Because the verb and noun indicate the address of the data that will be used, there won't be any relation between two consecutive values. I think for this one bruteforcing is the (only?) way to go.
-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
Advent of Code 2019...save Santa and Xmas with programming!
- Thread starter ElFly
- Start date
Day 7: https://github.com/Randdalf/aoc19/blob/master/d07.py
More intcode! Had to reread the puzzle a few times before I fully understood what it was asking.
More intcode! Had to reread the puzzle a few times before I fully understood what it was asking.
For day 3 part 1, I have a question to ask for clarification. I have written lines of codes that will check for intersections between wires. Now I am starting at a central port of (0, 0) and then extending the wires from there. I haven noticed for the first practice data set of:
Given a (0, 0) starting point, there appears to be an intersection at (158, -12) which created a distance of 146. That's shorter than the accepted distance of 159. Are we to assume that the wires cannot go in the left or below the starting point? Given the example graphs they provided, it's not clear whether they are allowing the wire to go left/down the starting point. I do not believe I made a mistake coming up with the intersecting points given all the other test datas did not give me this problem. Is that also something you guys faced when doing day 3 part 1?
Given a (0, 0) starting point, there appears to be an intersection at (158, -12) which created a distance of 146. That's shorter than the accepted distance of 159. Are we to assume that the wires cannot go in the left or below the starting point? Given the example graphs they provided, it's not clear whether they are allowing the wire to go left/down the starting point. I do not believe I made a mistake coming up with the intersecting points given all the other test datas did not give me this problem. Is that also something you guys faced when doing day 3 part 1?
Your code for calculating Manhattan distance is wrong. That point is 170 from the central port.
Oh my you're right... haha!!! Wow it's late here and my brain is not working...
I rewrote my day 6 part 2 and it's way simpler now: https://github.com/Randdalf/aoc19/blob/master/d06.py
I ended up with something similar to what you described.
I tried two ways of finding the path: inverting my orbits dictionary then working backwards, and recursive DFS. The latter seems just as fast and is less code, so it won out for me.
My abomination took a bit of time to calculate the Manhattan distance from the real data set but it worked :O Now I just need to optimize it... but scanning part B's problem set makes me think my current set up is best suited to tackle that lol
OP
OP
did day 7 part a, may do part b later cause it sounds complex
even part a was complex to figure out. had skipped a rule and gotten v wrong answers
edit: did part b. It was hard
in the end I was sharing state between the _amplifiers_ cause I had not properly copied the array of the state, just a reference somehow
after that I had to put a try / except in the code cause some of the inputs caused out of bounds index reads. Then finally got it right
woop woop!
even part a was complex to figure out. had skipped a rule and gotten v wrong answers
edit: did part b. It was hard
in the end I was sharing state between the _amplifiers_ cause I had not properly copied the array of the state, just a reference somehow
after that I had to put a try / except in the code cause some of the inputs caused out of bounds index reads. Then finally got it right
woop woop!
Last edited:
Fuck me, I'm stuck on part 2 :-/
So I'm having a big while loop until one of the thrusters ends his intcode application. I have different intcode instances for each of the thrusters, and whenever a thruster asks for input that isn't available, that thruster pauses execution, and I pass the last given output as input to the next one. Whenever I'm in the second feedback loop, I just continue from the address that asked for input with the last output. So either my logic is wrong, or I have made a mistake somewhere.
Does anybody have some output to compare for the first example? (Input: 3,26,1001,26,-4,26,3,27,1002,27,2,27,1,27,26,27,4,27,1001,28,-1,28,1005,28,6,99,0,0,5)
So I'm having a big while loop until one of the thrusters ends his intcode application. I have different intcode instances for each of the thrusters, and whenever a thruster asks for input that isn't available, that thruster pauses execution, and I pass the last given output as input to the next one. Whenever I'm in the second feedback loop, I just continue from the address that asked for input with the last output. So either my logic is wrong, or I have made a mistake somewhere.
