A simplified iterative implementation of NestedLoops

TL;DR

Let’s look at an iterative evolution of A simplified recursive implementation of NestedLoops for what NestedLoops does.

Having an iterative counterpart of a recursive function is often useful to gain some performance, especially in non-functional programming languages (and in general where calling a function can be considered expensive). This goes at the expense of the programmer’s time (usually) so it should really worth the effort, e.g. by doing some profiling. Also… this might also end up with a less-efficient implementation!

In this case, though, we go for an iterative solution for two reasons:

  • we’re studying!
  • the iterative solution paves the way to support the iterator-based alterantive, i.e. give back an iterator (which might be used by the caller completely, or only partially).

Here’s our iterative take on it:

The big difference here is that we have to simulate the stack of the functions call explicitly. To this extent, we leverage the @indexes array to keep track at which “depth” level we are in the simulated recursion.

At each stage, two things might happen:

  • we reached a point in our simulated call stack where we are beyond the available dimensions (line 27). This is the time where we can consume the current state and provide the inputs to the callback function (line 28), then exit from the call because there’s no dimension to iterate here (line 29);
  • we increment the index for the current depth level (line 33) and figure out whether:
    • we exhausted the items in this specific level (i.e. our index $i went past the last available index for the specific dimension, line 34). This means that we have to exit the current frame and backtrack (line 35);
    • we are still within the iteration, so we put the updated value in the @accumulator and do the simulated recursive call (line 39).

Each simulated recursive call is done by pushing a -1 index in @indexes, because we do a pre-increment (line 33) so we will always start getting items from index 0. This is the advantage of leveraging integer indexes to access the arrays in the input $dims.


Comments? Octodon, , GitHub, Reddit, or drop me a line!