Python: Copy a Row Value and Add to Another Row where a cell is empty

gandycraby1234703 0
Viewed 48

Trying to figure out how to do this in pandas but currently can't figure it out.

I would like to copy a value from Col A where a cell in Col B is blank and add it to the below rows, only until it reaches the next blank cell in Col B where it will then start again.

My Python isn't the strongest so any pointers would be appreciated as i currently haven't got a clue where to start with this one! I have included the below as an example as to how it currently is and as to how I'd like it. I'm currently just manipulating and cleaning the data in Pandas.

A B
Supply Voltage BLANK
Rated Value 10
Limit 20
Size BLANK
Height 10
Width 20
A B
Supply Voltage BLANK
Supply Voltage - Rated Value 10
Supply Voltage - Limit 20
Size BLANK
Size - Height 10
Size - Width 20

Alessandro answers the original question perfectly however in my case the data is something like this in my format. Where there are boolean Yes/No and unique values mixed in. Would a groupby and fill still work in this case?

A B
Supply Voltage BLANK
Rated Value 10
Limit 20
Work Yes
Size BLANK
Height 11
Depth 14
Width 55
Description BLANK
Time 1432
Date 10/12/2022
Quote Hello World
pandas-loc pandas copy python
1 Answers

Below you can find a working example:

import pandas as pd
import numpy as np

# Recreate example DataFrame
df = pd.DataFrame({
    'A': ['Supply Voltage', 'Rated Value', 'Limit', 'Size', 'Height', 'Width'],
    'B': [np.nan, 10, 20, np.nan, 10, 20]
})


# Add helper column (UPDATE)
l = []
c = 0
for i in df['B']:
    if pd.isnull(i):
        c = c+1
    l.append(c)
df['C'] = l

# Extract names associated with blank cells
blank_names = df.loc[df['B'].isna(), ['A', 'C']]
blank_names.columns = ['BLANK_NAME', 'C']

# Add names associated with blank cells to original DataFrame
df = df.merge(blank_names, on='C', how='left')
df['A'] = np.where(df['B'].notna(), df['BLANK_NAME'] + ' - ' + df['A'], df['A'])

# Display final output
df = df.drop(columns=['C', 'BLANK_NAME'])
df
userVmba10
Related