data-model

Javascript data schemas. Validate data structures, ensure data types, assign virtual methods, generate virtual properties for a data set.

Installing

Via Bower

bower install data-model --save

Via NPM

npm install data-model --save

If you are using Node.js

var datamodel = require('data-model');

Creating a Dataset Model

The dataset schema is to define what the expected format of a dataset is to be. There are three patterns that are used: Collections, Schemas (and nested schemas), and primitives. Further, there are two ways to define a schema: Caching, or instance.

Primitive Pattern

Primitives are the most basic element, and end recursive checks. You use primitives to define the type pattern required for proper use.

Example:A string

datamodel('personName')
.setTemplate(String);

Example:An array of strings

datamodel('personName')
.setTemplate([String]);

Example:An array objects with a string property

datamodel('personName')
.setTemplate([{ name:String }]);

Schema Pattern

A schema definition has a similar meaning to what what is understood as the Object javascript primitive. It is a definition of a key/value data structure, and uses the Object primitive syntax for defining the pattern. Nested schemas are supported for complex objects.

Example: A basic schema

datamodel('person')
.setTemplate({
	name:String,
	age:Number
});

Example: A nested Schema

datamodel('person')
.setTemplate({
	name:String,
	details:{
		age:Number
	}
});

Collection Pattern

A collection defines repetition in a data set. The collection is denoted with the array primitive syntax.

Example: Describing an array of simple objects

datamodel('pageHits')
.setTemplate([{
	name:String,
	hits:[Number]
}]);

Example: Describe a simple object with an array

datamodel('myHits')
.setTemplate({
	name:String,
	hits:[Number]
});

Validating Against Models

Examples

Example 1:

datamodel('personName')
.setTemplate(String);

var ValidData = "Jordan";
var InvalidData = 23; //wrong type...

datamodel('personName')
.validate(ValidData); //true

datamodel('personName')
.validate(InvalidData); //false

Example 2:

datamodel('personName')
.setTemplate([String]);

var ValidData = ["Jordan", "Chapian"];
var InvalidData = [23, "Jordan"]; //wrong type

datamodel('personName')
.validate(ValidData); //true

datamodel('personName')
.validate(InvalidData); //false

Example 3:

datamodel('personName')
.setTemplate([{name:String}]);

var ValidData = [{name:"Jordan"}, {name:"Chapian"}];
var InvalidData = [{name:"Jordan"}, "Jordan"]; //wrong type

datamodel('personName')
.validate(ValidData); //true

datamodel('personName')
.validate(InvalidData); //false

Example 4:

datamodel('person')
.setTemplate({
	name:String,
	age:Number
});

var ValidData = {name:'Jordan', age:10};
var InvalidData = {name:'Jordan'}; //missing age..

datamodel('person')
.validate(ValidData); //true

datamodel('person')
.validate(InvalidData); //false

Example 5:

datamodel('person')
.setTemplate({
	name:String,
	details:{
		age:Number
	}
});

var ValidData = {name:'Jordan', details:{age:10}};
var InvalidData = {name:'Jordan', details:{age:"10"}}; //age, wrong type..

datamodel('person')
.validate(ValidData); //true

datamodel('person')
.validate(InvalidData); //false

Example 6:

datamodel('pageHits')
.setTemplate([{
	name:String,
	hits:[Number]
}]);

var ValidDataset = [{name:"Jordan", hits:[5,7,6]}, {name:"Chapian", hits:[]}];
var InvalidDataset = [{name:"Jordan", hits:["5", 7, 6]}, {name:"Chapian", hits:[8]}];

datamodel('pageHits')
.validate(ValidDataset); //true

datamodel('pageHits')
.validate(InvalidDataset); //false

Example 7:

datamodel('myHits')
.setTemplate({
	name:String,
	hits:[Number]
});

var ValidDataset = {name:"Jordan", hits:[5, 10, 15, 20]};
var InvalidDataset =  {name:"Jordan", hits:12};

datamodel('myHits')
.validate(ValidDataset); //true

datamodel('myHits')
.validate(InvalidDataset); //false

Internal Representation

A graph is used to represent a data structure, where each node is represented by one of three subclasses of a TemplateNode. Recursive patterns are used to traverse the structures. Consider the below model specification.

datamodel('users')
.setTemplate([{
    name:String,
    email:String,
    posts:[{
    	title:String,
    	views:Number,
    	body:Number,
    	posted:Date
    }],
    loginPosts:[[Number]]
}]);

This is represented by the following graph structure: