用于将2D列表缩小为其唯一元素，保持顺序的算法

Question

Input:

[[102, 106], [106, 107], [101, 106], [104, 105, 106], [105], [103, 105], [104]]

Desired output:

[102, 107, 101, 106, 105, 103, 104]

Each index of the original array is essentially the list of possibilities for the index in the result array. The goal being to represent each number one time in the result array. I'm having difficulty finding an appropriate type of algorithm for this task. What general approach would one take to a problem of this type?

Answer 1

Here's a simple solution that just uses length of the set of all the solutions as a key to max :

>>> import itertools as it
>>> data = [[102, 106], [106, 107], [101, 106], [104, 105, 106], [105], [103, 105], [104]]
>>> max(it.product(*data), key=lambda x: len(set(x)))
(102, 107, 101, 106, 105, 103, 104)

Answer 2

You could use unique_everseen after you have flattened the list:

# sample input
data = [[102, 106], [106, 107], [101, 106], [104, 105, 106], [105], [103, 105], [104]]

from  more_itertools import unique_everseen
data = list(unique_everseen([a for sub in data for a in sub]))