class FuzorDigest(object):
def __init__(self, msg):
self.debug = []
self.digest = None
self.predigest = None
self.bodytext_size = 0
self.filter = SuspectFilter(None)
self.logger = logging.getLogger('fuglu.plugins.fuzor.Digest')
# digest config
self.LONG_WORD_THRESHOLD = 10 # what is considered a long word
self.REPLACE_LONG_WORD = '[LONG]' # Replace long words in pre-digest with... None to disable
self.REPLACE_EMAIL = '[EMAIL]' # Replace email addrs in pre-digest with... None to disable
self.REPLACE_URL = '[LINK]' # Replace urls in pre-digest with... None to disable
self.INCLUDE_ATTACHMENT_CONTENT = False # should non-text attachment contents be included in digest (not recommended, there are better attachment hash systems)
self.INCLUDE_ATTACHMENT_COUNT = True # should the number of non-text-attachments be included in the digest
self.MINIMUM_PREDIGEST_SIZE = 27 # if the predigest is smaller than this, ignore this message
self.MINIMUM_UNMODIFIED_CONTENT = 27 # minimum unmodified content after stripping, eg. [SOMETHING] removed from the predigest (27>'von meinem Iphone gesendet')
self.MINIMUM_BODYTEXT_SIZE = 27 # if the body text content is smaller than this, ignore this message
self.STRIP_WHITESPACE = True # remove all whitespace from the pre-digest
self.STRIP_HTML_MARKUP = True # remove html tags (but keep content)
self.REMOVE_HTML_TAGS = [
'script',
'style'] # strip tags (including content)
self.predigest = self._make_predigest(msg)
self.digest = self._make_hash(self.predigest)
def _make_hash(self, predigest):
if self.bodytext_size < self.MINIMUM_BODYTEXT_SIZE:
return None
predigest = predigest.strip()
if len(predigest) < self.MINIMUM_PREDIGEST_SIZE:
return None
unmodified = re.sub(r'\[[A-Z0-9:]+\]', '', predigest)
if len(unmodified) < self.MINIMUM_UNMODIFIED_CONTENT:
return None
predigest = predigest.encode('utf-8', errors='ignore')
return hashlib.sha1(predigest).hexdigest()
def _handle_text_part(self, part):
payload = part.get_payload(decode=True)
charset = part.get_content_charset()
errors = "ignore"
if not charset:
charset = "ascii"
elif charset.lower().replace("_", "-") in ("quopri-codec", "quopri", "quoted-printable", "quotedprintable"):
errors = "strict"
try:
payload = payload.decode(charset, errors)
except (LookupError, UnicodeError, AssertionError):
payload = payload.decode("ascii", "ignore")
if self.STRIP_HTML_MARKUP:
payload = self.filter.strip_text(
payload,
remove_tags=self.REMOVE_HTML_TAGS,
use_bfs=True)
if self.REPLACE_EMAIL is not None:
payload = re.sub(r'\S{1,50}@\S{1,30}', self.REPLACE_EMAIL, payload)
if self.REPLACE_URL is not None:
payload = re.sub(r'[a-z]+:\S{1,100}', self.REPLACE_URL, payload)
if self.REPLACE_LONG_WORD is not None:
patt = r'\S{%s,}' % self.LONG_WORD_THRESHOLD
payload = re.sub(patt, self.REPLACE_LONG_WORD, payload)
if self.STRIP_WHITESPACE:
payload = re.sub(r'\s', '', payload)
payload = payload.strip()
return payload
def _make_predigest(self, msg):
attachment_count = 0
predigest = ''
for part in msg.walk():
if part.is_multipart():
continue
if part.get_content_maintype() == "text":
try:
normalized_text_part = self._handle_text_part(part)
predigest += normalized_text_part
self.bodytext_size += len(normalized_text_part)
except Exception as e:
self.logger.warn(e)
else:
attachment_count += 1
if self.INCLUDE_ATTACHMENT_CONTENT:
predigest += "[ATTH:%s]" % hashlib.sha1(
part.get_payload()).hexdigest()
if self.INCLUDE_ATTACHMENT_COUNT and attachment_count:
predigest += "[ATTC:%s]" % attachment_count
if self.STRIP_WHITESPACE:
predigest = re.sub(r'\s', '', predigest)
