Skip to content

Commit

Permalink
Add anonymization features, enhanced system msg cleanup and added cor…
Browse files Browse the repository at this point in the history 
…responding tests
  • Loading branch information
@zoharbabin
zoharbabin committed Dec 26, 2024
1 parent b49ad9b commit d515c32
Show file tree
Hide file tree
Showing 4 changed files with 640 additions and 5 deletions.
314 changes: 314 additions & 0 deletions anonymizer.py
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,314 @@
import re
import random
from typing import Optional, Tuple

class PhoneAnonymizer:
# Fun word lists for generating random usernames
ADJECTIVES = [
"Happy", "Bouncy", "Cosmic", "Dancing", "Electric", "Fluffy", "Glowing",
"Hyper", "Jazzy", "Magical", "Nifty", "Quirky", "Sparkly", "Whimsical",
"Zesty", "Bubbly", "Cheerful", "Dazzling", "Energetic", "Fantastic"
]

NOUNS = [
"Penguin", "Unicorn", "Dragon", "Phoenix", "Wizard", "Ninja", "Panda",
"Robot", "Dolphin", "Koala", "Raccoon", "Tiger", "Llama", "Octopus",
"Platypus", "Narwhal", "Giraffe", "Kangaroo", "Hedgehog", "Chameleon"
]

# Regex pattern for cleaning unicode control characters
UNICODE_CONTROL_CHARS = re.compile(r'[\u200e\u200f\u202a-\u202f\xa0]+')

# Common international country codes and their lengths
COUNTRY_CODE_LENGTHS = {
'1': 1, # North America
'20': 2, # Egypt
'27': 2, # South Africa
'30': 2, # Greece
'31': 2, # Netherlands
'32': 2, # Belgium
'33': 2, # France
'34': 2, # Spain
'36': 2, # Hungary
'39': 2, # Italy
'40': 2, # Romania
'41': 2, # Switzerland
'43': 2, # Austria
'44': 2, # UK
'45': 2, # Denmark
'46': 2, # Sweden
'47': 2, # Norway
'48': 2, # Poland
'49': 2, # Germany
'51': 2, # Peru
'52': 2, # Mexico
'53': 2, # Cuba
'54': 2, # Argentina
'55': 2, # Brazil
'56': 2, # Chile
'57': 2, # Colombia
'58': 2, # Venezuela
'60': 2, # Malaysia
'61': 2, # Australia
'62': 2, # Indonesia
'63': 2, # Philippines
'64': 2, # New Zealand
'65': 2, # Singapore
'66': 2, # Thailand
'81': 2, # Japan
'82': 2, # South Korea
'84': 2, # Vietnam
'86': 2, # China
'90': 2, # Turkey
'91': 2, # India
'92': 2, # Pakistan
'93': 2, # Afghanistan
'94': 2, # Sri Lanka
'95': 2, # Myanmar
'98': 2, # Iran
'212': 3, # Morocco
'213': 3, # Algeria
'216': 3, # Tunisia
'218': 3, # Libya
'220': 3, # Gambia
'221': 3, # Senegal
'222': 3, # Mauritania
'223': 3, # Mali
'234': 3, # Nigeria
'249': 3, # Sudan
'250': 3, # Rwanda
'251': 3, # Ethiopia
'252': 3, # Somalia
'253': 3, # Djibouti
'254': 3, # Kenya
'255': 3, # Tanzania
'256': 3, # Uganda
'257': 3, # Burundi
'258': 3, # Mozambique
'260': 3, # Zambia
'261': 3, # Madagascar
'263': 3, # Zimbabwe
'264': 3, # Namibia
'265': 3, # Malawi
'266': 3, # Lesotho
'267': 3, # Botswana
'268': 3, # Swaziland
'269': 3, # Comoros
'297': 3, # Aruba
'298': 3, # Faroe Islands
'299': 3, # Greenland
'350': 3, # Gibraltar
'351': 3, # Portugal
'352': 3, # Luxembourg
'353': 3, # Ireland
'354': 3, # Iceland
'355': 3, # Albania
'356': 3, # Malta
'357': 3, # Cyprus
'358': 3, # Finland
'359': 3, # Bulgaria
'370': 3, # Lithuania
'371': 3, # Latvia
'372': 3, # Estonia
'373': 3, # Moldova
'374': 3, # Armenia
'375': 3, # Belarus
'376': 3, # Andorra
'377': 3, # Monaco
'378': 3, # San Marino
'380': 3, # Ukraine
'381': 3, # Serbia
'382': 3, # Montenegro
'385': 3, # Croatia
'386': 3, # Slovenia
'387': 3, # Bosnia and Herzegovina
'389': 3, # Macedonia
'420': 3, # Czech Republic
'421': 3, # Slovakia
'423': 3, # Liechtenstein
'500': 3, # Falkland Islands
'501': 3, # Belize
'502': 3, # Guatemala
'503': 3, # El Salvador
'504': 3, # Honduras
'505': 3, # Nicaragua
'506': 3, # Costa Rica
'507': 3, # Panama
'509': 3, # Haiti
'590': 3, # Guadeloupe
'591': 3, # Bolivia
'592': 3, # Guyana
'593': 3, # Ecuador
'594': 3, # French Guiana
'595': 3, # Paraguay
'596': 3, # Martinique
'597': 3, # Suriname
'598': 3, # Uruguay
'599': 3, # Netherlands Antilles
'670': 3, # East Timor
'672': 3, # Norfolk Island
'673': 3, # Brunei
'674': 3, # Nauru
'675': 3, # Papua New Guinea
'676': 3, # Tonga
'677': 3, # Solomon Islands
'678': 3, # Vanuatu
'679': 3, # Fiji
'680': 3, # Palau
'681': 3, # Wallis and Futuna
'682': 3, # Cook Islands
'683': 3, # Niue
'685': 3, # Samoa
'686': 3, # Kiribati
'687': 3, # New Caledonia
'688': 3, # Tuvalu
'689': 3, # French Polynesia
'690': 3, # Tokelau
'691': 3, # Micronesia
'692': 3, # Marshall Islands
'850': 3, # North Korea
'852': 3, # Hong Kong
'853': 3, # Macau
'855': 3, # Cambodia
'856': 3, # Laos
'880': 3, # Bangladesh
'886': 3, # Taiwan
'960': 3, # Maldives
'961': 3, # Lebanon
'962': 3, # Jordan
'963': 3, # Syria
'964': 3, # Iraq
'965': 3, # Kuwait
'966': 3, # Saudi Arabia
'967': 3, # Yemen
'968': 3, # Oman
'970': 3, # Palestinian Territory
'971': 3, # United Arab Emirates
'972': 3, # Israel
'973': 3, # Bahrain
'974': 3, # Qatar
'975': 3, # Bhutan
'976': 3, # Mongolia
'977': 3, # Nepal
'992': 3, # Tajikistan
'993': 3, # Turkmenistan
'994': 3, # Azerbaijan
'995': 3, # Georgia
'996': 3, # Kyrgyzstan
'998': 3, # Uzbekistan
}

