Anonymize without LangChain

anonymize.py

import re
import spacy
import hashlib
from faker import Faker

fake = Faker()
nlp = spacy.load("en_core_web_sm")

DATA = []
DATA_ANON = []

email_map = {}
name_map = {}
phone_map = {}

# Function to encrypt any string (names, emails, phone numbers)
def encrypt_string(text):
    return hashlib.sha256(text.encode()).hexdigest()

# Generate fake data
for _ in range(2):
    name = fake.name()
    phone = fake.phone_number()
    email = fake.email()
    DATA.append(f"My name is {name}, call me at {phone} or email me at {email}. I say it again, my name is {name}, call me at {phone} or email me at {email}")

# Anonymize data
for sentence in DATA:
    doc = nlp(sentence)
    anon_sentence = sentence

    # Detect and replace names
    for ent in doc.ents:
        if ent.label_ == "PERSON":
            encrypted_name = encrypt_string(ent.text)
            
            if encrypted_name not in name_map:
                new_name = fake.name()
                name_map[encrypted_name] = new_name
            
            anon_sentence = anon_sentence.replace(ent.text, name_map[encrypted_name])

    # Detect and replace phone numbers
    for phone in re.findall(r'\(\d{3}\)\d{3}-\d{4}', sentence):
        encrypted_phone = encrypt_string(phone)
        
        if encrypted_phone not in phone_map:
            new_phone = fake.phone_number()
            phone_map[encrypted_phone] = new_phone
        
        anon_sentence = anon_sentence.replace(phone, phone_map[encrypted_phone])

    # Detect and replace emails
    for email in re.findall(r'\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Z|a-z]{2,}\b', sentence):
        encrypted_email = encrypt_string(email)
        
        if encrypted_email not in email_map:
            new_email = fake.email()
            email_map[encrypted_email] = new_email
        
        anon_sentence = anon_sentence.replace(email, email_map[encrypted_email])

    DATA_ANON.append(anon_sentence)

print("Original Data:")
for sentence in DATA:
    print(sentence)

print("\nAnonymized Data:")
for sentence in DATA_ANON:
    print(sentence)