Does anybody have some output to compare for the first example? (Input: 3,26,1001,26,-4,26,3,27,1002,27,2,27,1,27,26,27,4,27,1001,28,-1,28,1005,28,6,99,0,0,5)
Engine 0 - phase 9
> Input: 9, 0
> Demands more input
Engine 1 - phase 8
> Input: 8, 5
> Demands more input
Engine 2 - phase 7
> Input: 7, 14
> Demands more input
Engine 3 - phase 6
> Input: 6, 31
> Demands more input
Engine 4 - phase 5
> Input: 5, 64
> Demands more input
> Output: 129
Engine 0 - phase 9
> Input: 129
> Demands more input
Engine 1 - phase 8
> Input: 263
> Demands more input
Engine 2 - phase 7
> Input: 530
> Demands more input
Engine 3 - phase 6
> Input: 1063
> Demands more input
Engine 4 - phase 5
> Input: 2128
> Demands more input
> Output: 4257
Engine 0 - phase 9
> Input: 4257
> Demands more input
Engine 1 - phase 8
> Input: 8519
> Demands more input
Engine 2 - phase 7
> Input: 17042
> Demands more input
Engine 3 - phase 6
> Input: 34087
> Demands more input
Engine 4 - phase 5
> Input: 68176
> Demands more input
> Output: 136353
Engine 0 - phase 9
> Input: 136353
> Demands more input
Engine 1 - phase 8
> Input: 272711
> Demands more input
Engine 2 - phase 7
> Input: 545426
> Demands more input
Engine 3 - phase 6
> Input: 1090855
> Demands more input
Engine 4 - phase 5
> Input: 2181712
> Demands more input
> Output: 4363425
Engine 0 - phase 9
> Input: 4363425
> End
> Input: 9, 0
> Demands more input
Engine 1 - phase 8
> Input: 8, 5
> Demands more input
Engine 2 - phase 7
> Input: 7, 14
> Demands more input
Engine 3 - phase 6
> Input: 6, 31
> Demands more input
Engine 4 - phase 5
> Input: 5, 64
> Demands more input
> Output: 129
Engine 0 - phase 9
> Input: 129
> Demands more input
Engine 1 - phase 8
> Input: 263
> Demands more input
Engine 2 - phase 7
> Input: 530
> Demands more input
Engine 3 - phase 6
> Input: 1063
> Demands more input
Engine 4 - phase 5
> Input: 2128
> Demands more input
> Output: 4257
Engine 0 - phase 9
> Input: 4257
> Demands more input
Engine 1 - phase 8
> Input: 8519
> Demands more input
Engine 2 - phase 7
> Input: 17042
> Demands more input
Engine 3 - phase 6
> Input: 34087
> Demands more input
Engine 4 - phase 5
> Input: 68176
> Demands more input
> Output: 136353
Engine 0 - phase 9
> Input: 136353
> Demands more input
Engine 1 - phase 8
> Input: 272711
> Demands more input
Engine 2 - phase 7
> Input: 545426
> Demands more input
Engine 3 - phase 6
> Input: 1090855
> Demands more input
Engine 4 - phase 5
> Input: 2181712
> Demands more input
> Output: 4363425
Engine 0 - phase 9
> Input: 4363425
> End
Okay, finally got part 2 as working well. Turns out that (for that example) the fifth time it gets to thruster A, the intcode executed an output operation and then a bit later stopped at the ending opcode. For me that meant that the feedback loop should stop and I should use the previous output value of thruster E. But apparently whenever an output was made, I should've moved immediately to the next thruster...
And here I was, thinking that finding all possible combinations of the phases would've been the hardest part today :')
And here I was, thinking that finding all possible combinations of the phases would've been the hardest part today :')
Randdalf seeing as you are doing this in python as well, I thought I'd share this with you and see what you think. So I started reworking how to detect day3's intersection puzzle by starting simple: Get a single wire to be able to detect its own intersections.
I used one of the example data set here which has three intersecting points of:
Now, I went online and searched around and found existing algorithm for solving lines intersection via four coordinates. Line 1's (x1, y1) to (x2, y2) and Line2's (x1, y1) to (x2, y2). Anyways, the result I got from using the algorithm gave me these results.
