def msg_train(h, msg, is_spam, force):
"""Train bayes with a single message."""
# XXX: big hack -- why is email.Message unable to represent
# multipart/alternative?
try:
mboxutils.as_string(msg)
except TypeError:
# We'll be unable to represent this as text :(
return False
if is_spam:
spamtxt = options["Headers", "header_spam_string"]
else:
spamtxt = options["Headers", "header_ham_string"]
oldtxt = msg.get(options["Headers", "trained_header_name"])
if force:
# Train no matter what.
if oldtxt != None:
del msg[options["Headers", "trained_header_name"]]
elif oldtxt == spamtxt:
# Skip this one, we've already trained with it.
return False
elif oldtxt != None:
# It's been trained, but as something else. Untrain.
del msg[options["Headers", "trained_header_name"]]
h.untrain(msg, not is_spam)
h.train(msg, is_spam)
msg.add_header(options["Headers", "trained_header_name"], spamtxt)
return True
def msg_train(h, msg, is_spam, force):
"""Train bayes with a single message."""
# XXX: big hack -- why is email.Message unable to represent
# multipart/alternative?
try:
mboxutils.as_string(msg)
except TypeError:
# We'll be unable to represent this as text :(
return False
if is_spam:
spamtxt = options["Headers", "header_spam_string"]
else:
spamtxt = options["Headers", "header_ham_string"]
oldtxt = msg.get(options["Headers", "trained_header_name"])
if force:
# Train no matter what.
if oldtxt != None:
del msg[options["Headers", "trained_header_name"]]
elif oldtxt == spamtxt:
# Skip this one, we've already trained with it.
return False
elif oldtxt != None:
# It's been trained, but as something else. Untrain.
del msg[options["Headers", "trained_header_name"]]
h.untrain(msg, not is_spam)
h.train(msg, is_spam)
msg.add_header(options["Headers", "trained_header_name"], spamtxt)
return True
def _calc_response(self, switches, body):
switches = switches.split()
actions = []
opts, args = getopt.getopt(switches, 'fgstGS')
h = self.server.hammie
for opt, arg in opts:
if opt == '-f':
actions.append(h.filter)
elif opt == '-g':
actions.append(h.train_ham)
elif opt == '-s':
actions.append(h.train_spam)
elif opt == '-t':
actions.append(h.filter_train)
elif opt == '-G':
actions.append(h.untrain_ham)
elif opt == '-S':
actions.append(h.untrain_spam)
if actions == []:
actions = [h.filter]
from spambayes import mboxutils
msg = mboxutils.get_message(body)
for action in actions:
action(msg)
return mboxutils.as_string(msg, 1)
def main(profiling=False):
h = HammieFilter()
actions = []
opts, args = getopt.getopt(sys.argv[1:], 'hvxd:p:nfgstGSo:P',
['help', 'version', 'examples', 'option='])
create_newdb = False
do_profile = False
for opt, arg in opts:
if opt in ('-h', '--help'):
usage(0)
elif opt in ('-v', '--version'):
version()
elif opt in ('-x', '--examples'):
examples()
elif opt in ('-o', '--option'):
Options.options.set_from_cmdline(arg, sys.stderr)
elif opt == '-f':
actions.append(h.filter)
elif opt == '-g':
actions.append(h.train_ham)
elif opt == '-s':
actions.append(h.train_spam)
elif opt == '-t':
actions.append(h.filter_train)
elif opt == '-G':
actions.append(h.untrain_ham)
elif opt == '-S':
actions.append(h.untrain_spam)
elif opt == '-P':
do_profile = True
if not profiling:
try:
import cProfile
except ImportError:
