Heres some tips for rails https.

First – use the plugin SslRequirement. But REMEMBER this ONLY works on controllers that are listed in the routes.rb file as RESOURCES

map.resources :blahblah

IE they only work on resources. Period. For non resource controllers, you will have to stick it in the links.

<%= button_to "Checkout", {:action => :checkout,:controller=>"catalog",:protocol=>"https://",:only_path=>false },{}%>

Here is how to use the ssl plugin

ruby script/plugin install ssl_requirement


class ApplicationController < ActiveRecord::Base
    include SslRequirement
end


class AccountController < ApplicationController
    ssl_required :signup, :payment
    ssl_allowed :index

    def signup
      # Non-SSL access will be redirected to SSL
    end

    def payment
      # Non-SSL access will be redirected to SSL
    end

    def index
      # This action will work either with or without SSL
    end

    def other
      # SSL access will be redirected to non-SSL
    end
end

Next Here is a working apache / mongrel config for a https ssl virtualhost. REMEMBER enable proxy, rewrite, and headers in the apache modules stanza

<IfDefine SSL>
<IfDefine !NOSSL>
<VirtualHost 192.168.1.42:443>
    DocumentRoot /srv/www/site-name/current/public
    ServerName www.site-name.com:443
    ServerAdmin webmaster@site-name.com
    ErrorLog /var/log/apache2/site-name.com-error_log
    CustomLog  /var/log/apache2/site-name.com-access_log combined
    HostnameLookups Off
    UseCanonicalName Off
    ServerSignature Off

    SSLEngine on
    SSLCipherSuite ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP:+eNULL 
    SSLCertificateFile    /etc/apache2/site-name-ssl/www.site-name.com.crt
    SSLCertificateKeyFile /etc/apache2/site-name-ssl/site-name.key
    SSLCertificateChainFile  /etc/apache2/site-name-ssl/gd_intermediate_bundle.crt
    SetEnvIf User-Agent ".*MSIE.*" nokeepalive ssl-unclean-shutdown downgrade-1.0 force-response-1.0

    <Directory "/srv/www/site-name/current/public">
        Options FollowSymLinks
        AllowOverride None
        Order allow,deny
        Allow from all
    </Directory>

    RewriteEngine On


    # Don't do forward proxying
    ProxyRequests Off

    # Enable reverse proxying
    <Proxy *>
        Order deny,allow
        Allow from all
    </Proxy>


    RequestHeader set X_FORWARDED_PROTO 'https'

    RewriteCond %{DOCUMENT_ROOT}/system/maintenance.html -f
    RewriteCond %{SCRIPT_FILENAME} !maintenance.html
    RewriteRule ^.*$ /system/maintenance.html [L]

    RewriteRule ^/$ /index.html [QSA] 
    RewriteRule ^([^.]+)$ $1.html [QSA]


    # Redirect all non-static requests to cluster
    RewriteCond %{DOCUMENT_ROOT}%{REQUEST_FILENAME} !-f
    RewriteRule .* http://127.0.0.1:3000%{REQUEST_URI} [L,P,QSA]

    # Deflate
    AddOutputFilterByType DEFLATE text/html text/plain text/css text/xml application/xml application/xhtml+xml text/javascript 
    BrowserMatch ^Mozilla/4 gzip-only-text/html
    BrowserMatch ^Mozilla/4.0[678] no-gzip
    BrowserMatch \bMSIE !no-gzip !gzip-only-text/html

</VirtualHost>                                  

</IfDefine>
</IfDefine>

My pain is your gain.

I needed to get at a nested java class from jruby.

It’s tricky. Here’s How I did it

Java class ‘Fonzie’ has a nested public class ‘Hair’

include_class 'com.als.Fonzie'
Hishair = Fonzie::Hair
folicle = Hishair.new(nil)   # if the constructor takes no arguments you must include nil here.

# view that it's really a nested class

folicle.java_class 

# returns com.als.Fonzie$Hair

I had to send large files with rails, so huge that it was killing rails.

Here is the solution, note the allow-x-send-file NOT allow-x-sendfile, this wasted a few hours of mine.

