Neighborhood search is a very trivial algorithm: starting from the node provided, follow every edge and return what you find. In our case, the degree of the relationship is important.
Just like for the tree algorithms shown previously, recursion is a great choice for solving this problem. While an iterative solution will often be more memory-efficient (no stack overflows), recursion is way more descriptive once you get the hang of it. Additionally, some compilers (and partly rustc, but not guaranteed) will expand the recursion into a loop, providing the best of both worlds (look for tail call optimization)! Obviously, the most important thing is to have a projected growth in mind; 100,000 recursive calls are likely to fill up the stack.
However, the function to run the neighborhood is implemented two-fold. First, the public-facing function takes care of validating input data and sees whether the node actually exists:
pub fn connected(&self, from: KeyType, degree...
