| Package | Stoic |
| Description | Schemas for Redis |
| Node Version | >= 0.6 |
Duck Types? No way! Goose Types are for structured data.
Ideally we want to define a schema as follows, and have the models/data accessors generated for us:
# NOTE: not supported yet, you must define an interface
module.exports =
'user:!{id}':
name: 'Cache'
email: 'Cache'This would allow you to write your domain logic like so:
should = require 'should'
stoic = require 'stoic'
schema = require './schema'
{User} = stoic.models
u = User.get(1)
u.name.set 'Frank', (err, result) ->
u.name.get (err, result) ->
result.should.eql 'Frank'We can support that syntax today, but you have to define an interface to go with your schema. Here is a full document description with schema and interface that would allow the above program.coffee to run:
face = (decorators) ->
{user: {name, email}} = decorators
get: (id) ->
user =
id: id
name user
email user
return user
schema =
'user:!{id}':
name: 'Cache'
email: 'Cache'
module.exports = ['User', face, schema]Well, looking at the schema above, there are a few things to notice. The schema is a heiarchical namespace. Each key extends the namespace within redis, and each value describes the type of data held there. When Stoic parses your schema, it will construct a set of decorators. Decorators:
- Are specific to a data type
- Are used to attach Accessors to an object
The accessors they attach:
- Are aware of their key space
- Fill in tokens in the key with properties from the parent object
These decorators are the building blocks needed to assemble an interface. Let's take a deeper look into that interface code.
# we are passed the decorators generated from parsing the schema
face = (decorators) ->
# let's pull them out of the object into local variables
{user: {name, email}} = decorators
# the get method will create an object that knows its id and can access user-stuff
# note that it's not actually querying redis, that won't happen until an accessor is called
get: (id) ->
# the base object
# the accessors will use this id to fill in !{id} in their key space whenever they are called
user =
id: id
# decorate the user object with 'name' and 'email' accessors
name user
email user
# return the completed object
return userThis should make it very easy to construct the interface that you want to access your data. Soon, we will support a sensible default interface, and then you can kiss your boilerplate goodbye, and say hello to sweet declarative goodness.
- Cache
- List
- Set
Other data types will be trivial to implement, it's just a few lines of boilerplate. In the future this will be extensible as well, so you could add custom decorators for your application.
Redback was not usable for this purpose due to the way they are handling namespace definitions and model creation. The approach we are taking should lead to more power and flexibility in your data model.
Hit me up on twitter: @qbitmage
(MIT License)
Copyright (c) 2012 Torchlight Software info@TorchlightSoftware.com
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.