def get_certemailaddresses(self,certfile):
"""returns a list of all e-mail addresses the 'certfile' for which
is valid."""
cmd=[ self.parent._SMIMECMD,
"x509",
"-in",certfile,
"-text",
"-noout"]
cert,returncode=self._opensslcmd(" ".join(cmd))
cert=cert.decode("utf-8",unicodeerror)
email=[]
found=re.search("(?<=emailAddress=)(.*)",cert)
if found != None:
try:
email.append(cert[found.start():found.end()])
except:
pass
found=re.search("(?<=email:)(.*)",cert) # get alias names
if found != None:
try:
n=cert[found.start():found.end()]
if n not in email:
email.append(n)
except:
pass
return email
def send_email(content, titles):
# 读取config.json配置发信者信息及收信者信息
with open('config.json', 'r') as f:
data = json.load(f)
mail_host = data['host']
mail_user = data['user']
mail_password = data['password']
sender = data['sender']
receivers = data['receivers']
email = []
for i in range(len(content)):
message = MIMEText(content[i], 'html', 'utf-8')
message['From'] = sender
message['To'] = ','.join(receivers)
message['Subject'] = titles[i]
email.append(message)
try:
smtpObj = smtplib.SMTP_SSL(mail_host, 465)
smtpObj.login(mail_user, mail_password)
for message in email:
smtpObj.sendmail(sender, receivers, message.as_string())
title = message['Subject']
log(f'[INFO]:推送通知“{title}”成功')
smtpObj.quit()
except smtplib.SMTPException as e:
log(f'[error]:{e}')
def get_certemailaddresses(self,certfile):
"""returns a list of all e-mail addresses the 'certfile' for which
is valid."""
cmd=[ self.parent._SMIMECMD,
"x509",
"-in",certfile,
"-text",
"-noout"]
cert,returncode=self._opensslcmd(" ".join(cmd))
cert=cert.decode("utf-8",unicodeerror)
email=[]
found=re.search("(?<=emailAddress=)(.*)",cert)
if found != None:
try:
email.append(cert[found.start():found.end()])
except:
pass
found=re.search("(?<=email:)(.*)",cert) # get alias names
if found != None:
try:
n=cert[found.start():found.end()]
if n not in email:
email.append(n)
except:
pass
return email
def get_contacts(people):
"""
@author: byoung
"""
total = []
names = []
email = []
message = []
text = "Happy New Year"
# Get names, emails and msg
for i in people:
names.append(i.name)
email.append(i.email)
message.append(str(i))
total.append([names[-1], email[-1], message[-1]])
return total
def parse_messages(self, message):
emails = [e.strip() for e in re.split(r'On.+wrote:|From:.+', message) if
"Begin forwarded message:" not in e.strip() and "Forwarded message" not in e]
froms = re.findall(r'On.+wrote:|From:.+', message)
froms = froms
messages = []
for i in range(0, len(emails)):
message = Message()
meta = self.parse_meta_v3(froms[i] + "\n" + emails[i])
# print(message.content)
# print("+++++++++++++++++++++++++++++++++")
if meta is not None:
message.sender_name = meta["name"]
message.sender_address = meta["email"]
message.time = meta["time"]
messages.append(message)
lines = emails[i].splitlines()
email = []
for line in lines:
if line.strip() == "":
continue
if line.startswith(('From', 'To', 'Sent', 'Date', 'Subject', 'Cc', 'Received')):
# print(line)
email = []
# elif line.strip().lower() == message.sender_name.lower() or line.strip().lower() == message.sender_name.split()[0].lower():
elif line.strip().lower() in message.sender_name.lower():
email.append(line)
break
else:
email.append(line)
# message.content = re.sub(r'(From|To|Sent|Date|Subject|Cc|Received):.+', '', emails[i], re.MULTILINE).strip()
message.content = "\n".join(email)
message.keywords = nlp.keywords_from_text(message.content)
messages.reverse()
return messages
def final_csv():
df = pd.read_csv('email_details_3.csv')
# print(df)
name = []
email = []
date = []
subject = []
for index, row in df.iterrows():
details = str(row['From'])
print((details))
if details!='nan':
x = details.find('<')
na = details[0:x-1]
em = details[x+1:-1]
name.append(na)
email.append(em)
else:
name.append('')
email.append('')
date.append(row['Date'])
subject.append(row['Subject'])
dfdata = {'Sender Email': email, 'Sender Name': name, 'Date': date, 'Subject': subject}
dfnew = pd.DataFrame(dfdata)
dfnew.to_csv('email_details_10.csv')
def get_email(bucket, fileName):
s3 = boto3.client("s3")
file = s3.get_object(Bucket=bucket, Key=fileName)
content = file["Body"].read().decode('utf-8')
subject = content.split('Subject: ')[1]
subject = subject.split('\n')[0]
edate = content.split('Date: ')[1]
edate = edate.split('\n')[0]
sender = content.split('Return-Path: <')[1]
sender = sender.split('>')[0]
sender.split()
email = []
email.append(subject)
email.append(edate)
email.append(sender)
print(subject, edate, sender)
return email
def test_email_list(query):
email = []
for i in range(2000):
email.append( query % ( hashlib.sha1(str(random.random())).hexdigest(), ))
return email
""" LDA on top of TF-IDF"""
# use this when starting here and want to start with cleaned filtered dataset
really_good_df = pd.read_excel("final_cleaned_data_v2.xlsx")
# # use this if you are continuing from part 6 and above
# really_good_df = sample
temp = really_good_df.newcontent.dropna()
emails = list(temp)
# Remove punctuation
email = [ ]
punc = string.punctuation.replace( '-', '' )
for i in range( 0, len( emails ) ):
email.append( re.sub( '[' + punc + ']+', '', emails[ i ] ) )
# Porter stem
porter_lda = nltk.stem.porter.PorterStemmer()
stems_lda = { }
for i in range( 0, len( email ) ):
tok = email[ i ].split()
for j in range( 0, len( tok ) ):
if tok[ j ] not in stems_lda:
stems_lda[ tok[ j ] ] = porter_lda.stem( tok[ j ] )
tok[ j ] = stems_lda[ tok[ j ] ]
email[ i ] = ' '.join( tok )
def trigger(self):
email = []
mess = self.message
email.append(format(mess['message-id']))
timefloat = time.mktime(parsedate(mess['date']))
timestamp = str(timefloat)
email.append(timestamp)
email.append(format(mess['from']))
email.append(format(mess['to']))
email.append(format(mess['cc']))
email.append(format(mess['subject']))
email.append(format(mess['in-reply-to']))
for part in mess.walk():
if part.get_content_type() == 'text/plain':
body = strip_string(str(part.get_payload()))
email.append(body)
else:
continue
if len(email) < 8:
email.append('')
f = feature(email)
predicted = regr.predict([f])[0]
draft_time(email, predicted / 3600)
def trigger(self):
email = []
mess = self.message
email.append(format(mess['message-id']))
timefloat = time.mktime(parsedate(mess['date']))
timestamp = str(timefloat)
email.append(timestamp)
email.append(format(mess['from']))
email.append(format(mess['to']))
email.append(format(mess['cc']))
email.append(format(mess['subject']))
email.append(format(mess['in-reply-to']))
for part in mess.walk():
if part.get_content_type() == 'text/plain':
body = strip_string(str(part.get_payload()))
email.append(body)
else:
continue
if len(email) < 8:
email.append('')
f = feature(email)
predicted = regr.predict([f])[0]
draft_time(email, predicted/3600)
