1. Insert the headless gem in your Gemfile

group :test do
  gem 'headless', :require => false
end

2. Install Xvfb if you don’t have it already

sudo apt-get install xvfb

3. Configure cucumber in features/support/env.rb

#Headless gem
if Capybara.current_driver == :selenium
  require 'headless'

  headless = Headless.new
  headless.start
  at_exit do
    headless.destroy
  end
end

4. BONUS: Configure cucumber for minimal output. In lib/tasks/cucumber.rake:

t.cucumber_opts = "--format progress"

It is like a bag of tricks for customizing the routes generated by devise, where you can pass a lot of ‘hidden’ options to the devise_for helper. I found specially nice these:

 * :skip => tell which controller you want to skip routes from being created:

     devise_for :users, :skip => :sessions

 * :only => the opposite of :skip, tell which controllers only to generate routes to:

     devise_for :users, :only => :sessions

 * :skip_helpers => skip generating Devise url helpers like new_session_path(@user).
   This is useful to avoid conflicts with previous routes and is false by default.
   It accepts true as option, meaning it will skip all the helpers for the controllers
   given in :skip but it also accepts specific helpers to be skipped:

     devise_for :users, :skip => [:registrations, :confirmations], :skip_helpers => true
     devise_for :users, :skip_helpers => [:registrations, :confirmations]

unless Rails.env.production?
  module Resque
    def self.enqueue(klass, *args)
      klass.perform(*args)
    end
  end
end

 config.stretches = Rails.env.test? ? 1 : 10 

Also, if you are having problems with the sign_out route and testing, maybe it could help to make it a GET route (instead of a DELETE one) whilst testing:

 config.sign_out_via = Rails.env.test? ? :get : :delete 

In linux, just write the testing database to /dev/shm in config/database.yml:

database: /dev/shm/test.sqlite3

Recent linux kernels (1) offer us a nice place to quickly interact with other apps ‘in memory‘ thanks to /dev/shm. Anyway, if you like to have some power over what you are really doing, here is the line to create a ramdisk:

sudo mount -t tmpfs -o size=64M,nr_inodes=10k,mode=0777 tmpfs /path/to/ramdisk

In macosx, it’s a little bit harder, because you have to manually create the disk:

#!/bin/bash
ramfs_size_mb=1024
mount_point=~/volatile

ramfs_size_sectors=$((${ramfs_size_mb}*1024*1024/512))

ramdisk_dev=`hdid -nomount ram://${ramfs_size_sectors}`
newfs_hfs -v 'Volatile' ${ramdisk_dev}
mkdir -p ${mount_point}
mount -o noatime -t hfs ${ramdisk_dev} ${mount_point}
echo "remove with:"

echo "umount ${mount_point}"
echo "diskutil eject ${ramdisk_dev}"

Thanks to Elias for the linux version and Salvatore Sanfilippo for the mac script.

Here are some typical numbers you can use to save yourself spreadsheet and research time:

10% conversion home page to free trial
10% conversion from free trial to paid plan
1% conversion from free forever to paid plan
$100,000/year cost per employee
30% of registered users (or number of downloads if a mobile app) will use the service each month
10% of registered users (or number of downloads if a mobile app) will use the service each day
10% of daily users is equal to the max number of concurrent users of a real-time service
The last three are from Fred Wilson.

from thoughtbot’s playbook

We wanted something quick and easy, a past-time but mind-challenging for someone with no technical background, so we came up with a simple algorithm:

Given a message M and N postcards, put each letter of M in a different postcard. When you run out of cards, start again.

This way, a message such us “I AM HAVING A GREAT TIME” in 2 postcards will look like:

1. IMAIGGETIE
2. AHVNARATM

class String
  def postcardify(number_of_postcards=1)
    bucket = Array.new(number_of_postcards) { Array.new }
    self.upcase.gsub(" ","").split(//).each_with_index do |letter, index|
      i = index.modulo number_of_postcards
      bucket[i] << letter
    end
    bucket.map { |a| a.join }
  end
end

The use is quite straightforward:

"This is an example text for postcardify".postcardify(3)
=> ["TSAXPTTRSAI", "HINALEFPTRF", "ISEMEXOOCDY"]

There you go, fun for everyone with ruby!

Estaba un poco harto de darle para arriba a la flechita en el
terminal… y he buscado algun truquillo ideal para unix hackers, a
ver si os gusta:

~/.bashrc:

HISTCONTROL=erasedups # Remove duplicates:
PROMPT_COMMAND="history-n; $PROMPT_COMMAND" # Save history each time a command is typed
HISTSIZE=1000
shopt -s histappend

~/.inputrc

"\e[A": history-search-backward
"\e[B": history-search-forward
set show-all-if-ambiguous on
set completion-ignore-case on

• Elimina duplicados del historial del bash (para que valen??)
• Incrementa el tamaño para que guarde 1000 comandos (por defecto son 500, ademas al quitar repetidos guarda más)
• En el historial ahora se guardan todos los comandos de todos los terminales abiertos (antes solo se guardaba lo del ultimo que cerrabas)
• Ahora al autocompletado le da igual mayusculas o minusculas (ahorramos de pulsar shifts)
• Ahora cuando le das para arriba tiene en cuenta lo que ya has escrito. (Escribes gcc, y cada vez que le das para arriba te aparecen las opciones de gcc distintas que has usado).

Espero que os guste!

Muchas gracias Mario, muy útil y bien pensado :-)

# Set git autocompletion and PS1 integration
if [ -f /usr/local/git/contrib/completion/git-completion.bash ]; then
  . /usr/local/git/contrib/completion/git-completion.bash
fi

if [ -f /opt/local/etc/bash_completion ]; then
    . /opt/local/etc/bash_completion
fi

export GIT_PS1_SHOWDIRTYSTATE=true
export GIT_PS1_SHOWUNTRACKEDFILES=true

PS1='\u@\h:\w\[\033[32m\]$(__git_ps1)\[\033[00m\]\$ '

Seen here and here.

1. http://abdullin.com/cqrs/
2. http://martinfowler.com/bliki/CQRS.html
3. http://codebetter.com/gregyoung/2010/02/16/cqrs-task-based-uis-event-sourcing-agh/
4. http://www.udidahan.com/2011/04/22/when-to-avoid-cqrs/
5. http://www.udidahan.com/2009/12/09/clarified-cqrs/
6. http://cqrsinfo.com/documents/