将字符串拆分为内部列表而不影响关系

Question

I have a list of lists, lol :

[ ['filiabus', 'filia +N +Abl +Sg', 'filia +N +Dat +Sg'], 
  ['canēs', 'canis +N +Acc +Pl', 'canis +N +Abl +Pl'], ...] 

Each of the inner lists has 3 elements, all of which are currently strings. What I want to do is split the second and third items by a space character to create something like this:

[ 
['filiabus', ['filia', '+N', '+Abl', '+Sg'], ['filia', '+N', '+Dat', '+Sg'] ], 
...
] 

It's important that these new nested lists be a part of the same list which has a first item (eg filiabus ). The first element can be a list unto itself if it makes it easier.

I feel like something like this should work

test=[]
for i in lol:
    for j in i:
        test.append([j[0],j[1].split(' '), j[2].split(' ')])

but it just produces:

>>> test
[['f', ['i'], ['l']], ['f', ['i'], ['l']], ['f', ['i'], ['l']], ['c', ['a'], ['n']], ['c', ['a'], ['n']], ['c', ['a'], ['n']]]

Thanks!

Answer 1

As your initial list, contains 3-length lists, you can directly unpack them in 3 variables like

 for name, v1, v2 in values:

Then, the result is just the first value, and the 2 other both splitted (no separator splits on 'consecutive spaces')

values = [['filiabus', 'filia +N +Abl +Sg', 'filia +N +Dat +Sg'],
          ['canēs', 'canis +N +Acc +Pl', 'canis +N +Abl +Pl']]

result = [[name, v1.split(), v2.split()] for name, v1, v2 in values]

print(result)  # [['filiabus', ['filia', '+N', '+Abl', '+Sg'], ['filia', '+N', '+Dat', '+Sg']], 
                # ['canēs', ['canis', '+N', '+Acc', '+Pl'], ['canis', '+N', '+Abl', '+Pl']]]

Answer 2

You can do it like this, for lists of any size:

lol = [['filiabus', 'filia +N +Abl +Sg', 'filia +N +Dat +Sg'], 
       ['canēs', 'canis +N +Acc +Pl', 'canis +N +Abl +Pl']]


def transform(sublist):
    first, *others = sublist
    return [first, *(item.split() for item in others)]

out = [transform(sublist) for sublist in lol]

print(out)
# [['filiabus', ['filia', '+N', '+Abl', '+Sg'], ['filia', '+N', '+Dat', '+Sg']], 
#  ['canēs', ['canis', '+N', '+Acc', '+Pl'], ['canis', '+N', '+Abl', '+Pl']]]

Answer 3

I believe this is what you are trying to do.

lol = [ ['filiabus', 'filia +N +Abl +Sg', 'filia +N +Dat +Sg'], 
  ['canēs', 'canis +N +Acc +Pl', 'canis +N +Abl +Pl']] 

# Iterate through each list in lol
for i in range(len(lol)):
    # Iterate through each string in the list
    for j in range(len(lol[i])):
        # Only split if string contains a space
        if " " in lol[i][j]:
            # Reassign position
            lol[i][j] = lol[i][j].split(" ")
        
print(lol)
# Prints 
#[['filiabus', ['filia', '+N', '+Abl', '+Sg'], ['filia', '+N', '+Dat', '+Sg']], 
#['canēs', ['canis', '+N', '+Acc', '+Pl'], ['canis', '+N', '+Abl', '+Pl']]]

The critical thing to do is, instead of iterating through each element, iterate through each INDEX of each element. This allows you to reassign to the element's position.

Also, note that when you tried to access j[i] in your example, you are accessing each letter, not each word. Each word would be i[x] .

This answer is the more readable version; for a shorter list comprehension version, check out azro's answer.

Answer 4

The second for is in excess. Explanation: the variable i will cycle through the lists of lol , while the variable j will cycle through the words of each list.

So for example, in the first iteration you will have:

i=['filiabus', 'filia +N +Abl +Sg', 'filia +N +Dat +Sg']
j='filiabus'

You can now see that j[0] = 'f' and j[1] = 'i' (so j[1].split(' ') = ['i'] ).

Solution: remove the inner loop:

test = []
for i in lol:
    test.append([i[0],i[1].split(' '), i[2].split(' ')])

or a more elegant (and pythonic) way:

test = [[i[0], i[1].split(), i[2].split()] for i in lol]