dymanic-rule-runner v1.1.0

Dynamic rule runer

Make dynamic behaviour without slowdown of js proxies

It is light and blazing fast

Writtern on typescript

About

Utility library to dinamicly define behavior on existing objects can be used in dev-environment to hook existing object, provide different behavior and etc

Sample of usage

install

yarn add dymanic-rule-runner or npm i dymanic-rule-runner

Suppose we have next structure of objects

type ID = number
type DTO = {
  id?: ID
  name: string
  age?: number
  birthDate?: string
  fullName?: string
  secondName?: string
  lastName?: string
  active?: boolean
  createdAt?: string
  updatedAt?: string
  deleted?: boolean
}

const user1: DTO = {
  name: 'John',
}
const user2: DTO = {
  name: 'Michel',
}
const user3: DTO = {
  name: 'Trevis',
}

for example this objects is typical DTO

Now we need to attach to some objects different behaviour:

autoIncrement

let id = 0
const autoIncrementId = Rule.createSetter<DTO>({
  field: 'id',
  condition: (obj) => obj != null,
  run: (obj) => {
    obj.id = id++
  },
})

fullname calculation

const fullName = Rule.createSetter<DTO>({
  field: 'fullName',
  subscribesTo: ['lastName', 'name', 'secondName'],
  condition: (obj) => !!(obj.name || obj.secondName || obj.lastName),
  run: (obj) => {
    const { name, lastName, secondName } = obj
    obj.fullName = `${name ? name : ''}${secondName ? ' ' + secondName : ''}${
      lastName ? ' ' + lastName : ''
    }`
  },
})

Method to activate user

const activate = Rule.createMethod<DTO>({
  name: 'activate user',
  condition: (obj) => !obj.active,
  run: (obj) => (obj.active = true),
})

make other stuff like

const create = Rule.createAction<DTO>({
  on: ['after:create','after:clone'],
  run: (obj) => {
    obj.createdAt = new Date().toJSON()
  },
})

const deleteDto = Rule.createAction<DTO>({
  on: 'before:delete',
  run: (obj) => {
    obj.deleted = true
  },
})

const patch = Rule.createAction<DTO>({
  on: ['after:patch', 'after:update'],
  run: (obj) => {
    obj.createdAt = new Date().toJSON()
  },
})

const clone = Rule.createAction<DTO>({
  on: 'after:clone',
  run: (obj) => {
    obj.id = id++
    obj.createdAt = new Date().toJSON()
  },
})

just write down this simple rule and use it

let runner = new RuleRunner<DTO>([
  ...autoIncrementId,
  ...fullName,
  ...activate,
  ...create,
  ...patch,
  ...deleteDto,
  ...clone,
])

then just use objects

runner.create({ name: 'Ivan', lastName: 'Gorky' })

this code is compared to other libraries

will look pretty different

mobx code

class User {
  id?: ID
  name: string = ''
  age?: number
  birthDate?: string
  secondName?: string = ''
  lastName: string = ''
  active?: boolean
  createdAt?: string
  updatedAt?: string
  deleted?: boolean
  constructor(name, lastName) {
    makeObservable(this, {
      name: observable,
      lastName: observable,
      secondName: observable,
      fullName: computed,
    })
    this.name = name
    this.lastName = lastName
    this.createdAt = new Date().toJSON()
    this.id = id++
  }

  get fullName() {
    const { name, lastName, secondName } = this
    return `${name ? name : ''}${secondName ? ' ' + secondName : ''}${
      lastName ? ' ' + lastName : ''
    }`
  }
}

benchmark results

suite
  .add('mobx', function () {
    const user = new User('Ivan', 'Gorky')
    const fullName = user.fullName
    const id = user.id
    const createdAt = user.createdAt
  })
  .add('runner', function () {
    runner.create({ name: 'Ivan', lastName: 'Gorky' })
  })
  .add('static', function () {
    let user: Partial<DTO> = { name: 'Ivan', lastName: 'Gorky' }
    user.id = id++
    const { name, lastName, secondName } = user
    user.fullName = `${name ? name : ''}${secondName ? ' ' + secondName : ''}${
      lastName ? ' ' + lastName : ''
    }`
    user.createdAt = new Date().toJSON()
  })
  .on('cycle', function (event) {
    console.log(String(event.target))
  })
  .on('complete', function () {
    console.log('Fastest is ' + this.filter('fastest').map('name'))
  })
suite.run()

The performance is on my macbook (not pro) 2016

mobx x 69,978 ops/sec ±7.05% (74 runs sampled)
runner x 359,202 ops/sec ±6.20% (78 runs sampled)
static x 591,505 ops/sec ±5.28% (82 runs sampled)