fossil

ActiveRecord-style persistence system for Node JS projects. Fossil provides a Model base class that may be used to create your own Model types. Fossil's Model class provides ActiveRecord functions such as create, save, find and where. Fossil will intelligently link Model types to their corresponding database table using an ActiveRecord-style naming convention, however you can override both default table and property name mappings if needed.

Fossil also provides an inbuilt validation system that's invoked when an attempt is made to save a Model object. Failed validation will result in an exception being thrown.

How to Install

npm install fossil --save

How to Use

Fossil uses knex to connect to your database. When initializing fossil, you'll need to provide a knex-style configuration object. Fossil query methods return promises which resolve to the desired results or reject if an error occurs.

Example

var Model = require( "fossil" )( {
    client: 'mysql',
    connection: {
        host     : '127.0.0.1',
        user     : 'your_database_user',
        password : 'your_database_password',
        database : 'myapp_test'
    }
} );

class MyModelClass extends Model
{}

MyModelClass.validations = {    id: [ "int", { minimum: 1 } ],
                                name: [ "string", function( v ) { return v[0] == "."; }, { length: { maximum: 50 } ] };

let first   = MyModelClass.first.then( ( obj ) => // do something with retrieved object );
let second  = MyModelClass.find( { id: 2 } ).then( ( obj ) => // do something with retrieved object );
let all     = MyModelClass.all.then( ( objects ) => // do something with retrieved objects );

let saved   = new MyModelClass;
saved.id    = 0;


saved.save.then( ( obj ) => {
    console.log( "We won't reach this point" );
} ).error( ( failures ) => {
    console.log( "Save will fail due to invalid id field value" );
} );

The first time any model is accessed, fossil will read the database and generate a map of tables, columns and column types. This map is used to generate model objects when queries are executed, so you don't need to defined properties on your models; these will be populated at query time