It seems very funny to me that when I search something related ruby, all ruby on rails related results popped up. So nobody using raw ruby anymore?
However, I am new to
The more commonly used way of running a ruby website is passenger: http://www.modrails.com/ It is not really hard to install and you use, here is he doc for apache: http://www.modrails.com/documentation/Users%20guide%20Apache.html#_deploying_a_ruby_on_rails_application
Your application must be a valid rack application, here is a minimal hello world (let's say /app is your application's root folder):
/app/config.ru
require 'rack'
require 'app'
run(app)
/app/app.rb
app = proc do |env|
[
# http status code
200,
# headers
{'Content-Type' => 'text/html'},
# html body
["<head><title>Test Page</title></head><body>Hello World !</body>"]
]
end
Save the files above and create a subfolder /app/public (required by passenger to detect a ruby/rails/sinatra application) and use /app/public as DocumentRoot in your apache config.
This may look scary but this is for production deployment, in development your really don't want to mess with a real server.
All you need to run the config.ru file I gave above is:
$ gem install rack
$ rackup config.ru
Or if you want to be closer to your production system:
$ gem install passenger
$ cd /app
$ passenger start
which will install you an nginx server with passenger and run your application.
In most case you will never use rack directly but instead use ruby on rails, sinatra or another framework to generate the html for you (they all use rack below now to provide a common api with the webservers).
I've heard mod_ruby is good. Unlike, #!/path/to/your/ruby
, mod_ruby won't spawn a new ruby interpreter.
https://github.com/shugo/mod_ruby
You could configure Apache (for example) to run .rb files as CGI scripts, and then add a shebang line (#!/path/to/your/ruby
or maybe #!/usr/bin/env ruby
) at the top of the script. It's not optimal, though, as it'd start a new interpreter for each request.
Ruby 1.9.2+ simple command.
ruby -run -e httpd . -p 5000
from this article http://til.justincampbell.me/start-an-http-server-with-ruby-run/ other article https://gist.github.com/willurd/5720255
Run this from your app root.
ruby -r webrick -e "s = WEBrick::HTTPServer.new(:Port => 8000, :DocumentRoot => Dir.pwd); trap('INT') { s.shutdown }; s.start"
#!/usr/bin/env ruby //shebang line to indicate path to ruby.
require 'cgi' //cgi file to create a simple cgi object.
cgi = CGI.new //instantiating a cgi object.
puts cgi.header //thats telling the server about the type(html).
puts "hello" // thats the output on the browser.