@staticmethod
def _generate_fun_username() -> str:
"""Generate a random fun username from adjective + noun combination."""
adjective = random.choice(PhoneAnonymizer.ADJECTIVES)
noun = random.choice(PhoneAnonymizer.NOUNS)
return f"{adjective}{noun}"

@staticmethod
def _clean_phone(phone: str) -> str:
"""Clean phone number by removing unicode control characters and normalizing spaces."""
# Remove unicode control characters and normalize spaces
cleaned = PhoneAnonymizer.UNICODE_CONTROL_CHARS.sub('', phone)
# Replace various types of hyphens/dashes with standard hyphen
cleaned = re.sub(r'[‑–—]', '-', cleaned)
# Remove any remaining non-digit characters except plus
cleaned = ''.join(c for c in cleaned if c.isdigit() or c == '+')
return cleaned

@staticmethod
def _extract_country_code(phone: str) -> Tuple[str, str]:
"""
Extract country code and remaining digits from phone number.
Uses a comprehensive list of international country codes.
"""
if not phone.startswith('+'):
if len(phone) == 10: # Standard North American number
return '+1', phone
return '', phone

# Remove the plus sign for checking
digits = phone[1:]

# Try matching country codes from longest to shortest
for length in [3, 2, 1]:
if len(digits) >= length:
potential_code = digits[:length]
if potential_code in PhoneAnonymizer.COUNTRY_CODE_LENGTHS and \
PhoneAnonymizer.COUNTRY_CODE_LENGTHS[potential_code] == length:
return f'+{potential_code}', digits[length:]

# If no known country code is found, assume first digit is country code
return f'+{digits[0]}', digits[1:]

@staticmethod
def anonymize(phone: str, username: Optional[str] = None) -> Tuple[str, str]:
"""
Anonymize a phone number and return both the anonymized number and display name.
Args:
phone: The phone number to anonymize. Can be in any format:
- International with country code (e.g., +1234567890)
- North American format (e.g., (123) 456-7890)
- Local format (e.g., 123-4567)
- Any other format with digits and common separators
username: Optional real username associated with the number
Returns:
Tuple of (anonymized_phone, display_name)
The display_name will be prefixed with [👤] for real usernames
or [🎭] for generated usernames
"""
# Clean and format the phone number
cleaned_phone = PhoneAnonymizer._clean_phone(phone)

if len(cleaned_phone) <= 5: # Handle short numbers
# For short numbers, keep only the last digit
last_digit = cleaned_phone[-1]
stars = '*' * (len(cleaned_phone) - 1)
anonymized = f"{stars}{last_digit}"
else:
# Extract country code and remaining digits
country_code, remaining = PhoneAnonymizer._extract_country_code(cleaned_phone)

# Keep last 4 digits
last_four = remaining[-4:] if len(remaining) >= 4 else remaining[-1:]
# Fill middle with stars
stars = '*' * 4 # Always use 4 stars for consistency

anonymized = f"{country_code}{stars}{last_four}"

# Generate display name
if username:
# Clean the username of any control characters
clean_username = PhoneAnonymizer.UNICODE_CONTROL_CHARS.sub('', username)
display_name = f"[👤] {clean_username}" # Real username
else:
random_name = PhoneAnonymizer._generate_fun_username()
display_name = f"[🎭] {random_name}" # Generated username

return anonymized, display_name

# Example usage
if __name__ == "__main__":
# Test with various phone numbers and usernames
test_cases = [
("+1 (587) 998‑1598", None),
("+44 7464 758875", "John"),
("+91 80885 89616", "Alice"),
("+234 811 635 8644", None),
("‪+1 (443) 739‑6663‬", "Bob"),
# Additional international formats
("+86 123 4567 8901", None), # China
("+971 50 123 4567", None), # UAE
("+81 90 1234 5678", None), # Japan
("+7 916 123 45 67", None), # Russia
("+972 50 123 4567", None), # Israel
]

for phone, name in test_cases:
anon_phone, display = PhoneAnonymizer.anonymize(phone, name)
print(f"Original: {phone}")
print(f"Anonymized: {anon_phone}")
print(f"Display: {display}")
print("-" * 40)
Loading

0 comments on commit d515c32

Please sign in to comment.