A complete web framework for Ruby
Lotus combines small yet powerful frameworks:
- Lotus::Utils - Ruby core extentions and class utilities
- Lotus::Router - Rack compatible HTTP router for Ruby
- Lotus::Validations - Validation mixin for Ruby objects
- Lotus::Model - Persistence with entities, repositories and data mapper
- Lotus::View - Presentation with a separation between views and templates
- Lotus::Controller - Full featured, fast and testable actions for Rack
These components are designed to be used independently or together in a Lotus application. If you aren't familiar with them, please take time to go through their READMEs.
- Home page: http://lotusrb.org
- Mailing List: http://lotusrb.org/mailing-list
- API Doc: http://rdoc.info/gems/lotusrb
- Bugs/Issues: https://github.com/lotus/lotus/issues
- Support: http://stackoverflow.com/questions/tagged/lotus-ruby
- Chat: https://gitter.im/lotus/chat
Lotus supports Ruby (MRI) 2+
% gem install lotusrb% lotus new bookshelf
% cd bookshelf && bundle
% bundle exec lotus server # visit http://localhost:2300Lotus is a modular web framework. It scales from single file HTTP endpoints to multiple applications in the same Ruby process.
Unlike other Ruby web frameworks, Lotus has flexible conventions for code structure. Developers can arrange the layout of their projects as they prefer. There is a suggested architecture that can be easily changed with a few settings.
Lotus encourages the use of Ruby namespaces. This is based on the experience of working on dozens of projects. By using Ruby namespaces, as your code grows it can be split with less effort. In other words, Lotus is providing gentle guidance for avoiding monolithic applications.
Lotus has a smart mechanism of duplication of its frameworks. It allows multiple copies of the framework and multiple applications to run in the same Ruby process. In other words, Lotus applications are ready to be split into smaller parts, but these parts can coexist in the same heap space.
All this adaptability can be helpful to bend the framework for your advanced requirements, but we recognize the need for guidance in standard architectures. For this reason Lotus is shipped with code generators.
TL;DR: Develop your application like a gem. Implement use cases in lib/. Use one or more Lotus applications in apps/.
This is the default architecture. Use it when you are ready to begin a new project.
The core of this architecture lives in lib/, where developers should build features independently from the delivery mechanism.
Imagine you are building a personal finance application, and you have a feature called "register expense." This functionality involves Money and Expense Ruby objects and the need for persisting data into a database. You can have those classes living in lib/pocket/money.rb and lib/pocket/expense.rb and use Lotus::Model to persist them.
It's based on a few simple concepts: use cases and applications.
Use cases (features) should be implemented in lib/ with a combination of pure objects and the needed Ruby gems.
One or more Lotus applications live in apps/. They are isolated from each other, and depend only on the code in lib/.
Each application should serve only one purpose: user-facing web application, administrative backend, JSON API, metrics dashboard, etc.
This architecture has important advantages:
- Code reusability. You can consume a feature from the Web UI or from a HTTP API. Each can be a different Lotus application or simple Rack-based endpoints.
- Decoupled components. The core of your application depends only on a few gems and it doesn't need to worry about the Web/HTTP/Console/Background jobs.
- Applications are built like gems. This eases the process of packaging and sharing them among projects, without the need for many dependencies.
- Avoid monoliths. Each Lotus application under
apps/is a candidate for later extraction into a separated microservice.
The last point is crucial. In the early days of a new project it is convenient to build and deploy all the code together. However, as time passes, it can become nearly impossible to extract sets of cohesive functionalities into separated deliverables. Lotus helps to plan those things ahead of time, but without the burden that is required by those choices, because it supports multiple applications natively.
Here's the name container: a Lotus "shell" that can run multiple micro applications in the same process.
% lotus new pocket --arch=container
% lotus new pocket # --arch=container is the defaultRead more about this architecture.
upcoming
upcoming
- Lotus expects controllers, actions and views to have a specific pattern (see Configuration for customizations)
- All the commands must be run from the root of the project. If this requirement cannot be satisfied, please hardcode the path with
Configuration#root. - The template name must reflect the name of the corresponding view:
Bookshelf::Views::Dashboard::Indexfordashboard/index.html.erb. - All static files are served by the internal Rack middleware stack.
- The application expects to find static files under
public/(seeConfiguration#assets) - If the public folder doesn't exist, it doesn't serve static files.
- The application structure can be organized according to developer needs.
- No file-to-name convention: modules and classes can live in one or multiple files.
- No autoloading paths. They must be explicitly configured.
