no]: () => 2,
    [Enum.unknown]: () => 3,
  })
  .exhaustive() // .exhaustive can only be called if all cases are defined

GroupsBy

Groups an array of objects by multiple criteria and maps the results.

const data = [
  {id: 1, category: 'A', type: 'X', value: 10},
  {id: 2, category: 'B', type: 'Y', value: 20},
  {id: 3, category: 'A', type: 'Y', value: 15},
  {id: 4, category: 'B', type: 'X', value: 25},
  {id: 5, category: 'A', type: 'X', value: 5},
]

const result = groupsBy({
  data,
  groups: [{by: item => item.category}, {by: item => item.type}],
  finalTransform: items => items.reduce((sum, item) => sum + item.value, 0),
})

// result → {
//   A: { X: 15, Y: 15 },
//   B: { X: 25, Y: 20 },
// }

Chunkify

Processes an array of items in batches, applying an asynchronous function fn to each chunk. It supports optional concurrency control for optimized parallel execution.

await chunkify({
  size: 10,
  data: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],
  fn: async chunk => {
    await db.insert(chunk) // Simulate a database with a batch insert limit
    return chunk.map(_ => _ * 2)
  },
})
// → [[2, 4, 6, 8, 10], [12, 14, 16, 18, 20], [22, 24]]

Lazy

Caches function results based on parameters to avoid redundant calls.

const findByUserId = lazy((id: number) => {
  return users.find(_ => _.id === id)
})
findByUserId(1) // Actual call
findByUserId(1) // Retrieved from cache
findByUserId(2) // Actual call

Gzip

[!IMPORTANT]
Note: Only available on Node environment.

await gunzipFile(`${fixturePath}/zipped.gz`)
await gzipFile(`${fixturePath}/notzipped`)

FileSplitter

[!IMPORTANT]
Note: Only available on Node environment.

const splitFiles = await fileSplitter({
  filepath: 'large.txt',
  maxFileSizeMB: 5,
  outputDirPath: '.', // Optional
})