functional-way | Write small programs in a functional way | Functional Programming library

Note that languages usually do not specify the (possible) optimizations to be applied. Optimizations are delegated to compilers and runtime systems.

Functional languages, such as Haskell, provide many function-based composition features enabling the targetted optimizations. You can look at function composition, lazy evaluation, partial application and higher order functions.

Please note that a lot of work has been done regarding the specific case of sequence-based computations. See generators in Python or the more general/abstract concept called ranges. Ranges are available for example in the D and C++20 programming languages. The composition and the optimization of ranges are achieved at a rather low level and compilers are generally not responsible for their specific optimizations.

Finally, several compilers are able to perform rather clever high-level transformations (see polyhedral model for loops, inter-procedural optimization for functions).

Use aperm as a generalized transpose (where a is as in the Note at the end).

It's not tail recursive because the last call is not to go, but to foldLeft. There's no way it could be even mutually tail recursive, as foldLeft calls go multiple times. It's hard to make DFS tail recursive, as the recursive algorithm relies heavily on the call stack to keep track of your position in the tree. I would suggest not bothering if you can guarantee that your tree is shallow. Otherwise, you will need to pass around an explicit stack (List is a good choice here) and completely rewrite your code.

You can use Function constructor

The Function constructor creates a new Function object. Calling the constructor directly can create functions dynamically, but suffers from security and similar (but far less significant) performance issues similar to eval. However, unlike eval, the Function constructor allows executing code in the global scope, prompting better programming habits and allowing for more efficient code minification.