Skip to content

Latest commit

 

History

History
99 lines (78 loc) · 2.74 KB

README.md

File metadata and controls

99 lines (78 loc) · 2.74 KB
Raw

PiiRegex Build Status

This wouldn't have been possible without CommonRegex. Thanks!

Attempt to find PII in regex either using a specific PII type, or search through everything available.

Pull requests welcome!

Install via pip.

pip install piiregex

Tests are available through pytest.

pip install -r dev_requirements.text
pytest -vv

Usage

>>> from piiregex import PiiRegex
>>> parsed_text = PiiRegex("""John, please get that article on www.linkedin.com to me by 5:00PM 
                               on Jan 9th 2012. 4:00 would be ideal, actually. If you have any 
                               questions, You can reach me at (519)-236-2723x341 or get in touch with
                               my associate at harold.smith@gmail.com""")
>>> parsed_text.times
['5:00PM', '4:00']
>>> parsed_text.dates
['Jan 9th 2012']
>>> parsed_text.phones
['(519)-236-2727']
>>> parsed_text.phones_with_exts
['(519)-236-2723x341']
>>> parsed_text.emails
['harold.smith@gmail.com']

Alternatively, you can generate a single PiiRegex instance and use it to parse multiple segments of text.

>>> parser = PiiRegex()
>>> parser.times("When are you free?  Do you want to meet up for coffee at 4:00?")
['4:00']

Finally, all regular expressions used are publicly exposed.

>>> from piiregex import email
>>> import re
>>> text = "...get in touch with my associate at harold.smith@gmail.com"
>>> re.sub(email, "anon@example.com", text)
'...get in touch with my associate at anon@example.com'
>>> from piiregex import time
>>> for m in time.finditer("Does 6:00 or 7:00 work better?"):
>>>     print(m.start(), m.group())
5 6:00 
13 7:00

Most importantly (for our use case) any_match iterates through all regexes to match anything.

>>> from piiregex import PiiRegex
>>> parsed_text = PiiRegex("07123 123123") # should match a UK phone number. 
>>> parsed_text.any_match()
True

Please note that this module is currently English/US and UK specific. Due to the European nature of GDPR though this is being expanded. PRs are welcome.

Supported Methods/Attributes

  • obj.dates, obj.dates()
  • obj.times, obj.times()
  • obj.phones, obj.phones()
  • obj.phones_with_exts, obj.phones_with_exts()
  • obj.emails, obj.emails()
  • obj.ips, obj.ips()
  • obj.ipv6s, obj.ipv6s()
  • obj.credit_cards, obj.credit_cards()
  • obj.btc_addresses, obj.btc_addresses()
  • obj.street_addresses, obj.street_addresses()
  • obj.postcodes, obj.postcodes()
  • obj.ukphones, obj.ukphones()