monkey_patch_template.rb

# If you have one or two monkey patches, you can easily stick them in an initializer, like this:
# ./config/initializers/monkey_patch_*.rb
# Or if you have some sort of load order requirement, prepend numbers to put them before other initializers:
# ./config/initializers/000_monkey_patch_*.rb
# If you have a lot, you might want to put them in ./app/monkey_patches or some other clearly marked location.

# Require the dependency to ensure it's loaded/available
require 'path/to/class/we/patch' # TODO

# TODO: Describe the problem this monkey patch solves. If possible refer to issue threads, PRs and commits with more details.

# Code in here is executed before every request in development mode, but only once in production mode.
# Here it makes sure the contained monkey patch is always applied, even if the patched gem's code is reloaded.
# `to_prepare` is available from Rails 5, and is intended to play nice with Zeitwerk (Rails 6.0)
Rails.configuration.to_prepare do
  checked_version = '1.2.3' # TODO: put the current version here
  current_version = TheGemWePatchHere::VERSION # TODO: point it at the gem's version method/constant
  if Gem::Version.new(current_version) > Gem::Version.new(checked_version)
    raise "Please review this monkey patch still applies to version #{current_version} (last checked for #{checked_version})"
  end

  # This opens up an existing class, which will explode if the require above is not working as intended.
  # This is a feature, because it helps you spot classes which no longer exist.
  Namespace::To::ClassWithIssues.class_eval do
    # TODO: Describe intention of the patch and where we deviate from the original code
    # Source: TODO (link to file in source repository at tag we last checked it)
    def some_method(arg1, arg2)
      # TODO: upstream code goes in, then mark your changes so the next person looking at this
      #   can evaluate what you did and if it's still relevant for the next version.
    end
  end
end