Make sure that “mod_fastcgi” is in the server.modules stanza of the lighttpd.conf file

Here is the vhosts.conf file

$HTTP["host"] =~ "www.yoursite.com" {
    server.document-root        = base + "/domains/yoursite.com/web/public/"
    server.error-handler-404 = "/dispatch.fcgi"
    compress.filetype        = ( "text/plain", "text/html", "text/css", "text/javascript" )
    compress.cache-dir       = base + "/domains/yoursite.com/web/tmp/cache"
    fastcgi.server = ( ".fcgi" =>
        ( "localhost" =>
            ( "socket" => base + "/var/run/yoursite-0.socket",
                "allow-x-send-file" => "enable",
                "bin-environment"   => ("RAILS_ENV" => "production")
            )
        )
    )
}

rails controller code

def send_this_file
    # this will force rails to read the entire file then stream it out BAD
    #send_file "#{RAILS_ROOT}/public/pdf/#{params[:id]}" , :type => mime_for(params[:id]), :disposition => 'inline', :stream => false

    # this will pass the headers to lighttpd and force it to handle the large static file, rails doesn't read the large file.
    filename = "#{RAILS_ROOT}/public/pdf/#{params[:id]}"
    response.headers['Content-Type'] = "application/force-download"
    response.headers['Content-Disposition'] = "attachment; filename=\"#{File.basename(filename)}\"" 
    response.headers["X-LIGHTTPD-send-file"] = filename
    response.headers['Content-length'] = File.size(filename)
    render :nothing => true
end

Set up restful authentication for rails

cd site-root/vendors/plugins
svn export http://svn.techno-weenie.net/projects/plugins/restful_authentication/
cd ../../
script/generate authenticated user sessions --include-activation

# add this line to /config/environment.rb
config.active_record.observers = :user_observer

# add the resource routes in config/routes.rb.  
map.resources :users, :sessions

# activate the changes in the database
rake db:migrate

I am making a site which uses an object datastore cache and rails. The only feasable way of getting data to rails is via XML. The simplest way is to use ActiveResource , which is still way way beta.

Here’s how I did it.

from http://reprocessed.org/blog/archives/2006/08/01/active_resource_on_edge.html

You may be tempted to use edge rails

rake rails:freeze:edge

This won’t work, ActiveResource is not included in Edge, only Trunk. Here is how to set up a rails instance that uses ActiveResource.

mkdir -p some_name/vendor; cd some_name
svn co http://dev.rubyonrails.org/svn/rails/trunk vendor/rails
rails .

Then, add this line to your config/environment.rb file, somewhere inside the Rails::Initializer.run do |config| block:

config.load_paths += %W( #{RAILS_ROOT}/vendor/rails/activeresource/lib )

You are running on trunk now, this is the latest version of rails, and may be flakey.

Here is how to use it

From http://weblog.techno-weenie.net/2006/12/13/taking-ares-out-for-a-test-drive

I created two rails sites on my local machine, one is the server. One is the client. ActiveResource needs to be on the client. On the server I made a RESTful site which hooked up to MySQL

Here is the Migration for the database.

class CreateParts < ActiveRecord::Migration
  def self.up
    create_table :parts do |t|
      t.column :date, :date
      t.column :customer, :string
      t.column :customer_part_number, :string
      t.column :manufacturer, :string
      t.column :factory_part_number, :string
      t.column :description, :string
      t.column :cost, :string
      t.column :selling_price, :string
      t.column :notes, :text
    end
  end

  def self.down
    drop_table :parts
  end
end

This command generates the restful scaffolding to view the data.

ruby script/generate scaffold_resource part date:date customer:string customer_part_number:string manufacturer:string factory_part_number:string description:string cost:string selling_price:string notes:text

Then on the client site I did the same. Except there is no database, just the previous classes.

In the Models section on the client, use this as your model for the part
class PartResource < ActiveResource::Base
  self.site = 'http://localhost:3000'
  # site.user = 'username'
  # site.password = 'secret_sauce'
end

class Part < PartResource
end

When I started each server up on a different port, the client can query the server and the server queries the database.

script/server lighttpd -p3000    # server
script/server lighttpd -p3001    # client

Pull up a page at http://localhost:3001/parts/1000 and check it out.

Enjoy

I am playing with the various applications in the new rails book. and came upon this issue.

rake db:migrate doesn't work

Solution ( from site root )

rake rails:freeze:edge
rake db:migrate

Yay! it works now.

Here are some code examples:

I had difficulty consuming a webservice from a currently running server. ( maybe edit concurrent connections might fix this )

Make the site” test” and code it up.

