HappyCoder - Home

Railsconf 2008

2008-05-12T15:00:00Z

I will be at Railsconf 2008 at Portland. I still haven’t finalized the sessions I’m attending but I’m sure it will be great.

Notes on migrations

2007-11-13T15:49:00Z

If you add columns to a model, and you want to use that model, call Base#reset_column_information.

If you save a model in a migration, and if you do not want the timestamps (created_at, updated_at) to be updated, put ActiveRecord::Base.record_timestamps = false.

class AddLastCommentAtToTopic < ActiveRecord::Migration
  class Topic < ActiveRecord::Base
    has_many :posts
  end
  
  class Post < ActiveRecord::Base
  end
  
  def self.up
    ActiveRecord::Base.record_timestamps = false
    add_column :topics :last_comment_at, :datetime
    Topic.reset_column_information
    Topic.find(:all).each do |topic|
      topic = topic.posts.find(:first, :order => "last_comment_at DESC")
      if topic
        topic.last_comment_at  = post.last_comment_at
        topic.save
      end
    end
    ActiveRecord::Base.record_timestamps = true
  end

  def self.down
    remove_column :topics, :last_comment_at
  end
end

Send emails in development environment using Gmail

2007-11-11T16:38:00Z

If you don’t have a mail server in your development machine, you can use Gmail. Put the following code in RAILS_ROOT/lib/smtp_tls.rb

require "openssl"
require "net/smtp"

Net::SMTP.class_eval do
  private
  def do_start(helodomain, user, secret, authtype)
    raise IOError, 'SMTP session already started' if @started
    check_auth_args user, secret, authtype if user or secret

    sock = timeout(@open_timeout) { TCPSocket.open(@address, @port) }
    @socket = Net::InternetMessageIO.new(sock)
    @socket.read_timeout = 60 #@read_timeout
    #@socket.debug_output = STDERR #@debug_output

    check_response(critical { recv_response() })
    do_helo(helodomain)

    if starttls
      raise 'openssl library not installed' unless defined?(OpenSSL)
      ssl = OpenSSL::SSL::SSLSocket.new(sock)
      ssl.sync_close = true
      ssl.connect
      @socket = Net::InternetMessageIO.new(ssl)
      @socket.read_timeout = 60 #@read_timeout
      #@socket.debug_output = STDERR #@debug_output
      do_helo(helodomain)
    end

    authenticate user, secret, authtype if user
    @started = true
  ensure
    unless @started
      # authentication failed, cancel connection.
      @socket.close if not @started and @socket and not @socket.closed?
      @socket = nil
    end
  end

  def do_helo(helodomain)
     begin
      if @esmtp
        ehlo helodomain
      else
        helo helodomain
      end
    rescue Net::ProtocolError
      if @esmtp
        @esmtp = false
        @error_occured = false
        retry
      end
      raise
    end
  end

  def starttls
    getok('STARTTLS') rescue return false
    return true
  end

  def quit
    begin
      getok('QUIT')
    rescue EOFError, OpenSSL::SSL::SSLError
    end
  end
end

Then, add these at the bottom of RAILS_ROOT/config/environments/development.rb

ActionMailer::Base.server_settings = {
  :address => "smtp.gmail.com",
  :port => 587,
  :domain => "mycompany.com",
  :authentication => :plain,
  :user_name => "username",
  :password => "password"
}

Add your username and password, restart the server and you’re good to go. Note that the emails your app sends might end up in the bulk folder.

Link How to use GMail SMTP server to send emails in Rails ActionMailer

Capistrano 2.0

2007-07-22T03:01:00Z

Capistrano 2.0 is out. Check the announcement here.

Capistrano is great for automating tasks via SSH on remote servers, like software installation, application deployment, configuration management, ad hoc server monitoring, and more.

Learn more about capistrano on it’s website.

I like to share my deployment recipe. After running “capify .” on the app directory, I got the file config/deploy.rb. Initially it looks like this

set :application, "set your application name here" 
set :repository,  "set your repository location here" 

# If you aren't deploying to /u/apps/#{application} on the target
# servers (which is the default), you can specify the actual location
# via the :deploy_to variable:
# set :deploy_to, "/var/www/#{application}" 

# If you aren't using Subversion to manage your source code, specify
# your SCM below:
# set :scm, :subversion

role :app, "your app-server here" 
role :web, "your web-server here" 
role :db,  "your db-server here", :primary => true

After putting in the values and adding some tasks, my deploy.rb looks like this

set :application, "topsecret"
set :repository,  "http://svn.example.com/topsecret/trunk"