pass
else:
return cProfile.run("main(True)")
elif opt == "-n":
create_newdb = True
h.dbname, h.usedb = storage.database_type(opts)
if create_newdb or not os.path.exists(h.dbname):
h.newdb()
print("Created new database in", h.dbname, file=sys.stderr)
if create_newdb:
sys.exit(0)
if actions == []:
actions = [h.filter]
if not args:
args = ["-"]
for fname in args:
mbox = mboxutils.getmbox(fname)
for msg in mbox:
for action in actions:
action(msg)
if args == ["-"]:
unixfrom = msg.get_unixfrom() is not None
else:
unixfrom = True
result = mboxutils.as_string(msg, unixfrom=unixfrom)
sys.stdout.write(result)
def mhdir_train(h, path, is_spam, force):
"""Train bayes with an mh directory"""
if loud: print " Reading as MH mailbox"
import glob
counter = 0
trained = 0
for fn in glob.glob(os.path.join(path, "[0-9]*")):
counter += 1
cfn = fn
tfn = os.path.join(path, "spambayes.tmp")
if loud and counter % 10 == 0:
sys.stdout.write("\r%6d" % counter)
sys.stdout.flush()
f = file(fn, "rb")
msg = get_message(f)
f.close()
if not msg:
print "Malformed message: %s. Skipping..." % cfn
continue
msg_train(h, msg, is_spam, force)
trained += 1
if not options["Headers", "include_trained"]:
continue
f = file(tfn, "wb")
f.write(mboxutils.as_string(msg))
f.close()
shutil.copystat(cfn, tfn)
os.rename(tfn, cfn)
if loud:
sys.stdout.write("\r%6d" % counter)
sys.stdout.write("\r Trained %d out of %d messages\n" %
(trained, counter))
def _calc_response(self, switches, body):
switches = switches.split()
actions = []
opts, args = getopt.getopt(switches, 'fgstGS')
h = self.server.hammie
for opt, arg in opts:
if opt == '-f':
actions.append(h.filter)
elif opt == '-g':
actions.append(h.train_ham)
elif opt == '-s':
actions.append(h.train_spam)
elif opt == '-t':
actions.append(h.filter_train)
elif opt == '-G':
actions.append(h.untrain_ham)
elif opt == '-S':
actions.append(h.untrain_spam)
if actions == []:
actions = [h.filter]
from spambayes import mboxutils
msg = mboxutils.get_message(body)
for action in actions:
action(msg)
return mboxutils.as_string(msg, 1)
def mbox_train(h, path, is_spam, force):
"""Train bayes with a Unix mbox"""
if loud:
print " Reading as Unix mbox"
import mailbox
import fcntl
# Open and lock the mailbox. Some systems require it be opened for
# writes in order to assert an exclusive lock.
f = file(path, "r+b")
fcntl.flock(f, fcntl.LOCK_EX)
mbox = mailbox.PortableUnixMailbox(f, get_message)
outf = os.tmpfile()
counter = 0
trained = 0
for msg in mbox:
if not msg:
print "Malformed message number %d. I can't train on this mbox, sorry." % counter
return
counter += 1
if loud and counter % 10 == 0:
sys.stdout.write("\r%6d" % counter)
sys.stdout.flush()
if msg_train(h, msg, is_spam, force):
trained += 1
if options["Headers", "include_trained"]:
# Write it out with the Unix "From " line
outf.write(mboxutils.as_string(msg, True))
if options["Headers", "include_trained"]:
outf.seek(0)
try:
os.ftruncate(f.fileno(), 0)
f.seek(0)
except:
# If anything goes wrong, don't try to write
print "Problem truncating mbox--nothing written"
raise
try:
for line in outf.xreadlines():
f.write(line)
except:
print >> sys.stderr ("Problem writing mbox! Sorry, "
"I tried my best, but your mail "
"may be corrupted.")
