class TestDates(unittest.TestCase):
def setUp(self):
self.parser = CommonRegex()
def test_numeric(self):
formats = ["1-19-14", "1.19.14", "1.19.14", "01.19.14"]
for f in formats:
self.assertEqual(self.parser.dates(f), [f])
def test_verbose(self):
formats = ["January 19th, 2014", "Jan. 19th, 2014", "Jan 19 2014", "19 Jan 2014"]
for f in formats:
self.assertEqual(self.parser.dates(f), [f])
class PiiAnalyzer(object):
def __init__(self, filepath):
self.filepath = filepath
self.parser = CommonRegex()
self.standford_ner = StanfordNERTagger(
'classifiers/english.conll.4class.distsim.crf.ser.gz')
def analysis(self):
people = []
organizations = []
locations = []
emails = []
phone_numbers = []
street_addresses = []
credit_cards = []
ips = []
data = []
with open(self.filepath, 'rU') as filedata:
reader = csv.reader(filedata)
for row in reader:
data.extend(row)
for text in row:
emails.extend(self.parser.emails(text))
phone_numbers.extend(
self.parser.phones("".join(text.split())))
street_addresses.extend(self.parser.street_addresses(text))
credit_cards.extend(self.parser.credit_cards(text))
ips.extend(self.parser.ips(text))
for title, tag in self.standford_ner.tag(set(data)):
if tag == 'PERSON':
people.append(title)
if tag == 'LOCATION':
locations.append(title)
if tag == 'ORGANIZATION':
organizations.append(title)
return {
'people': people,
'locations': locations,
'organizations': organizations,
'emails': emails,
'phone_numbers': phone_numbers,
'street_addresses': street_addresses,
'credit_cards': credit_cards,
'ips': ips
}
def addresses(data):
st_address = []
loc = ''
loc_list = []
parsed_text = CommonRegex(data)
st_address = parsed_text.street_addresses
words = nltk.word_tokenize(data)
taggsets = nltk.pos_tag(words)
namedEnt = nltk.ne_chunk(taggsets, binary=False)
for subtree in namedEnt.subtrees():
if subtree.label() == 'GPE':
l = []
# print((subtree.leaves()))
for leaf in subtree.leaves():
l.append(leaf[0])
loc = ' '.join(l)
if loc not in loc_list:
loc_list.append(loc)
loc_list.extend(st_address)
# for add in loc_list:
# data = data.replace(add, block)
return data, loc_list
def check_for_address_format(addresses):
final = []
for item in addresses:
parsed = CommonRegex(str(item))
if parsed.street_addresses:
final.append(parsed)
return len(final)
def dates(data):
parsed_text = CommonRegex(data)
dates_list = parsed_text.dates
# for date in dates_list:
# data = data.replace(date, block)
return data, dates_list
class TestTimes(unittest.TestCase):
def setUp(self):
self.parser = CommonRegex()
def test_times(self):
formats = ["09:45", "9:45", "23:45", "9:00am", "9am", "9:00 A.M.", "9:00 pm"]
for f in formats:
self.assertEqual(self.parser.times(f), [f])
def getFields(filename):
filenamer = filename
text = open(filenamer).read()
text = text.upper()
#print text
#parser.parse(text)
#print Parser.contacttest_address =
parsed_text = CommonRegex(tex)
print addresses
def index():
if request.method == 'GET':
return 'OK'
elif request.method == 'POST':
# Store the IP address of the requester
request_ip = ipaddress.ip_address(u'{0}'.format(request.remote_addr))
# If GHE_ADDRESS is specified, use it as the hook_blocks.
if os.environ.get('GHE_ADDRESS', None):
hook_blocks = [os.environ.get('GHE_ADDRESS')]
