Reduce array to set number of items? [closed] - ruby

Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 8 years ago.
Improve this question
I have a Ruby array of 300 items. I want to reduce that array down to a set number of items, evenly picked from the array.
The number of items in the array will not be the same every time, nor will the number of items needed.
Something like this:
arr = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
num_of_items = 4
final_arr = [0, 5, 10, 15]

You can use Enumerable#each_slice
arr = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
#=> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
num_of_items = 4
#=> 4
arr.each_slice(arr.size/num_of_items + 1).map(&:first)
#=> [0, 5, 10, 15]
arr = (0..16).to_a
#=> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16]
num_of_items = 5
#=> 5
arr.each_slice(arr.size/num_of_items + 1).map(&:first)
#=> [0, 4, 8, 12, 16]
OR
Numeric#step
arr = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
#=> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
num_of_items = 4
#=> 4
arr.first.step(arr.size, arr.size/num_of_items + 1).map { |i| arr[i] }
#=> [0, 5, 10, 15]
arr = (0..16).to_a
#=> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16]
num_of_items = 5
#=> 5
arr.first.step(arr.size, arr.size/num_of_items + 1).map { |i| arr[i] }
#=> [0, 4, 8, 12, 16]

If you're picking a given number of items at random from this array, use sample -
$ arr = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
$ arr.sample(5)
=> [1, 3, 5, 4, 12]
$ arr.sample(5)
=> [15, 6, 13, 5, 11]

If the goal of your algorithm is exactly as shown above, you could use in_groups_of, then grab the first element of each child array:
1.9.3p484 :014 > num_of_items = 5
=> 5
1.9.3p484 :011 > arr = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
=> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
1.9.3p484 :012 > arr.in_groups_of(num_of_items)
=> [[0, 1, 2, 3, 4], [5, 6, 7, 8, 9], [10, 11, 12, 13, 14], [15, nil, nil, nil, nil]]
1.9.3p484 :013 > arr.in_groups_of(num_of_items).map(&:first)
=> [0, 5, 10, 15]

Related

split an array which comtains partially relatively order into two sorted array in O(n) time

Assume I have two arrays, both of them are sorted, for example:
A: [1, 4, 5, 8, 10, 24]
B: [3, 6, 9, 29, 50, 65]
And then I merge these two array into one array and keep original relative order of both two array
C: [1, 4, 3, 5, 6, 9, 8, 29, 10, 24, 50, 65]
Is there any way to split C into two sorted array in O(n) time?
note: not necessarily into the original A and B
Greedily assign your integers to list 1 if they can go there. If they can't, assign them to list 2.
Here's some Ruby code to play around with this idea. It randomly splits the integers from 0 to n-1 into two sorted lists, then randomly merges them, then applies the greedy approach.
def f(n)
split1 = []
split2 = []
0.upto(n-1) do |i|
if rand < 0.5
split1.append(i)
else
split2.append(i)
end
end
puts "input 1: #{split1.to_s}"
puts "input 2: #{split2.to_s}"
merged = []
split1.reverse!
split2.reverse!
while split1.length > 0 && split2.length > 0
if rand < 0.5
merged.append(split1.pop)
else
merged.append(split2.pop)
end
end
merged += split1.reverse
merged += split2.reverse
puts "merged: #{merged.to_s}"
merged.reverse!
greedy1 = [merged.pop]
greedy2 = []
while merged.length > 0
if merged[-1] >= greedy1[-1]
greedy1.append(merged.pop)
else
greedy2.append(merged.pop)
end
end
puts "greedy1: #{greedy1.to_s}"
puts "greedy2: #{greedy2.to_s}"
end
Here's sample output:
> f(20)
input 1: [2, 3, 4, 5, 8, 9, 10, 18, 19]
input 2: [0, 1, 6, 7, 11, 12, 13, 14, 15, 16, 17]
merged: [2, 0, 1, 6, 3, 4, 5, 8, 9, 7, 10, 11, 18, 12, 13, 19, 14, 15, 16, 17]
greedy1: [2, 6, 8, 9, 10, 11, 18, 19]
greedy2: [0, 1, 3, 4, 5, 7, 12, 13, 14, 15, 16, 17]
> f(20)
input 1: [1, 3, 5, 6, 8, 9, 10, 11, 13, 15]
input 2: [0, 2, 4, 7, 12, 14, 16, 17, 18, 19]
merged: [0, 2, 4, 7, 12, 14, 16, 1, 3, 5, 6, 8, 17, 9, 18, 10, 19, 11, 13, 15]
greedy1: [0, 2, 4, 7, 12, 14, 16, 17, 18, 19]
greedy2: [1, 3, 5, 6, 8, 9, 10, 11, 13, 15]
> f(20)
input 1: [0, 1, 2, 6, 7, 9, 11, 14, 15, 18]
input 2: [3, 4, 5, 8, 10, 12, 13, 16, 17, 19]
merged: [3, 4, 5, 8, 10, 12, 0, 13, 16, 17, 1, 19, 2, 6, 7, 9, 11, 14, 15, 18]
greedy1: [3, 4, 5, 8, 10, 12, 13, 16, 17, 19]
greedy2: [0, 1, 2, 6, 7, 9, 11, 14, 15, 18]
Let's take your example.
[1, 4, 3, 5, 6, 9, 8, 29, 10, 24, 50, 65]
In time O(n) you can work out the minimum of the tail.
[1, 3, 3, 5, 6, 8, 8, 10, 10, 24, 50, 65]
And now the one stream is all cases where it is the minimum, and the other is the cases where it isn't.
[1, 3, 5, 6, 8, 10, 24, 50, 65]
[ 4, 9, 29, ]
This is all doable in time O(n).
We can go further and now split into 3 streams based on which values in the first stream could have gone in the last without changing it being increasing.
[ 3, 5, 6, 8, 10, 24, ]
[1, 5, 6, 8, 50, 65]
[ 4, 9, 29, ]
And now we can start enumerating the 2^6 = 64 different ways of splitting the original stream back into 2 increasing streams.