raise
fcntl.flock(f, fcntl.LOCK_UN)
f.close()
if loud:
sys.stdout.write("\r%6d" % counter)
sys.stdout.write("\r Trained %d out of %d messages\n" %
(trained, counter))
def mbox_train(h, path, is_spam, force):
"""Train bayes with a Unix mbox"""
if loud:
print " Reading as Unix mbox"
import mailbox
import fcntl
# Open and lock the mailbox. Some systems require it be opened for
# writes in order to assert an exclusive lock.
f = file(path, "r+b")
fcntl.flock(f, fcntl.LOCK_EX)
mbox = mailbox.PortableUnixMailbox(f, get_message)
outf = os.tmpfile()
counter = 0
trained = 0
for msg in mbox:
if not msg:
print "Malformed message number %d. I can't train on this mbox, sorry." % counter
return
counter += 1
if loud and counter % 10 == 0:
sys.stdout.write("\r%6d" % counter)
sys.stdout.flush()
if msg_train(h, msg, is_spam, force):
trained += 1
if options["Headers", "include_trained"]:
# Write it out with the Unix "From " line
outf.write(mboxutils.as_string(msg, True))
if options["Headers", "include_trained"]:
outf.seek(0)
try:
os.ftruncate(f.fileno(), 0)
f.seek(0)
except:
# If anything goes wrong, don't try to write
print "Problem truncating mbox--nothing written"
raise
try:
for line in outf.xreadlines():
f.write(line)
except:
print >> sys.stderr("Problem writing mbox! Sorry, "
"I tried my best, but your mail "
"may be corrupted.")
raise
fcntl.flock(f, fcntl.LOCK_UN)
f.close()
if loud:
sys.stdout.write("\r%6d" % counter)
sys.stdout.write("\r Trained %d out of %d messages\n" %
(trained, counter))
def msg_train(h, msg, is_spam, force):
"""Train bayes with a single message."""
try:
mboxutils.as_string(msg)
except TypeError:
return False
if is_spam:
spamtxt = options["Headers", "header_spam_string"]
else:
spamtxt = options["Headers", "header_ham_string"]
oldtxt = msg.get(options["Headers", "trained_header_name"])
if force:
if oldtxt != None:
del msg[options["Headers", "trained_header_name"]]
elif oldtxt == spamtxt:
return False
elif oldtxt != None:
del msg[options["Headers", "trained_header_name"]]
h.untrain(msg, not is_spam)
h.train(msg, is_spam)
msg.add_header(options["Headers", "trained_header_name"], spamtxt)
return True
def maildir_train(h, path, is_spam, force, removetrained):
"""Train bayes with all messages from a maildir."""
if loud:
print " Reading %s as Maildir" % (path, )
import time
import socket
pid = os.getpid()
host = socket.gethostname()
counter = 0
trained = 0
for fn in os.listdir(path):
cfn = os.path.join(path, fn)
tfn = os.path.normpath(
os.path.join(path, "..", "tmp",
"%d.%d_%d.%s" % (time.time(), pid, counter, host)))
if (os.path.isdir(cfn)):
continue
counter += 1
if loud and counter % 10 == 0:
sys.stdout.write("\r%6d" % counter)
sys.stdout.flush()
f = file(cfn, "rb")
msg = get_message(f)
f.close()
if not msg:
print "Malformed message: %s. Skipping..." % cfn
continue
if not msg_train(h, msg, is_spam, force):
continue
trained += 1
if not options["Headers", "include_trained"]:
continue
f = file(tfn, "wb")
f.write(mboxutils.as_string(msg))
f.close()
shutil.copystat(cfn, tfn)
# XXX: This will raise an exception on Windows. Do any Windows
# people actually use Maildirs?
os.rename(tfn, cfn)
if (removetrained):
os.unlink(cfn)
if loud:
sys.stdout.write("\r%6d" % counter)
sys.stdout.write("\r Trained %d out of %d messages\n" %
(trained, counter))
def maildir_train(h, path, is_spam, force, removetrained):
"""Train bayes with all messages from a maildir."""
