def spam_detector(data):
# We secure the opening of the directory
try:
os.chdir(data)
files = os.listdir(".")
spam_detected = 0
mails =[]
print "Beginning of the analysis of " + data
#We put the mails in the RAM
print "We put the mails in the RAM..."
for mail in files:
text = open(mail, "r").read()
mails.append(text)
#We examine each mail
print "We analyze each mail ..."
for mail in mails:
if isspam.is_spam(mail) == "spam":
spam_detected += 1
print "The number of spams detected in the directory " + data +" is " + str(spam_detected) + " over " + str(files.__len__()) + " mails"
return spam_detected
except OSError:
print "No such directory: " + data
def spam_detector(data):
# We secure the opening of the directory
try:
os.chdir(data)
files = os.listdir(".")
# We create the dictionary of spam words
cursor.execute("SELECT * FROM words_spam")
table1 = [element[1] for element in cursor.fetchall()]
cursor.execute("SELECT * FROM words_spam")
table2 = [element[2] for element in cursor.fetchall()]
table_word_spam = dict(zip(table1, table2))
# We create the dictionary of ham words
cursor.execute("SELECT * FROM words_ham")
table1 = [element[1] for element in cursor.fetchall()]
cursor.execute("SELECT * FROM words_ham")
table2 = [element[2] for element in cursor.fetchall()]
table_word_ham = dict(zip(table1, table2))
spam_detected = 0
# counter = 0
mails = []
print "Beginning of the analysis of " + data
# We put the mails in the RAM
print "We put the mails in the RAM..."
for mail in files:
text = open(mail, "r").read()
mails.append(text)
# We analyze each mails
print "We analyze each mail..."
for mail in mails:
if isspam.is_spam(mail, table_word_spam, table_word_ham) == "spam":
spam_detected += 1
# counter += 1
# print counter
print "The number of spams detected in this directory is " + str(spam_detected) + " over " + str(
files.__len__()
) + " mails"
return spam_detected
except OSError:
print "No such directory: " + data
except sqlite3.OperationalError:
print "We need to analyze mails with the learning files in order to detect spam"
