Repeat per group row pattern n times for each group and save it

I have a pandas dataframe like this:

df = [[1, 'a1 - d1'], 
      [1, 'b1 - c1'], 
      [1, 'd1 - c1'], 
      [1, 'a1 - b1'], 
      [1, 'b1 - d1'], 
      [1, 'c1 - a1'], 
      [2, 'y1 - z1'],
      [2, 'x1 - z1'],
      [2, 'z1 - x1']]
# Create the pandas DataFrame
df = pd.DataFrame(df, columns=['group_id', 'combinations'])
df

• Everyday I want to use 1 combination of each group (represented by group_id) for n = 30 days. I want the code to have variable n = 30, so that I can vary this number later if required.
• Some groups might have more than n = 30 combinations and some groups may have less than n = 30 combinations to begin with.

Expected output:

A simple

pd.concat([x]*2)

repeats the whole dataframe twice of n times, which is not the output that I am expecting. Is there any way to repeat such a pattern??

Update code dump that i have tried so far.

Basically querying each a every group and repeating sequence of that and you know somehow merging them all in the end. Probably the most inefficient solution. But am not sure how else to do this...

group_list = set(df['group_id'])

for i in group_list:
    df1 = df.query(f'group_id == {i}')
    # assuming the worst case scenario to be a group containing only 1 combination
    df1 = pd.concat([df1]*30) 
    df1 = df1.head(30)

Here, this is not so bad. You're looking to create a sequence that repeats of the length of the number of rows in each group. We'll first split our df into groups, then iterate through each group. We'll use some tricks from itertools (list(islice(cycle(range(4)),7)) -> [0, 1, 2, 3, 0, 1, 2]) to get our sequence to repeat and get indices of the rows we want. We'll then select those rows and set the index to the day of the month.

import pandas as pd
from itertools import cycle,islice
df = [[1, 'a1 - d1'],
[1, 'b1 - c1'],
[1, 'd1 - c1'],
[1, 'a1 - b1'],
[1, 'b1 - d1'],
[1, 'c1 - a1'],
[2, 'y1 - z1'],
[2, 'x1 - z1'],
[2, 'z1 - x1']]
# Create the pandas DataFrame
df = pd.DataFrame(df, columns=['group_id', 'combinations'])
grouped = df.groupby('group_id')
out = {}
for grp_id, grp_data in grouped:  # 1,2
indexes = islice(cycle(range(len(grp_data))), 30)
result = pd.DataFrame(grp_data.iloc[indexes])
result.index = range(1,31)
out[grp_id] = result
print(out)

{1:     group_id combinations
1          1      a1 - d1
2          1      b1 - c1
3          1      d1 - c1
4          1      a1 - b1
5          1      b1 - d1
6          1      c1 - a1
7          1      a1 - d1
8          1      b1 - c1
9          1      d1 - c1
10         1      a1 - b1
11         1      b1 - d1
12         1      c1 - a1
13         1      a1 - d1
14         1      b1 - c1
15         1      d1 - c1
16         1      a1 - b1
17         1      b1 - d1
18         1      c1 - a1
19         1      a1 - d1
20         1      b1 - c1
21         1      d1 - c1
22         1      a1 - b1
23         1      b1 - d1
24         1      c1 - a1
25         1      a1 - d1
26         1      b1 - c1
27         1      d1 - c1
28         1      a1 - b1
29         1      b1 - d1
30         1      c1 - a1, 2:     group_id combinations
1          2      y1 - z1
2          2      x1 - z1
3          2      z1 - x1
4          2      y1 - z1
5          2      x1 - z1
6          2      z1 - x1
7          2      y1 - z1
8          2      x1 - z1
9          2      z1 - x1
10         2      y1 - z1
11         2      x1 - z1
12         2      z1 - x1
13         2      y1 - z1
14         2      x1 - z1
15         2      z1 - x1
16         2      y1 - z1
17         2      x1 - z1
18         2      z1 - x1
19         2      y1 - z1
20         2      x1 - z1
21         2      z1 - x1
22         2      y1 - z1
23         2      x1 - z1
24         2      z1 - x1
25         2      y1 - z1
26         2      x1 - z1
27         2      z1 - x1
28         2      y1 - z1
29         2      x1 - z1
30         2      z1 - x1}