Converting tuple occurences to string in dataframe column

Question

I have a dataframe column that has the occasional tuple inserted into it. I would like to join all of these tuples into one string separated by ','.

EX

Data      People
 A        XYZ
 B        ABX,LMN
 C       ('OPP', 'GGG')
 D        OAR

I am only trying to 'target' the tuple here and convert it to a string giving the following dataframe:

Data      People
 A        XYZ
 B        ABX,LMN
 C        OPP,GGG
 D        OAR

df['People'] = df['People'].apply(','.join)

I tried this, but it ends up inserting commas in between every character in all of the 'OK' strings.

I'd look into first fixing this upstream -- what monstrosity of a process causes data to be written like this? Fix that first. No need for hacky solutions, especially when dealing with mixed object columns. — cs95
– cs95, Commented Jun 25, 2019 at 17:59

harpan · Accepted Answer · 2019-06-25 18:10:09Z

3

If you must, you can do something like below.

df['People'] = df['People'].apply(lambda x: ', '.join(x) if isinstance(x,tuple) else x)

Output:

  Data  People
0   A   XYZ
1   B   ABX, LMN
2   C   OPP, GGG
3   D   QAR

edited Jun 25, 2019 at 18:10

answered Jun 25, 2019 at 18:03

harpan

8,6412 gold badges22 silver badges40 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

MaxB Over a year ago

This gives a type error 'type' object is not subscriptable

ALollz Over a year ago

Probably better to have ', '.join(x) to deal with tuples with any number of elements

TPereira Over a year ago

Also probably better to use ', '.join(str(x)) in case any of the elements is an integer

harpan Over a year ago

@TPereira, that will separate out each character in a string.

harpan Over a year ago

@MaxB, have you used tuple as a variable name in your script?

piyush daga · Accepted Answer · 2019-06-25 18:10:07Z

1

This works may not be the most elegant solution:

df = pd.DataFrame({'A': [1, 2, 3, 4], 'B': ['AA', 'ABC, LMN', ('XYZ', 'PQR'), 'OLA']})

# Output

    A   B
0   1   AA
1   2   ABC, LMN
2   3   (XYZ, PQR)
3   4   OLA

df['B'].apply(lambda x: ','.join([val for val in x]) if isinstance(x, tuple) else x)

# Output

0          AA
1    ABC, LMN
2     XYZ,PQR
3         OLA
Name: B, dtype: object

answered Jun 25, 2019 at 18:10

piyush daga

5016 silver badges19 bronze badges

Comments

Andy L. · Accepted Answer · 2019-06-25 18:27:26Z

0

You may avoid apply by using map to create mask True on tuple. Use this mask to slice on rows having tuple and use str.join directly on it.

m = df.People.map(type).eq(tuple)
df.loc[m, 'People'] = df.loc[m, 'People'].str.join(',')


Out[2206]:
  Data   People
0    A      XYZ
1    B  ABX,LMN
2    C  OPP,GGG
3    D      OAR

answered Jun 25, 2019 at 18:27

Andy L.

25.3k4 gold badges20 silver badges30 bronze badges

Collectives™ on Stack Overflow

Converting tuple occurences to string in dataframe column

3 Answers 3

5 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

5 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related