#! /usr/bin/env python3
import re
from functools import reduce
from dataclasses import dataclass
from typing import Any
from typing import Dict
from typing import Iterable
from typing import Set


food_parser = re.compile(r"([a-z ]+) \(contains ([a-z, ]+)\)")


@dataclass
class Food:
    items: Set[str]
    alergens: Set[str]


def read_input(filename: str) -> Iterable[Food]:
    with open(filename) as f:
        for line in f:
            line = line.strip()
            if line == "":
                continue
            match = food_parser.match(line)
            if not match:
                raise ValueError(f"cannot parse line {line}")
            items = match.group(1).split(" ")
            alergens = match.group(2).split(", ")
            yield Food(set(items), set(alergens))


def extract_alergen_info(foods: Iterable[Food]) -> Dict[str, Set[str]]:
    alergen_defs: Dict[str, Set[str]] = {}

    for food in foods:
        for alergen in food.alergens:
            if alergen not in alergen_defs:
                alergen_defs[alergen] = food.items.copy()
            else:
                alergen_defs[alergen] &= food.items

    return alergen_defs


def reduce_union(items: Iterable[Set[Any]]) -> Set[Any]:
    return reduce(lambda x, y: x | y, items)


def reduce_intersecion(items: Iterable[Set[Any]]) -> Set[Any]:
    return reduce(lambda x, y: x & y, items)


def narrow_results(alergen_defs: Dict[str, Set[str]]):
    sorted_alergens = sorted(alergen_defs.items(), key=lambda x: len(x[1]))
    for i, alergen_info in enumerate(sorted_alergens):
        if len(alergen_info[1]) == 1:
            # Remove from others
            if i+1 < len(sorted_alergens):
                for other_info in sorted_alergens[i+1:]:
                    other_info[1].difference_update(alergen_info[1])


def part1():
    all_foods = list(read_input("input.txt"))
    alergen_defs = extract_alergen_info(all_foods)

    print("Alergen defs", alergen_defs)

    safe_items = reduce_union((food.items for food in all_foods))
    all_possible_alergens = reduce_union(alergen_defs.values())

    print("All possible alergens", all_possible_alergens)

    safe_items -= all_possible_alergens

    print("All safe items", safe_items)

    count_safe = sum(
        map(
            lambda food: len(food.items & safe_items),
            (food for food in all_foods),
        ),
    )

    print("Count of safe items appeared is", count_safe)

    print("Part 2")

    narrow_results(alergen_defs)
    narrow_results(alergen_defs)
    narrow_results(alergen_defs)

    print("Narrowed alergen list", alergen_defs)

    canon_list = ",".join((
        a[1].pop() for a in sorted(alergen_defs.items(), key=lambda x: x[0])
    ))

    print("Canon list of ingredients", canon_list)


if __name__ == "__main__":
    part1()