Copy the site to "test2" 

cd test1
script/server --port=3000

cd test2
script/server --port=3001

open http://localhost:3001/ws

Publishing soap webservices from rails

File app:apis:subscription_api

class SubscriptionApi < ActionWebService::API::Base
  api_method :subscribe,:expects=>[:string],:returns=>[:string]
end

File app:controllers:subscription_controller

class SubscriptionController < ApplicationController wsdlservicename ‘subscription’

def subscribe(blib) return “You sent: ” + blib end end

Consuming soap webservices from rails

Here are 3 examples of consuming web services. make a controller called ws The wsdl and api are automaticly generated for you by rails at runtime

require 'soap/wsdlDriver'
class WsController < ApplicationController
    def index
        factory = SOAP::WSDLDriverFactory.new("http://localhost:3000/subscription/wsdl")
        service = factory.create_rpc_driver
        render :text => service.subscribe("hey there")
    end
end

# no require needed
class WsController < ApplicationController
    web_client_api :subscription, :soap, "http://localhost:3000/subscription/api" 
    def index
        render :text => subscription.subscribe("in")
    end
end


# no require needed
class WsController < ApplicationController
    def index
        subscription = ActionWebService::Client::Soap.new(SubscriptionApi,"http://localhost:3000/subscription/api")
        render :text => subscription.subscribe("in")
    end
end

Consuming soap webservices from ruby.

require 'soap/wsdlDriver'
factory = SOAP::WSDLDriverFactory.new("http://localhost:3000/subscription/wsdl")
service = factory.create_rpc_driver
p service.subscribe("hey there")

I’ve been finishing this rather large, heavy on the ajax rails site. And been nagged by this persistent untraceable bug for the last while.

So I went out for most excellent Thai food. When I came back refreshed, the solution was apparent. It turns out to be a combination of a few things. First off, the bug only appears in IE windows, not Firefox, not Mac. Damn majority of browser users. Hmph.

def get_agent_info
    # by converting the :id to an integer I strip out the letters. if the text_field returns "100 joel jensen" the id becomes 100
    @agent = Agent.find(params[:id].to_i)

    # this section does an inline .rjs template
    # I am sending back javascript that will update the required fields to the new values.
    if request.xhr? 
        render :update do |page|
            page << "document.getElementById('appointment_office_phone').value = '#{number_to_phone(@agent.office.office_phone, :area_code => true)}'"
            page << "document.getElementById('appointment_showing_agent_id').value = '#{@agent.id}'"
            page << "document.getElementById('appointment_showingagent').value = '#{@agent.fullname}'"
        end
    else
        render :text => 'javascript update failed.'
    end
end

def get_agent_info
    # by converting the :id to an integer I strip out the letters. if the text_field returns "100 joel jensen" the id becomes 100
    @agent = Agent.find(params[:id].to_i)

    # this section does an inline .rjs template
    # I am sending back javascript that will update the required fields to the new values.
    if request.xhr? 
        render :update do |page|
            page << "appointment_office_phone'.value = '#{number_to_phone(@agent.office.office_phone, :area_code => true)}'"
            page << "appointment_showing_agent_id'.value = '#{@agent.id}'"
            page << "appointment_showingagent'.value = '#{@agent.fullname}'"
        end
    else
        render :text => 'javascript update failed.'
    end
end

Don’t install the fcgi gem. If you did, run:

gem uninstall fcgi

Install FastCGI

wget http://www.fastcgi.com/dist/fcgi-2.4.0.tar.gz
cd fcgi-2.4.0
./configure
make
sudo make install

Go to the homepage http://raa.ruby-lang.org/project/fcgi

Download the tarball read the README and follow the install instructions

wget http://www.moonwolf.com/ruby/archive/ruby-fcgi-0.8.7.tar.gz
ruby install.rb config -- --with-fcgi-include=/usr/local/include --with-fcgi-lib=/usr/local/lib
ruby install.rb setup
ruby install.rb install

Get lighttpd installed

yum install lighttpd-fastcgi

Set your security level

If your were unfortunate enough to install SeLinux ( Security Enhanced ), I feel your pain.

Go to the “Security level and firewall” settings, disable SeLinux

Configure Lighttpd

Go to rails root. Run “script/server lighttpd” it should fail the first time but it generates a /config/lighttpd.conf file. Use the config/lighttpd.conf, modify it and put in /etc/lighttpd/lighttpd.conf. It didn’t like the cwd part so I hardcoded all paths.

Set lighttpd to start automatically in the “Systems/Services” application