i still don't really understand this

2020/21/21.py

@@ -1,16 +1,58 @@
 def main():
-	with open("input_sample.txt") as f:
+	with open("input.txt") as f:
 		lines = f.read().split('\n')
+	foods = [s[:-1].split(' (contains ') for s in lines]
+	# map allergens to possible ingredients
+	possible = [
+		(allergen, set(food[0].split(' ')))
+		for food in foods
+		for allergen in food[1].split(', ')
+		]
+	# get sets of all allergens and all ingredients
+	allergens = {p[0] for p in possible}
+	ingredients = set.union(*[p[1] for p in possible])
+	# find which ingredients are always listed for a given allergen
+	possible = {
+		a:set.intersection(*[
+			s[1]
+			for s in possible
+			if s[0] == a]
+		) for a in allergens
+		}
+	# reduce
+	while any({i for i in possible if len(possible[i]) > 1}):
+		for b in {j for j in possible if len(possible[j]) > 1}:
+			for a in {i for i in possible if len(possible[i]) == 1}:
+				possible[b] -= possible[a]
+	# find sets of allergenic and non-allergenic
+	unsafe = set.union(*list(possible.values()))
+	safe = ingredients - unsafe
+	#
+	print('|non-allergen occurrences| =', sum(
+		food[0].split().count(s)
+		for food in foods
+		for s in safe)
+		)
+	print('allergens:', ','.join(
+		[
+			list(possible[a])[0]
+			for a in sorted(list(possible))
+		])
+		)
+
+	"""
 	all_ingredients = set()
 	all_allergens = set()
 	may_contain = {}
 	unsafe_ingredients = {}
+	foods = []
 	for line in lines:
 		ingredients, contains = line.split(' (contains ')
 		ingredients = ingredients.split(' ')
 		contains = contains[:-1].split(', ') # remove trailing ")" then split
 		all_ingredients = all_ingredients | set(ingredients)
 		all_allergens = all_allergens | set(contains)
+	"""
 
 if __name__ == "__main__":
 	main()