Notice that the 3 real intersecting points are indeed in that larger list. I know the results got a lot of extra intersecting points and some are from the fact that the purpose of this data sample have the same "starting point" at each turn. Given the way that I am checking for a single wire intersecting itself, which is that every time it moves, it is checking for intersection from its past moves. Eliminating that, I still got some other unexpected points that I am still trying to figure out where they came from. I figured, maybe you wouldn't mind taking a look at it if you got a moment and see what you think?
The sources of the algorithms are from here:
I used one of the example data set here which has three intersecting points of:
(158, 4), (158, 11), (146, 11)Now, I went online and searched around and found existing algorithm for solving lines intersection via four coordinates. Line 1's (x1, y1) to (x2, y2) and Line2's (x1, y1) to (x2, y2). Anyways, the result I got from using the algorithm gave me these results.
(0, 0), (75, -30), (0, 0), (30, -30), (75, 53), (158, 53), (0, 0), (-51, -30), (-13, 53), (75, 4), (158, 4), (146, 4), (0, 0), (355, -30), (90, 53), (-41, 4), (75, 11), (158, 11), (146, 11), (217, 11)Notice that the 3 real intersecting points are indeed in that larger list. I know the results got a lot of extra intersecting points and some are from the fact that the purpose of this data sample have the same "starting point" at each turn. Given the way that I am checking for a single wire intersecting itself, which is that every time it moves, it is checking for intersection from its past moves. Eliminating that, I still got some other unexpected points that I am still trying to figure out where they came from. I figured, maybe you wouldn't mind taking a look at it if you got a moment and see what you think?
The sources of the algorithms are from here:
- https://www.topcoder.com/community/...cepts-line-intersection-and-its-applications/
- https://stackoverflow.com/questions/20677795/how-do-i-compute-the-intersection-point-of-two-lines
Python:
class WireObject(object):
START_POS = (0, 0)
X_MOVE, Y_MOVE = range(2)
def __init__(self):
self.posX = WireObject.START_POS[0]
self.posY = WireObject.START_POS[-1]
self.history = [(self.posX, self.posY)]
self.intersectList = []
def Move(self, x=0, y=0):
self.posY += y
self.posX += x
self.GetIntersect(self.history[-1], (self.posX, self.posY))
self.history.append(self.Position)
return self.Position
def _getIntersect(self, posA, posB, posC, posD):
def _getLineEquation(pos1, pos2):
a = pos2[1] - pos1[1]
b = pos1[0] - pos2[0]
c = (a * pos1[0]) + (b * pos1[1])
return a, b, c
a1, b1, c1 = _getLineEquation(posA, posB)
a2, b2, c2 = _getLineEquation(posC, posD)
det = (a1 * b2) - (a2 * b1)
if det != 0:
x = (b2*c2 - b1*c2)/det
y = (a1*c2 - a2*c1)/det
return (x, y)
return None
def GetIntersect(self, from_pos, to_pos):
# Loop through previous positions and play through it to find if there
# is any line that intersects
for count, histPos in enumerate(self.history):
if count == 0:
continue
intersect = self._getIntersect(from_pos, to_pos, self.history[count - 1], histPos)
if intersect:
self.intersectList.append(intersect)
return self.intersectList
@property
def Intersects(self):
return self.intersectList
@property
def Distance(self):
x, y = self.Position
return int(abs(x) + abs(y))
@property
def History(self):
return self.history
@property
def Position(self):
return (self.posX, self.posY)
class MoveMachine(object):
RIGHT = 'R'
