The second argument of python replace function

I would like to sort words in roots, and then replace the sentence by each root. The below code throw me an error

TypeError: replace() argument 2 must be str, not list

def replace_words(roots, sentence):

    for word in sentence.split():
        matches = [root for root in roots if word.startswith(root)]
    #print(matches)
    
        if len(matches) >0:
           sentence = sentence.replace(word, sorted(matches, key=lambda x:len(x)))

    return sentence.strip()   

But if we add [0] behind sorted( ), it will fix the error:

Looks like [0] converted the second argument of sorted function from list to string, but I really have trouble understanding what is the underlying process. Could anyone kindly explain it?

>Solution :

You’re overthinking the error…

Does it make sense if you write it like this?

sorted_roots = sorted(matches, key=lambda x:len(x))
sentence = sentence.replace(word, sorted_roots)

sorted() returns a list. str.replace() needs 2 string arguments, not a string, and a list.

By adding [0] to the end of sorted(), you’ve gotten the first element of the sorted list (the shortest root)