How to prove this josephus problem variation is a np-complete problem?

I have a problem that is a Josephus problem variation. It is described below:
There are m cards with number from 1 to m,and each of them has a unique number. The cards are dispatched to n person who sit in a circle. Note that m >= n.
Then we choose the person "A" who sits at the position "p" to out of the circle, just like the Josephus problem does. Next step we skip "k" person at the right of p while k is the number of the card toked by the person "A", and we do the same thing until only one person left in the circle.
Question is given n person and m cards, can we choose n cards and allocate them to the n person, to make that whether start at which position(exclude the first position), the person survival at the end is always the first person in the circle.
For example, m = n = 5, the only solution is (4, 1, 5, 3, 2).
I think this problem is a np-complete problem, but I can't prove it. Anybody has a good idea to find a polynomial time solution or prove it's np-hard?
--- example solutions ---
2: [ 1, 2]
2: [ 2, 1]
3: [ 1, 3, 2]
3: [ 3, 1, 2]
4: [ 4, 1, 3, 2]
5: [ 4, 1, 5, 3, 2]
7: [ 5, 7, 3, 1, 6, 4, 2]
9: [ 2, 7, 3, 9, 1, 6, 8, 5, 4]
9: [ 3, 1, 2, 7, 6, 5, 9, 4, 8]
9: [ 3, 5, 1, 8, 9, 6, 7, 4, 2]
9: [ 3, 9, 2, 7, 6, 1, 5, 4, 8]
9: [ 6, 1, 8, 3, 7, 9, 4, 5, 2]
10: [ 3, 5, 6, 10, 1, 9, 8, 7, 4, 2]
10: [ 4, 5, 2, 8, 7, 10, 6, 1, 9, 3]
10: [ 5, 1, 9, 2, 10, 3, 7, 6, 8, 4]
10: [ 6, 3, 1, 10, 9, 8, 7, 4, 5, 2]
10: [ 8, 5, 9, 10, 1, 7, 2, 6, 4, 3]
10: [10, 5, 2, 1, 8, 7, 6, 9, 3, 4]
11: [ 2, 1, 10, 11, 9, 3, 7, 5, 6, 8, 4]
11: [ 3, 7, 11, 10, 9, 8, 1, 6, 5, 4, 2]
11: [ 3, 11, 10, 9, 8, 1, 7, 2, 4, 5, 6]
11: [ 4, 1, 10, 2, 9, 8, 7, 5, 11, 3, 6]
11: [ 4, 2, 7, 11, 5, 1, 10, 9, 6, 3, 8]
11: [ 4, 7, 2, 3, 1, 10, 9, 6, 11, 5, 8]
11: [ 4, 7, 3, 9, 11, 10, 1, 8, 6, 5, 2]
11: [ 4, 11, 7, 2, 1, 10, 9, 6, 5, 3, 8]
11: [ 5, 11, 3, 9, 8, 7, 6, 1, 10, 4, 2]
11: [ 6, 1, 10, 2, 9, 8, 7, 5, 11, 3, 4]
11: [ 6, 2, 7, 11, 5, 1, 10, 9, 4, 3, 8]
11: [ 6, 11, 1, 3, 10, 2, 7, 5, 4, 9, 8]
11: [ 9, 5, 3, 1, 10, 2, 8, 7, 11, 6, 4]
12: [ 1, 7, 11, 10, 4, 9, 2, 12, 6, 5, 8, 3]
12: [ 3, 7, 12, 2, 11, 10, 9, 1, 6, 5, 4, 8]
12: [ 3, 8, 11, 2, 12, 9, 1, 7, 5, 10, 4, 6]
12: [ 4, 2, 5, 1, 11, 10, 9, 8, 12, 7, 3, 6]
12: [ 4, 3, 7, 6, 1, 11, 10, 9, 8, 12, 5, 2]
12: [ 5, 1, 6, 11, 9, 2, 10, 7, 12, 8, 3, 4]
12: [ 5, 2, 3, 12, 9, 10, 7, 6, 1, 11, 4, 8]
12: [ 5, 7, 12, 2, 10, 9, 8, 11, 1, 4, 6, 3]
12: [ 7, 1, 2, 3, 5, 9, 10, 8, 11, 6, 12, 4]
12: [ 8, 7, 1, 11, 9, 3, 5, 10, 6, 4, 12, 2]
12: [ 8, 7, 11, 10, 12, 3, 1, 9, 6, 5, 4, 2]
12: [12, 3, 11, 5, 1, 10, 8, 7, 6, 4, 9, 2]
12: [12, 7, 11, 1, 9, 3, 2, 10, 6, 5, 4, 8]
13: [ 2, 1, 4, 7, 11, 6, 3, 10, 13, 5, 8, 12, 9]
13: [ 2, 5, 13, 12, 4, 11, 3, 1, 9, 7, 8, 6, 10]
13: [ 2, 13, 12, 11, 3, 1, 9, 4, 8, 7, 10, 5, 6]
13: [ 3, 5, 2, 1, 12, 9, 11, 10, 7, 6, 13, 4, 8]
13: [ 3, 5, 13, 1, 11, 2, 9, 8, 7, 12, 6, 4, 10]
13: [ 4, 13, 3, 1, 12, 11, 10, 9, 7, 2, 5, 6, 8]
13: [ 6, 4, 3, 1, 10, 11, 13, 5, 9, 12, 7, 8, 2]
13: [ 6, 4, 13, 7, 5, 1, 12, 11, 10, 9, 8, 3, 2]
13: [ 6, 7, 3, 13, 12, 11, 10, 2, 1, 9, 5, 4, 8]
13: [ 6, 7, 13, 11, 2, 10, 9, 1, 8, 12, 5, 3, 4]
