A Knex.js service adapter for FeathersJS
npm install feathers-knex --saveYou can create a *SQL Knex service like this:
var knex = require('feathers-knex');
var todos = knex('todos', {
dialect: 'sqlite3',
connection: {
filename: './data.db'
}
});
app.use('/todos', todos);This will create a todos endpoint and connect to a local todos table on an SQLite database in data.db.
Here's a complete example of a Feathers server with a todos SQLite service. We are using the Knex schema builder
// server.js
var feathers = require('feathers');
var bodyParser = require('body-parser');
var knex = require('feathers-knex');
var todos = knex('todos', {
dialect: 'sqlite3',
connection: {
filename: './data.db'
}
});
// This drops and creates table every time
todos.knex.schema.dropTableIfExists('todos').then(() => {
people.knex.schema.createTable('todos', function(table) {
table.increments('id');
table.string('text');
table.boolean('complete');
});
});
// Create a feathers instance.
var app = feathers()
// Setup the public folder.
.use(feathers.static(__dirname + '/public'))
// Enable Socket.io
.configure(feathers.socketio())
// Enable REST services
.configure(feathers.rest())
// Turn on JSON parser for REST services
.use(bodyParser.json())
// Turn on URL-encoded parser for REST services
.use(bodyParser.urlencoded({extended: true}))
// Connect to the db, create and register a Feathers service.
app.use('/todos', todos);
// Start the server.
var port = 8080;
app.listen(port, function() {
console.log('Feathers server listening on port ' + port);
});You can run this example by using node server and going to localhost:8080/todos. You should see an empty array. That's because you don't have any Todos yet but you now have full CRUD for your new todos service!
// Todo
To extend the basic service there are two options. Either through using Uberproto's inheritance mechanism or by using feathers-hooks.
The basic Knex service is implemented using Uberproto, a small EcmaScript 5 inheritance library so you can use the Uberproto syntax to add your custom functionality.
For example, you might want update and patch to behave the same (the basic implementation of update replaces the entire object instead of merging it) and add an updatedAt and createdAt flag to your data:
// myservice.js
var knex = require('feathers-knex');
var Proto = require('uberproto');
var TimestampPatchService = knex.Service.extend({
create: function(data, params, callback) {
data.createdAt = new Date();
// Call the original `create`
return this._super(data, params, callback);
},
update: function() {
// Call `patch` instead so that PUT calls merge
// instead of replace data, too
this.patch(id, data, params, callback);
},
patch: function(id, data, params, callback) {
data.updatedAt = new Date();
// Call the original `patch`
this._super(id, data, params, callback);
}
});
// Export a simple function that instantiates this new service like
// var myservice = require('myservice');
// app.use('/users', myservice(options));
module.exports = function(options) {
// We need to call `Proto.create` explicitly here since we are overriding
// the original `create` method
return Proto.create.call(TimestampPatchService, options);
}
module.exports.Service = TimestampPatchService;Another option is to weave functionality into your existing services using feathers-hooks, for example the above createdAt and updatedAt functionality:
var feathers = require('feathers');
var hooks = require('feathers-hooks');
var knex = require('feathers-knex');
// Initialize a MongoDB service with the users collection on a local MongoDB instance
var app = feathers()
.configure(hooks())
.use('/users', knex('users', {
dialect: 'sqlite3',
connection: {
filename: './data.db'
}
}));
app.lookup('users').before({
create: function(hook, next) {
hook.data.createdAt = new Date();
next();
},
update: function(hook, next) {
hook.data.updatedAt = new Date();
next();
}
});
app.listen(8080);The find API allows the use of $limit, $skip, $sort, and $select in the query. These special parameters can be passed directly inside the query object:
// Find all recipes that include salt, limit to 10, only include name field.
{"ingredients":"salt", "$limit":10, "$select": { "name" :1 } } // JSON
GET /?ingredients=salt&$limit=10&$select[name]=1 // HTTPAs a result of allowing these to be put directly into the query string, you won't want to use $limit, $skip, $sort, or $select as the name of fields in your document schema.
$limit will return only the number of results you specify:
// Retrieves the first two records found where age is 37.
query: {
age: 37,
$limit: 2
}
$skip will skip the specified number of results:
// Retrieves all except the first two records found where age is 37.
query: {
age: 37,
$skip: 2
}
$sort will sort based on the object you provide:
// Retrieves all where age is 37, sorted ascending alphabetically by name.
query: {
age: 37,
$sort: { name: 1 }
}
// Retrieves all where age is 37, sorted descending alphabetically by name.
query: {
age: 37,
$sort: { name: -1}
}
$select support in a query allows you to pick which fields to include or exclude in the results.
// Only retrieve name.
query: {
name: 'Alice',
$select: {'name': 1}
}
// Retrieve everything except age.
query: {
name: 'Alice',
$select: {'age': 0}
}
In addition to sorting and pagination, properties can also be filtered by criteria. Standard criteria can just be added to the query. For example, the following find all users with the name Alice:
query: {
name: 'Alice'
}Additionally, the following advanced criteria are supported for each property.
Find all records where the property does ($in) or does not ($nin) contain the given values. For example, the following query finds every user with the name of Alice or Bob:
query: {
name: {
$in: ['Alice', 'Bob']
}
}Find all records where the value is less ($lt) or less and equal ($lte) to a given value. The following query retrieves all users 25 or younger:
query: {
age: {
$lte: 25
}
}Find all records where the value is more ($gt) or more and equal ($gte) to a given value. The following query retrieves all users older than 25:
query: {
age: {
$gt: 25
}
}Find all records that do not contain the given property value, for example anybody not age 25:
query: {
age: {
$ne: 25
}
}Find all records that match any of the given objects. For example, find all users name Bob or Alice:
query: {
$or: [
{ name: 'Alice' },
{ name: 'Bob' }
]
}1.1.0
- Compatibility with latest service tests
1.0.0
- Initial release
Copyright (c) 2015
Licensed under the MIT license.