if loud:
print " Reading %s as Maildir" % (path,)
import time
import socket
pid = os.getpid()
host = socket.gethostname()
counter = 0
trained = 0
for fn in os.listdir(path):
cfn = os.path.join(path, fn)
tfn = os.path.normpath(os.path.join(path, "..", "tmp",
"%d.%d_%d.%s" % (time.time(), pid,
counter, host)))
if (os.path.isdir(cfn)):
continue
counter += 1
if loud and counter % 10 == 0:
sys.stdout.write("\r%6d" % counter)
sys.stdout.flush()
f = file(cfn, "rb")
msg = get_message(f)
f.close()
if not msg:
print "Malformed message: %s. Skipping..." % cfn
continue
if not msg_train(h, msg, is_spam, force):
continue
trained += 1
if not options["Headers", "include_trained"]:
continue
f = file(tfn, "wb")
f.write(mboxutils.as_string(msg))
f.close()
shutil.copystat(cfn, tfn)
# XXX: This will raise an exception on Windows. Do any Windows
# people actually use Maildirs?
os.rename(tfn, cfn)
if (removetrained):
os.unlink(cfn)
if loud:
sys.stdout.write("\r%6d" % counter)
sys.stdout.write("\r Trained %d out of %d messages\n" %
(trained, counter))
def main():
h = HammieFilter()
actions = []
opts, args = getopt.getopt(sys.argv[1:], 'hxd:p:nfgstGSo:',
['help', 'examples', 'option='])
create_newdb = False
for opt, arg in opts:
if opt in ('-h', '--help'):
usage(0)
elif opt in ('-x', '--examples'):
examples()
elif opt in ('-o', '--option'):
Options.options.set_from_cmdline(arg, sys.stderr)
elif opt == '-f':
actions.append(h.filter)
elif opt == '-g':
actions.append(h.train_ham)
elif opt == '-s':
actions.append(h.train_spam)
elif opt == '-t':
actions.append(h.filter_train)
elif opt == '-G':
actions.append(h.untrain_ham)
elif opt == '-S':
actions.append(h.untrain_spam)
elif opt == "-n":
create_newdb = True
h.dbname, h.usedb = storage.database_type(opts)
if create_newdb:
h.newdb()
sys.exit(0)
if actions == []:
actions = [h.filter]
if not args:
args = ["-"]
for fname in args:
mbox = mboxutils.getmbox(fname)
for msg in mbox:
for action in actions:
action(msg)
if args == ["-"]:
unixfrom = msg.get_unixfrom() is not None
else:
unixfrom = True
result = mboxutils.as_string(msg, unixfrom=unixfrom)
sys.stdout.write(result)
def mbox_train(h, path, is_spam, force):
"""Train bayes with a Unix mbox"""
if loud:
print(" Reading as Unix mbox")
import mailbox
import fcntl
f = file(path, "r+b")
fcntl.flock(f, fcntl.LOCK_EX)
mbox = mailbox.PortableUnixMailbox(f, get_message)
outf = os.tmpfile()
counter = 0
trained = 0
for msg in mbox:
if not msg:
print("Malformed message number %d. I can't train on this mbox, sorry." % counter)
return
counter += 1
if loud and counter % 10 == 0:
sys.stdout.write("\r%6d" % counter)
sys.stdout.flush()
if msg_train(h, msg, is_spam, force):
trained += 1
if options["Headers", "include_trained"]:
outf.write(mboxutils.as_string(msg, True))
if options["Headers", "include_trained"]:
outf.seek(0)
try:
os.ftruncate(f.fileno(), 0)
f.seek(0)
except:
print("Problem truncating mbox--nothing written")
raise
try:
for line in outf:
f.write(line)
except:
print(file=sys.stderr ("Problem writing mbox! Sorry, "
"I tried my best, but your mail "
"may be corrupted."))