13: [ 6, 11, 7, 13, 1, 10, 2, 12, 9, 8, 5, 4, 3]
13: [ 7, 3, 2, 1, 11, 10, 9, 8, 13, 5, 12, 4, 6]
13: [ 7, 5, 13, 3, 10, 11, 2, 9, 1, 6, 8, 4, 12]
13: [ 7, 5, 13, 3, 11, 2, 9, 8, 1, 6, 12, 4, 10]
13: [ 7, 5, 13, 3, 11, 12, 2, 1, 9, 8, 6, 4, 10]
13: [ 7, 9, 1, 11, 3, 13, 2, 10, 12, 6, 5, 4, 8]
13: [ 8, 3, 5, 11, 13, 9, 10, 7, 1, 6, 4, 12, 2]
13: [ 8, 3, 13, 1, 5, 11, 10, 9, 12, 7, 6, 4, 2]
13: [ 9, 3, 13, 2, 10, 4, 1, 7, 6, 5, 12, 11, 8]
13: [ 9, 4, 7, 5, 1, 11, 13, 10, 12, 8, 6, 3, 2]
13: [ 9, 5, 4, 13, 2, 11, 8, 10, 1, 7, 12, 3, 6]
13: [ 9, 5, 13, 4, 11, 1, 8, 3, 7, 12, 6, 10, 2]
13: [10, 4, 3, 5, 13, 1, 9, 11, 7, 6, 8, 12, 2]
13: [11, 2, 7, 3, 12, 1, 10, 9, 6, 5, 13, 4, 8]
13: [11, 13, 5, 2, 10, 9, 8, 7, 1, 6, 4, 3, 12]
13: [11, 13, 7, 1, 12, 9, 2, 3, 10, 5, 4, 6, 8]
13: [12, 1, 3, 5, 11, 13, 4, 10, 9, 8, 7, 6, 2]
13: [12, 7, 13, 3, 11, 1, 9, 8, 6, 5, 10, 4, 2]
13: [12, 13, 7, 11, 2, 5, 1, 9, 10, 6, 4, 3, 8]
13: [13, 3, 1, 12, 11, 2, 9, 10, 7, 6, 4, 5, 8]
13: [13, 3, 7, 1, 5, 12, 4, 10, 9, 8, 11, 6, 2]
14: [ 3, 5, 13, 14, 1, 12, 11, 10, 9, 8, 7, 6, 4, 2]
14: [ 3, 9, 1, 13, 11, 10, 2, 4, 7, 14, 6, 8, 5, 12]
14: [ 3, 14, 4, 12, 11, 1, 9, 8, 2, 13, 7, 5, 10, 6]
14: [ 4, 11, 1, 13, 7, 10, 12, 2, 14, 9, 8, 5, 6, 3]
14: [ 4, 14, 2, 5, 13, 1, 12, 11, 7, 6, 10, 9, 3, 8]
14: [ 5, 7, 1, 13, 12, 11, 10, 2, 9, 8, 14, 6, 4, 3]
14: [ 6, 3, 14, 5, 11, 13, 2, 12, 9, 1, 7, 4, 8, 10]
14: [ 6, 14, 1, 12, 5, 13, 2, 11, 9, 7, 8, 4, 3, 10]
14: [ 7, 5, 13, 12, 1, 11, 4, 10, 2, 14, 9, 8, 6, 3]
14: [ 7, 11, 5, 13, 1, 3, 2, 4, 10, 9, 14, 6, 8, 12]
14: [ 7, 14, 1, 13, 2, 5, 11, 12, 10, 9, 8, 4, 3, 6]
14: [ 8, 7, 5, 13, 2, 11, 3, 9, 10, 12, 1, 14, 4, 6]
14: [11, 2, 10, 5, 8, 7, 9, 1, 13, 14, 12, 4, 3, 6]
14: [11, 3, 14, 2, 13, 1, 10, 8, 9, 7, 5, 12, 4, 6]
14: [11, 5, 3, 14, 2, 1, 13, 10, 8, 7, 6, 12, 4, 9]
14: [11, 14, 5, 3, 13, 1, 10, 2, 9, 4, 7, 8, 12, 6]
14: [12, 1, 14, 3, 13, 4, 10, 9, 2, 7, 6, 5, 11, 8]
14: [12, 11, 7, 5, 13, 3, 2, 14, 1, 9, 8, 4, 6, 10]
14: [12, 14, 7, 13, 6, 5, 11, 1, 10, 9, 8, 4, 3, 2]
14: [13, 1, 7, 2, 11, 3, 9, 14, 8, 6, 5, 10, 4, 12]
14: [13, 11, 3, 1, 4, 2, 7, 10, 9, 6, 14, 12, 5, 8]
14: [14, 1, 13, 3, 11, 5, 10, 9, 2, 6, 8, 7, 4, 12]
14: [14, 5, 1, 13, 12, 2, 11, 3, 7, 9, 6, 8, 4, 10]
--- possibly helpful for a mathematical solution ---
I noticed that starting with length 9, at least one solution for every length has a longish sequence of integers that decrement by 1.
9: [3, 1, 2, 7, 6, 5, 9, 4, 8]
10: [6, 3, 1, 10, 9, 8, 7, 4, 5, 2]
11: [3, 7, 11, 10, 9, 8, 1, 6, 5, 4, 2]
11: [3, 11, 10, 9, 8, 1, 7, 2, 4, 5, 6]
11: [5, 11, 3, 9, 8, 7, 6, 1, 10, 4, 2]
12: [4, 2, 5, 1, 11, 10, 9, 8, 12, 7, 3, 6]
12: [4, 3, 7, 6, 1, 11, 10, 9, 8, 12, 5, 2]
13: [6, 4, 13, 7, 5, 1, 12, 11, 10, 9, 8, 3, 2]
14: [3, 5, 13, 14, 1, 12, 11, 10, 9, 8, 7, 6, 4, 2]
I noticed that for every length I tested except the very small, at least one solution contains a relatively long run of descending
numbers. So far this answer only considers m = n. Here are a few examples; note that excess is n - run_len:
n = 3, run_len = 2, excess = 1: [1] + [3-2] + []
n = 4, run_len = 2, excess = 2: [4, 1] + [3-2] + []
