I am currently working on a kattis problem: Treasure Hunt, Link. The goal is to find the path which will cost the least amount of days to reach the treasure.
I currently use Dijkstra's algorithm with weighted vertexes to calculate the shortest route to the treasure. I have defined a class 'Node' in which I define it's weight and the previous Node if assigned. I use a heapq and needed to override the lt method in my Node class.
When the shortest route is defined I try to calculate the number of days that it will take to complete this shortest route.
I know it is not the fines code, I am sorry.
Defining neighbor nodes and the pathfinding
def createNode(row):
rl = list()
for x in row:
rl.append(Node(x))
return rl
rows, columns, stamina = map(int, input().split(' '))
treasure_map = list()
for row in range(rows):
treasure_map.append(list(input()))
treasure_map = list(map(createNode, treasure_map))
for x in range(len(treasure_map)):
for y in range(len(treasure_map[x])):
tile = treasure_map[x][y]
# add tile to south link
if x - 1 >= 0 and treasure_map[x - 1][y] is not None:
tile.add_neighbour(treasure_map[x - 1][y])
if y + 1 < len(treasure_map[x]) and treasure_map[x][y + 1] is not None:
tile.add_neighbour(treasure_map[x][y + 1])
if x+1 < len(treasure_map) and treasure_map[x + 1][y] is not None:
tile.add_neighbour(treasure_map[x + 1][y])
if y - 1 >= 0 and treasure_map[x][y - 1] is not None:
tile.add_neighbour(treasure_map[x][y - 1])
visited = list()
nodes = list()
for x in treasure_map:
for y in x:
if y.name is not '#':
nodes.append(y)
heapq.heapify(nodes)
endpoint = None
if len(nodes) < 2:
print(-1)
# Search route with minimum load
while nodes:
curr = heapq.heappop(nodes)
if curr.name is 'G':
endpoint = curr
break
for node in curr.get_neighbours():
if node not in visited and not node.load > stamina:
if curr.weight + curr.load < node.weight:
node.add_previous(curr)
visited.append(curr)
heapq.heapify(nodes)
The code of calculating the days it will take to get from the start to end: (once again I know it can be better)
if endpoint is not None:
nexNode = endpoint.previous
found = False
stamina_to_use = list()
while nexNode:
if nexNode.name == "S":
# Maybe count the day here
found = True
stamina_to_use.append(nexNode.load)
nexNode = nexNode.previous
days = 1
curr = stamina
counter = 0
stamina_to_use.reverse()
# Count the number of days needed for finishing
while stamina_to_use:
tocheck = stamina_to_use.pop()
# print("Current days: {}, current stamina: {},stamina to withdraw {}".format(
# days, curr, tocheck))
if curr > stamina:
print(-1)
break
if (curr - tocheck) < 0:
days += 1
curr = stamina
curr -= tocheck
if found:
print(days)
else:
print(-1)
else:
print(-1)
The results are actually as I expected, I get the shortest path and get the right number of days according to my own test cases and also to the ones on kattis. But for some reason when I submit the project to kattis the first 8 or so test cases pass and then I suddenly get: "Wrong answer", I don't know where the mistake in my thinking or code is. Is my approach the right one or should I use a different one. Or is there just a simple mistake made in counting the days?
Thanks in advance
Here is a solution that almost works, there is a small check missing.
This way you need to figure out what it does :)
BIG_NUMBER = 9999
STAMINA_COST = {'F': 2, 'M': 3}
def maybe_update(field1, field2, maze, time_taken, n, m, k, updated_fields):
i1, j1 = field1
i2, j2 = field2
if not (0 <= i2 < n and 0 <= j2 < m):
# Out of bounds
return
field = maze[i2][j2]
if field == '#':
# Can not walk on river
return
days_taken, stamina_taken = time_taken[i1][j1]
stamina_to_move = STAMINA_COST.get(field, 1)
stamina_taken += stamina_to_move
if k < stamina_taken:
days_taken += 1
stamina_taken = stamina_to_move
new_time_taken = (days_taken, stamina_taken)
if new_time_taken < time_taken[i2][j2]:
time_taken[i2][j2] = new_time_taken
updated_fields.add((i2, j2))
def main():
# Read input
n, m, k = map(int, input().split())
maze = []
for i in range(n):
line = input()
maze.append(line)
# Create map of how long it takes to get somewhere
# Each field has (days_spent, stamina_spent)
time_taken = []
for i in range(n):
time_taken.append([(BIG_NUMBER, BIG_NUMBER) for j in range(m)])
# Look for the start and mark it as (1, 0), also look for the gold
updated_fields = set()
for i in range(n):
for j in range(m):
if maze[i][j] == 'S':
time_taken[i][j] = (1, 0)
updated_fields.add((i, j))
elif maze[i][j] == 'G':
gold_at = (i, j)
# BFS to propagate time_taken
while updated_fields:
i, j = updated_fields.pop()
maybe_update((i, j), (i + 1, j), maze, time_taken, n, m, k, updated_fields)
maybe_update((i, j), (i - 1, j), maze, time_taken, n, m, k, updated_fields)
maybe_update((i, j), (i, j + 1), maze, time_taken, n, m, k, updated_fields)
maybe_update((i, j), (i, j - 1), maze, time_taken, n, m, k, updated_fields)
# Print days taken to get to the gold
i, j = gold_at
days_taken = time_taken[i][j][0]
print(-1 if days_taken == BIG_NUMBER else days_taken)
if __name__ == '__main__':
main()