# Otherwise get the hook address blocks from the API.
else:
hook_blocks = requests.get(
'https://api.github.com/meta').json()['hooks']
if request.headers.get('X-GitHub-Event') == "ping":
return json.dumps({'msg': 'Hi!'})
if request.headers.get('X-GitHub-Event') == 'pull_request':
merge_state = request.json['pull_request']['state']
merge_body = request.json['pull_request']['body']
if (merge_state == 'closed'):
print('Merge state closed')
print('Merge Body: ' + merge_body)
parsed_bounty_issue = re.findall(r"#(\w+)", merge_body)[0]
repository_path_encode = str(repository_path)
repository_path_encode = repository_path.encode('utf-8')
bounty_issue_encode = str(parsed_bounty_issue)
bounty_issue_encode = bounty_issue_encode.encode('utf-8')
passphrase = hashlib.sha256(repository_path +
issue_title).hexdigest()
addresses = CommonRegex(merge_body).btc_addresses[0]
parsed_bounty_issue = re.findall(r"#(\w+)", merge_body)
bounty_address = github.get_address_from_issue(
parsed_bounty_issue)
amount = utils.get_address_balance(bounty_address)
with open(DEFAULT_WALLET_PATH, 'r') as f:
json_data = json.load(f)
issue_name = json_data[parsed_bounty_issue]
multisig_wallet.send_bitcoin(str(issue_name), str(addresses),
int(amount * 1e8),
str(passphrase))
return json.dumps({'message': 'Pull request received'})
return json.dumps({'message': 'Pull request payout failed'})
if request.headers.get('X-GitHub-Event') == 'issue_comment':
comment_data = {
'url': request.json['comment']['issue_url'],
'payout_address': request.json['issue']['labels'][0]['name'],
'payout_amount': request.json['issue']['labels'][1]['name'],
'body': request.json['comment']['body']
}
print(comment_data)
return json.dumps({'message': 'Issue comment received'})
class TestLinks(unittest.TestCase):
def setUp(self):
self.parser = CommonRegex()
def test_links(self):
formats = ["www.google.com", "http://www.google.com", "www.google.com/?query=dog"
"sub.example.com", "http://www.google.com/%&#/?q=dog"]
for f in formats:
self.assertEqual(self.parser.links(f), [f])
def contains_personal_info(text):
parsed_text = CommonRegex(text)
return any([
bool(parsed_text.links),
bool(parsed_text.emails),
bool(parsed_text.ips),
bool(parsed_text.ipv6s),
bool(parsed_text.credit_cards),
bool(parsed_text.btc_addresses),
])
class TestPhones(unittest.TestCase):
def setUp(self):
self.parser = CommonRegex()
def test_phones(self):
formats = ["12345678900", "1234567890", "1 234 567 8900", "234-567-8900",
"1-234-567-8900", "1.234.567.8900", "5678900", "567-8900"]
for f in formats:
self.assertEqual(self.parser.phones(f), [f])
def extract_email(values):
emaillist = []
perm_email = []
for item in values:
Email = CommonRegex(str(item))
if Email.emails:
emaillist.append(Email.emails)
for i in emaillist:
for j in i:
perm_email.append(j)
return perm_email
def extract_phonenumbers(values):
phonelist = []
perm_phone = []
for item in values:
phonenumbers = CommonRegex(str(item))
if phonenumbers.phones:
phonelist.append(phonenumbers.phones)
for i in phonelist:
for j in i:
perm_phone.append(j)
return perm_phone
class PiiAnalyzer(object):
def __init__(self, filepath):
self.filepath = filepath
self.parser = CommonRegex()
self.standford_ner = StanfordNERTagger('classifiers/english.conll.4class.distsim.crf.ser.gz')
def analysis(self):
people = []
organizations = []
locations = []
emails = []
phone_numbers = []
street_addresses = []
credit_cards = []
ips = []
data = []
with open(self.filepath, 'rU') as filedata:
reader = csv.reader(filedata)
for row in reader:
data.extend(row)
for text in row:
emails.extend(self.parser.emails(text))
phone_numbers.extend(self.parser.phones("".join(text.split())))
street_addresses.extend(self.parser.street_addresses(text))
credit_cards.extend(self.parser.credit_cards(text))
ips.extend(self.parser.ips(text))
for title, tag in self.standford_ner.tag(set(data)):
if tag == 'PERSON':
people.append(title)
if tag == 'LOCATION':
locations.append(title)
if tag == 'ORGANIZATION':
organizations.append(title)
return {'people': people, 'locations': locations, 'organizations': organizations,
'emails': emails, 'phone_numbers': phone_numbers, 'street_addresses': street_addresses,
'credit_cards': credit_cards, 'ips': ips
}
def find_numbers(data):
numbers = []
data1 = CommonRegex(data)
if data1.phones != 0:
for n in data1.phones:
numbers.append(n)
size = len(numbers)
stats = (
"The number of unique phone numbers replaced in the given file is %d \n"
% size)
return numbers, stats
def find_address(data):
addresses = []
data1 = CommonRegex(data)
if data1.street_addresses != 0:
for n in data1.street_addresses:
addresses.append(n)
size = len(addresses)
stats = (
"The number of unique addresses replaced in the given file is %d \n" %
size)
return addresses, stats
def detect(self, column: CatColumn, datum: str) -> Optional[PiiType]:
"""Scan the text and return an array of PiiTypes that are found"""
regex_result = CommonRegex(datum)
if regex_result.phones: # pylint: disable=no-member
return Phone()
if regex_result.emails: # pylint: disable=no-member
return Email()
if regex_result.credit_cards: # pylint: disable=no-member
return CreditCard()
if regex_result.street_addresses: # pylint: disable=no-member
return Address()
return None
def find_dates(data):
data1 = CommonRegex(data)
date5 = []
if data1.dates != 0:
for n in data1.dates:
date5.append(n)
for n in date5:
data = data.replace(n, "█" * len(n))
size = len(date5)
stats = ("The number of unique dates replaced in the given file is %d \n" %
size)
return date5, stats
def scan(self, text):
"""Scan the text and return an array of PiiTypes that are found"""
regex_result = CommonRegex(text)
types = []
if regex_result.phones: # pylint: disable=no-member
types.append(PiiTypes.PHONE)
if regex_result.emails: # pylint: disable=no-member
types.append(PiiTypes.EMAIL)
if regex_result.credit_cards: # pylint: disable=no-member
types.append(PiiTypes.CREDIT_CARD)
if regex_result.street_addresses: # pylint: disable=no-member
types.append(PiiTypes.ADDRESS)
return types
def get_address_from_issue(issue_number):
github_url = "https://api.github.com/repos/" + repository['path'] + '/issues/' + issue_number
headers = { "Authorization": "token " + GITHUB_TOKEN, "Content-Type": "application/json" }
# Setup the request
req = urllib.request.Request(github_url, headers=headers)
# Make the request, capture the response
res = urllib.request.urlopen(req).read()
res = json.loads(res.decode())
body = res['body']
address = CommonRegex(body).btc_addresses[0]
return address
def find_info(text, attribute):
if text is None:
return []
parsed_text = CommonRegex(text)
if attribute == Information.link:
return parsed_text.links
if attribute == Information.email:
return parsed_text.emails
if attribute == Information.time:
return parsed_text.times
if attribute == Information.date:
return parsed_text.dates
if attribute == Information.phone:
return parsed_text.phones
if attribute == Information.price:
return parsed_text.prices
if attribute == Information.address:
return parsed_text.street_addresses
def dates(data):
text_parsed = CommonRegex(data)
list_of_dates = text_parsed.dates
#alternately, if i use regular expression, to find dates.The regular expressions can be de#defined as we want.
