def test_claim_overlap():
c1 = Claim.new_from_string("#1 @ 1,3: 4x4")
c2 = Claim.new_from_string("#2 @ 3,1: 4x4")
c3 = Claim.new_from_string("#3 @ 5,5: 2x2")
c4 = Claim.new_from_string("#4 @ 1,1: 1x1")
assert c1.overlaps(c2)
assert c2.overlaps(c1)
assert c1.overlaps(c3) == False
assert c4.overlaps(c4)
def part2_v3(file_name: str) -> str:
"""
Find the one claim that doesn't overlap any other claim. Return its id.
This version is similar to part2_v2(), but we handle the filtering of the
claim ID's differently. In my haphazard benchmarking, it seems marginally
slower.
"""
with open(file_name) as f:
claims = [Claim.new_from_string(line.rstrip()) for line in f]
# Each value is a claim ID.
set_claim_ids = set()
# Each key is a sq inch tuple. Each value is a list of claim ID's.
d_sq_inches: DefaultDict[Tuple[int, int], List[str]] = defaultdict(list)
# set_claim_ids will end up with all claim ID's.
# d_sq_inches will end up with all square inches that have claims on them,
# and a list of the claim ID's that have the claims.
for claim in claims:
set_claim_ids.add(claim.id)
for sq_inch in claim.sq_inches:
d_sq_inches[sq_inch].append(claim.id)
# Now for all square inches with more than one claim,
# discard the claim ID's from set_claim_ids.
for claim_list in d_sq_inches.values():
if len(claim_list) > 1:
for claim_id in claim_list:
set_claim_ids.discard(claim_id)
# The only claim ID left is one we want.
return set_claim_ids.pop()
def part2_v2(file_name: str) -> str:
"""
Find the one claim that doesn't overlap any other claim. Return its id.
This version is much faster than part2_v1().
"""
with open(file_name) as f:
claims = [Claim.new_from_string(line.rstrip()) for line in f]
# Each key is a claim ID. Each value is a bool.
d_claim_ids = {}
# Each key is a sq inch tuple. Each value is a list of claim ID's.
d_sq_inches: DefaultDict[Tuple[int, int], List[str]] = defaultdict(list)
# Load both dictionaries. d_claim_ids will end up with all claim ID's.
# d_sq_inches will end up with all square inches that have claims on them,
# and a list of the claim ID's that have the claims.
for claim in claims:
d_claim_ids[claim.id] = True
for sq_inch in claim.sq_inches:
d_sq_inches[sq_inch].append(claim.id)
# Now for all square inches with more than one claim,
# flip d_claim_ids_no_overlap to False.
for claim_list in d_sq_inches.values():
if len(claim_list) > 1:
for claim_id in claim_list:
d_claim_ids[claim_id] = False
# Now return the one claim ID that's still True.
for claim_id, b in d_claim_ids.items():
if b:
return claim_id
raise Exception(NOT_FOUND)
def test_claim_new_from_string():
s = "#14 @ 690,863: 12x20"
claim = Claim.new_from_string(s)
assert claim.id == "14"
assert claim._x == 690
assert claim._y == 863
assert claim._width == 12
assert claim._height == 20
def part2_v1(file_name: str) -> str:
"""
Find the one claim that doesn't overlap any other claim. Return its id.
"""
with open(file_name) as f:
claims = [Claim.new_from_string(line.rstrip()) for line in f]
for claim in claims:
if _no_overlap(claim, claims):
return claim.id
raise Exception(NOT_FOUND)
def part1(file_name: str) -> int:
"""
How many square inches of fabric are within two or more claims?
"""
with open(file_name) as f:
claims = [Claim.new_from_string(line.rstrip()) for line in f]
d: DefaultDict[Tuple[int, int], int] = defaultdict(lambda: 0)
# Count how many times each square inch appears in a claim
for claim in claims:
for sq_inch in claim.sq_inches:
d[sq_inch] += 1
# Count the # of values in d that are > 1
return len([v for v in d.values() if v > 1])
def test_claim_sq_inches():
c1 = Claim.new_from_string("#1 @ 1,3: 4x4")
assert c1.sq_inches == [
(1, 3),
(1, 4),
(1, 5),
(1, 6),
(2, 3),
(2, 4),
(2, 5),
(2, 6),
(3, 3),
(3, 4),
(3, 5),
(3, 6),
(4, 3),
(4, 4),
(4, 5),
(4, 6),
]