LEFT = 'L'
UP = 'U'
DOWN = 'D'
def __init__(self,):
self.allWires = []
def AddWire(self):
self.wireObj = WireObject()
self.allWires.append(self.wireObj)
return self.wireObj
def MoveWire(self, wireObj, direction, movement):
if direction == MoveMachine.RIGHT:
wireObj.Move(movement, 0)
if direction == MoveMachine.LEFT:
wireObj.Move(-movement, 0)
if direction == MoveMachine.UP:
wireObj.Move(0, movement)
if direction == MoveMachine.DOWN:
wireObj.Move(0, -movement)
return wireObj.Position
def MoveWireByData(self, wireObj, data):
for i in data:
direction = i[0]
move = int(i.replace(direction, ''))
self.MoveWire(wireObj, direction, move)
return wireObj
def Wires(self):
return self.allWires
if __name__ == '__main__':
dataList = [["R75", "D30", "R83", "U83", "L12", "D49", "R71", "U7", "L72"]]
machine = MoveMachine()
for data in dataList:
wire = machine.AddWire()
machine.MoveWireByData(wire, data)
print 'Intersects', wire.Intersects
Last edited:
The line intersection algorithm you're using is for lines that are infinitely long. Try this one instead:
Code:
def _getIntersect(self, posA, posB, posC, posD):
if (posA[0] == posB[0]):
tmp = posA
posA = posC
posC = tmp
tmp = posB
posB = posD
posD = tmp
if (posC[0] == posD[0]):
if (posA[0] == posB[0]):
return None
if (posA[0] > posB[0]):
tmp = posA
posA = posB
posB = tmp
if (posC[1] > posD[1]):
tmp = posC
posC = posD
posD = tmp
if (posC[0] > posA[0] and posC[0] < posB[0] and posA[1] > posC[1] and posA[1] < posD[1]):
return (posC[0], posA[1])
return None
Got my first wrong answer with part 2 of the image decoding. Not because I failed to decode the image, because I failed to read the image (hint: if it's anything like mine, it's easier to read with black and white reversed).
Find myself wondering whether it's worth creating a generic image decoder in case they're used in future tasks. Current implementation is crazily hardcoded.
Find myself wondering whether it's worth creating a generic image decoder in case they're used in future tasks. Current implementation is crazily hardcoded.
Last edited:
Day 8: https://github.com/Randdalf/aoc19/blob/master/d08.py
Not really that interesting, this one.
Not really that interesting, this one.
I guess if you really wanted to you could write something which would read the digits from the image.
Solving that puzzle using line segment intersection is a very challenging way of solving that puzzle. Particularly when it comes to the second part of the puzzle. But yeah, like RiOrius suggested above: you want line segment intersection not infinite line intersection.
Wow I was close to flipping a table on Day 3, then I figured it out. NUMPY AND SCIPY DON'T LIE
[[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 2. 2. 2. 2. 2. 2. 2. 0. 0. 0.]
[0. 2. 0. 0. 0. 0. 0. 2. 0. 0. 0.]
[0. 2. 0. 0. 1. 1. 1. 8. 1. 1. 0.]
[0. 2. 0. 0. 1. 0. 0. 2. 0. 1. 0.]
[0. 2. 0. 2. 8. 2. 2. 2. 0. 1. 0.]
[0. 2. 0. 0. 1. 0. 0. 0. 0. 1. 0.]
[0. 2. 0. 0. 0. 0. 0. 0. 0. 1. 0.]
[0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]] L4
(8, 1) [3 5] [7 4] {4: (3, 7), 5: (5, 4)}
manhattan_distance = [[6.]]
[[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 2. 2. 2. 2. 2. 2. 2. 0. 0. 0.]
[0. 2. 0. 0. 0. 0. 0. 2. 0. 0. 0.]
[0. 2. 0. 0. 1. 1. 1. 8. 1. 1. 0.]
[0. 2. 0. 0. 1. 0. 0. 2. 0. 1. 0.]
[0. 2. 0. 2. 8. 2. 2. 2. 0. 1. 0.]
[0. 2. 0. 0. 1. 0. 0. 0. 0. 1. 0.]
[0. 2. 0. 0. 0. 0. 0. 0. 0. 1. 0.]
[0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]] L4
(8, 1) [3 5] [7 4] {4: (3, 7), 5: (5, 4)}
manhattan_distance = [[6.]]
