[FIXED] Splitting a column based on character transition in a dataframe

Issue

I want to make a basic split (see image below) with Python/Pandas.

For people who use Excel/PowerQuery, there is a nice function that allow them to do that, Splitter.SplitTextByCharacterTransition.

I tried to make an equivalent in Python using itertools but unfortunately I get a copy of the column "Col1":

import pandas as pd
import itertools

df = pd.DataFrame({'Col1': ['304', '321A', '14', '319B', '315', '88', '242C', '243']})

df['Col2'] = [''.join(a) for b, a in itertools.groupby(df['Col1'])]

>>> df

    Col1    Col2
0   304     304
1   321A    321A
2   14      14
3   319B    319B
4   315     315
5   88      88
6   242C    242C
7   243     243

Do you have any suggestions/propositions to fix that, please ?

Solution

Let's try .str.extractall to extract both the number and character then get the first non value in group

out = df.join(df.pop('Col1')
              .str.extractall('(\d+)|([a-zA-Z]+)')
              .groupby(level=0).first()
              .fillna('')
              .set_axis(['Col1', 'Col2'], axis=1))

print(out)

  Col1 Col2
0  304
1  321    A
2   14
3  319    B
4  315
5   88
6  242    C
7  243

Answered By - Ynjxsjmh