"""
init_file=[]
total_data=data
dates_red=[]
for i in range(len(total_data)):
red = total_data[i]
red_dates = re.findall(r"([A-Z]\w\w\w+\s\d+,\s\d\d\d\d)", red)
dates_red.append(red_dates)
for i in range(len(dates_red)):
for j in dates_red:
red = red.replace(j,'██')
init_file.append(red)
return init_file
"""
return data, list_of_dates
def is_stop_sentence(sentence):
words = {
'copyright', 'unsubscribe', 'instagram', 'twitter', 'facebook',
'youtube', 'style='
}
tokens = nltk.word_tokenize(sentence)
for token in tokens:
if token.lower() in words:
return True
parsed_text = CommonRegex(sentence)
if len(parsed_text.street_addresses) > 0 or len(
parsed_text.emails) > 0 or len(parsed_text.phones) > 0 or len(
parsed_text.times) > 0 or len(parsed_text.dates) > 0 or len(
parsed_text.links) > 0 or len(parsed_text.zip_codes) > 0:
return True
return False
def parsedContent(this):
openObject = open(this.path, "r")
this.contents = openObject.read()
this.parsed_text = CommonRegex(contents)
def parsedContent(self):
openObject = open(self.path, "r")
self.contents = openObject.read()
self.parsed_text = CommonRegex(self.contents)
def extractEntries(zipCode, page):
entries = driver.find_elements_by_css_selector("td")
countNewEntries = 0
countDupes = 0
countElements = 0
pageDupes = set()
for entry in entries:
children = entry.find_elements_by_css_selector("div b a")
if len(children) != 1:
continue
countElements += 1
nameElements = entry.find_elements_by_css_selector("div b a")
name = None
phone = None
address = None
email = None
url = None
if nameElements:
possibleNames = [elem.text for elem in nameElements if elem.text]
if possibleNames:
name = possibleNames[0]
lines = entry.text.splitlines()
for line in lines:
parsed_text = CommonRegex(line)
valid_urls = []
if hasattr(parsed_text.links, '__call__'):
if parsed_text.links():
valid_urls = [link for link in parsed_text.links() if 'gmail' not in link and 'yahoo' not in link and 'hotmail' not in 'link']
else:
if parsed_text.links:
valid_urls = [link for link in parsed_text.links if 'gmail' not in link and 'yahoo' not in link and 'hotmail' not in 'link']
if valid_urls:
url = valid_urls[0]
if hasattr(parsed_text.emails, '__call__'):
if parsed_text.emails():
email = parsed_text.emails()[0]
else:
if parsed_text.emails:
email = parsed_text.emails[0]
if hasattr(parsed_text.phones, '__call__'):
if parsed_text.phones():
phone = parsed_text.phones()[0]
else:
if parsed_text.phones:
phone = parsed_text.phones[0]
if hasattr(parsed_text.street_addresses, '__call__'):
if parsed_text.street_addresses():
address = parsed_text.street_addresses()[0]
else:
if parsed_text.street_addresses:
address = parsed_text.street_addresses[0]
dataText = entry.text.replace("\n", " --- ")
if name or phone or email:
data = {
"name": name,
"phone": phone,
"address": address,
"email": email,
"zip": zipCode,
"url": url,
"data": dataText
}
dedupeKey = str(data['name']) + str(data['email'])
if dedupeKey not in dedupeKeys:
countNewEntries += 1
extracted.append(data)
dedupeKeys.add(dedupeKey)
elif dedupeKey not in pageDupes:
countDupes += 1
pageDupes.add(dedupeKey)
print(f" {zipCode}@{page}: Added {countNewEntries} new entries. Had {countDupes} dupes. Examined {countElements} elements")
return countNewEntries > 0
def index():
if request.method == 'GET':
return 'OK'
elif request.method == 'POST':
# Store the IP address of the requester
request_ip = ipaddress.ip_address(u'{0}'.format(request.remote_addr))
# If GHE_ADDRESS is specified, use it as the hook_blocks.
if os.environ.get('GHE_ADDRESS', None):
hook_blocks = [os.environ.get('GHE_ADDRESS')]
# Otherwise get the hook address blocks from the API.
else:
hook_blocks = requests.get(
'https://api.github.com/meta').json()['hooks']
if request.headers.get('X-GitHub-Event') == "ping":
return json.dumps({'msg': 'Hi!'})
if request.headers.get('X-GitHub-Event') == 'pull_request':
if (request.json['pull_request']['user']['site_admin'] == 'false'):
return json.dumps(
{'message': 'Pull request not submitted by site admin'})
merge_state = request.json['pull_request']['state']
merge_body = request.json['pull_request']['body']
if (merge_state == 'closed'):
print('Merge state closed')
print('Merge Body: ' + merge_body)
parsed_bounty_issue = re.findall(r"#(\w+)", merge_body)[0]
addresses = CommonRegex(merge_body).btc_addresses[0]
bounty_address = github.get_address_from_issue(
parsed_bounty_issue)
amount = multisig_wallet.get_address_balance(bounty_address)
try:
# use username to look up wallet Id
with open(DEFAULT_WALLET_PATH, 'r') as wallet:
data = json.loads(wallet.read())
for user in data:
try:
if (user['issue_number'] == int(
parsed_bounty_issue)):
print('Wallet found')
wallet_name = user['wallet_name']
walletId = user[wallet_name]['walletId']
except:
print('Loading wallet..')
except:
print('Wallet not found, creating new user...')
# Set up sending of the bounty
issue_title = wallet_name
repository_path_encode = repository_path.encode('utf-8')
issue_title_encode = issue_title.encode('utf-8')
passphrase = hashlib.sha256(repository_path_encode +
issue_title_encode).hexdigest()
multisig_wallet.send_bitcoin_simple(walletId, str(addresses),
amount, passphrase)
# Set up sending of the tweet
usd_per_btc = requests.get(
'https://bitpay.com/api/rates/usd').json()['rate']
bounty_in_btc = round((int(bounty_in_satoshi) / 10**8), 3)
bounty_in_usd = round(bounty_in_btc * usd_per_btc, 2)
url = 'https://github.com/21hackers/git-money/issues/' + parsed_bounty_issue
twitter.send('Bounty Granted (' + amount + ' bits ~ $' +
bounty_in_usd + '): ' + issue_title + ' ' + url)
return json.dumps({'message': 'Pull request received'})
return json.dumps({'message': 'Pull request payout failed'})
if request.headers.get('X-GitHub-Event') == 'issue_comment':
comment_data = {
'url': request.json['comment']['issue_url'],
'payout_address': request.json['issue']['labels'][0]['name'],
'payout_amount': request.json['issue']['labels'][1]['name'],
'body': request.json['comment']['body']
}
print(comment_data)
return json.dumps({'message': 'Issue comment received'})
def setUp(self):
self.parser = CommonRegex()
def echo():
# Getting the data
InputData = request.args.get('echoValue') #unicode
if InputData:
EncodedInputData = InputData.encode("utf-8")
else:
return jsonify(Empty='Empty')
SpacySmallData = SpacySmall(InputData)
SpacyMediumData = SpacyMedium(InputData)
places1 = GeoText(EncodedInputData)
# Email Regex expression library
email = []
parsed_text = CommonRegex(EncodedInputData)
email.append(parsed_text.emails)
# URL Finding(URL Regex)
URLMatcher = re.search(
r'\(?\b(http://|www[.])[-A-Za-z0-9+&@#/%?=~_()|!:,.;]*[-A-Za-z0-9+&@#/%=~_()|]',
EncodedInputData)
url = []
if URLMatcher:
url.append(URLMatcher.group(0))
places = GeoText(EncodedInputData)
# Finding name using the Spacy library
name = []
for ent in SpacySmallData.ents:
if ent.label_ == 'PERSON':
name.append((ent.text))
names = list(set(name) - set(places1.countries) - set(places1.cities))
name1 = [x for x in names if not any(x1.isdigit() for x1 in x)]
org = []
# Finding the Organisation using the Spacy library
for ent in SpacyMediumData.ents:
if ent.label_ == 'ORG':
org.append((ent.text))
if not org:
for ent in SpacySmallData.ents:
if ent.label_ == 'ORG':
org.append((ent.text))
places2 = GeoText(EncodedInputData)
language = ""
# Detecting the language
try:
language = detect(InputData)
except:
pass
# Distinguishing between Phone number,Fax number and Mobile Number
TrainedSpacy = TrainedSpacyData(InputData)
mob = []
fax = []
tel = []
for ent in TrainedSpacy.ents:
if ent.label_ == 'TEL':
tel.append((ent.text))
if ent.label_ == 'MOB':
mob.append(ent.text)
if ent.label_ == 'FAX':
fax.append(ent.text)
# Finding the word infront of the Phone number and classifying based on the word as Phone or Fax or Mob --- This is done using the Phonenumber Library.
tel1 = []
fax1 = []
mob1 = []
for match in phonenumbers.PhoneNumberMatcher(InputData, None):
cnew = NaiveBayesClassifier(train)
a = match.start
start = match.start
end = match.end
# print a
my = InputData
# if (a > 1):
try:
word = my[:(a - 1)].split()[-1]
# print word
if (any(x.isalpha() for x in word)):
if cnew.classify(word) is 'tel':
tel1.append(my[start:end])
elif cnew.classify(word) is 'fax':
fax1.append(my[start:end])
elif cnew.classify(word) is 'mob':
mob1.append(my[start:end])
else:
word2 = my[:(a - 1)].split()[-2]
if cnew.classify(word2) is 'tel':
tel1.append(my[start:end])
elif cnew.classify(word2) is 'fax':
fax1.append(my[start:end])
elif cnew.classify(word2) is 'mob':
mob1.append(my[start:end])
except IndexError:
tel1.append(my[start:end])
# Find the word infront of the Phone NO & classifying based on the word as Phone/Fax/Mob - This is done using the Trained Model
phone = TrainedSpacyData(InputData)
for ent in phone.ents:
if (ent.label_ == 'TEL' or ent.label_ == 'FAX' or ent.label_ == 'MOB'):
cnew = NaiveBayesClassifier(train)
a = ent.start_char
# print a
my = InputData
try:
word = my[:(a - 1)].split()[-1]
if (any(x.isalpha() for x in word)):
# print (word)
if cnew.classify(word) is 'tel':
tel1.append(ent.text)
elif cnew.classify(word) is 'fax':
fax1.append(ent.text)
elif cnew.classify(word) is 'mob':
mob1.append(ent.text)
else:
word2 = my[:(a - 1)].split()[-2]
# print (word2)
if cnew.classify(word2) is 'tel':
tel1.append(ent.text)
elif cnew.classify(word2) is 'fax':
fax1.append(ent.text)
elif cnew.classify(word2) is 'mob':
mob1.append(ent.text)
except IndexError:
tel1.append(ent.text)
tel1 = [x for x in tel1 if sum(c.isdigit() for c in x) > 9]
fax1 = [x for x in fax1 if sum(c.isdigit() for c in x) > 9]
mob1 = [x for x in mob1 if sum(c.isdigit() for c in x) > 9]
tel2 = list(set(tel1))
fax2 = list(set(fax1))
mob2 = list(set(mob1))
# Remove Alphabetic in Telephone NO
tel2 = removeAplhabet(tel2)
fax2 = removeAplhabet(fax2)
mob2 = removeAplhabet(mob2)
# Title detetction
# Finding Position entity by using library and model
pos = []
data = EncodedInputData.decode("utf-8")
data = data.title()
title = finder.findall(data)
y = 0
for x in title:
pos.append(InputData[title[y][0]:title[y][1]])
y += 1
if not pos:
for ent in TrainedSpacy.ents:
if ent.label_ == 'POS':
pos.append(ent.text)
# Continuation of the Organisation to remove duplicate from the position
for x in pos:
for y in org:
if x in y:
org.remove(y)
org1 = list(
set(org) - set(places1.countries) - set(places1.cities) - set(name1))
orgg = []
for x in org1:
orgg.append((x))
# Address Detection
AddressList = []
# Using a PYAP library
addresses = pyap.parse(EncodedInputData, country='US')
for address in addresses:
AddressList.append(str(address))
# Identifying Address using Trained Spacy Model
if not AddressList:
for ent in phone.ents:
if ent.label_ == 'add':
AddressList.append(ent.text)
# Identifying Address by finding the line where it has City
if not AddressList:
for line in EncodedInputData.splitlines():
for city in places2.cities:
if city in line:
AddressList.append(line)
AddressList = [x for x in AddressList
if "Mobile" not in x] # Limitation
AddressList = [x for x in AddressList
if "MOBILE" not in x] # Limitation
# Identifying Address using Regex Expression
if not AddressList:
if places2.cities:
r2 = re.compile(
r'([(\d|-|/|/s|(A-Z)?){1-7}]+[,|-|\s]+[A-zZ]+[Aa-zZ]+.*)')
add = r2.findall(EncodedInputData)
# print add
for text in add:
for text2 in places.cities:
if text2 in text:
AddressList.append(text)
wotex = list(set(AddressList))
# Passing Address into the Geocoder-GOOGLE Library to extract into components
import geocoder
add_1 = []
city_1 = []
country_1 = []
code_1 = []
zip_1 = []
county_1 = []
state_1 = []
for x in wotex:
try:
g1 = geocoder.google(x)
if g1.postal:
zip_1.append(g1.postal)
add1 = ""
if g1.housenumber:
add1 = g1.housenumber
if g1.street:
add1 = add1 + " " + g1.street
add_1.append(add1)
if g1.country:
code_1.append(g1.country)
# city_1.append(g1.city)
if g1.city:
city_1.append(g1.city)
if g1.country_long:
country_1.append(g1.country_long)
if g1.state_long:
state_1.append(g1.state_long)
if g1.county:
county_1.append(g1.county)
if not g1.city:
place_new = GeoText(x)
y = str(place_new.cities)
g2 = geocoder.google(y)
if g2.country:
code_1.append(g2.country)
if g2.country_long:
country_1.append(g2.country_long)
if g2.city:
city_1.append(g2.city)
if g2.postal:
zip_1.append(g2.postal)
if g2.county:
county_1.append(g2.county)
except Exception:
pass
# Passing Address into the Geocoder-OSM Library to extract into components
add_OSM = []
city_OSM = []
zip_OSM = []
code_OSM = []
country_OSM = []
county_OSM = []
state_OSM = []
for y in wotex:
try:
g21 = geocoder.osm(y)
add2 = ""
if g21.json and 'housenumber' in g21.json:
add2 = g21.json['housenumber']
if g21.json and 'street' in g21.json:
add2 = add2 + ' ' + g21.json['street']
add_OSM.append(add2)
if g21.json and 'city' in g21.json:
city_OSM.append(g21.json['city'])
if g21.json and 'country' in g21.json:
country_OSM.append(g21.json['country'])
if g21.json and 'postal' in g21.json:
zip_OSM.append(g21.json['postal'])
if g21.json and g21.json['raw']['address']['country_code']:
code_OSM.append(g21.json['raw']['address']['country_code'])
if g21.json and 'county' in g21.json:
county_OSM.append(g21.json['county'])
if g21.json and 'state' in g21.json:
state_OSM.append(g21.json['state'])
if not city_OSM:
placess = GeoText(y)
x = str(placess.cities)
print(x)
g3 = geocoder.osm(x)
print(g3.json)
if g3.json and 'city' in g3.json:
city_OSM.append(g3.json['city'])
if g3.json and 'country' in g3.json:
country_OSM.append(g3.json['country'])
if g3.json and 'postal' in g3.json:
zip_OSM.append(g3.json['postal'])
if g3.json and g3.json['raw']['address']['country_code']:
code_OSM.append(g3.json['raw']['address']['country_code'])
if g3.json and 'county' in g3.json:
county_OSM.append(g3.json['county'])
if g3.json and 'state' in g3.json:
state_OSM.append(g3.json['state'])
except Exception:
pass
email_N = ','.join(map(unicode, parsed_text.emails))
url_N = ','.join(map(unicode, url))
name1_N = ','.join(map(unicode, name1))
orgg_N = ','.join(map(unicode, orgg))
try:
wotex_N = ','.join(map(unicode, wotex))
except Exception:
wotex_N = ','.join(map(str, wotex))
pos_N = ','.join(map(unicode, pos))
tel_N = ','.join(map(unicode, tel))
mob_N = ','.join(map(unicode, mob))
fax_N = ','.join(map(unicode, fax))
tel2_N = ','.join(map(unicode, tel2))
fax2_N = ','.join(map(unicode, fax2))
mob2_N = ','.join(map(unicode, mob2))
add_1_N = ','.join(map(unicode, add_1))
zip_1_N = ','.join(map(unicode, zip_1))
code_1_N = ','.join(map(unicode, code_1))
add_OSM_N = ','.join(map(unicode, add_OSM))
zip_OSM_N = ','.join(map(unicode, zip_OSM))
code_OSM_N = ','.join(map(unicode, code_OSM))
city_1_N = ','.join(map(unicode, city_1))
city_OSM_N = ','.join(map(unicode, city_OSM))
country_1_N = ','.join(map(unicode, country_1))
country_OSM_N = ','.join(map(unicode, country_OSM))
state_1_N = ','.join(map(unicode, state_1))
county_1_N = ','.join(map(unicode, county_1))
# state_OSM_N = ','.join(map(unicode, state_OSM))
# county_OSM_N = ','.join(map(unicode, county_OSM))
return jsonify(Email=email_N,
Www=url_N,
Name=name1_N,
Organization=orgg_N,
FullAddress=wotex_N,
Role=pos_N,
Tel1=tel_N,
Mob1=mob_N,
Fax1=fax_N,
Phone=tel2_N,
Fax=fax2_N,
Mobile=mob2_N,
Address1Google=add_1_N,
ZipCodeGoogle=zip_1_N,
CountryCodeGoogle=code_1_N,
Address1Osm=add_OSM_N,
ZipCodeOsm=zip_OSM_N,
CountryCodeOsm=code_OSM_N,
CityGoogle=city_1_N,
CityOsm=city_OSM_N,
CountryGoogle=country_1_N,
CountryOsm=country_OSM_N,
StateGoogle=state_1_N,
CountyGoogle=county_1_N,
Language=language)
entities_list = []
from chunk import chunker
for sent in nltk.sent_tokenize(text):
#print pos_tag(tokenize(sent))
for chunk in chunker.parse(pos_tag(tokenize(sent))):
# print chunk
entities_list.append(chunk)
with open('out.txt', 'w') as file:
file.write(str(entities_list))
#print entities_list
import unicodecsv as csv
from commonregex import CommonRegex
from nltk.tag.stanford import StanfordNERTagger
parser = CommonRegex()
standford_ner = StanfordNERTagger(
'classifiers/english.conll.4class.distsim.crf.ser.gz', 'stanford-ner.jar')
people = []
organizations = []
locations = []
emails = []
phone_numbers = []
street_addresses = []
credit_cards = []
ips = []
data = []
with open('sample-data.csv', 'r') as filedata:
reader = csv.reader(filedata)
for row in reader:
def __init__(self, filepath):
self.filepath = filepath
self.parser = CommonRegex()
self.standford_ner = StanfordNERTagger('classifiers/english.conll.4class.distsim.crf.ser.gz')
def is_other(df):
other_dic = {}
parser = CommonRegex()
find_name = NameDataset()
other_label, other_count = is_person_name(df)
other_dic[other_label] = other_count
other_label, other_count = is_business_name(df)
other_dic[other_label] = other_count
other_label, other_count = is_phone_number(df)
other_dic[other_label] = other_count
other_label, other_count = is_address(df)
other_dic[other_label] = other_count
other_label, other_count = is_street_name(df)
other_dic[other_label] = other_count
other_label, other_count = is_city_agency(df)
other_dic[other_label] = other_count
other_label, other_count = is_city(df)
other_dic[other_label] = other_count
other_label, other_count = is_neighborhood(df)
other_dic[other_label] = other_count
other_label, other_count = is_lat_lon_cord(df)
other_dic[other_label] = other_count
other_label, other_count = is_zip_code(df)
other_dic[other_label] = other_count
other_label, other_count = is_borough(df)
other_dic[other_label] = other_count
other_label, other_count = is_school_name(df)
other_dic[other_label] = other_count
other_label, other_count = is_color(df)
other_dic[other_label] = other_count
other_label, other_count = is_car_make(df)
other_dic[other_label] = other_count
other_label, other_count = is_area_of_study(df)
other_dic[other_label] = other_count
other_label, other_count = is_subject_in_school(df)
other_dic[other_label] = other_count
other_label, other_count = is_school_level(df)
other_dic[other_label] = other_count
other_label, other_count = is_college_name(df)
other_dic[other_label] = other_count
other_label, other_count = is_website(df)
other_dic[other_label] = other_count
other_label, other_count = is_building_classification(df)
other_dic[other_label] = other_count
other_label, other_count = is_vehicle_type(df)
other_dic[other_label] = other_count
other_label, other_count = is_location_type(df)
other_dic[other_label] = other_count
other_label, other_count = is_park_playground(df)
other_dic[other_label] = other_count
other_label, other_count = is_letter(df)
other_dic[other_label] = other_count
other_label_list = []
other_label_count = []
k = Counter(other_dic)
top_three = k.most_common(3)
for i in top_three:
other_label_list.append(i[0])
other_label_count.append(i[1])
return other_label_list, other_label_count
for i, d in enumerate(dictionary):
if d[0] == word:
wordID = i
features_matrix[0, wordID] = words.count(word)
return features_matrix
test_doc = 'travel-nontravel/tr3.txt'
doc_matrix = extract_features_for_single_doc(test_doc)
result3 = model1.predict(doc_matrix)
if result3 == 0:
print "non travel"
else:
print "travel"
print str(result3) + "\n"
if result3 == 1:
f = open(test_doc, "r")
if f.mode == 'r':
contents = f.read()
print contents
#matches = datefinder.find_dates(contents)
#for match in matches:
#print match
parsed_text = CommonRegex(contents)
print(parsed_text.times)
print(parsed_text.dates)
print(parsed_text.street_addresses)
print(parsed_text.btc_addresses)
def setUp(self):
self.parser = CommonRegex()
import requests
import json
from commonregex import CommonRegex
from .exceptions import *
from .helper import make_request
from sentry_sdk import capture_message, capture_exception
from urllib.parse import urlencode
import html
parser = CommonRegex()
class SongParser(object):
base_url = "https://songwhip.com/api/"
headers = {
'Origin': 'https://songwhip.com',
'User-Agent':
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.122 Safari/537.36 OPR/67.0.3575.53',
'Content-Type': 'application/json'
}
# payload = "{\"url\":\"{}\",\"country\":\"IN\"}"
payload = {"country": "IN", "url": ""}
SONG_SERVICES_MAP = {
"tidal": "Tidal",
"deezer": "Deezer",
"itunes": "iTunes",
"pandora": "Pandora",
"spotify": "Spotify",
"youtube": "YouTube",
"googleplay": "Google Play",
"itunesStore": "iTunes Store",
def __mark_merchant_line(text_lines, line_rect_list):
"""
Check and mark the Merchant lines
"""
merchant_rect = None
name_line_list = []
page_rect = line_rect_list['rect'][0]
list_no_name = ['welcome', 'thank you', 'customer', 'copy', 'only', '*', 'ticket',
'(', ')', ':', 'invoice', '!', 'more', 'congratulation', 'bill']
for i in range(len(text_lines)):
# pre-processing of text line
for j in range(i + 1, len(line_rect_list['text'])):
if text_lines[i] == line_rect_list['text'][j]:
break
line_rect = line_rect_list['rect'][j]
text_lines[i] = text_lines[i].replace('Welcome to', '')
text_lines[i] = text_lines[i].strip('-')
# check contains of key in list_no_name
f_check_no_list = False
for j in range(len(list_no_name)):
if text_lines[i].lower().__contains__(list_no_name[j]):
f_check_no_list = True
break
if f_check_no_list:
continue
# check validation of key
if len(text_lines[i]) <= 2:
continue
elif len(name_line_list) > 0 and name_line_list[-1] + 1 != i:
break
elif len(name_line_list) > 0 and text_lines[i].__contains__(text_lines[name_line_list[-1]]):
continue
elif len(name_line_list) > 2:
continue
elif len(name_line_list) > 1 and not text_lines[i].isupper():
continue
elif text_lines[i][0] == '#':
continue
elif len(CommonRegex(text_lines[i]).dates) > 0:
continue
elif len(CommonRegex(text_lines[i]).phones) > 0:
continue
elif len(CommonRegex(text_lines[i]).links) > 0:
continue
elif len(text_lines[i].replace('@', '').replace('&', '').split()) > 5:
continue
elif len(text_lines[i].split()) > 3 and text_lines[i].__contains__('-'):
continue
elif text_lines[i].replace('-', '').replace(' ', '').isdigit(): # '305337 - 1'
continue
elif len(name_line_list) > 0 and line_rect[1] > 2 * merchant_rect[3] - merchant_rect[1]:
continue
elif (line_rect[0] + line_rect[2]) > (page_rect[0] + page_rect[2]) * 1.3: # check the position
continue
name_line_list.append(i)
merchant_rect = line_rect
return name_line_list
def sanitize(text, sanitized_filename):
# Remove web URLs
sanitized = re.sub(r"http\S+", "", text)
# Remove HTML tags
sanitized = re.sub('<[^<]+?>', '', sanitized)
# Remove HTML characters
sanitized = re.sub('&[^ ]+;', '', sanitized)
# Remove state abbreviations
sanitized = re.sub(
'(?<!\w)(?:,\s+)?(?:A[LKZR]|C[AOT]|DE|FL|GA|HI|I[ADLN]|K[SY]|LA|M[EDAINSOT]|N[EVHJMYCD]|O[HKR]|PA|RI|S[CD]|T[NX]|UT|V[AT]|W[AIVY]),?\s?(?!\w)',
'', sanitized)
fo = open('unique_senders.txt')
try:
text_ = fo.read()
finally:
fo.close()
# Load natural language processor and add custom organizations
nlp = spacy.load("en_core_web_lg")
ruler = nlp.add_pipe("entity_ruler")
unique_senders = set(text_.split('\n'))
patterns = list()
for unique_sender in unique_senders:
pattern = {"label": "ORG", "pattern": [{"LOWER": unique_sender}]}
patterns.append(pattern)
ruler.add_patterns(patterns)
lemmatizer = WordNetLemmatizer()
sentences = nltk.sent_tokenize(sanitized)
counter = 0
num_sentences = len(sentences)
res = ''
print('There are %d sentences' % (num_sentences))
# Write the result
try:
sanitized_file = open(sanitized_filename, "w")
for sentence in sentences:
counter += 1
if counter % 1000 == 0 or counter == num_sentences:
print('%d sentences sanitized (%.2f%% complete)' %
(counter, float(counter / num_sentences) * 100))
if is_stop_sentence(sentence):
continue
#print('1 ' + sentence)
doc = nlp(sentence)
for ent in doc.ents:
#print('ent: %s ent.label: %s' % (ent.text, ent.label_))
if ent.label_ == 'ORG' or ent.label_ == 'NORP':
sentence = sentence.replace(ent.text, 'BUSINESSNAME')
elif ent.label_ == 'PRODUCT':
sentence = sentence.replace(ent.text, 'PRODUCTNAME')
elif ent.label_ == 'PERSON' and ent.label_ == 'GPE':
sentence = sentence.replace(ent.text, '')
#print('2 ' + sentence)
# Remove times, dates, and prices
parsed_text = CommonRegex(sentence)
for price in parsed_text.prices:
sentence = sentence.replace(price, '')
#print('3 ' + sentence)
# Remove non-dictionary words
words = set(nltk.corpus.words.words())
puncts = {'.', '"', "'", ',', '-', '%', '!', '?'}
etcs = {'email', 'online'}
words = words.union(puncts).union(etcs)
tokens = nltk.wordpunct_tokenize(sentence)
pos_tags = nltk.pos_tag(tokens)
#print(pos_tags)
for token_idx, token in enumerate(tokens):
syntactic_category = get_syntactic_category(
pos_tags[token_idx][1])
if syntactic_category != None:
lemmatized_token = lemmatizer.lemmatize(
token.lower(), syntactic_category)
#print('lemmatized: %s original: %s' % (lemmatized_token, token))
if lemmatized_token.lower(
) not in words and token != 'BUSINESSNAME' and token != 'PRODUCTNAME' and not token.isnumeric(
):
sentence = sentence.replace(token, '')
else:
if token.lower(
) not in words and token != 'BUSINESSNAME' and token != 'PRODUCTNAME' and not token.isnumeric(
):
sentence = sentence.replace(token, '')
#print('4 ' + sentence)
# Remove extra whitespaces
sentence = re.sub(r'\s+', ' ', sentence)
sentence = re.sub(r' , ', ' ', sentence)
sentence = re.sub(r' \.', '.', sentence)
sentence = re.sub(r' !', '!', sentence)
sentence = re.sub(r' \?', '?', sentence)
#print('5 ' + sentence)
sanitized_file.write(sentence + '\n')
finally:
sanitized_file.close()