How to split one row into multiple rows in a generic way in pandas?

Question

name           text       group
a|b            a test     m|l|n

I have a DataFrame like above. If there is a delimiter in a column value, I want to split it and put it in a separate line.

columns = ['name', 'text', 'group']            
for column in columns:
   if column == 'name' and column in df:
      df = df.assign(name=df.name.str.split(delimiter)).explode(column)

The problem with this code is that, I have to use multiple if to test the actual column name string, i.e. 'name'. I want to a general way like below:

if column in df:
   df = df.assign(column=df.column.str.split(delimiter)).explode(column)

But this is invalid. Any walk-around to do this?

jezrael · Accepted Answer · 2022-10-17 09:32:10Z

2

Use [] instead dot notation:

delimiter = '|'
column = 'group'

if column in df:
    df = df.assign(**{column:df[column].str.split(delimiter)}).explode(column)
    print (df)
  name    text group
0  a|b  a test     m
0  a|b  a test     l
0  a|b  a test     n

Another idea if need exploding multiple columns:

#get values from columns list if exist in df.columns
cols = df.columns.intersection(columns)         
print (cols)

#assign back splitted columns by dict comprehension and explode by all columns in list cols
df = df.assign(**{x: df[x].str.split(delimiter) for x in cols}).explode(cols)

edited Oct 17, 2022 at 9:32

answered Oct 17, 2022 at 6:05

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

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

8 Comments

marlon Over a year ago

Let me try it! If the first way works, that would be simplest.

marlon Over a year ago

But I got this error message after I made the change: df = df.assign(column=df[column].str.split(delimiter)).explode(column) TypeError: unhashable type: 'list'

jezrael Over a year ago

@marlon - Problem is use old pandas version, is possible upgrade to last?

marlon Over a year ago

Yes, I will try.

marlon Over a year ago

So that requires python3.8, right? I am using Python3.7 and will upgrade it first. Also, would your second way of exploding multiple columns be more efficient? The first method is quite slow when I have a lot of columns that need to be split.

|

Collectives™ on Stack Overflow

How to split one row into multiple rows in a generic way in pandas?

1 Answer 1

8 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

8 Comments

Your Answer

Sign up or log in

Post as a guest

Related