Rails3教程系列之六：从Rails2到Rails3

趁着上班前写一段 @@

那么从前面的教程中我们学习了如何创建一个简单的博客应用, 我个人觉得无论你是新手还是从rails2过来, rails3还是比较容易上手的, 现在我们就来看下rails3相比rails2, 进步在哪里, 优势又在什么地方. (本来这章打算写ujs的, 无奈工作繁忙只能推到周日了)

1. 脚本命令

旧的命令                                      新的用法

script/generate                          rails g

script/console                            rails c

script/server                              rails s

script/dbconsole                        rails db

2. 配置文件

rails2: config/environment.rb

Rails::Initializer.run do |config|
    config.load_paths += %W( #{RAILS_ROOT}/extras )
    config.gem "bj"
    config.gem "sqlite3-ruby",  :lib => "sqlite3"
    config.gem "aws-s3", :lib => "aws/s3"
    config.plugins = [ :exception_notification ]
    config.time_zone = 'UTC'
end

Rails::Initializer.run do |config|
    config.load_paths += %W( #{RAILS_ROOT}/extras )
    config.gem "bj"
    config.gem "sqlite3-ruby", :lib => "sqlite3"
    config.gem "aws-s3", :lib => "aws/s3"
    config.plugins = [ :exception_notification ]
    config.time_zone = 'UTC'
end

rails3:config/application.rb

module APP_NAME
    class Application < Rails::Application
        config.load_paths += %W( #{RAILS_ROOT}/extras )
        config.plugins = [ :exception_notification ]
        config.time_zone = 'UTC'
    end
end

module APP_NAME
    class Application < Rails::Application
        config.load_paths += %W( #{RAILS_ROOT}/extras )
        config.plugins = [ :exception_notification ]
        config.time_zone = 'UTC'
    end
end

这样就变成了一种架构式的应用, 我们可以根据方便的对config进行操作

3. 路由

在rails3中, 已经的路由可以继续工作, 而新的路由方式更加简洁.

在 rails2 中:

map.resources :posts  do |post|
    post.resources :comments
end

map.resources :posts do |post|
    post.resources :comments
end

而在rails3中, 表达更为形象:

resources :posts  do
    resources :comments
end

resources :posts do
    resources :comments
end

对于一些复杂的路由, rails2:

post.resources :comments,
                 :member => {  :preview => :post },
                 :collection => {  :archived => :get }

post.resources :comments,
                 :member => { :preview => :post },
                 :collection => { :archived => :get }

在rails3中可以这样表达:

resources :comments  do
    member do
        post :preview
    end
    collection do
        get :archived
    end
end

resources :comments do
    member do
        post :preview
    end
    collection do
        get :archived
    end
end

不够简洁? 我们还可以这样做:

resources :comments  do
    post :preview, :on => :member
    get :archived, :on => :collection
end

resources :comments do
    post :preview, :on => :member
    get :archived, :on => :collection
end

对于基本路由, rails2:

map.connect 'login',  :controller => 'session', :action => 'new'

map.connect 'login', :controller => 'session', :action => 'new'

那么在rails3中:

match 'login' =>  'session#new'

match 'login' => 'session#new'

对于具名路由, rails2:

map.login 'login',  :controller => 'session', :action => 'new'

map.login 'login', :controller => 'session', :action => 'new'

在rails3中:

match 'login' =>  'session#new', :as => :login

match 'login' => 'session#new', :as => :login

对于程序根路由, rails2:

map.root :controller => 'users', :action => 'index'

map.root :controller => 'users', :action => 'index'

rails3:

root :to => 'users#index'

root :to => 'users#index'

对于遗留路由, rails2:

map.connect ':controller/:action/:id'
map.connect ':controller/:action/:id.:format'

map.connect ':controller/:action/:id'
map.connect ':controller/:action/:id.:format'

那么在rails3中写法更优雅:

match ':controller(/:action(/:id(.:format)))'

match ':controller(/:action(/:id(.:format)))'

对于路由参数, rals2:

map.connect '/articles/:year/:month/:day', :controller => 'posts', :action => 'index'

map.connect '/articles/:year/:month/:day', :controller => 'posts', :action => 'index'

rails3:

match '/articles/:year/:month/:day' => "posts#index"

match '/articles/:year/:month/:day' => "posts#index"

那么对于存档请求, 比如rails2:

map.connect '/articles/:year/:month/:day', :controller => 'posts', :action => 'index'
map.connect '/articles/:year/:month', :controller => 'posts', :action => 'index'
map.connect '/articles/:year',  :controller => 'posts', :action => 'index'

map.connect '/articles/:year/:month/:day', :controller => 'posts', :action => 'index'
map.connect '/articles/:year/:month', :controller => 'posts', :action => 'index'
map.connect '/articles/:year', :controller => 'posts', :action => 'index'

