Moving window thru list

Question

Let say I have list [1, 2, 3, 4, 5, 6, 7] I need produce output like (moving window):

when 5-row-height window get to last element of list (7 here) we need start 1, and I need first column be generated like i * 8.

I came with something like this (but not quite):

def w(l):
    global lines
    for i in range(lines - 1, lines + 4):
        location = (i % 5) * 8
        print(f'{location} {l[i]}')
        if lines >= len(l) - 4:
            lines = 1


if __name__ == '__main__':
    ll = [1, 2, 3, 4, 5, 6, 7]
    for i in range(1, 4):
        lines = i
        print(f'-- {i} --')
        w(ll)

Output is:

Maybe deque is an answer?

A `deque` would be a good way to implement one, see this [answer](https://stackoverflow.com/a/6822761/355230). — martineau, Jun 16 '22 at 21:19

score 2 · Answer 1 · answered Jun 16 '22 at 21:18

2

To fix the location part, you can just add enumerate and multiply with 8.

def w(l):
    global lines
    for p,i in enumerate(range(lines - 1, lines + 4)):
        location = p * 8
...

answered Jun 16 '22 at 21:18

Ritwick Jha

340
2
10

Nice (forgot about `enumerate`), however I got problem when `list` is shorter then window. – emcek Jun 17 '22 at 11:01
There are a few holes in the logic right now, especially because it is more complicated than need be. It would be easier to rotate the list and take the first n elements rather than calculating the index each time. You could just put conditions on how much to rotate the list by so it's appropriate. Also it's very easy to rotate a list, just by using slices – Ritwick Jha Jun 17 '22 at 14:05

score 1 · Answer 2 · answered Jun 16 '22 at 21:15

You can use sliding_window_view() from numpy:

import numpy as np
sliding_window = np.lib.stride_tricks.sliding_window_view([1, 2, 3, 4, 5, 6, 7], 5)
for i in sliding_window:
    for j, k in enumerate(i):
        print(j*8, k)

Output:

Moving window thru list

2 Answers2