Lens Library

Part 1

So we’re doing some math on ASCII values. We’ll need to use ord to get the ASCII value of a character; with this in hand, we can just do exactly what it says in the prompt.

def holiday_hash(st: str) -> int:
    result = 0
    for c in st:
        result += ord(c)
        result *= 17
        result %= 256
    return result

class Solution(StrSplitSolution):
    separator = ","

    def part_1(self) -> int:
        return sum(map(holiday_hash, self.input))

Part 2

This puzzle really reminded me of my Data Structures and Programming course in college. If you’re not aware, the HASHMAP described by the prompt is one way to implement a hash map¹ — the Python equivalent of which would be a dict — and the HASH algorithm from Part 1 is its hash function!

And in fact, specifically because we’re using Python, it’s convenient here to use dicts to represent our boxes of lenses. As of Python 3.7, dicts preserve insertion order² — which is exactly the behavior we want our boxes of lenses to have.

The order of the boxes themselves within the line of boxes doesn’t matter for the answer, so we can use a defaultdict instead of a list to make our lives a bit easier. So our line of lens boxes will be a defaultdict of dicts (which themselves map str labels to int focal lengths).

The code for parsing the input and updating the lenses is simple from here.

from collections import defaultdict

class Solution(StrSplitSolution):
    ...

    def part_2(self) -> int:
        boxes: dict[int, dict[str, int]] = defaultdict(dict)
        for step in self.input:
            # If step looks like "abc-"
            if "-" in step:
                label = step[:-1]
                boxes[holiday_hash(label)].pop(label, None)
            # If step looks like "abc=6"
            elif "=" in step:
                label, val_str = step.split("=")
                boxes[holiday_hash(label)][label] = int(val_str)
            else:
                raise ValueError(f"unrecognized step: {step}")

Doing a sum over a nested generator gives us our answer. (Note that I add 1 to the box_id and lens_pos, because I want their indices starting from 1 instead of 0.)

...

class Solution(StrSplitSolution):
    ...

    def part_2(self) -> int:
        ...
        return sum(
            (box_id + 1) * (lens_pos + 1) * focal_length
            for box_id, box in boxes.items()
            for lens_pos, focal_length in enumerate(box.values())
        )

Today definitely wasn’t as hard as I thought it’d be, but that’s probably because of Python’s dict behavior being so convenient. I would imagine that some more code would be required in any other language.