Dooter

Library for creating descriptive schemas from typescript interfaces and validating JSON object according to same type of schemas

Restrictions

!!!INTERFACES MUST BE NAMED WITH PREFIXED "I", NAME IN PASCAL CASE AND SUFFIXED "DTO" EG: "book" = "IBookDTO"!!!

!!!INTERFACE FILES MUST BE NAMED IN KEBAB CASE PREFIXED WITH ".dto". EG: "bookStore" = "book-store.dto.ts"!!!

This will most likely be updated with custom naming overide possibilities

Examples

Create schema from interface

// test/dto/author.dto.ts

// Swedish authors
export interface IAuthorDTO {
  // Authors lastname
  lastname: string;
  // Authors firstname (if exists)
  firstname?: string;
}

With typescript:

import { fileToSchema } from "@agilearchitects/dooter";
const schema = fileToSchema("./test/dto/author.dto.ts", "IAuthorDTO");
console.log(JSON.stringify(schema, null, 2))

Outputs:

{
  "name": "author",
  "description": "Swedish authors",
  "properties": {
    "lastname": {
      "type": "string",
      "description": "Authors lastname"
    },
    "firstname": {
      "type": "string",
      "description": "Authors firstname (if exists)",
      "optional": true
    }
  }
}

Comments on the line right before a schema or prop definition maps to corresponding description in schema

Or with multiple files

import { filesToSchemas } from "@agilearchitects/dooter";
const schemas = filesToSchemas("./test/dto", ["book", "author"]);
console.log(JSON.stringify(schemas, null, 2))

Outputs:

{
  "book": {
    "description": "Swedish books",
    "properties": {
      "title": "string",
      "author": {
        "type": "author[]",
        "description": "Book title"
      },
      "store": {
        "type": "string | string[]",
        "description": "Book author"
      },
      "binding": {
        "type": "'hardcover' | 'spiral'",
        "description": "Book store name(s)",
        "optional": true
      }
    }
  },
  "author": {
    "description": "Swedish authors",
    "properties": {
      "lastname": {
        "type": "string",
        "description": "Authors lastname"
      },
      "firstname": {
        "type": "string",
        "description": "Authors firstname (if exists)",
        "optional": true
      }
    }
  }
}

Parse object

The parseObject allows validation of an object as type

Simple validation of object and type

import { parseObject } from "@agilearchitects/dooter";
const obj = "hello world";
const type = "string";
parseObject(obj, type));

parseObject will either return the object itself or throw an error if not valid

Union type is also acceptable

import { parseObject } from "@agilearchitects/dooter";
const obj1 = "hello world";
const obj2 = 2;
const type = "string | number";
parseObject(obj1, type));
parseObject(obj2, type));

As well as array type

import { parseObject } from "@agilearchitects/dooter";
const obj = ["hello", "world"];
const type = "string[]";
parseObject(obj, type));

Union type IS NOT allowed. Eg. "(string, number)[]"

Allowed types are: string, number, boolean, null, undefined, true and false. String and numeric literals are also allowed for enum behavior.

import { parseObject } from "@agilearchitects/dooter";
const type = "'blue' | 'green' | 'yellow'";
parseObject("blue", type));
parseObject("green", type));
parseObject("brown", type)); // Will throw error

String literals are encased with single quotes

Parse schema

parseObject also allowes schema input for parsing nested objects mapping to schema names

import { filesToSchemas, parseObject } from "@agilearchitects/dooter";

const schemas = filesToSchemas("./test/dto", ["book", "author"]);
const author = {
    firstName: "Jon"
    lastName: "Doe",
  };
const book = [{
  title: "My Book",
  author: [author, {
    lastName: "Walker"
  }],
  store: "The bookStore",
  binding: "hardcover"
}, {
  title: "My Second Book",
  author: [author],
  store: ["The Bookstore", "Amazing"]
}]
parseObject(books, "book", schemas);
parseObject(author, "author", schemas);