Enumerate in nested for loop with list comprehension for inner increment

Question

Following up on this question on how to implement enumerate within a list comprehension for an outer increment, I am wondering if there is a way to adjust the accepted answer in order to create an inner increment this time; I am looking to generate this output:

['D_50_1-1_1',
 'D_50_2-1_2',
 'D_50_3-1_3',
 'D_80_1-2_1',
 'D_80_2-2_2',
 'D_80_3-2_3',
 'D_20_1-3_1',
 'D_20_2-3_2',
 'D_20_3-3_3',
 'D_60_1-4_1',
 'D_60_2-4_2',
 'D_60_3-4_3',
 'E_35_1-5_1',
 'E_35_2-5_2',
 'E_35_3-5_3',
 'E_25_1-6_1',
 'E_25_2-6_2',
 'E_25_3-6_3']

where the inner increment is the second to last column ( 1,1,1,2,2,3,3,3...6,6,6 ).

At the moment, I achieve this by hard-coding the increment (eg. list(range(1, 4+1)) ) depending on the length of levelB , which is obviously prone to errors:

[f'{levelA}_{levelB}_{sub}-{inner_increment}_{sub}'
 for levelA, pairs in
          [('D',  dict(zip([50,80,20,60], list(range(1, 4+1))))),
           ('E',  dict(zip([35, 25],      list(range(5, 7+1)))))]
                                               
 for levelB,inner_increment in pairs.items()
 for sub in range(1, 3+1)
]

Is there a way to produce the same output by introducing enumerate while keeping the list comprehension approach and ideally wihhout any imports, similar to the accepted answer in the other question?

Answer 1

The tricky part is the partitioning of levelb. We have to enumerate over the whole thing once, but have a way to separate it so only parts of it get matched with each levela. Here's a version with some preliminary steps:

levela = 'D', 'E'
levelb = (50, 80, 20, 60), (35, 25)
levelc = 1, 2, 3

pairs = [
    (a, b)
    for a, partition in zip(levela, levelb)
    for b in partition
]

result = [
    (a, b, n, i)
    for i, (a, b) in enumerate(pairs, start=1)
    for n in (1, 2, 3)
]

actual = [f'{a}_{b}_{n}-{i}_{n}'.format(a, b, i, n) for a, b, n, i in result]

desired = ['D_50_1-1_1',
 'D_50_2-1_2',
 'D_50_3-1_3',
 'D_80_1-2_1',
 'D_80_2-2_2',
 'D_80_3-2_3',
 'D_20_1-3_1',
 'D_20_2-3_2',
 'D_20_3-3_3',
 'D_60_1-4_1',
 'D_60_2-4_2',
 'D_60_3-4_3',
 'E_35_1-5_1',
 'E_35_2-5_2',
 'E_35_3-5_3',
 'E_25_1-6_1',
 'E_25_2-6_2',
 'E_25_3-6_3']

for a, b in zip(result, desired):
    if a == b:
        print (a)
    else:
        print (a, b)

Not being able to use itertools is a pain.

Answer 2

I eventually managed to solve it myself. Posting for completeness.

[f'{levelA}_{levelB}_{sub}-{increment}_{sub}' 
      for increment, (levelA, levelB) in enumerate(
      (   
           (levelA,levelB)
           for levelA,lists in 
                  [('D',  [50,80,20,60]), 
                   ('E',  [35, 25])]
      for levelB in lists
      ), start=1,
      )
      for sub in range(1,3+1)
      ]
Out[67]: 
['D_50_1-1_1',
 'D_50_2-1_2',
 'D_50_3-1_3',
 'D_80_1-2_1',
 'D_80_2-2_2',
 'D_80_3-2_3',
 'D_20_1-3_1',
 'D_20_2-3_2',
 'D_20_3-3_3',
 'D_60_1-4_1',
 'D_60_2-4_2',
 'D_60_3-4_3',
 'E_35_1-5_1',
 'E_35_2-5_2',
 'E_35_3-5_3',
 'E_25_1-6_1',
 'E_25_2-6_2',
 'E_25_3-6_3']