I'm working with an API right now that is absolutely based on duplicated code. They have a system for querying items and the API has 3 different ways depending

I'm working with an API right now that is absolutely based on duplicated code. They have a system for querying items and the API has 3 different ways depending on the endpoint. I just found a new one the other day and I hated it -- why is this one API unnecessarily different from all the rest! I'm building a library to call this API and I've abstracted over all these differences so my callers never have to know how messed up the underlying API is -- they get a consistent experience regardless.