python script to split csv file month wise in separate folders year wise with selected columns in the output files

split_csv_year_and_month_wise.py

import os
import pandas as pd
from pytz import timezone

# Read the original CSV file
csv_file_path = input('enter csv path ')
df = pd.read_csv(csv_file_path)

# Extract month and year from 'created_at'
df['created_at'] = pd.to_datetime(df['created_at'])
df['month'] = df['created_at'].dt.month
df['year'] = df['created_at'].dt.year

# Convert UTC to IST
utc_timezone = timezone('UTC')
ist_timezone = timezone('Asia/Kolkata')
df['created_at'] = df['created_at'].dt.tz_localize(utc_timezone).dt.tz_convert(ist_timezone)

# Ensure data types for customer_id and product_id
df['customer_id'] = pd.to_numeric(df['customer_id'], errors='coerce').astype(pd.Int64Dtype())
df['product_id'] = pd.to_numeric(df['product_id'], errors='coerce').astype(pd.Int64Dtype())

# Create folders for each year
output_dir = 'output_month_wise_data'
os.makedirs(output_dir, exist_ok=True)

# Write data to separate CSV files
for year, group in df.groupby('year'):
    year_folder = os.path.join(output_dir, str(year))
    os.makedirs(year_folder, exist_ok=True)
    for month, month_data in group.groupby('month'):
        month_name = pd.to_datetime(f'2022-{month}-01').strftime('%B').lower()
        month_csv_path = os.path.join(year_folder, f'{month_name}.csv')
        selected_columns = ['customer_id', 'product_id', 'created_at']
        month_data[selected_columns].to_csv(month_csv_path, index=False)

print("CSV files created successfully!")