# If you aren't deploying to /u/apps/#{application} on the target
# servers (which is the default), you can specify the actual location
# via the :deploy_to variable:
set :deploy_to, "/home/topher/apps/#{application}"

role :app, "topher.88-mph.net"
role :web, "topher.88-mph.net"
role :db,  "topher.88-mph.net", :primary => true

set :mongrel_config, "/etc/mongrel_cluster/#{application}.yml" 

namespace :deploy do
  desc "Link in the *.yml files and rails."
  task :after_update_code do
    run "ln -nfs #{deploy_to}/shared/config/database.yml #{release_path}/config/database.yml" 
    run "ln -nfs #{deploy_to}/shared/config/mongrel_cluster.yml #{release_path}/config/mongrel_cluster.yml" 
    run "ln -nfs /home/topher/rails_src/rails_1_2_stable #{release_path}/vendor/rails"
  end

  desc "Restart the mongrels."
  task :restart do
    sudo "mongrel_rails cluster::restart -C #{mongrel_config}" 
  end
end

:application is the app name
:repository is the svn repository
:deploy_to is the path to the app in the server
roles :app, :web, and :db is the server. I’m using one server for the app, web, and db so they’re all the same.
:mongrel_config is config file to be used by the mongrel_cluster gem
I create symlinks to config/database.yml and config/mongrel_cluster.yml.
I use rails 1.2.3 for this app but the server has 1.1.6 installed. You can check in vendor/rails to svn (or svn externals) but I decided to put rails to a directory outside of the app so it can be shared by multiple apps. I just create a symlink.
I restart my mongrel cluster using the cluster::restart. The mongrel_cluster gem includes a recipe but so far it only works with capistrano 1.4. (Thanks to topfunky for this.)

To deploy I just run “cap deploy” on my development machine. To deploy with migrations run “cap deploy:migrations”.

Plugin: attachment_fu

2007-07-19T02:41:00Z

attachment_fu is a rails plugin written by Rick Olson. From the README,

attachment_fu facilitates file uploads in Ruby on Rails. There are a few storage options for the actual file data, but the plugin always at a minimum stores metadata for each file in the database.

Check out Mike Clark’s tutorial.

On windows, when I submit the form with the file upload, I get an error “Size not included in list”. Upon inspection, the size is zero. (The default minimum size is 1 byte.) The size of the file uploaded isn’t set properly. I found somewhere on the beast forum that you need to add “sleep 1” on your controller. I haven’t tried to use attachment_fu on my server (ubuntu) and I’m hoping this is only a problem with windows.

My friend Mon, also a rails developer, encountered another problem with attachment_fu. He added a user_id in the model which uses attachment_fu. When he saves the uploaded file, he gets an error saying that user_id is null. We’re absolutely sure there’s a user_id.

Here’s what happened: attachment_fu can create a thumbnail(s) for you when you save the uploaded image (you will need an image processor like image magick). If you don’t specify a thumbnail_class, attachment_fu will use the same model for the thumbnail. So if you have a Mugshot model, after submitting the form, 1 model will be created for the uploaded image, and another one will be created for the thumbnail. So even if you add user_id to your model, the thumbnail won’t have a user_id. You need to add a callback before_thumbnail_saved on your model.

123	before_thumbnail_saved do \|record, thumbnail\| thumbnail.user_id = record.user_idend

Install the plugin with

ruby script/plugin install http://svn.techno-weenie.net/projects/plugins/attachment_fu/

Editing models in migrations

2007-07-13T11:30:00Z

In migrations, you can also edit your models. I have a Topic and a Post. A topic has many posts. The models are already created using previous migrations. Now I want to add 3 new fields to Topic—posts_count, replied_by, and replied_at.

Here is the migration:

class AddCachedFieldsOnTopic < ActiveRecord::Migration
  class Topic < ActiveRecord::Base
    has_many :posts, :order => "posts.created_at"
  end
  class Post < ActiveRecord::Base; end
  
  def self.up
    add_column :topics, :posts_count, :integer, :default => 0
    add_column :topics, :replied_by, :integer
    add_column :topics, :replied_at, :datetime
    
    Topic.find(:all, :include => :posts).each do |topic|
      last_post = topic.posts.last
      posts_count = topic.posts.size
      Topic.update_all(["posts_count = ?, replied_by = ?, replied_at = ?", posts_count, last_post.user_id, last_post.created_at], ["id = ?", topic.id]) if last_post
    end
  end

  def self.down
    remove_column :topics, :posts_count
    remove_column :topics, :replied_by
    remove_column :topics, :replied_at
  end
end

Notes

posts_count will be used by rails for caching. In the post model, I’ll put :belongs_to :topic, :counter_cache => true. Set the default to 0.
You can actually get these fields through the association—topic.posts.count, topic.posts.last.user, topic.posts.last.created_at – but I’m adding it to Topic to avoid the additional queries.
I’m looking at (copying) the beast application.