Finished part 1 and it's almost 3am. As long as you're comfortable working with numpy arrays that almost exceed base64 - it's easy lol
OP
OP
did day 8 part 1,it was easy
part 2 has me stuck cause I get non sense images
gotta fiddle with it a little
edit: mm, inverting colors ain't helping
edit2: I wasn't using all my layers (wtf?). inverting did help once I did that
edit3: Oh, I lucked out in part 1; I was looking in the first few layers, and somehow the minimum value was in there. goddamnit
part 2 has me stuck cause I get non sense images
gotta fiddle with it a little
edit: mm, inverting colors ain't helping
edit2: I wasn't using all my layers (wtf?). inverting did help once I did that
edit3: Oh, I lucked out in part 1; I was looking in the first few layers, and somehow the minimum value was in there. goddamnit
Last edited:
OP
OP
the first iteration ends in 5, 5, 5 instead of 0, 0, 5
especifically:
[3, 26, 1001, 26, -4, 26, 3, 27, 1002, 27, 2, 27, 1, 27, 26, 27, 4, 27, 1001, 28, -1, 28, 1005, 28, 6, 99, 5, 5, 5]
[3, 26, 1001, 26, -4, 26, 3, 27, 1002, 27, 2, 27, 1, 27, 26, 27, 4, 27, 1001, 28, -1, 28, 1005, 28, 6, 99, 5, 263, 4]
yeah, you gotta store both the state AND where you were on that state, but be careful to add to the address just before the amplifier "returns", otherwise when you restart it it will return immediately
other possible problem that i got is that I was sharing state _between_ amplifiers instead of keeping them separate per amplifier. Same with the suspended address
edit:to be clear, you gotta keep the states separated per amplifier. I am saying I got in trouble for not doing that
Last edited:
You're sharing the IntCode program between multiple amplifiers that modify it? That would be a big problem, yes. Each amplifier must be running its own copy of the code.
OP
OP
Yeah, I fucked up cause I assumed
states = [initial_state, initial_state, initial_state, initial_state, initial_state]
would get me independent copies of the states, instead of
states = [initial_state.copy(), initial_state.copy(), initial_state.copy(), initial_state.copy(), initial_state.copy()]
states = [initial_state, initial_state, initial_state, initial_state, initial_state]
would get me independent copies of the states, instead of
states = [initial_state.copy(), initial_state.copy(), initial_state.copy(), initial_state.copy(), initial_state.copy()]
I "wrote" an intcode computer ... in intcode: https://www.reddit.com/r/adventofco...code/?utm_medium=android_app&utm_source=share
Surely we need to go back and solve some of the other problems in IntCode...
Okay, here's a hand-crafted solution to part one of day 1 written in IntCode (as per the day 5 version of the spec):
Takes as input the values from input data, with a -1 for end of input.
Code:
3,45,1008,45,-1,7,1105,7,42,1101,0,0,46,1001,45,-3,45,1007,45,0,22,1105,22,31,101,1,46,46,1105,1,13,1,46,47,47,101,-2,47,47,1105,1,0,4,47,99,0,0,0Takes as input the values from input data, with a -1 for end of input.
Finally finish the rewrite. I haven't looked into the reason why, but the _getIntersect() from Randdalf would work on the wire checking if it is intersecting itself while drawing. However, it won't work when I am doing line-with-line comparison. So I basically wrote a simpler way for that with some inspiration Randdalf's solution. Did part B as well and it all seems to work now. Feels like it's put together by duct tapes though, haha! So who here feel like they can totally apply to SpaceX or NASA now and become a rocket scientist? š Just kidding!