n = 5, run_len = 2, excess = 3: [4, 1, 5] + [3-2] + []
n = 6, no solution
n = 7, run_len = 1, excess = 6: [5] + [7-7] + [3, 1, 6, 4, 2]
n = 8, no solution
n = 9, run_len = 3, excess = 6: [3, 1, 2] + [7-5] + [9, 4, 8]
n = 10, run_len = 4, excess = 6: [6, 3, 1] + [10-7] + [4, 5, 2]
n = 11, run_len = 4, excess = 7: [3, 7] + [11-8] + [1, 6, 5, 4, 2]
n = 12, run_len = 4, excess = 8: [4, 2, 5, 1] + [11-8] + [12, 7, 3, 6]
n = 13, run_len = 5, excess = 8: [6, 4, 13, 7, 5, 1] + [12-8] + [3, 2]
n = 14, run_len = 7, excess = 7: [3, 5, 13, 14, 1] + [12-6] + [4, 2]
n = 15, run_len = 8, excess = 7: [3, 15, 2] + [13-6] + [1, 5, 4, 14]
n = 16, run_len = 6, excess = 10: [6, 3, 1, 10] + [16-11] + [2, 9, 7, 4, 5, 8]
n = 17, run_len = 8, excess = 9: [2, 5, 17, 15, 14, 1] + [13-6] + [4, 3, 16]
n = 18, run_len = 10, excess = 8: [6, 3, 17, 18, 1] + [16-7] + [5, 4, 2]
n = 19, run_len = 10, excess = 9: [4, 19, 3, 17, 18, 1] + [16-7] + [5, 6, 2]
n = 20, no solution found with run_length >= 10
n = 21, run_len = 14, excess = 7: [3, 21, 2] + [19-6] + [1, 5, 4, 20]
n = 22, run_len = 14, excess = 8: [22, 3, 2, 1] + [20-7] + [5, 21, 4, 6]
n = 23, run_len = 14, excess = 9: [7, 1, 23, 3] + [21-8] + [6, 5, 22, 4, 2]
n = 24, run_len = 16, excess = 8: [6, 5, 24, 2] + [22-7] + [3, 1, 23, 4]
n = 25, run_len = 17, excess = 8: [25, 3, 2, 1] + [23-7] + [5, 24, 4, 6]
n = 26, run_len = 17, excess = 9: [26, 3, 25, 2, 1] + [23-7] + [5, 24, 4, 6]
n = 27, run_len = 20, excess = 7: [3, 27, 2] + [25-6] + [1, 5, 4, 26]
n = 28, run_len = 18, excess = 10: [28, 1, 27, 2, 3] + [25-8] + [6, 5, 7, 4, 26]
n = 29, run_len = 20, excess = 9: [2, 5, 29, 27, 26, 1] + [25-6] + [4, 3, 28]
n = 30, run_len = 23, excess = 7: [30, 5, 2, 1] + [28-6] + [29, 3, 4]
n = 31, run_len = 24, excess = 7: [5, 31, 3] + [29-6] + [1, 30, 4, 2]
n = 32, run_len = 23, excess = 9: [7, 32, 31, 2, 1] + [30-8] + [5, 4, 3, 6]
n = 33, run_len = 26, excess = 7: [3, 33, 2] + [31-6] + [1, 5, 4, 32]
n = 34, run_len = 27, excess = 7: [3, 5, 33, 34, 1] + [32-6] + [4, 2]
n = 35, run_len = 27, excess = 8: [5, 35, 3, 33, 34, 1] + [32-6] + [4, 2]
n = 36, run_len = 26, excess = 10: [35, 7, 3, 1, 36, 2] + [34-9] + [6, 5, 4, 8]
n = 37, run_len = 29, excess = 8: [6, 5, 2, 1] + [35-7] + [36, 37, 3, 4]
n = 38, run_len = 29, excess = 9: [3, 7, 37, 38, 1] + [36-8] + [6, 4, 5, 2]
n = 39, run_len = 32, excess = 7: [3, 39, 2] + [37-6] + [1, 5, 4, 38]
n = 40, run_len = 31, excess = 9: [5, 2, 1] + [38-8] + [3, 7, 40, 4, 6, 39]
n = 41, run_len = 33, excess = 8: [3, 5, 1, 40, 2] + [38-6] + [41, 39, 4]
n = 42, run_len = 33, excess = 9: [42, 3, 41, 2, 1] + [39-7] + [5, 4, 40, 6]
n = 43, run_len = 34, excess = 9: [6, 5, 7, 43, 1] + [41-8] + [42, 4, 3, 2]
n = 44, run_len = 35, excess = 9: [5, 3, 2, 1] + [42-8] + [43, 7, 4, 44, 6]
n = 45, run_len = 38, excess = 7: [3, 45, 2] + [43-6] + [1, 5, 4, 44]
n = 50, run_len = 43, excess = 7: [50, 5, 2, 1] + [48-6] + [49, 3, 4]
n = 100, run_len = 91, excess = 9: [5, 2, 1] + [98-8] + [3, 7, 100, 4, 6, 99]
n = 201, run_len = 194, excess = 7: [3, 201, 2] + [199-6] + [1, 5, 4, 200]
20 is missing from the above table because the run length is at most 10, and is taking a long time to compute. No larger value that I've tested has such a small max run length relative to n.