A Lotus application can be configured with a DSL that determines its behavior.
require 'lotus'
module Bookshelf
class Application < Lotus::Application
configure do
########################
# BASIC CONFIGURATIONS #
########################
# Determines the root of the application (optional)
# Argument: String, Pathname, defaults to Dir.pwd
#
root 'path/to/root' # or __root__
# The relative load paths where the application will recursively load the code (mandatory)
# Argument: String, Array<String>, defaults to empty set
#
load_paths << [
'controllers',
'views'
]
# Handle exceptions with HTTP statuses (true) or don't catch them (false)
# Argument: boolean, defaults to true
#
handle_exceptions true
#######################
# HTTP CONFIGURATIONS #
#######################
# The route set (mandatory)
# Argument: Proc with the routes definition
#
routes do
get '/', to: 'home#index'
end
# The route set (mandatory) (alternative usage)
# Argument: A relative path where to find the routes definition
#
routes 'config/routes'
# URI scheme used by the routing system to generate absolute URLs (optional)
# Argument: A string, default to "http"
#
scheme 'https'
# URI host used by the routing system to generate absolute URLs (optional)
# Argument: A string, default to "localhost"
#
host 'bookshelf.org'
# URI port used by the routing system to generate absolute URLs (optional)
# Argument: An object coercible to integer, default to 80 if the scheme is http and 443 if it's https
# This SHOULD be configured only in case the application listens to that non standard ports
#
port 2323
# Toggle cookies (optional)
# Argument: A [`TrueClass`, `FalseClass`], default to `FalseClass`.
#
cookies true
# Toggle sessions (optional)
# Argument: Symbol the Rack session adapter
# A Hash with options
#
sessions :cookie, secret: ENV['SESSIONS_SECRET']
# Default format for the requests that don't specify an HTTP_ACCEPT header (optional)
# Argument: A symbol representation of a mime type, default to :html
#
default_format :json
# Rack middleware configuration (optional)
#
middleware.use Rack::Protection
# HTTP Body parsers (optional)
# Parse non GET responses body for a specific mime type
# Argument: Symbol, which represent the format of the mime type (only `:json` is supported)
# Object, the parser
#
body_parsers :json, MyXMLParser.new
###########################
# DATABASE CONFIGURATIONS #
###########################
# Configure a database adapter (optional)
# Argument: A Hash with the settings
# type: Symbol, :file_system, :memory and :sql
# uri: String, 'file:///db/bookshelf'
# 'memory://localhost/bookshelf'
# 'sqlite:memory:'
# 'sqlite://db/bookshelf.sqlite3'
# 'postgres://localhost/bookshelf'
# 'mysql://localhost/bookshelf'
#
adapter type: :file_system, uri: ENV['DATABASE_URL']
# Configure a database mapping (optional)
# Argument: Proc
#
mapping do
collection :users do
entity User
repository UserRepository
attribute :id, Integer
attribute :name, String
end
end
# Configure a database mapping (optional, alternative usage)
# Argument: A relative path where to find the mapping definitions
#
mapping 'config/mapping'
############################
# TEMPLATES CONFIGURATIONS #
############################
# The layout to be used by all the views (optional)
# Argument: A Symbol that indicates the name, default to nil
#
layout :application # Will look for Bookshelf::Views::ApplicationLayout
# The relative path where to find the templates (optional)
# Argument: A string with the relative path, default to the root of the app
#
templates 'templates'
#########################
# ASSETS CONFIGURATIONS #
#########################
# Specify sources for assets (optional)
# Argument: String, Array<String>, defaults to 'public'
#
assets << [
'public',
'vendor/assets'
]
# Enabling serving assets (optional)
# Argument: boolean, defaults to false
#
serve_assets true
#############################
# FRAMEWORKS CONFIGURATIONS #
#############################
# Low level configuration for Lotus::View (optional)
# The given block will be yielded every time `Lotus::View` is included.
# This is helpful to share logic between views
# See the related documentation
# Argument: Proc
#
view.prepare do
include MyCustomRoutingHelpers # included by all the views
end
# Low level configuration for Lotus::Controller (optional)
# Argument: Proc
controller.prepare do
include Authentication # included by all the actions
before :authenticate! # run auth logic before each action
end
end
###############################
# ENVIRONMENTS CONFIGURATIONS #
###############################
configure :development do
# override the general configuration only for the development environment
handle_exceptions false
serve_assets true
end
configure :test do
# override the general configuration only for the test environment
host 'test.host'
end
end
endLotus provides a few command line utilities:
It looks at the config.ru file in the root of the application, and starts the Rack server defined in your Gemfile (eg. puma, thin, unicorn). It defaults to WEBRick.
It supports code reloading feature by default, useful for development purposes.
% bundle exec lotus serverIt starts a REPL, by using the engine defined in your Gemfile. It defaults to IRB. Run it from the root of the application.
% bundle exec lotus consoleIt supports code reloading via the reload! command.
It prints the routes defined by the Lotus application(s).
% bundle exec lotus routesIt prints the current Lotus version.
% bundle exec lotus version- Fork it ( https://github.com/lotus/lotus/fork )
- Create your feature branch (
git checkout -b my-new-feature) - Commit your changes (
git commit -am 'Add some feature') - Push to the branch (
git push origin my-new-feature) - Create a new Pull Request
Lotus uses Semantic Versioning 2.0.0
Copyright © 2014-2015 Luca Guidi – Released under MIT License