aflatter.de

Testing custom headers and ssl with Cucumber and Capybara

2010-06-21T22:00:00Z

Today I ran into yet another issue where a little hack is required to get the job done. I was testing our application and came to a point where ssl is required. In plain old webrat you could do:

header "HTTPS", "on"

Even simpler, when using Rack::Test directly, you can just use the third parameter of your favorite request method to pass additional stuff:

get :index, {}, { :https => 'on' }

If you try to test for custom headers with capybara, you will run into trouble. There is simply no method to get access to the request environment, so things become a little more difficult. The following solution will work only if you use the :rack_test driver. Any other drivers do not support setting headers. Simply put the following code into features/support/headers_hack.rb which will automatically be loaded by cucumber.

module RackTestMixin

  def self.included(mod)
    mod.class_eval do
      # This is where we save additional entries.
      def hacked_env
        @hacked_env ||= {}
      end
      
      # Alias the original method for further use.
      alias_method  :original_env, :env

      # Override the method to merge additional headers.
      # Plus this implicitly makes it public.
      def env
        original_env.merge(hacked_env)
      end
    end
  end

end

Capybara::Driver::RackTest.send :include, RackTestMixin

module HeadersHackHelper
  
  def add_headers(headers)
    page.driver.hacked_env.merge!(headers)
  end

end

World(HeadersHackHelper)

It will give you access to an additional attribute of the driver, #hacked_env. Use this hash to set additional entries as you like. Some examples:

# To test ssl
Given /^I use ssl$/ do
  add_headers('HTTPS', 'on')
end

# To set the remote ip
Given /^my IP is (\d{1,3}\.){3}\d{1,3}$/ do |ip|
  add_headers('REMOTE_ADDR', ip)
end

# If you check for user agent
Given /^my user agent is "(.+)"$/ do |agent|
  add_headers('User-Agent', agent)
end

Github Pages and nanoc

2010-05-26T22:00:00Z

I just found out that my old shabby weblog running Wordpress has been exploited by some spam bots. This is not unusual - no matter if the blog is maintained or not. To deal with this whole mess, I finally abandoned Wordpress in favor of nanoc and Github pages. This is how I set up my repository:

Github pages will serve your site from the master branch, so we will use the nanoc output for our initial commit. Compile your site with nanoc3 co and move the output directory outside of your site directory. Change to the directory, initialize a new git repository with git init . and add the github remote with git remote add origin git@github.com:username/username.github.com.git. Now stage all the files with git add . and commit them with git commit -m 'Initial commit'. All the steps listed again:

nanoc3 co
mv output ..
cd ../output
git init .
git remote add origin git@github.com:username/username.github.com.git
git add .
git commit -m 'Initial commit'
git push origin master

This is not new to you, I know. :)

But of course you do want to keep your source in the repository too, so here are the steps to achieve that (found on the pages documentation): Create a new root branch (named source in my case) and remove your old index. git clean removes all untracked files - which every file is because you just removed the index.

git symbolic-ref HEAD refs/heads/source
rm .git/index
git clean -fdx

Now you can add your site to the repository and push it to the source branch.

cp -R /path/to/your/site/ .
git add .
git commit -m 'Initial commit of source'
git push origin source

That’s it.

To deploy, you can now issue the following commands:

rm -rf output
git checkout source
nanoc3 co
git checkout master
cp -R output/* .
git add .
git commit -a -m 'I did something cool to my site'
git push origin master

git commit -a adds all files that are untracked and removes files that are not present in the working tree, so you don’t have to worry about abandoned files.

To automate this process, I use the following bash script:

#!/bin/bash

rm -rf output
git checkout source
nanoc3 co
git checkout master
cp -R output/* .
git add .
git commit -a -m "Updated site on `date`"
git push origin master
git checkout source

Cheating Facebook's Biotronic App

2009-07-21T22:00:00Z

Okay, so this is crazy. There’s this nice little game called Biotronic on Facebook:

A puzzle game that features colorful biotechnology. Easy mouse controls, exploding combos and beautiful animations. Challenge yourself and your friends in this exciting game.

I sometimes played that and achieved a score of about 35k. I already thought about hacking the game via image processing etc. - and being curious, I just used google.

And guess what? Yes, some guy already did that. It’s cross platform and all kind of stuff and open source:

http://gitorious.org/web-automation/biocheat

Interesting project!