Python:
from math import sqrt, pow
class WireObject(object):
START_POS = (0, 0)
X_MOVE, Y_MOVE = range(2)
def __init__(self):
self.posX = WireObject.START_POS[0]
self.posY = WireObject.START_POS[-1]
self.history = [(self.posX, self.posY)]
self.intersectList = []
self.steps = []
def Move(self, x=0, y=0):
if x:
self.steps.append(abs(x))
else:
self.steps.append(abs(y))
self.posY += y
self.posX += x
self.GetIntersect(self.history[-1], (self.posX, self.posY))
self.history.append(self.Position)
return self.Position
def _getIntersect(self, posA, posB, posC, posD):
if (posA[0] == posB[0]):
tmp = posA
posA = posC
posC = tmp
tmp = posB
posB = posD
posD = tmp
if (posC[0] == posD[0]):
if (posA[0] == posB[0]):
return None
if (posA[0] > posB[0]):
tmp = posA
posA = posB
posB = tmp
if (posC[1] > posD[1]):
tmp = posC
posC = posD
posD = tmp
if (posC[0] > posA[0] and posC[0] < posB[0] and posA[1] > posC[1] and posA[1] < posD[1]):
return (posC[0], posA[1])
return None
def GetIntersect(self, from_pos, to_pos):
# Loop through previous positions and play through it to find if there
# is any line that intersects
for count, histPos in enumerate(self.history):
if count < 0:
continue
intersect = self._getIntersect(
from_pos, to_pos, self.history[count - 1], histPos)
if intersect:
self.intersectList.append(intersect)
return self.intersectList
@property
def Intersects(self):
return self.intersectList
@property
def Distance(self):
x, y = self.Position
return int(abs(x) + abs(y))
@property
def History(self):
return self.history
@property
def Position(self):
return (self.posX, self.posY)
def GetSteps(self, stop=None):
if stop:
return sum(self.steps[:stop])
return sum(self.steps)
class MoveMachine(object):
RIGHT = 'R'
LEFT = 'L'
UP = 'U'
DOWN = 'D'
def __init__(self,):
self.allWires = []
self.intersectList = []
def AddWire(self):
self.wireObj = WireObject()
self.allWires.append(self.wireObj)
self.intersectList = []
self.stepsList = []
return self.wireObj
def MoveWire(self, wireObj, direction, movement):
if direction == MoveMachine.RIGHT:
wireObj.Move(movement, 0)
if direction == MoveMachine.LEFT:
wireObj.Move(-movement, 0)
if direction == MoveMachine.UP:
wireObj.Move(0, movement)
if direction == MoveMachine.DOWN:
wireObj.Move(0, -movement)
return wireObj.Position
def MoveWireByData(self, wireObj, data):
for i in data:
direction = i[0]
move = int(i.replace(direction, ''))
self.MoveWire(wireObj, direction, move)
return wireObj
def _getIntersect(self, posA, posB, posC, posD):
line1x = posB[0] - posA[0]
line1y = posB[1] - posA[1]
line2x = posD[0] - posC[0]
line2y = posD[1] - posC[1]
x = None
y = None
if line1x and line2y:
if line2y > 0:
yRange = range(posC[1], posD[1] + 1)
else:
yRange = range(posD[1], posC[1] + 1)
if posA[1] in yRange:
if line1x > 0:
xRange = range(posA[0], posB[0] + 1)
else:
xRange = range(posB[0], posA[0 + 1])
if posC[0] in xRange:
x = posD[0]
y = posB[1]
if line1y and line2x:
if line2x > 0:
xRange = range(posC[0], posD[0] + 1)
else:
xRange = range(posD[0], posC[0] + 1)
if posA[0] in xRange:
if line1y > 0:
yRange = range(posA[1], posB[1] + 1)
else:
yRange = range(posB[1], posA[1] + 1)
if posC[1] in yRange:
x = posB[0]
y = posD[1]
if x and y:
return (x, y)
return None
def GetIntersects(self, wireObj):
wireAsteps = 0
wireBsteps = 0
for allWire in self.allWires:
if allWire != wireObj:
for count, thisWireHistPos in enumerate(wireObj.History):
if count < 0:
continue
from_pos = wireObj.History[count - 1]
to_pos = thisWireHistPos
for i, histPos in enumerate(allWire.History):
if i < 0:
continue
intersect = self._getIntersect(
from_pos, to_pos, allWire.History[i - 1], histPos)
if intersect:
self.intersectList.append(intersect)
wireAsteps = wireObj.GetSteps(count - 1)
wireAsteps += self._getDistanceBetweenTwoPoints(
from_pos, intersect)
wireBsteps = allWire.GetSteps(i - 1)
wireBsteps += self._getDistanceBetweenTwoPoints(
allWire.History[i - 1], intersect)
self.stepsList.append(int(wireAsteps + wireBsteps))
return self.intersectList
def _getDistanceBetweenTwoPoints(self, from_pos, to_pos):
x1 = from_pos[0]
y1 = from_pos[1]
x2 = to_pos[0]
y2 = to_pos[1]
distance = sqrt(pow((x2-x1), 2) + pow((y2-y1), 2))
return distance
def Wires(self):
return self.allWires
def DistanceToStart(self, x, y):
return int(abs(x) + abs(y))
def GetShortestDistanceToStart(self):
distances = []
if not self.intersectList:
for wireObj in self.allWires:
self.GetIntersects(wireObj)
for intersect in self.intersectList:
