Python: Compare last elements in a list

Question

I am working with a column containing lists of strings, and would like to compare the last element in each row. If the final elements do not match, I want to create a new variable that would have the first and last elements concatenated like this: element[0].element[-1]

If they do match, I'd like to differentiate between them by appending the next element in the list: element[0].element[-2].element[-1]

I have made this column a list from its original format. Here is a snippet of the original variable from the pandas dataframe:

apple.banana.pear
apple.starfruit.grape
apple.kiwi.orange.pear
apple.durian.coconut
Name: original, Length: 4, dtype: string

mylist = df['original'].apply(lambda x: x.split('.'))

My current list:

[apple, banana, pear]
[apple, starfruit, grape]
[apple, kiwi, orange, pear]
[apple, durian, coconut]

Desired output:

apple.banana.pear
apple.grape
apple.orange.pear
apple.coconut

I'm not sure if making it into a list is optimal, but figured it would be easier to access each portion as an element. That may not be the case. Here is what I've tried:

l = 0
j = l + 1
for l in mylist:
    for j in mylist:
        if mylist[l][-1] == mylist[j][-1]:
            newvar = mylist[l][0] + '.' + mylist[l][-2] + '.' + mylist[l][-1]
        else:
            newvar = mylist[l][0] + '.' + mylist[l][-1]

KeyError: "None of [Index(['apple', 'banana', 'pear'], dtype='object')] are in the [index]"

Any suggestions are greatly appreciated.

Answer 1

We can do

s=df.original.str.split('.')
df['new']=np.where(s.str[-1].duplicated(keep=False),
                   s.str[0]+'.'+s.str[-2]+'.'+s.str[-1],
                   s.str[0]+'.'+s.str[-1])
df
Out[47]: 
                 original                new
0       apple.banana.pear  apple.banana.pear
1   apple.starfruit.grape        apple.grape
2  apple.kiwi.orange.pear  apple.orange.pear
3    apple.durian.coconut      apple.coconut