raise
fcntl.flock(f, fcntl.LOCK_UN)
f.close()
if loud:
sys.stdout.write("\r%6d" % counter)
sys.stdout.write("\r Trained %d out of %d messages\n" %
(trained, counter))
def mhdir_train(h, path, is_spam, force):
"""Train bayes with an mh directory"""
if loud:
print " Reading as MH mailbox"
import glob
counter = 0
trained = 0
for fn in glob.glob(os.path.join(path, "[0-9]*")):
counter += 1
cfn = fn
tfn = os.path.join(path, "spambayes.tmp")
if loud and counter % 10 == 0:
sys.stdout.write("\r%6d" % counter)
sys.stdout.flush()
f = file(fn, "rb")
msg = get_message(f)
f.close()
if not msg:
print "Malformed message: %s. Skipping..." % cfn
continue
msg_train(h, msg, is_spam, force)
trained += 1
if not options["Headers", "include_trained"]:
continue
f = file(tfn, "wb")
f.write(mboxutils.as_string(msg))
f.close()
shutil.copystat(cfn, tfn)
# XXX: This will raise an exception on Windows. Do any Windows
# people actually use MH directories?
os.rename(tfn, cfn)
if loud:
sys.stdout.write("\r%6d" % counter)
sys.stdout.write("\r Trained %d out of %d messages\n" %
(trained, counter))
def main(profiling=False):
h = HammieFilter()
actions = []
opts, args = getopt.getopt(sys.argv[1:], 'hvxd:p:nfgstGSo:P',
['help', 'version', 'examples', 'option='])
create_newdb = False
do_profile = False
for opt, arg in opts:
if opt in ('-h', '--help'):
usage(0)
elif opt in ('-v', '--version'):
version()
elif opt in ('-x', '--examples'):
examples()
elif opt in ('-o', '--option'):
Options.options.set_from_cmdline(arg, sys.stderr)
elif opt == '-f':
actions.append(h.filter)
elif opt == '-g':
actions.append(h.train_ham)
elif opt == '-s':
actions.append(h.train_spam)
elif opt == '-t':
actions.append(h.filter_train)
elif opt == '-G':
actions.append(h.untrain_ham)
elif opt == '-S':
actions.append(h.untrain_spam)
elif opt == '-P':
do_profile = True
if not profiling:
try:
import cProfile
except ImportError:
pass
else:
return cProfile.run("main(True)")
elif opt == "-n":
create_newdb = True
h.dbname, h.usedb = storage.database_type(opts)
if create_newdb or not os.path.exists(h.dbname):
h.newdb()
print >> sys.stderr, "Created new database in", h.dbname
if create_newdb:
sys.exit(0)
if actions == []:
actions = [h.filter]
if not args:
args = ["-"]
for fname in args:
mbox = mboxutils.getmbox(fname)
for msg in mbox:
for action in actions:
action(msg)
if args == ["-"]:
unixfrom = msg.get_unixfrom() is not None
else:
unixfrom = True
result = mboxutils.as_string(msg, unixfrom=unixfrom)
sys.stdout.write(result)
def score_and_filter(self,
msg,
header=None,
spam_cutoff=None,
ham_cutoff=None,
debugheader=None,
debug=None,
train=None):
"""Score (judge) a message and add a disposition header.
msg can be a string, a file object, or a Message object.
Optionally, set header to the name of the header to add, and/or
spam_cutoff/ham_cutoff to the probability values which must be met
or exceeded for a message to get a 'Spam' or 'Ham' classification.
An extra debugging header can be added if 'debug' is set to True.
The name of the debugging header is given as 'debugheader'.
If 'train' is True, also train on the result of scoring the
message (ie. train as ham if it's ham, train as spam if it's
spam). If the message already has a trained header, it will be
untrained first. You'll want to be very dilligent about
retraining mistakes if you use this option.
All defaults for optional parameters come from the Options file.
Returns the score and same message with a new disposition header.
