Search code examples
pythonconcatenationmultiple-columnsstring-length

Python - Column concatenation based on length of data inside the column

Need help on column concatenation based on length size .

Column3= df["column1"] + "_" + df["column2"]

data = {'column1':['af28912368', 'Nan', '234671', 'asr61239'], 'column2':[701, Nan, 761, 312]}

df = pd.DataFrame(data)

df :

column1 column2
af28912368 701
NaN Nan
234671 761
asr61239 312

If length of the data in column1 is greater than 8 , then need to Insert last 8 symbols of column1 If length of the data in column1 <8 & >0 , Insert value of df['column1'] + (8-Len(df['column1'])) of Blank Spaces If NaN on Column1 , Column3 can remain as NaN

expected result as shown on column3

column1 column2 column3
aa289123sf 701 289123sf_701
NaN 723 Nan
234671 761 234671 _761
asr61239 312 asr61239_312

I tried this :

df["column3"] = df["column1"].str[-8:] + '_' + df["column2"].astype(str) But not working with different length size of df["column1"] . Please help on this one.

Solution

  • you're first row column1 value keeps changing so I'm assuming this is a typo and not part of the exercise

    this script worked for me:

    ───────┬─────────────────────────────────────────────────────────────────────
       │ File: test-so.py
───────┼─────────────────────────────────────────────────────────────────────
   1   │ import pandas as pd
   2   │ from pprint import pprint
   3   │
   4   │ data = {'column1':['af28912368', None, '234671', 'asr61239'], 'column2':[701, None, 761, 312]}
   5   │ df = pd.DataFrame(data)
   6   │
   7   │ def funk(col1,col2):
   8   │     try:
   9   │         tmp = len(col1)
  10   │         if tmp > 8:
  11   │             return col1[:8] +'_'+ str(int(col2))
  12   │         elif tmp <= 8 and tmp > 0:
  13   │             return (int(tmp % 8))*' ' + col1 +'_'+ str(int(col2))
  14   │         else:
  15   │             return None
  16   │     except TypeError as e:
  17   │         return None
  18   │
  19   │ def main():
  20   │     df['column3'] = df.apply(lambda row: funk(row['column1'],row['column2']),axis=1)
  21   │     pprint(df)
  22   │
  23   │ if __name__ == '__main__':
  24   │     main()

    from command line:

    python test-so.py

    yields:

    pic