Generating circular shifts / reduced Latin Squares in Python

Question

Was just wondering what\'s the most efficient way of generating all the circular shifts of a list in Python. In either direction. For example, given a list [1, 2, 3, 4

Answer 1

For the first part, the most concise way probably is

a = [1, 2, 3, 4]
n = len(a)
[[a[i - j] for i in range(n)] for j in range(n)]
# [[1, 2, 3, 4], [4, 1, 2, 3], [3, 4, 1, 2], [2, 3, 4, 1]]

and for the second part

[[a[i - j] for i in range(n)] for j in range(n, 0, -1)]
# [[1, 2, 3, 4], [2, 3, 4, 1], [3, 4, 1, 2], [4, 1, 2, 3]]

These should also be much more efficient than your code, though I did not do any timings.

Answer 2

You can use collections.deque:

from collections import deque

g = deque([1, 2, 3, 4])

for i in range(len(g)):
    print list(g) #or do anything with permutation
    g.rotate(1) #for right rotation
    #or g.rotate(-1) for left rotation

It prints:

 [1, 2, 3, 4]
 [4, 1, 2, 3]
 [3, 4, 1, 2]
 [2, 3, 4, 1]

To change it for left rotation just replace g.rotate(1) with g.rotate(-1).

Answer 3

variation on slicing "conservation law" a = a[:i] + a[i:]

ns = list(range(5))
ns
Out[34]: [0, 1, 2, 3, 4]

[ns[i:] + ns[:i] for i in range(len(ns))]
Out[36]: 
[[0, 1, 2, 3, 4],
 [1, 2, 3, 4, 0],
 [2, 3, 4, 0, 1],
 [3, 4, 0, 1, 2],
 [4, 0, 1, 2, 3]]


[ns[-i:] + ns[:-i] for i in range(len(ns))]
Out[38]: 
[[0, 1, 2, 3, 4],
 [4, 0, 1, 2, 3],
 [3, 4, 0, 1, 2],
 [2, 3, 4, 0, 1],
 [1, 2, 3, 4, 0]]

Answer 4

This will be my solution.

#given list
a = [1,2,3,4]
#looping through list
for i in xrange(len(a)):
    #inserting last element at the starting
    a.insert(0,a[len(a)-1])
    #removing the last element
    a = a[:len(a)-1]
    #printing if you want to
    print a

This will output the following:

[4, 1, 2, 3]
[3, 4, 1, 2]
[2, 3, 4, 1]
[1, 2, 3, 4]

You can also use pop instead of using list slicing but the problem with pop is that it will return something.

Also the above code will work for any length of list. I have not checked for performance of the code. I am assuming that it will work better.

You should have a look at Python docs for getting a good understanding of List slicing.

Answer 5

Using itertools to avoid indexing:

x = itertools.cycle(a)
[[x.next() for i in a] for j in a]

Answer 6

The answer by @Bruno Lenzi does not seem to work:

In [10]: from itertools import cycle

In [11]: x = cycle('ABCD')

In [12]: print [[x.next() for _ in range(4)] for _ in range(4)]
[['A', 'B', 'C', 'D'], ['A', 'B', 'C', 'D'], ['A', 'B', 'C', 'D'], ['A', 'B', 'C', 'D']]

I give a correct version below, however the solution by @f5r5e5d is faster.

In [45]: def use_cycle(a):
    x=cycle(a)
    for _ in a:
        x.next()
        print [x.next() for _ in a]
   ....:         

In [46]: use_cycle([1,2,3,4])
[2, 3, 4, 1]
[3, 4, 1, 2]
[4, 1, 2, 3]
[1, 2, 3, 4]

In [50]: def use_slice(a):
    print [ a[n:] + a[:n] for n in range(len(a)) ]
  ....:     

In [51]: use_slice([1,2,3,4])
[[1, 2, 3, 4], [2, 3, 4, 1], [3, 4, 1, 2], [4, 1, 2, 3]]

In [54]: timeit.timeit('use_cycle([1,2,3,4])','from __main__ import use_cycle',number=100000)
Out[54]: 0.4884989261627197

In [55]: timeit.timeit('use_slice([1,2,3,4])','from __main__ import use_slice',number=100000)
Out[55]: 0.3103291988372803

In [58]: timeit.timeit('use_cycle([1,2,3,4]*100)','from __main__ import use_cycle',number=100)
Out[58]: 2.4427831172943115

In [59]: timeit.timeit('use_slice([1,2,3,4]*100)','from __main__ import use_slice',number=100)
Out[59]: 0.12029695510864258

I removed the print statement in use_cycle and use_slice for timing purposes.