I found these by checking run lengths from n-1 descending, with all possible starting values and permutations of the run & surrounding elements. This reduces the search space immensely.
For a given n, if the max run in any solution to n is length n-k, then this will find it in O(k! * n). While this looks grim, if k has a constant upper bound (e.g. k <= some threshold for all sufficiently large n) then this is effectively O(n). 'Excess' is what I'm calling k in the examples above. I haven't found any greater than 10, but I don't have a solution yet to n = 20. If it has a solution then its excess will exceed 10.
UPDATE: There are a lot of patterns here.
If n mod 6 is 3 and n >= 9, then [3, n, 2, [n-2, n-3, ..., 6], 1, 5, 4, n-1] is valid.
If n mod 12 is 5 and n >= 17 then [2, 5, n, n-2, n-3, 1, [n-4, n-5, ..., 6], 4, 3, n-1] is valid.
If n mod 20 is 10, then [n, 5, 2, 1, [n-2, n-3, ..., 6], n-1, 3, 4] is valid.
If n mod 60 is 7, 11, 31, or 47, then [5, n, 3, [n-2, n-3, ..., 6], 1, n-1, 4, 2] is valid.
If n mod 60 is 6 or 18 and n >= 18 then [6, 3, n-1, n, 1, [n-2, n-3, ..., 7], 5, 4, 2] is valid.
If n mod 60 is 1, 22, 25 or 52 and n >= 22 then [n, 3, 2, 1], [n-2, n-3, ..., 7], 5, n-1, 4, 6] is valid.
If n mod 60 is 23 then [7, 1, n, 3, [n-2, n-3, ..., 8], 6, 5, n-1, 4, 2] is valid.
If n mod 60 is 14 or 34 then [3, 5, n-1, n, 1, [n-2, n-3, ..., 6], 4, 2] is valid.
If n mod 60 is 24 then [6, 5, n, 2, [n-2, n-1, ..., 7], 3, 1, n-1, 4] is valid
If n mod 60 is 2, 6, 26, 42 and n >= 26 then [n, 3, n-1, 2, 1, [n-3, n-4, ..., 7], 5, n-2, 4, 6] is valid.
If n mod 60 is 16 or 28 then [n, 1, n-1, 2, 3, [n-3, n-4, ..., 8], 6, 5, 7, 4, n-2] is valid.
If n mod 60 is 32 then [7, n, n-1, 2, 1, [n-2, n-3, ..., 8], 5, 4, 3, 6] is valid.
If n mod 60 is 35 or 47 then [5, n, 3, n-2, n-1, 1, [n-3, n-4, ..., 6], 4, 2] is valid.
If n mod 60 is 37 then [6, 5, 2, 1, [n-2, n-1, ..., 7], n-1, n, 3, 4]
If n mod 60 is 38 then [3, 7, n-1, n, 1] + [n-2, n-3, ..., 8] + [6, 4, 5, 2]
If n mod 60 is 40 then [5, 2, 1, [n-2, n-3, ..., 8], 3, 7, n, 4, 6, n-1] is valid
If n mod 60 is 0 and n >= 60 then [3, 5, n, 2, [n-2, n-3, ..., 7], 1, 6, n-1, 4] is valid
If n mod 60 is 7, 19, or 31 and n >= 19 then [4, n, 3, n-2, n-1, 1, [n-3, n-4, ..., 7], 5, 6, 2] is valid
If n mod 60 is 23, 38, or 43 then [7, 3, n, 1, [n-2, n-3, ..., 8], 6, 5, n-1, 4, 2] is a valid solution
If n mod 60 is 14 or 44 and n >= 74 then [3, 5, n-1, n, 1, [n-3, n-4, ..., 6], n-2, 4, 2] is valid.
If n mod 60 is 1 or 49 and n >= 49 then [3, 5, n, 1, [n-2, n-3, ..., 7], 2, n-1, 4, 6] is valid.
If n mod 60 is 6, 18, 30, 42, or 54 and n >= 18 then [n, 3, n-1, 2, 1, [n-3, n-4, ..., 7], 5, 4, n-2, 6] is valid.
If n mod 60 is 10, 18, 38 or 58 and n >= 18 then [n-1, 7, 5, n, 1, [n-2, n-3, ..., 8], 2, 6, 4, 3] is valid.
Currently solved for n mod 60 is any of the following values:
0, 1, 2, 3, 5, 6, 7, 9,
10, 11, 14, 15, 16, 17, 18, 19,
21, 22, 23, 24, 25, 26, 27, 28, 29,
30, 31, 32, 33, 34, 35, 37, 38, 39,
40, 41, 42, 43, 44, 45, 47, 49,
50, 51, 52, 53, 54, 57, 58
Also,
If n mod 42 is 31 then [n, 3, 2, 1, [n-2, n-3, ..., 8], n-1, 5, 4, 7, 6] is valid.