Server Setup

2007-07-05T14:06:00Z

I have a VPS from Rimuhosting and I’m using nginx and a cluster of mongrel for my rails applications. I followed this tutorial which guides you on setting up everything – ruby, rails, nginx, mongrel, postgresql, subversion and capistrano. I’m sharing my vps with some friends and ruby, rails, and mysql were already installed. I’m using Dreamhost for subversion. Setting up nginx and mongrel is easy. I just followed the tutorial and didn’t encounter any problem.

Mongrel is a fast HTTP library and server for Ruby that is intended for hosting Ruby web applications of any kind using plain HTTP rather than FastCGI or SCGI. It is framework agnostic and already supports Ruby On Rails, Og+Nitro, Camping, and IOWA frameworks.

I use nginx for this reason (taken from the faq page in mongrel website):

Ruby on Rails is not thread safe so there is a synchronized block around the calls to Dispatcher.dispatch. This means that everything is threaded right before and right after Rails runs. While Rails is running there is only one controller in operation at a time. This is why people typically have to run a small set of Mongrel processes (a “Pack of Mongrels”) to get good concurrency.

For a rails application, I run 2 mongrel servers. I use the gem mongrel_cluster to manage these mongrels (this is in the tutorial). For example, mongrels are running in port 7500 and 7501. When a request is made, nginx passes that request to one of the mongrels. For bigger applications, use more mongrels.

Since I’m sharing the VPS with my friends, I can’t change the web server easily. Besides, they are running php and though it’s possible to run php in nginx I haven’t researched that yet. We are using Apache 2.0 which is not easy to set up with mongrel. Apache 2.2 is recommended. I have to run nginx on a different port, say 8088. Then on the apache conf file, I put this

<VirtualHost *:80>
        ServerName topher.88-mph.net
        ServerAdmin crigor@gmail.com
        ProxyRequests Off
        ProxyPreserveHost on
        <Location />
          ProxyPass http://topher.88-mph.net:8088/
          ProxyPassReverse http://topher.88-mph.net:8088/
        </Location>
</VirtualHost>

When you go to topher.88-mph.net, apache gets the request then passes it to ngingx which in turns passes it to one of the mongrels. Apache is not really needed on this set up.

Ruby script for creating a rails app

2007-06-21T12:00:00Z

A few months ago I saw this ruby script by Akhil Bansal which creates a rails app and imports it to svn, ignoring and removing files like database.yml, tmp directory and log files.

I modified it to use the trunk, tags, and branches directories and updated some of the code. Here’s what the script does:

create a rails app in your local machine
create trunk, tags and branches directories in the svn repository
svn import the rails app
delete the app in your local machine
check out from svn
remove log files from svn
ignore log files
ignore tmp directory
remove tmp directory from svn
move database.yml to database.yml.sample
ignore database.yml

Download the script here.

Run it with “ruby create_rails_with_subversion.rb” on the command line. You need svn installed on your local machine and a subversion repository.

django.contrib.databrowse

2007-05-12T07:12:00Z

There’s a new django.contrib add-on called databrowse. From the documentaion,

Databrowse is a Django application that lets you browse your data. As the Django admin dynamically creates an admin interface by introspecting your models, Databrowse dynamically creates a rich, browsable Web site by introspecting your models.

You need to:

1. Add ‘django.contrib.databrowse’ to your INSTALLED_APPS setting.

You don’t need to run syncdb.

2. Register models on the Databrowse site.

1234	from django.contrib import databrowsedatabrowse.site.register(SomeModel)databrowse.site.register(SomeOtherModel)

You can put this on urls.py.

3. Add to your URLConf

(r'^databrowse/(.*)', databrowse.site.root),

Test databroswe by going to /databrowse.

Here’s a screenshot:

Full screen

Using named URL patterns in Django views

2007-04-27T14:03:00Z

Named URL patterns have been added to Django recently but the documentation isn’t complete yet.

In urls.py, use url()

12	urlpatterns = patterns('', url(r'^$', index, name="app_index"),

On the templates, use url()

<a href="{% url app_index %}">Home</a>

On the views, use django.core.urlresolvers.reverse()

123	from django.core.urlresolvers import reversedef some_method(request): return HttpResponseRedirect(reverse('app_index'))

Exporting to CSV in Rails

2007-04-26T13:54:00Z

To export data to CSV, I use CSV::Writer.

def report
  CSV::Writer.generate(output="") do |csv|
    csv << %w(Name Price)
    @items.each do |item|
       csv << [item.name, item.price]
    end
  end

  send_data(output,
    :type => "text/csv",
    :filename => 'report.csv')
  end
end

You can also checkout FasterCSV.