views:

606

answers:

3

Here is one more newbie question:

require 'tasks/rails'

I saw this line in Rakefile in the root path of every rails project. I guess this line is used to require vendor/rails/railties/lib/tasks/rails.rb to get all rake tasks loaded:

$VERBOSE = nil
# Load Rails rakefile extensions
Dir["#{File.dirname(__FILE__)}/*.rake"].each { |ext| load ext }
# Load any custom rakefile extensions
Dir["#{RAILS_ROOT}/lib/tasks/**/*.rake"].sort.each { |ext| load ext }
Dir["#{RAILS_ROOT}/vendor/plugins/*/**/tasks/**/*.rake"].sort.each { |ext| load ext }

My question is why only 'tasks/rails' is specified for the require method, but not the full path of the file?

Thanks in advance.

A: 

I believe because your paths are set up in your /config/environment.rb file:

require File.join(File.dirname(__FILE__), 'boot')
The Pied Pipes
I don't understand, could you please be more specific?
eric2323223
This gets the filename of the script currently running, it gets the directory that file lives in using File.dirname and then appends a new file to it (i.e. requiring a file that you know is in the same directory but not in the $LOAD_PATH). Then File.join safely makes the new filename.
Lolindrath
+6  A: 

When you start your rails app it runs config/boot.rb which calls Rails::Initializer.set_load_path and thatsets up the $LOAD_PATH.

Ruby uses that list of directories to find the files specified on a require line. If you give it an absolute path like require '/home/lolindrath/ruby/lib.rb' it will skip that search.

This is roughly analogous to #include <stdlib.h> in C/C++ where it searches the include path you give the compiler to find that header file.

Lolindrath
A: 

Sure. In /config/boot.rb (called in environment.rb) the RAILS_ROOT is set up as so:

RAILS_ROOT = "#{File.dirname(__FILE__)}/.." unless defined?(RAILS_ROOT)

Which allows you to require things from the root I believe. Hope that's the answer anyway!

The Pied Pipes