EDIT: My question is not a duplicate as someone has marked. The other question is incorrect and does not even work.
I have tried a few ways to group the results of iter
An implementation using a collections.deque
based on the Scheduling_algorithm in the linked question:
from collections import deque
from itertools import islice
def fixtures(teams):
if len(teams) % 2:
teams.append("Bye")
ln = len(teams) // 2
dq1, dq2 = deque(islice(teams, None, ln)), deque(islice(teams, ln, None))
for _ in range(len(teams)-1):
yield zip(dq1, dq2) # list(zip.. python3
# pop off first deque's left element to
# "fix one of the competitors in the first column"
start = dq1.popleft()
# rotate the others clockwise one position
# by swapping elements
dq1.appendleft(dq2.popleft())
dq2.append(dq1.pop())
# reattach first competitor
dq1.appendleft(start)
Output:
In [37]: teams = ["team1", "team2", "team3", "team4"]
In [38]: list(fixtures(teams))
Out[38]:
[[('team1', 'team3'), ('team2', 'team4')],
[('team1', 'team4'), ('team3', 'team2')],
[('team1', 'team2'), ('team4', 'team3')]]
In [39]: teams = ["team1", "team2", "team3", "team4","team5"]
In [40]: list(fixtures(teams))
Out[40]:
[[('team1', 'team4'), ('team2', 'team5'), ('team3', 'Bye')],
[('team1', 'team5'), ('team4', 'Bye'), ('team2', 'team3')],
[('team1', 'Bye'), ('team5', 'team3'), ('team4', 'team2')],
[('team1', 'team3'), ('Bye', 'team2'), ('team5', 'team4')],
[('team1', 'team2'), ('team3', 'team4'), ('Bye', 'team5')]]