def recognizers():
rec_map = {
"IP_ADDRESS": IpRecognizer(),
"US_SSN": UsSsnRecognizer(),
"PHONE_NUMBER": UsPhoneRecognizer(),
"US_ITIN": UsItinRecognizer(),
"US_DRIVER_LICENSE": UsLicenseRecognizer(),
"US_BANK_NUMBER": UsBankRecognizer(),
"US_PASSPORT": UsPassportRecognizer(),
"FIN": SgFinRecognizer(),
}
return rec_map
def load_predefined_recognizers(self):
# TODO: Change the code to dynamic loading -
# Task #598: Support loading of the pre-defined recognizers
# from the given path.
# Currently this is not integrated into the init method to speed up
# loading time if these are not actually needed (SpaCy for example) is
# time consuming to load
self.recognizers.extend([
CreditCardRecognizer(),
CryptoRecognizer(), DomainRecognizer(),
EmailRecognizer(), IbanRecognizer(),
IpRecognizer(), NhsRecognizer(),
UsBankRecognizer(), UsLicenseRecognizer(),
UsItinRecognizer(), UsPassportRecognizer(),
UsPhoneRecognizer(), UsSsnRecognizer(),
SpacyRecognizer(), SgFinRecognizer()])
import pytest
from presidio_analyzer import PatternRecognizer, Pattern
from presidio_analyzer.predefined_recognizers import CreditCardRecognizer, \
UsPhoneRecognizer, DomainRecognizer, UsItinRecognizer, \
UsLicenseRecognizer, UsBankRecognizer, UsPassportRecognizer, \
IpRecognizer, UsSsnRecognizer, SgFinRecognizer
from presidio_analyzer.nlp_engine import NlpArtifacts
from tests import TESTS_NLP_ENGINE
ip_recognizer = IpRecognizer()
us_ssn_recognizer = UsSsnRecognizer()
phone_recognizer = UsPhoneRecognizer()
us_itin_recognizer = UsItinRecognizer()
us_license_recognizer = UsLicenseRecognizer()
us_bank_recognizer = UsBankRecognizer()
us_passport_recognizer = UsPassportRecognizer()
sg_fin_recognizer = SgFinRecognizer()
@pytest.fixture(scope="class")
def sentences_with_context(request):
""" Loads up a group of sentences with relevant context words
"""
path = os.path.dirname(__file__) + '/data/context_sentences_tests.txt'
f = open(path, "r")
if not f.mode == 'r':
return []
content = f.read()
f.close()
def recognizer():
return UsBankRecognizer()