Ruby load path

Where to look for all the things

Ruby is software, and software stores things somewhere on your file system. In order to define places on your computer where interesting stuff is stored, software often has the concept of a “load path”.

If you are using a Unix based operating system such as Linux or Mac OS X, you may have seen the environment variable $PATH in installation instructions. This variable defines all the directories where executable files are stored.

Ruby has a load path, too. Inside your Ruby program you can print it out using:


This will print out the array that is defined as the $LOAD_PATH when Ruby starts your program. puts is smart enough to put each string in that array on a separate line.

Each of these lines represents a directory on your computer where Ruby files are stored. If you use require anywhere in your application (e.g. require "digest") then Ruby will look for a Ruby file with the same name (e.g. digest.rb) in each of these directories. It will load the first file with this name that it can find.

If you are curious, you can quickly check the default load path of your Ruby installation like this:

ruby -e 'puts $LOAD_PATH'

The -e flag is a way to run some Ruby code without having to store it in a file.

For me, this prints:


From this you can see that I am using rbenv to manage my Ruby versions, that my currently active Ruby version is 2.2.1, and that I am running this on Mac OS X (“darwin”). All these paths are directories somewhere in the directory where Ruby 2.2.1 is installed on my computer.

Whenever there’s a require "something" statement in some Ruby code that I run on my computer, Ruby will check all these directories for a file something.rb.

Now let’s do a few exercises on Rubygems and Bundler.