JinnDb is a pure javascript persistent key-value store for Node.js that supports out-of-core data access.
About
What can Jinn do for me?
Jinn allows you to store and retrieve JSON data using NoSQL style queries from an embedded database.
Jinn is designed to be a lightweight database for local node applications that need a data storage solution. If you are a server or cloud-based application with heavy usage, a dedicated server like PostgresSQL or MongoDB might be a better choice.
Is this another in-memory database?
No. I got frustrated with the abundance of in-memory only databases for node, so I wrote this for another project I was working on. In-memory databases can only handle as much data as you can fit in memory which is very limiting for larger datasets.
How fast is Jinn?
Pretty fast. Jinn caches as many entries as it can in memory, so for smaller datasets, it will be as fast as any in-memory database. The difference is that it won't choke when the dataset is larger than that, and it won't starve the rest of your application from memory as the database grows.
Usage
Get Jinn
npm install jinn-db --save
Open/Load a Database
var Database = require('jinn-db');
var db = new Database('wishes.db'); // If this file doesn't exist, it will be created
db = new Database(); // If no fileName is provided, the database is created as a temporary file that will be deleted when the process exits.
db.load()
.then(function(db) {
// Done loading!
});For the examples below, our database is composed of the following entries:
[{"_id":"0", "person": "ali", "wishNum": 0, "wishedFor": "gold"},
{"_id":"1", "person": "ali", "wishNum": 1, "wishedFor": "gold"},
{"_id":"2", "person": "ali", "wishNum": 2, "wishedFor": "more gold"}]Find entries
// Find all of Ali's wishes
db.find({person: 'ali'})
.then(function(results) {
/*
* results -> [{_id: '0', person: 'ali', wishNum: 0, wishedFor: 'gold'},
* {_id: '1', person: 'ali', wishNum: 1, wishedFor: 'gold'},
* {_id: '2', person: 'ali', wishNUm: 2, wishedFor: 'more gold'}]
*/
});Credit to @louischatriot for NeDB, and by proxy MongoDB, off of which this API is largely based.
Benchmarks
TODO: