def anony_txt(txtFile):
anon_en = AnonymizerChain(Anonymization('en_US'))
anon_en.add_anonymizers(EmailAnonymizer, UriAnonymizer,
MacAddressAnonymizer, PhoneNumberAnonymizer,
INNAnonymizer, NamedEntitiesAnonymizer4('en'))
anon_ru = AnonymizerChain(Anonymization('ru_RU'))
anon_ru.add_anonymizers(EmailAnonymizer, UriAnonymizer,
MacAddressAnonymizer, PhoneNumberAnonymizer,
INNAnonymizer, NamedEntitiesAnonymizer4('ru'))
with open(txtFile, 'r', encoding='utf-8-sig') as f:
anfile = open(txtFile[:-4] + "_anonymized.txt", 'w', encoding='utf-8')
rejfile = open(txtFile[:-4] + "_rejected_anonymization.txt",
'w',
encoding='utf-8')
line = f.readline()
while line:
cfg.NERnumber = 0
anline_en = anon_en.anonymize(line.split("\t")[0])
NERnumber_en = cfg.NERnumber
cfg.NERnumber = 0
anline_ru = anon_ru.anonymize(line.split("\t")[1].strip())
NERnumber_ru = cfg.NERnumber
if NERnumber_en == NERnumber_ru:
anfile.write(anline_en + "\t" + anline_ru + "\n")
else:
anfile.write(line)
rejfile.write(anline_en + "\t" + anline_ru + "\n")
line = f.readline()
anfile.close()
rejfile.close()
def get_anonymize_text(self, query):
anon = AnonymizerChain(Anonymization('en_US'))
anon.add_anonymizers(FilePathAnonymizer,\
EmailAnonymizer, UriAnonymizer,MacAddressAnonymizer,Ipv4Anonymizer, Ipv6Anonymizer,\
NamedEntitiesAnonymizer('en'))
anonymizedText = anon.anonymize(query)
return anonymizedText
from anonymization import Anonymization
text = "This is a message to Marco"
class MarcoAnonymizer():
'''
Replace all occurences of Marco or marco with a star (*)
'''
def __init__(self, anonymization: Anonymization):
self.anonymization = anonymization
def anonymize(self, text: str) -> str:
return text.replace(r'Marco', '*')
anon = Anonymization(None)
marcoAnonymizer = MarcoAnonymizer(anon)
print(marcoAnonymizer.anonymize(text))
from anonymization import Anonymization, AnonymizerChain, EmailAnonymizer, NamedEntitiesAnonymizer text = "Hi John,\nthanks for you for subscribing to Superprogram, feel free to ask me any question at [email protected] \n Superprogram the best program!" anon = AnonymizerChain(Anonymization('en_US')) anon.add_anonymizers(EmailAnonymizer, NamedEntitiesAnonymizer('en_core_web_lg')) clean_text, patch = anon.pseudonymize(text) print(clean_text) print(patch) revert_text = anon.revert(clean_text, patch) print(text == revert_text)
from anonymization import Anonymization, PhoneNumberAnonymizer
text = "C'est bien le 0611223344 ton numéro ?"
anon = Anonymization('fr_FR')
phoneAnonymizer = PhoneNumberAnonymizer(anon)
print(phoneAnonymizer.anonymize(text))