在rails3中:

match '/articles(/:year(/:month(/:day)))' => "posts#index"

match '/articles(/:year(/:month(/:day)))' => "posts#index"

指定请求方式, rails2:

map.connect '/articles/:year', :controller => 'posts', :action => 'index',
:conditions => {:method => :get}

map.connect '/articles/:year', :controller => 'posts', :action => 'index',
:conditions => {:method => :get}

在rails3中:

match '/articles/:year' => "posts#index", :via => :get
#或者更简单的:
get '/articles/:year' => "posts#index"

match '/articles/:year' => "posts#index", :via => :get
#或者更简单的:
get '/articles/:year' => "posts#index"

对于跳转, rails3:

match 'signin', :to => redirect("/login")
match 'users/:name', :to => redirect {|params| "/#{params[:name]}" }
match 'google' => redirect('http://www.google.com/')

match 'signin', :to => redirect("/login")
match 'users/:name', :to => redirect {|params| "/#{params[:name]}" }
match 'google' => redirect('http://www.google.com/')

路由约束: rails2中实际上使用了 :requirements 符号

map.connect '/:year',  :controller => 'posts', :action => 'index',
                                    :requirements => { :year => /\d{4}/ }

map.connect '/:year', :controller => 'posts', :action => 'index',
                                    :requirements => { :year => /\d{4}/ }

在rails3中:

match '/:year' =>  "posts#index", :constraints => {:year => /\d{4}/}

match '/:year' => "posts#index", :constraints => {:year => /\d{4}/}

:constraints => { :user_agent => /iphone/ }
:constraints => { :ip => /192\.168\.1\.\d{1,3}/ }
constraints(:host => /localhost/)  do
    resources :posts
end
constraints IpRestrictor do
    get 'admin/accounts' => "queenbee#accounts"
end

:constraints => { :user_agent => /iphone/ }
:constraints => { :ip => /192\.168\.1\.\d{1,3}/ }
constraints(:host => /localhost/) do
    resources :posts
end
constraints IpRestrictor do
    get 'admin/accounts' => "queenbee#accounts"
end

对于Rack应用, rails3:

get 'hello' => proc { |env| [200, {}, "Hello Rack"] }

get 'rack_endpoint' => PostsController.action(:index)

get 'rack_app' => CustomRackApp

get 'hello' => proc { |env| [200, {}, "Hello Rack"] }

get 'rack_endpoint' => PostsController.action(:index)

get 'rack_app' => CustomRackApp

4. Bundler与ActionController

一个典型的rails应用, 我们一般需要在 environment.rb 指定你的 gems:

config.gem "haml"
config.gem "chronic", :version => '0.2.3'

config.gem "haml"
config.gem "chronic", :version => '0.2.3'

然后我们运行 $ rake gems:install, 该命令会取得并下载然后安装编译这些gems到你的系统RubyGems目录中.

之后我们运行 $ rake gems:unpack:dependencise, 把这些gem打包到你应用程序的vendor/gems目录中去.

这样做产生的问题:

1. 它直接绑定到Rails中

2. 没有从本质上解决依赖问题

3. 运行时容易发生冲突

在rails3中, 使用了 bundle 命令:

直接在你的 gemfile 中指定你的 gem

gem "haml"
gem "chronic", '0.2.3'

gem "haml"
gem "chronic", '0.2.3'

然后运行 $ bundle, 该命令会会取得并下载然后安装编译这些gems

然后运行 $ bundle package 把gem源移到/vendor/cache中去.

这样rails应用中的gem与系统中的gem就不会相冲突.

一般的控制器语法:

class UsersController < ApplicationController
    def index
        @users = User.all
        respond_to do |format|
            format.html
            format.xml { render :xml => @users.to_xml }
        end
    end

    def show
        @user = User.find(params[:id])
        respond_to do |format|
            format.html # show.html.erb
            format.xml { render :xml => @user }
        end
    end

...

class UsersController < ApplicationController
    def index
        @users = User.all
        respond_to do |format|
            format.html
            format.xml { render :xml => @users.to_xml } 
        end
    end

    def show
        @user = User.find(params[:id])
        respond_to do |format|
            format.html # show.html.erb
            format.xml { render :xml => @user }
        end
    end

...

改进的语法:

class UsersController < ApplicationController
    respond_to :html, :xml, :json
    def index
        @users = User.all
        respond_with(@users)
    end
    def show
        @user = User.find(params[:id])
        respond_with(@user)
    end
...

class UsersController < ApplicationController
    respond_to :html, :xml, :json
    def index
        @users = User.all
        respond_with(@users)
    end
    def show
        @user = User.find(params[:id])
        respond_with(@user)
    end
...

