Create a new column with the minimum difference (in days) between today's date and future dates.

min_future_delta_date.py

import pandas as pd
from datetime import datetime

def min_date_difference(df, date_columns):
    """
    Create a new column with the minimum difference (in days) between today's date and future dates.

    Parameters:
    - df: pandas DataFrame
    - date_columns: list of str, column names containing datetime values

    Returns:
    - pandas DataFrame with a new column 'min_date_difference' in days
    """
    today = datetime.today()

    # Initialize the new column with None
    df['min_date_difference'] = None

    # Iterate through specified date columns
    for column in date_columns:
        future_dates = pd.to_datetime(df[column])
        
        # Filter only future dates
        future_dates = future_dates[future_dates > today]
        
        if not future_dates.empty:
            differences = (future_dates - today).dt.days

            # Update 'min_date_difference' column with the minimum positive difference in days
            min_positive_difference = differences[differences > 0].min()
            if not pd.isnull(min_positive_difference):
                idx_min_difference = differences.idxmin()
                df.at[idx_min_difference, 'min_date_difference'] = min_positive_difference

    return df

# Example usage
data = {'date1': ['2022-01-01', '2023-02-01', '2023-03-01'],
        'date2': ['2022-04-01', '2024-05-01', '2023-06-01'],
        'date3': ['2022-07-01', '2023-08-01', '2023-09-01']}

df = pd.DataFrame(data)
date_columns = ['date1', 'date2', 'date3']

df = min_date_difference(df, date_columns)
print(df)