Random Unique ID

This gem will generate a random unique id for your active record records that you can use instead of their actual ID for all external interactions with users. The goal is for you to be able to hide how many records you have, for business purposes, but also to make IDs non-predictable.

This gem is built to work with Ruby 1.9, 2.0, 2.1 as well as with Rails 3.2, 4.0, 4.1 and 4.2. All of these cases are continuously tested for.

Add this line to your application's Gemfile:

gem "random_unique_id"

And then execute:

$ bundle

Or install it yourself as:

$ gem install random_unique_id

The usage is very simple. For each record where you want to have a random id generated, add the following line to the class:

has_random_unique_id

For example:

class Post < ActiveRecord::Base
  has_random_unique_id
end

You need to also add a column, called rid of type string/varchar. It is recommended that you also add a unique index on that column, for example:

def up
  add_column :posts, :rid, :string
  add_index :posts, :rid, :unique
end

The method to_param will be overridden to return the rid instead of the id. The method belongs_to gets extended to define _rid methods similar to the _id method, like: blog_rid and blog_rid=. If you don't want to define those pass define_rid_method as false, for example:

class Post
  belongs_to :blog, define_rid_method: false
end

Classes that have rids also get a method called populate_random_unique_ids to help you populate the rid of existing records. For example:

def up
  add_column :posts, :rid, :string
  add_index :posts, :rid, :unique
  say_with_time "Post.populate_random_unique_ids" do
    Post.reset_column_information
    Post.populate_random_unique_ids { print "."}
  end
end

There are 2 settings that are configurable for RandomUniqueId:

• min_rid_length: If you have a large table, the default of 5 characters for an initial length may be a problem, since it may lead to several "bounces" until a new unique RID gets generated. The gem scales well with this problem, since it makes each subsequent try longer than the one before, but it may still may 3 or 4 DB hits for each record creation. In that case, you may want to set it to start with a higher number directly.

• random_generation_method: If you have a very large table, and having very long IDs is not a problem, you can choose to generate UUIDs instead of short-ish Random IDs. This looks worse if you're displaying these IDs in a URL, but it allows to skip the check for existence, which in a large table can make a large difference. random_generation_method can be either :short (the default) or :uuid

Both of these settings can be specified on a per-model basis, when adding RandomUniqueId to the model:

has_random_unique_id(min_rid_length: 10)
has_random_unique_id(random_generation_method: :uuid)

Or globally in an initializer:

RandomUniqueId.config(min_rid_length: 10)

This gem is being used by:

• Watu
• MSTY
• You? please, let us know if you are using this gem.

• Added global configuration.
• Added UUID generation.
• Allow changing the name of the field used for the random id.

• Added support for Rails 4.2.

• Added RandomUniqueId::Util.set_initial_ids.
• Started testing Ruby 2.1.2 and 2.1.3.
• Started testing Rails 4.1 (fixed some deprecation warnings).
• Improved documentation.

• Internal refactorings.

• Added method populate_random_unique_ids.
• Improved documentation
• Started testing with Ruby 2.1.

• Initial release of the code extracted from Watu.

1. Fork it

2. Create your feature branch (git checkout -b my-new-feature)

3. Code your thing

4. Write and run tests:

    bundle install
    appraisal
    appraisal rake test
   

5. Write documentation and make sure it looks good: yard server --reload

6. Add items to the changelog, both in the README and the CHANGELOG file.

7. Commit your changes (git commit -am "Add some feature")

8. Push to the branch (git push origin my-new-feature)

9. Create new Pull Request