怎么样通过Ruby on Rails框架程序连接MongoDB?

怎么样通过Ruby on Rails框架程序连接MongoDB?

前边有介绍mongodb的安装以及ror项目的搭建,现在进行一下整合。

1.创建项目

创建项目时不再使用rails active_record支持

rails new todo -O

2.我们将要使用MongoMapper来驱动MongoDB到Rails

编辑GemFile,增加下面的内容

gem"mongo_mapper"

然后 执行 bundle install 安装gem

bundle install

3.添加数据库链接

在config/initializer下面新建一个mongo.rb文件,指定全局的数据库信息:

MongoMapper.connection = Mongo::Connection.new('localhost', 27017)


MongoMapper.database ='todo'#通过指定Rails运行环境参数,我们可以在不同的运行环境下创建互不干扰的数据,为了简单起见,没有为不同的环境指定不同的数据

 
if defined?(PhusionPassenger)

 PhusionPassenger.on_event(:starting_worker_process)do|forked|

  MongoMapper.connection.connectifforked

 end

end

完成以上步骤后,启动程序:

$ rails server

**Notice: C extension not loaded. This is required for optimum MongoDB Ruby driver performance.

You can install the extension as follows:

gem install bson_ext

If you continue to receive this message after installing, make sure that the

bson_ext gem is in your load path and that the bson_ext and mongo gems are of the same version.

=> Booting WEBrick

=> Rails 3.0.10 application starting in development on http://0.0.0.0:3000

=> Call with -d to detach

=> Ctrl-C to shutdown server

[2011-10-19 23:36:14] INFO WEBrick 1.3.1

[2011-10-19 23:36:14] INFO ruby 1.9.2 (2011-07-09) [x86_64-linux]

[2011-10-19 23:36:14] INFO WEBrick::HTTPServer#start: pid=19595 port=3000

从上面输出中可以看到bson_ext库没有加载。按照提示安装该库即可(别忘了在gemfile中添加gem):

再次启动程序,Notice提示消息消失,启动正常。在浏览器输入:http://127.0.0.1:3000,就可以看到如下页面

4.添加页面和处理逻辑

通过rails的generate命令来生成页面、控制器和模型层文件(个人还是喜欢自己手动创建,这里为了演示方便)

rails generate scaffold project name:string --orm=mongo_mapper

由于我们使用mongo作为数据库。那么,我们需要把ActiveRecord的model,改成MongoMapper的类型,也就是把继承关系从ActiveRecord::Base变成MongoMapper::Document。我们使用key这个方法标明该MongoMapper的字段属性。我们的属性是name,再加上这个字段的类型String,那么定义如下:

classProject

 include MongoMapper::Document

 key:name,String

end

通过以上的修改,我们就已经拥有了所有添加,更新,删除和列表的操作

5.数据查看

可以通过命令mongo进入mongodb数据库进行数据的查询

mongo //进入数据库

use todo //切换库

db.projects.find() //执行查询


 
6.其他

MongoMapper和ActiveRecord是完全相同的。甚至,MongoMapper还是支持ActiveRecord的验证方式如下

validates_presence_of:name

由于MongoDB没有schema-less(数据版本记录)我们可以非常容易的添加和更改model的属性,而不需要执行任何migrations的操作。比如,我们需要添加一个priority的属性,我们仅仅需要的是修改Project model如下:

classProject

 include MongoMapper::Document

 key:name,String,:required=>true

 key:priority,Integer

end

表之间的关联对于MongoDB这里稍微有点区别,我们需要ObjectId类型来存储所有id。

至于,处理不同表之前的关联,我们可以像ActiveRecord一样定义belongs_to,当然,稍微有点不同,在Project中我们需要定义has_many :tasks,在MongoMapper中需要用many代替。

我目前也就做到这里。有时间再去深入研究其他的功能。

PS:Ruby编写MongoDB备份脚本(fsync & lock)

#!/usr/local/bin/ruby
# date:  06-12-2014
# auther: lucifer
# use fsync and lock to the file-system before backup the file-system
# mongo-ruby-driver version > 1.10.0

require 'mongo'
require 'fileutils'
require 'date'

include Mongo
include BSON

# the members of replcation-set
# test mongodb server version 2.6.0
# host = "192.168.11.51"

# The port of members
# If the port is 27017 by default then otherport don't need to assignment
# otherport = ""
# port = otherport.length != 0 ? otherport : MongoClient::DEFAULT_PORT

# opts = {:pool_size => 5, :pool_timeout => 10}
# Create a new connection
# client = MongoClient.new(host, port, opts)

uri_string = "mongodb://caoqing:xxxxxxxx@x.x.x.x:27017/admin"
client = MongoClient.from_uri(uri = "#{uri_string}")
db = client['admin']

# fsync and lock the database
cmd = OrderedHash.new
cmd[:fsync] = 1
cmd[:lock] = true
# p cmd
db.command(cmd)

# datafile path
d = "/var/lib/mongo"

# dir = Dir.new("#{d}")
# entries = dir.entries
# entries.delete_if { |entry| entry =~ /^\./}
# convert the relative path to the full path
# entries.map! { |entry| File.join(dir.path, entry) }
# maintain only the type of file
# entries.delete_if { |entry| !File.file?(entry) }
# p entries

start = Date.today.to_s
prev = (Date.today - 7).to_s

dest = "/backup/#{start}"
sour = "/backup/#{prev}"

FileUtils.rm_rf("#{sour}") if File::exist?("#{sour}")

Dir.mkdir("#{dest}", 0755) unless File::exist?("#{dest}")

FileUtils.cp_r Dir.glob("#{d}/**"), dest if client.locked?

puts "*" * 20
puts "\tbackup complete"
puts "*" * 20

# DB::SYSTEM_COMMAND_COLLECTION
# unlock the database
db["$cmd.sys.unlock"].find_one
client.close