tim
Veteran
I had to write a sq-1 scrambler for cubemania, but couldn't find anyone on the web. Except for Jaap's Javascript scrambler. I first tried to convert his scrambler, but it was so boring and didn't make fun at all. So i wrote my own one.
i use up_layer and down_layer for the two layers and both arrays describe the permutation of corners/edges with the numbers 30/60. So each layer array has the sum 360.
Since i used my own notation during the scrambler ("(1,0) means turning the upper face by one piece"), i had to convert it back. That's what "humanize_sq_one_move" is for.
P.S.: Yeah, it looks ugly for ruby code.
i use up_layer and down_layer for the two layers and both arrays describe the permutation of corners/edges with the numbers 30/60. So each layer array has the sum 360.
Since i used my own notation during the scrambler ("(1,0) means turning the upper face by one piece"), i had to convert it back. That's what "humanize_sq_one_move" is for.
Code:
def square1_scramble
scramble = []
up_layer = (0..7).map{|i| i%2 == 0 ? 30 : 60}
down_layer = [up_layer].flatten!
length = 0
begin
up_moves = possible_moves up_layer
down_moves = possible_moves down_layer
up_move = up_moves.rand
down_moves.delete 0 if up_move == 0
down_move = down_moves.rand
scramble << [humanize_sq_one_move(up_layer, up_move), humanize_sq_one_move(down_layer, down_move) * -1]
do_move up_layer, up_move
do_move down_layer, down_move
length += up_move == 0 ? 0 : 1
length += down_move == 0 ? 0 : 1
do_slice(up_layer, down_layer)
length += 1
end while length <= scramble_length + 1
scramble.map {|s| "(#{s.join(',')})"}.join(' ')
end
def humanize_sq_one_move(layer, move)
move = layer[0..move - 1].inject(0){|sum, x| x == 30 ? sum + 1 : sum + 2} unless move == 0
move > 6 ? move - 12 : move
end
def possible_moves(layer)
layer_moves = []
layer.length.times do |start|
sum = 0
possible = false
layer.length.times do |i|
sum += layer[(start + i) % layer.length]
possible = true if sum == 180
end
layer_moves << start if possible
end
layer_moves
end
def do_move(layer, l)
l %= layer.length
l.times do
layer << layer.shift
end
layer
end
def do_slice(up, down)
sum = 0
small_up, big_up = up.partition {|n| sum += n; sum <= 180}
small_down, big_down = down.partition {|n| sum += n; sum <= 540}
up.replace(small_down.reverse + big_up)
down.replace(small_up.reverse + big_down)
endP.S.: Yeah, it looks ugly for ruby code.
