如何在Pandas DF列中对值进行排序并删除重复项

Question

This is probably a very basic question but I haven't been able to find the answer so here goes...

Question:

Is there an say way to sort the values alphabetically while also removing any duplicate instances?

Here's what I have:

data = ['Car | Book | Apple','','Book | Car | Apple | Apple']
df = pd.DataFrame(data,columns=['Labels']
print(df)

    Labels
0   Car | Book | Apple
1   
2   Book | Car | Apple | Apple

Desired Output:

    Labels
0   Apple | Book | Car
1   
2   Apple | Book | Car

Thanks!

Answer 1

str.join after str.split

df=df.replace({' ':''},regex=True)
df.Labels.str.split('|').apply(set).str.join('|')
Out[339]: 
0    Apple|Book|Car
1                  
2    Apple|Book|Car
Name: Labels, dtype: object

Base on the comment adding sorted

df.Labels.str.split('|').apply(lambda x : sorted(set(x),reverse=False)).str.join(' | ')

Answer 2

One way is to use pd.Series.map with sorted & set after splitting by | :

import pandas as pd

data = ['Car | Book | Apple','','Book | Car | Apple | Apple']
df = pd.DataFrame(data,columns=['Labels'])

df['Labels'] = df['Labels'].map(lambda x: ' | '.join(sorted(set(x.split(' | ')))))

#                Labels
# 0  Apple | Book | Car
# 1                    
# 2  Apple | Book | Car

Answer 3

df['Labels'].str.split('|') will split the string on | and return a list

#0             [Car ,  Book ,  Apple]
#1                                 []
#2    [Book ,  Car ,  Apple ,  Apple]
#Name: Labels, dtype: object

See that there are extra spaces in the resulting list elements. One way to remove those is by applying str.strip() to each element in the list:

df['Labels'].str.split('|').apply(lambda x: map(str.strip, x))
#0           [Car, Book, Apple]
#1                           []
#2    [Book, Car, Apple, Apple]
#Name: Labels, dtype: object

Finally we apply the set constructor to remove duplicates, sort the values, and join them back together using " | " as a separator:

df['Labels'] = df['Labels'].str.split('|').apply(
    lambda x: " | ".join(sorted(set(map(str.strip, x))))
)
print(df)
#               Labels
#0  Apple | Book | Car
#1                    
#2  Apple | Book | Car