If n mod 420 is 36 or 396 then [n-1, 7, 3, 1, n, 2, [n-2, n-3, ..., 9], 6, 5, 4, 8] is valid.
--- Example for n=21, using the first pattern listed above, and all starting indices.
1: [21, 2, 18, 19, 16, 17, 14, 15, 12, 13, 10, 11, 8, 9, 6, 7, 5, 4, 20, 1]
2: [ 2, 18, 21, 16, 19, 14, 17, 12, 15, 10, 13, 8, 11, 6, 9, 5, 1, 4, 20, 7]
3: [19, 21, 18, 2, 16, 17, 14, 15, 12, 13, 10, 11, 8, 9, 6, 7, 5, 4, 20, 1]
4: [18, 21, 19, 17, 2, 15, 16, 13, 14, 11, 12, 9, 10, 7, 8, 1, 5, 4, 20, 6]
5: [17, 21, 19, 18, 16, 2, 14, 15, 12, 13, 10, 11, 8, 9, 6, 7, 5, 4, 20, 1]
6: [16, 21, 19, 18, 17, 15, 2, 13, 14, 11, 12, 9, 10, 7, 8, 1, 5, 4, 20, 6]
7: [15, 21, 19, 18, 17, 16, 14, 2, 12, 13, 10, 11, 8, 9, 6, 7, 5, 4, 20, 1]
8: [14, 21, 19, 18, 17, 16, 15, 13, 2, 11, 12, 9, 10, 7, 8, 1, 5, 4, 20, 6]
9: [13, 21, 19, 18, 17, 16, 15, 14, 12, 2, 10, 11, 8, 9, 6, 7, 5, 4, 20, 1]
10: [12, 21, 19, 18, 17, 16, 15, 14, 13, 11, 2, 9, 10, 7, 8, 1, 5, 4, 20, 6]
11: [11, 21, 19, 18, 17, 16, 15, 14, 13, 12, 10, 2, 8, 9, 6, 7, 5, 4, 20, 1]
12: [10, 21, 19, 18, 17, 16, 15, 14, 13, 12, 11, 9, 2, 7, 8, 1, 5, 4, 20, 6]
13: [ 9, 21, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 8, 2, 6, 7, 5, 4, 20, 1]
14: [ 8, 21, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 7, 2, 1, 5, 4, 20, 6]
15: [ 7, 21, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 6, 2, 5, 4, 20, 1]
16: [ 6, 21, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 1, 5, 4, 20, 2]
17: [ 1, 5, 2, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 4, 19, 20, 21]
18: [ 5, 2, 18, 19, 16, 17, 14, 15, 12, 13, 10, 11, 8, 9, 6, 7, 4, 1, 20, 21]
19: [ 4, 2, 18, 19, 16, 17, 14, 15, 12, 13, 10, 11, 8, 9, 6, 7, 5, 20, 21, 1]
20: [20, 4, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 1, 5, 21, 2]
You can observe the same relationship between elements from the decrementing run and other elements for all values of n that the pattern applies to. This isn't a proof, but you can turn this into a proof, though I think the work would need to be done for each pattern separately and it's beyond the scope of what I'm going to spend time on for an S/O question.
--- We can fill in the blanks by using m > n. ---
The pattern [n-1, n, 1, [n-2, n-3, ..., 3], n+5] is valid for n mod 4 is 1 and n >= 9.
The pattern [n, 2, 1, [n-2, n-3, ..., 3], n+4] is valid for n mod 2 is 0 and n >= 6.
With these two, plus what we already found, we get nearly everything. I found these by checking a single replacement value in a limited range.
0, 1, 2, 3, 4, 5, 6, 7, 8, 9,
10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
20, 21, 22, 23, 24, 25, 26, 27, 28, 29,
30, 31, 32, 33, 34, 35, 36, 37, 38, 39,
40, 41, 42, 43, 44, 45, 46, 47, 48, 49,
50, 51, 52, 53, 54, 56, 57, 58
If n mod 30 is 29, then [3, n, 2, [n-2, n-3, ..., 4], n-1, n+15) is valid, giving us n mod 60 is 59. We're left with just one unknown: n mod 60 is 55.
...And finally! If n mod 12 is 7 (i.e. n mod 60 is 7, 19, 31, 43, or 55) then [n-1, n, 1, [n-2, n-3, ..., 6], 2, 5, 3, n+4] is valid for all n >= 19.
We now have solutions for all n mod 60, using m=n in most cases, and m=n+15 in the worst case.

