One of the values of having a full set of tests is that should you desire to refactor your code is that it is easier to do with the confidence that you will not change the behavior of the code in a way that you can’t understand.
Your tests can ensure that you can controll your code base before and after testing.
If you want to refactor your Erlang code check out Wrangler from Simon Thompson and Huiqing Li at the Univeristy of Kent. https://www.cs.kent.ac.uk/projects/wrangler/Wrangler/Home.html Wrangler provides a number of refactoring that a developer can use to work on their code base. In addition this also includes an API that allows developers and teams to create their own refactorings.
Uses of Wrangler include finding cloned code, extracting or inlining functions and more.
You can also use Wrangler to provide an automated way to update code to match API changes.
Wrangler can be used from Emacs or Eclipse via ErlIDE.
You can download wrangler from the project’s home page or from GitHub. To use Wrangler with Emacs you need to tell emacs where to find wrangler and tell emacs to load it. On my mac it looks like this, but the path will be different on your system.
(add-to-list 'load-path "/usr/local/Cellar/erlang/R15B03/lib/erlang/lib/wrangler-1.0/elisp") (require 'wrangler)
To start wrangler then run the command "M-x wrangler-menu-init", you will now have a set of menus showing a number of refactorings.
If you have a function that is overly long or you otherwise wish to turn a block of code into its own function Wrangler makes it quite easy just select the code in Emacs and chose "Extract function" from the wrangler menu, or hit "C-c C-w n f". Wrangler will then prompt you for the name of the new function. Wrangler will then take care of all of the details of creating the new function, making sure all the parameters are setup and return values are assigned.