new-jta

import pandas as pd
import re

# Load CSV file
df = pd.read_csv('input.csv')

# Remove rows where phone is empty
df = df[df['phone'].notna() & (df['phone'].astype(str).str.strip() != '')]

# Function to format and validate Indian phone numbers
def format_indian_number(number):
    number = re.sub(r'\D', '', str(number))  # Remove non-digit characters
    if number.startswith('0') and len(number) == 11:
        number = number[1:]
    if len(number) == 10:
        return '91' + number
    elif len(number) == 12 and number.startswith('91'):
        return number
    else:
        return None

# Function to format names
def format_name(name):
    name = re.sub(r'[^\w\s]', '', str(name))  # Remove punctuation
    name_parts = name.split()[:3]  # Keep only first 3 words
    return ' '.join(name_parts)

# Apply formatting and validation
df['phone'] = df['phone'].apply(format_indian_number)
df['name'] = df['name'].apply(format_name)

# Remove rows with invalid phone numbers
df = df[df['phone'].notna()]

# Keep only 'name' and 'phone' columns
df_final = df[['name', 'phone']]

# Save to new CSV file
df_final.to_csv('output.csv', index=False)

print('Filtered and formatted CSV saved as output.csv.')