Extract text from a series into new columns

Question

I've tries the following code with the following error message:

AttributeError: 'DataFrame' object has no attribute 'str'

I'm referencing a string series, then why am i getting a "Dataframe" error?

df['PO'] =  df['extract'].str.extract(r"(?:^(?=[A-Z\d/-]+$)|\bPO\W*)([A-Z\d/-]+)")

If check `print (df['extract'])` it return duplicated columns `extract` in df? — jezrael, Feb 10 '22 at 13:14

score 0 · Accepted Answer · answered Feb 10 '22 at 13:17

0

Problem is with duplicated columns names, so if select one column extract get all columns in DataFrame.

print (df['extract'])

For deduplicted is simpliest set columns names or some another solution:

df.columns= ['a','b']

answered Feb 10 '22 at 13:17

jezrael

i'm sorry, i'm not sure what you mean. – amit singh Feb 10 '22 at 13:22
@amitsingh - what is not clear? – jezrael Feb 10 '22 at 13:25
1) why is the columns name df['extract'] duplicated? Is that a data quality issue? and in my sitaution one of the extracts has data, whereasthe other has only NaN values. – amit singh Feb 10 '22 at 13:33
@amitsingh - `why is the columns name df['extract'] duplicated?` Because data, because some wrong pandas operation, because... not idea. – jezrael Feb 10 '22 at 13:34
@amitsingh - Maybe help me see your complete code before `df['PO'] = df['extract'].str.extract` – jezrael Feb 10 '22 at 13:35
1

i'm sorry for being annoying, but thank you for pointing me to the issue. Many thanks! – amit singh Feb 10 '22 at 13:42

1 Answers1