distances.append(self.DistanceToStart(intersect[0], intersect[1]))
if not distances:
return 0
return sorted(distances)[0]
def GetShortestDistanceFromStepsToStart(self):
stepDict = {}
if not self.intersectList:
for wireObj in self.allWires:
self.GetIntersects(wireObj)
for steps, intersect in zip(self.stepsList, self.intersectList):
stepDict[steps] = intersect
if not stepDict:
return 0
return sorted(self.stepsList)[0]
def SolveA(dataSet):
machine = MoveMachine()
for data in dataSet:
wire = machine.AddWire()
machine.MoveWireByData(wire, data)
return machine.GetShortestDistanceToStart()
def SolveB(dataSet):
machine = MoveMachine()
for data in dataSet:
wire = machine.AddWire()
machine.MoveWireByData(wire, data)
return machine.GetShortestDistanceFromStepsToStart()
def main():
dataSet = []
with open('day3-wires-data.txt', 'r') as f:
lines = f.readlines()
for line in lines:
dataSet.append([i.replace('\n', '') for i in line.split(',')])
f.close()
print SolveA(dataSet)
print SolveB(dataSet)
print 'Solved.'
if __name__ == '__main__':
main()
Last edited:
OP
OP
This is awesome.
As per today's spec, you can actually drop the trailing 0s.
I think building an own interpreter is very cool. I'm wondering if they have any further plans with it. I'm trying to do 25 different languages, so it would be best for me if they decided to drop the interpreter. I've got a version in Haskell, Ruby, Rust and D so far.
Good stuff. As PHP's integers are big enough by default, I only lost time because I forgot to take the relative argument mode into account when writing values as well. I'm guessing the intcode computer is now complete, and for day 11 they're gonna ask us to write some intcode ourselves, maybe?
What sort of idiot would forget to use relative mode for writes?
Yeah, me too...
The text says you now have a complete intcode computer, so seems like a good guess (unless we need to expand it, maybe with an FPU). Writing intcode seems unlikely, since an intcode program to do one thing can be written in a variety of ways - my intcode solution for day 1 part 1 could have been written with the same basic instructions but quite a lot of changes (switch position/immediate register types, initialise code that's going to be modified before use to different values etc...) and that's just for a basically trivial piece of code. Anything more interesting would be a pain to test.
Part 2 was a disappointment, though. Was expecting something more challenging than
run the same program again with no changes, just providing input as 2 instead of 1
Worst thing is that the application first threw an exception - one I wrote - telling me output could only be done in position mode. No bells were rung in my brain at that time and I just included relative mode in my check -_-
I'm guessing the creators have a working intcode processor and can easily check the output ;) But yeah, doesn't sound like the most fun idea
I'm guessing the creators have a working intcode processor and can easily check the output ;) But yeah, doesn't sound like the most fun idea
Running an interpreted program that might potentially never terminate on the server to check everyone's answer sounds a bit painful. ;) Also does away with the principle you can use whatever language you like. Though I've seen someone on Reddit using IntCode for at least most of them.
Phew, finally caught up. I am trying really hard to write clean, readable Python code while avoiding any additional libraries, but sometimes it still gets a bit messy.
I think the difficulty escalated a bit early with the Intcode interpreter on day 5. I spent a long time future proofing that one, because I feared what would come next. But it really paid off. Day 7 and 9 were barely a problem after that. Let's see how they are going to escalate things...
If anyone in this thread is interested:
I think the difficulty escalated a bit early with the Intcode interpreter on day 5. I spent a long time future proofing that one, because I feared what would come next. But it really paid off. Day 7 and 9 were barely a problem after that. Let's see how they are going to escalate things...
If anyone in this thread is interested:
Hidden content
That would suggest you have a bug in your implementation that doesn't break the test input. You might find part 2 doesn't work...
Ya, I just had an issue with immediate mode for the output opcode, all better now. At least it was broken early in the stuff and not further into the run.