88 lines
2.3 KiB
Python
Executable file
88 lines
2.3 KiB
Python
Executable file
#!/usr/bin/env python
|
|
|
|
import sys
|
|
from cmath import phase
|
|
from itertools import groupby
|
|
from math import gcd, pi
|
|
from pprint import pprint
|
|
from typing import NamedTuple, Set, Tuple
|
|
|
|
|
|
class Position(NamedTuple):
|
|
x: int
|
|
y: int
|
|
|
|
|
|
def pos_to_angle_dist(pos: Position) -> Tuple[float, float]:
|
|
cartesian = complex(*pos)
|
|
angle = phase(cartesian)
|
|
if angle < -pi / 2:
|
|
angle += 2.5 * pi
|
|
else:
|
|
angle += pi / 2
|
|
return (angle, abs(cartesian))
|
|
|
|
|
|
def main() -> None:
|
|
asteroids = [
|
|
Position(x, y)
|
|
for y, line in enumerate(sys.stdin.readlines())
|
|
for x, c in enumerate(line.rstrip())
|
|
if c == "#"
|
|
]
|
|
|
|
def count_spotted(x: int, y: int) -> int:
|
|
seen: Set[Position] = set()
|
|
ans = 0
|
|
radius = 1
|
|
|
|
while True:
|
|
|
|
def is_r_away(pos: Position) -> bool:
|
|
return max(abs(pos.x - x), abs(pos.y - y)) == radius
|
|
|
|
to_visit = list(filter(is_r_away, asteroids))
|
|
radius += 1
|
|
if len(to_visit) == 0:
|
|
break
|
|
for pos in to_visit:
|
|
rel = (pos.x - x, pos.y - y)
|
|
common = gcd(*rel)
|
|
rel = Position(*(a // common for a in rel))
|
|
if rel in seen:
|
|
continue # Already have an asteroid on this path
|
|
seen.add(rel)
|
|
ans += 1
|
|
|
|
return ans
|
|
|
|
# We need to find the observatory's position as a prerequisite
|
|
ans, orig = max((count_spotted(*pos), pos) for pos in asteroids)
|
|
print(f"({orig.x}, {orig.y}): {ans}")
|
|
|
|
def to_rel(p: Position) -> Position:
|
|
return Position(*(a - o for (a, o) in zip(p, orig)))
|
|
|
|
angle_dists = sorted(
|
|
(pos_to_angle_dist(to_rel(p)), p) for p in asteroids if p != orig
|
|
)
|
|
grouped_angle_dists = [
|
|
[val[1] for val in group]
|
|
for __, group in groupby(angle_dists, key=lambda x: x[0][0])
|
|
]
|
|
|
|
def find_n_th(n: int) -> Position:
|
|
assert 0 < n < len(asteroids) # Sanity check
|
|
while n >= len(grouped_angle_dists):
|
|
for group in grouped_angle_dists:
|
|
group.pop(0)
|
|
n -= 1
|
|
return grouped_angle_dists[n - 1][0]
|
|
|
|
x, y = find_n_th(200)
|
|
print(x * 100 + y)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|