🗄️ Datacasting

Make schemas and cast data declaratively

Installing

npm install --save datacasting
# or
yarn add datacasting

Basic usage

Working with an array of objects

import { scheme, toDate } from 'datacasting'

const userScheme = scheme({
  verified: Boolean,
  created_at: toDate('yyyy-MM-dd'),
})

const casted = userScheme.cast([
  {
    first_name: 'Jonh',
    last_name: 'Doe',
    verified: 0,
    created_at: '2020-01-23',
  }
])

console.log(casted)
// [
//   {
//     "first_name": "Jonh",
//     "last_name": "Doe",
//     "verified": false,
//     "created_at": Date(2020, 0, 23)
//   }
// ]

const rewrited = userScheme.rewrite([
  {
    first_name: 'Jonh',
    last_name: 'Doe',
    verified: 0,
    created_at: '2020-01-23',
  }
])

console.log(rewrited)
// [
//   {
//     "verified": false,
//     "created_at": Date(2020, 0, 23)
//   }
// ]

Working with a single object

import { scheme, toDate } from 'datacasting'

const userScheme = scheme({
  verified: Boolean,
  created_at: toDate('yyyy-MM-dd'),
})

const casted = userScheme.cast({
  first_name: 'Jonh',
  last_name: 'Doe',
  verified: 0,
  created_at: '2020-01-23',
})

console.log(casted)
// {
//  "first_name": "Jonh",
//  "last_name": "Doe",
//  "verified": false,
//  "created_at": Date(2020, 0, 23)
// }

const rewrited = userScheme.rewrite({
  first_name: 'Jonh',
  last_name: 'Doe',
  verified: 0,
  created_at: '2020-01-23',
})

console.log(rewrited)
// {
//  "verified": false,
//  "created_at": Date(2020, 0, 23)
// }

API

Scheme methods

• cast – converts data according to the schema, inheriting the passed fields, even if they were not declared in the schema.
• rewrite - converts data according to the schema without inheriting the original fields. The result will be an object only from the fields declared in the schema.

Available caster functions

• First, you can use the built-in type conversion: String, Number and Boolean

  import { scheme } from 'datacasting'
  
  const dataScheme = scheme({
    user_id: String,
    verified: Boolean,
    role: Number,
  })

• replace(searchValue: string | RegExp, replaceValue: string | RegExp): (value: string) => string

  import { scheme, replace } from 'datacasting'
  
  const dataScheme = scheme({
    name: replace('admin', 'nobody'),
  })
  
  const casted = dataScheme.cast({
    name: "John Doe - admin",
  })
  
  console.log(casted)
  // {
  //   "name": "John Doe - nobody"
  // }

• toArrayOf<K = StringConstructor | NumberConstructor | BooleanConstructor>(constructor: K): (value: any) => K[]

  import { scheme, toArrayOf } from 'datacasting'
  
  const dataScheme = scheme({
    skill: toArrayOf(String),
  })
  
  console.log(dataScheme.cast({
    skill: "javascript",
  }))
  // {
  //   "skill": ["javascript"]
  // }
  
  console.log(dataScheme.cast({
    skill: ["javascript", "vue", "react"],
  }))
  // {
  //   "skill": ["javascript", "vue", "react"]
  // }

• toDate(fromFormat: string): (value: string) => Date

  Argument fromFormat refers to date-fns tokens. See: date-fns parse tokens.

  import { scheme, toDate } from 'datacasting'
  
  const dataScheme = scheme({
    created_at: toDate('yyyy-MM-dd'),
  })
  
  console.log(dataScheme.cast({
    created_at: "2020-01-15",
  }))
  // {
  //   "created_at": Date(2020, 0, 15)
  // }

• toInteger(value: any): number

  import { scheme, toInteger } from 'datacasting'
  
  const dataScheme = scheme({
    balance: toInteger,
  })
  
  console.log(dataScheme.cast({
    balance: '150.33',
  }))
  // {
  //   "balance": 150
  // }

• toLowerCase(value: any): string

  import { scheme, toLowerCase } from 'datacasting'
  
  const dataScheme = scheme({
    name: toLowerCase,
  })
  
  console.log(dataScheme.cast({
    name: 'John Doe',
  }))
  // {
  //   "name": "john doe"
  // }

• toUpperCase(value: any): string

  import { scheme, toUpperCase } from 'datacasting'
  
  const dataScheme = scheme({
    name: toUpperCase,
  })
  
  console.log(dataScheme.cast({
    name: 'John Doe',
  }))
  // {
  //   "name": "JOHN DOE"
  // }

Write your own caster function

Arguments:

• {any} value
• {string} key
• {Object} data

Returns:

• {any} result

import { scheme } from 'datacasting'

const dataScheme = scheme({
  role_id: (value, key, data) => {
    return data.role.id
  })
})

const casted = dataScheme.cast({
  name: 'Jonh Doe',
  role: {
    id: 1,
    name: 'user',
  }
})

console.log(casted)
// {
//  "name": "John Doe",
//  "role_id": 1,
//  "role": {
//    "id": 1,
//    "name": "user"
//  },
// }

License

MIT © Daniel Slepov