Creating a nested dictionary comprehension for year and month in python

I would like to create a nested dictionary with dict comprehension but I am getting syntax error.
years = [2016, 2017, 2018, 2019]
months = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]
my_Dict = {i:{j: for j in months}, for i in years}
I am not sure how to declare this nested dict comprehension without getting a syntax error.
In this case, the correct way would be to use a dictionary with a nested list comprehension because if you use a nested dictionary, it will replace old values. The correct syntax, in this case, would be this one:
years = [2016, 2017, 2018, 2019]
months = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]
my_Dict = {
year: [month for month in months] for year in years
}
print(my_Dict)
>>> {2016: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],
2017: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],
2018: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],
2019: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]}
Made some slight changes to the code above and this works
key_years = {2016, 2017, 2018, 2019}
key_months = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12}
myDict = {i:{j for j in key_months} for i in key_years}
print (myDict)
Output: {2016: {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12}, 2017: {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12}, 2018: {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12}, 2019: {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12}}

How to duplicate value of array in ruby

I have two arrays of integers, e.g.
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
b = [7, 8, 9]
I would like to repeatedly duplicate the value of 'b' to get a perfectly matching array lengths like this:
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
b = [7, 8, 9, 7, 8, 9, 7, 8, 9, 7]
We can assume that a.length > b.length
Assuming you mean
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
b = [7, 8, 9]
then you can do:
b.cycle.take(a.length) #=> [7, 8, 9, 7, 8, 9, 7, 8, 9, 7]
<script src="//repl.it/embed/JJ3x/2.js"></script>
See Array#cycle and Enumerable#take for more details.
I would have used Array#cycle had it been available, but since it was taken I thought I'd suggest some alternatives (the first being my fav).
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
b = [7, 8, 9]
[*b*(a.size/b.size), *b[0, a.size % b.size]]
#=> [7, 8, 9, 7, 8, 9, 7, 8, 9, 7]
Array.new(a.size) { |i| b[i % b.size] }
#=> [7, 8, 9, 7, 8, 9, 7, 8, 9, 7]
b.values_at(*(0..a.size-1).map { |i| i % b.size })
#=> [7, 8, 9, 7, 8, 9, 7, 8, 9, 7]

When I make a second instance of a class in Ruby it changes the first instance of that class

