opennodedb

About

OpenNodeDB is an open source database implementation made by me (Matthew Snow). It's just a simple solution that uses an ORM approach to storing JSON data both in memory and in files (file system still in the works). Although the index system created doesn't use B or B+ trees, it uses a hash table for indexes (only for the id field), and each collection of record sets that the index points to has an incrementing id, we use a binary search method to quickly locate a record when searched by id.

Description

This database implementation will be best used for small/medium sized intranets, or apps, where performance is a must. More to come as this project starts to take shape.

How to Use

From a terminal, inside main project

npm install opennodedb

From inside a node.js project

// create new database instance
var OpenNodeDB = require('opennodedb');
var db = new OpenNodeDB();


// create a model w/ validators, getters and setters
db.createModel('Users', 1, { name: String, password: String },
	{
		validators: {
			name: function (name) {
				if (name.length < 5) throw new Error('Name of length needs to be at least 5 characters!')
			}
		},

		getters: {
			name: function (name) {
				return name + 'dbcombine!'
			}
		},

		setters: {
			name: function (name) {
				return name + ' I got appended to the name haha!.'
			}
		}
	})



// inserting an object: db.insert(modelname, object)
db.insert('Users', { name: 'Apple Pear', password: 'doublefruit123'})
.then(user => {
	// here you can see that apple pear will be assigned the id of 0
	// each subsequent id will be incremented
})
.catch(err => {
	//....
})


// finding a user or users: db.read(modelname, queryParamsObj)
db.read('Users', { name: 'Apple Pear'})
.then(foundData => {
	// foundData will be an array with our results
	// if an id was searched, there will only be 1 result at foundData[0]
})

Test Results

Test Type	Times
Inserting 100,000 items	198.723 ms
Reading by id (after 100000)	0.075 ms
Reading by name ""	3.9 ms

Goals

• Implement Getters functionality
• Create a file persistence mechanism using child processes.
• Creating multiple driver mechanisms (using a tcp socket) with auth.
• Create update functionality
• Create delete functionality