5. ActionMailer

rails2:  $ script/generate mailer UserMailer welcome forgot_password

这将创建 app/models/user_mailer.rb

那么在rails3中: $ rails g mailer UserMailer welcome forgot_password

这将创建 app/mailers/user_mailer.rb

在实现部分, rails2:

def welcome(user, subdomain)
    subject 'Welcome to TestApp'
    recipients user.email
    from '[email protected]'
    body :user => user, :subdomain => subdomain
end

def welcome(user, subdomain)
    subject 'Welcome to TestApp'
    recipients user.email
    from '[email protected]'
    body :user => user, :subdomain => subdomain
end

UserMailer.deliver_welcome(user, subdomain)

UserMailer.deliver_welcome(user, subdomain) 

在rails3中:

def welcome(user, subdomain)
    @user = user
    @subdomain = subdomain
    mail(:from => "[email protected]",
            :to => user.email,
            :subject => "Welcome to TestApp")
end

def welcome(user, subdomain)
    @user = user
    @subdomain = subdomain
    mail(:from => "[email protected]",
            :to => user.email,
            :subject => "Welcome to TestApp")
end

UserMailer.welcome(user, subdomain).deliver

UserMailer.welcome(user, subdomain).deliver

相比rails2, 我们在rails3下实现一个mail要简单的多:

class UserMailer < ActionMailer::Base
    default :from => "[email protected]",
                :reply_to => "[email protected]",
                "X-Time-Code" =>  Time.now.to_i.to_s
    def welcome(user, subdomain)
        @user = user
        @subdomain = subdomain
        attachments['test.pdf'] =  File.read("#{Rails.root}/public/test.pdf")
        mail(:to => @user.email, :subject => "Welcome to TestApp") do |format|
            format.html { render 'other_html_welcome' }
            format.text { render 'other_text_welcome' }
        end
    end
end

class UserMailer < ActionMailer::Base
    default :from => "[email protected]",
                :reply_to => "[email protected]",
                "X-Time-Code" => Time.now.to_i.to_s
    def welcome(user, subdomain)
        @user = user
        @subdomain = subdomain
        attachments['test.pdf'] = File.read("#{Rails.root}/public/test.pdf")
        mail(:to => @user.email, :subject => "Welcome to TestApp") do |format|
            format.html { render 'other_html_welcome' }
            format.text { render 'other_text_welcome' }
        end
    end
end 

6. ActiveRelation 以及 ActiveModel

在rails2中, 我们经常使用下面的方法来进行查询:

@posts = Post.find(:all, :conditions => {:published => true})

@posts = Post.find(:all, :conditions => {:published => true})

该方式将立即查询数据库然后返回Posts数组

而在rails3中:

@posts = Post.where(:published => true)

@posts = Post.where(:published => true)

该方法不会查询数据库, 仅仅返回一个 ActiveRecord::Relation 对象, 然后:

@posts = Post.where(:published => true)
if params[:order]
    @posts = @posts.order(params[:order])
end
@posts.each do |p|
    ...                 #在这里进行查询, 实现延迟加载
end

@posts = Post.where(:published => true)
if params[:order]
    @posts = @posts.order(params[:order])
end
@posts.each do |p|
    ...                 #在这里进行查询, 实现延迟加载
end

对于命名范围, 在rails2中:

class Post < ActiveRecord::Base
    default_scope :order => 'title'
    named_scope :published, :conditions => {:published => true}
    named_scope :unpublished,  :conditions => {:published => false}
end

class Post < ActiveRecord::Base
    default_scope :order => 'title'
    named_scope :published, :conditions => {:published => true}
    named_scope :unpublished, :conditions => {:published => false}
end

而在rails3中:

class Post < ActiveRecord::Base
    default_scope order('title')
    scope :published, where(:published => true)
    scope :unpublished, where(:published => false)
end

class Post < ActiveRecord::Base
    default_scope order('title')
    scope :published, where(:published => true)
    scope :unpublished, where(:published => false)
end

对于查找方法, rails2:

Post.find(:all, :conditions => {:author => "Joe"}, :includes => :comments, :order => "title", :limit => 10)

Post.find(:all, :conditions => {:author => "Joe"}, :includes => :comments, :order => "title", :limit => 10)

在rails3:

Post.where(:author =>  "Joe").include(:comments).order(:title).limit(10).<strong><span style="font-size: medium;">all</span></strong>

Post.where(:author => "Joe").include(:comments).order(:title).limit(10).all

7. 跨站点脚本(XSS)

在rails2中, 一般我们输入一段文本的时候, 我们往往会这样写: <%= h @post.body %>

那么在rails3中, <%= @post.body %> 默认输出的是一段safe html, 如果想输出XSS, 可以在前面加上 raw