I have a class that randomly creates a two dimensional array. I am testing this program in irb. When I create the first instance of a class by doing thing1 = CatanBoard.new everything works fine. When I create a second instance by doing thing2 = CatanBoard.new I run into problems. This causes thing1.board to be identical to thing2.board and it adds elements to the arrays in thing1.
Correct output of thing1.board after the initialization of thing1 looks like
[[2, 8, "wheat"], [4, 8, "forest"], [15, 6, "forest"], [12, 6, "stone"], [19, 12, "sheep"], [9, 11, "forest"], [17, 11, "stone"], [6, 10, "wheat"], [14, 10, "stone"], [18, 9, "wheat"], [11, 9, "sheep"], [10, 5, "forest"], [16, 5, "brick"], [1, 4, "sheep"], [13, 4, "brick"], [5, 3, "sheep"], [8, 3, "wheat"], [7, 2, "brick"], [3, nil, "desert"]]
Then when I initialize thing2 by doing thing2 = CatanBoard.new I get this as the value of thing2.board:
[[2, 8, "wheat", 8, 8, "wheat"], [4, 8, "forest", 10, 8, "stone"], [15, 6, "forest", 17, 6, "forest"], [12, 6, "stone", 19, 6, "brick"], [19, 12, "sheep", 14, 12, "brick"], [9, 11, "forest", 11, 11, "sheep"], [17, 11, "stone", 18, 11, "stone"], [6, 10, "wheat", 16, 10, "wheat"], [14, 10, "stone", 12, 10, "brick"], [18, 9, "wheat", 13, 9, "forest"], [11, 9, "sheep", 5, 9, "wheat"], [10, 5, "forest", 2, 5, "sheep"], [16, 5, "brick", 1, 5, "sheep"], [1, 4, "sheep", 6, 4, "stone"], [13, 4, "brick", 9, 4, "wheat"], [5, 3, "sheep", 15, 3, "forest"], [8, 3, "wheat", 4, 3, "forest"], [7, 2, "brick", 3, 2, "sheep"], [3, nil, "desert", 7, nil, "desert"]]
Then I check the value of thing1.board and it is now the same as thing2.board:
[[2, 8, "wheat", 8, 8, "wheat"], [4, 8, "forest", 10, 8, "stone"], [15, 6, "forest", 17, 6, "forest"], [12, 6, "stone", 19, 6, "brick"], [19, 12, "sheep", 14, 12, "brick"], [9, 11, "forest", 11, 11, "sheep"], [17, 11, "stone", 18, 11, "stone"], [6, 10, "wheat", 16, 10, "wheat"], [14, 10, "stone", 12, 10, "brick"], [18, 9, "wheat", 13, 9, "forest"], [11, 9, "sheep", 5, 9, "wheat"], [10, 5, "forest", 2, 5, "sheep"], [16, 5, "brick", 1, 5, "sheep"], [1, 4, "sheep", 6, 4, "stone"], [13, 4, "brick", 9, 4, "wheat"], [5, 3, "sheep", 15, 3, "forest"], [8, 3, "wheat", 4, 3, "forest"], [7, 2, "brick", 3, 2, "sheep"], [3, nil, "desert", 7, nil, "desert"]]
This post had a similar problem, but since I'm using Array.new I don't think it's the same problem, since I am creating deep copies. What do you think the problem is with my code?
Here is my code:
# The class CatanBoard represents a Catan Board with no expansions
# A Catan board has 19 hexagons. Each hexagon has a roll and a resource on it.
# Rolls that are 6's and 8's cannot be adjacent to other 6's and 8's
# The 'desert' square has no roll on it
RESOURCES = ['forest', 'forest', 'forest', 'forest', 'brick', 'brick', 'brick', 'wheat', 'wheat', 'wheat', 'wheat', 'sheep', 'sheep', 'sheep', 'sheep', 'stone', 'stone', 'stone'] # desert is left out because it isn't a resource and needs to be specially added
# note - there are 19 tiles, yet 18 rolls. This is because the desert tile does not get a roll.
SPECIAL_ROLLS = [6, 6, 8, 8]
PLAIN_ROLLS = [2, 3, 3, 4, 4, 5, 5, 9, 9, 10, 10, 11, 11, 12]
TILES = (1..19).to_a
EMPTY_BOARD = Array.new(19) {Array.new(0) {[]}}
HARBORS = ['brick', 'generic', 'generic', 'generic', 'generic', 'sheep', 'stone', 'wheat', 'wood']
# this function returns the adjacent tiles on the catan board.
# See the picture in "catan overview.odg" for details
def neighbors(loc)
case loc
when 1
return [2, 4, 5]
when 2
return [1, 3, 5, 6]
when 3
return [2, 6, 7]
when 4
return [1, 5, 8, 9]
when 5
return [1, 2, 4, 6, 9, 10]
when 6
return [2 , 3, 5, 7, 10, 11]
when 7
return [3, 6, 11, 12]
when 8
return [4, 9, 13]
when 9
return [4, 5, 8, 10, 13, 14]
when 10
return [5, 6, 9, 11, 14, 15]
when 11
return [6, 7, 10, 12, 15, 16]
when 12
return [7, 11, 16]
when 13
return [9, 14, 17]
when 14
return [9, 10, 13, 15, 17, 18]
when 15
return [10, 11, 14, 16, 18, 19]
when 16
return [11, 12, 15, 19]
when 17
return [13, 14, 18]
when 18
return [14, 15, 17, 19]
when 19
return [15, 16, 18]
else
return "error"
end
end
class CatanBoard
def initialize()
# #board and #harbors are tha arrays that represent the games
# The other variables are used to set up the board
#board = Array.new(EMPTY_BOARD)
#harbors = Array.new(HARBORS) # #harbors[0] corresponds with A, while #harbors[8] corresponds with I in the harbor diagram in "catan overview.odg"
#resources = Array.new(RESOURCES)
#special_rolls = Array.new(SPECIAL_ROLLS)
#plain_rolls = Array.new(PLAIN_ROLLS)
#tiles = Array.new(TILES)
# RANDOMIZE THE HARBORS #
#harbors = #harbors.shuffle
# PLACE THE SPECIAL ROLLS #
temp_tiles = #tiles
for i in (0..#special_rolls.length-1)
loc = temp_tiles.delete_at(rand(temp_tiles.length)) # chooses a random tile as the location and saves it
#temp_tiles.delete(loc) # I think this line isn't needed
temp_tiles = temp_tiles - neighbors(loc)
# puts the tile and the roll onto the board
#board[i] << loc
#board[i] << #special_rolls.pop
#tiles.delete(loc)
end
# THEN PLACE THE REST OF THE ROLLS #
for i in (0..#tiles.length-1)
loc = #tiles.delete_at(rand(#tiles.length))
#board[i+4] << loc # +4 because loctions 0 through 3 are filled
#board[i+4] << #plain_rolls.pop
end
# THEN PLACE THE RESOURCES #
#board[#board.length-1] << 'desert' # matches the desert tile to the nil roll
for r in (0..#resources.length-1)
# removes a random resource and pairs it with a location and a roll
# resources must be removed randomly otherwise the 6's and 8's are all on stone and sheep
#board[r] << #resources.delete_at(rand(#resources.length))
end
end
def board()
return #board
end
def harbors()
return #harbors
end
# This determines if the board is set up according to the game's rules
def is_legal?()
#TEST SPECIAL ROLLS#
# special rolls are in first four locations
for loc in (0..3)
for other_loc in (0..3)
if neighbors(#board[loc][0]).include?(!#board[other_loc][0])
return false
end
end
end
#TEST TOTAL ROLLS#
all_rolls = Array.new(SPECIAL_ROLLS + PLAIN_ROLLS)
# extracts the rolls from board
temp_rolls = []
# -2 because the desert square doesn't have a roll and it is last in the array
for i in (0..#board.length-2)
temp_rolls << #board[i][1]
end
temp_rolls = temp_rolls.sort
all_rolls = all_rolls.sort
if temp_rolls != all_rolls
temp_resources
return false
end
#TEST RESOURCE AMOUNT#
all_resources = Array.new(RESOURCES)
temp_resources = []
# -2 because the desert square isn't a resource and it is last in the array
for i in (0..#board.length-2)
temp_resources << #board[i][2]
end
temp_resources = temp_resources.sort
all_resources = all_resources.sort
if temp_resources != all_resources
return false
end
#TEST HARBORS#
temp_harbors = Array.new(HARBORS)
sorted_harbors = #harbors.sort
if temp_harbors != sorted_harbors
return false
end
return true
end
end
Array.new does not create a deep copy, so your boards are shared, since EMPTY_BOARD is two levels deep.
In addition, keeping EMPTY_BOARD outside of your CatanBoard class probably isn't a good idea anyway from a cohesion point of view. If you introduce a private method
def empty_board
Array.new(19) {Array.new(0) {[]}}
end
private :empty_board
then you can just reference this method during initialization:
#board = empty_board
Instead Array.new(EMPTY_BOARD) do a deep copy. Here is an example how you can do it since there is no standard way.
def deep_copy(obj)
Marshal.load(Marshal.dump(obj))
end
#board = deep_copy(EMPTY_BOARD)

Resources