"""
if header == None:
header = options["Headers", "classification_header_name"]
if spam_cutoff == None:
spam_cutoff = options["Categorization", "spam_cutoff"]
if ham_cutoff == None:
ham_cutoff = options["Categorization", "ham_cutoff"]
if debugheader == None:
debugheader = options["Headers", "evidence_header_name"]
if debug == None:
debug = options["Headers", "include_evidence"]
if train == None:
train = options["Hammie", "train_on_filter"]
msg = mboxutils.get_message(msg)
try:
del msg[header]
except KeyError:
pass
if train:
self.untrain_from_header(msg)
prob, clues = self._scoremsg(msg, True)
if prob < ham_cutoff:
is_spam = False
disp = options["Headers", "header_ham_string"]
elif prob > spam_cutoff:
is_spam = True
disp = options["Headers", "header_spam_string"]
else:
is_spam = False
disp = options["Headers", "header_unsure_string"]
if train:
self.train(msg, is_spam, True)
basic_disp = disp
disp += "; %.*f" % (options["Headers", "header_score_digits"], prob)
if options["Headers", "header_score_logarithm"]:
if prob <= 0.005 and prob > 0.0:
import math
x = -math.log10(prob)
disp += " (%d)" % x
if prob >= 0.995 and prob < 1.0:
x = -math.log10(1.0 - prob)
disp += " (%d)" % x
del msg[header]
msg.add_header(header, disp)
# Obey notate_to and notate_subject.
for header in ('to', 'subject'):
if basic_disp in options["Headers", "notate_" + header]:
orig = msg[header]
del msg[header]
msg[header] = "%s,%s" % (basic_disp, orig)
if debug:
disp = self.formatclues(clues)
del msg[debugheader]
msg.add_header(debugheader, disp)
result = mboxutils.as_string(msg,
unixfrom=(msg.get_unixfrom() is not None))
return prob, result
def score_and_filter(self, msg, header=None, spam_cutoff=None,
ham_cutoff=None, debugheader=None,
debug=None, train=None):
"""Score (judge) a message and add a disposition header.
msg can be a string, a file object, or a Message object.
Optionally, set header to the name of the header to add, and/or
spam_cutoff/ham_cutoff to the probability values which must be met
or exceeded for a message to get a 'Spam' or 'Ham' classification.
An extra debugging header can be added if 'debug' is set to True.
The name of the debugging header is given as 'debugheader'.
If 'train' is True, also train on the result of scoring the
message (ie. train as ham if it's ham, train as spam if it's
spam). If the message already has a trained header, it will be
untrained first. You'll want to be very dilligent about
retraining mistakes if you use this option.
All defaults for optional parameters come from the Options file.
Returns the score and same message with a new disposition header.
"""
if header == None:
header = options["Headers", "classification_header_name"]
if spam_cutoff == None:
spam_cutoff = options["Categorization", "spam_cutoff"]
if ham_cutoff == None:
ham_cutoff = options["Categorization", "ham_cutoff"]
if debugheader == None:
debugheader = options["Headers", "evidence_header_name"]
if debug == None:
debug = options["Headers", "include_evidence"]
if train == None:
train = options["Hammie", "train_on_filter"]
msg = mboxutils.get_message(msg)
try:
del msg[header]
except KeyError:
pass
if train:
self.untrain_from_header(msg)
prob, clues = self._scoremsg(msg, True)
if prob < ham_cutoff:
is_spam = False
disp = options["Headers", "header_ham_string"]
elif prob > spam_cutoff:
is_spam = True
disp = options["Headers", "header_spam_string"]
else:
is_spam = False
disp = options["Headers", "header_unsure_string"]
if train:
self.train(msg, is_spam, True)
basic_disp = disp
disp += "; %.*f" % (options["Headers", "header_score_digits"], prob)
if options["Headers", "header_score_logarithm"]:
if prob<=0.005 and prob>0.0:
import math
x=-math.log10(prob)
disp += " (%d)"%x
if prob>=0.995 and prob<1.0:
import math
x=-math.log10(1.0-prob)
disp += " (%d)"%x
del msg[header]
msg.add_header(header, disp)
for header in ('to', 'subject'):
if basic_disp in options["Headers", "notate_"+header]:
orig = msg[header]
del msg[header]
msg[header] = "%s,%s" % (basic_disp, orig)
if debug:
disp = self.formatclues(clues)
del msg[debugheader]
msg.add_header(debugheader, disp)
result = mboxutils.as_string(msg, unixfrom=(msg.get_unixfrom()
is not None))
return prob, result