return predigest
class FuzorDigest(object):
def __init__(self, msg):
self.debug = []
self.digest = None
self.predigest = None
self.bodytext_size = 0
self.filter = SuspectFilter(None)
self.logger = logging.getLogger('fuglu.plugins.fuzor.Digest')
# digest config
self.LONG_WORD_THRESHOLD = 10 # what is considered a long word
self.REPLACE_LONG_WORD = '[LONG]' # Replace long words in pre-digest with... None to disable
self.REPLACE_EMAIL = '[EMAIL]' # Replace email addrs in pre-digest with... None to disable
self.REPLACE_URL = '[LINK]' # Replace urls in pre-digest with... None to disable
self.INCLUDE_ATTACHMENT_CONTENT = False # should non-text attachment contents be included in digest (not recommended, there are better attachment hash systems)
self.INCLUDE_ATTACHMENT_COUNT = True # should the number of non-text-attachments be included in the digest
self.MINIMUM_PREDIGEST_SIZE = 27 # if the predigest is smaller than this, ignore this message
self.MINIMUM_UNMODIFIED_CONTENT = 27 # minimum unmodified content after stripping, eg. [SOMETHING] removed from the predigest (27>'von meinem Iphone gesendet')
self.MINIMUM_BODYTEXT_SIZE = 27 # if the body text content is smaller than this, ignore this message
self.STRIP_WHITESPACE = True # remove all whitespace from the pre-digest
self.STRIP_HTML_MARKUP = True # remove html tags (but keep content)
self.REMOVE_HTML_TAGS = [
'script',
'style'] # strip tags (including content)
self.predigest = self._make_predigest(msg)
self.digest = self._make_hash(self.predigest)
def _make_hash(self, predigest):
if self.bodytext_size < self.MINIMUM_BODYTEXT_SIZE:
return None
predigest = predigest.strip()
if isinstance(predigest, unicode):
predigest = predigest.encode('utf-8', 'ignore')
if len(predigest) < self.MINIMUM_PREDIGEST_SIZE:
return None
unmodified = re.sub(r'\[[A-Z0-9:]+\]', '', predigest)
if len(unmodified) < self.MINIMUM_UNMODIFIED_CONTENT:
return None
try:
return hashlib.sha1(predigest).hexdigest()
except:
return None
def _handle_text_part(self, part):
payload = part.get_payload(decode=True)
charset = part.get_content_charset()
errors = "ignore"
if not charset:
charset = "ascii"
elif (charset.lower().replace("_", "-") in ("quopri-codec",
"quopri", "quoted-printable", "quotedprintable")):
errors = "strict"
try:
payload = payload.decode(charset, errors)
except (LookupError, UnicodeError, AssertionError):
payload = payload.decode("ascii", "ignore")
if self.STRIP_HTML_MARKUP:
payload = self.filter.strip_text(
payload,
remove_tags=self.REMOVE_HTML_TAGS,
use_bfs=True)
if self.REPLACE_EMAIL is not None:
payload = re.sub(r'\S{1,50}@\S{1,30}', self.REPLACE_EMAIL, payload)
if self.REPLACE_URL is not None:
payload = re.sub(r'[a-z]+:\S{1,100}', self.REPLACE_URL, payload)
if self.REPLACE_LONG_WORD is not None:
patt = r'\S{%s,}' % self.LONG_WORD_THRESHOLD
payload = re.sub(patt, self.REPLACE_LONG_WORD, payload)
if self.STRIP_WHITESPACE:
payload = re.sub(r'\s', '', payload)
payload = payload.strip()
return payload
def _make_predigest(self, msg):
attachment_count = 0
predigest = ''
for part in msg.walk():
if part.is_multipart():
continue
if part.get_content_maintype() == "text":
try:
normalized_text_part = self._handle_text_part(part)
predigest += normalized_text_part
self.bodytext_size += len(normalized_text_part)
except Exception as e:
self.logger.warn(e)
else:
attachment_count += 1
if self.INCLUDE_ATTACHMENT_CONTENT:
predigest += "[ATTH:%s]" % hashlib.sha1(
part.get_payload()).hexdigest()
if self.INCLUDE_ATTACHMENT_COUNT and attachment_count:
predigest += "[ATTC:%s]" % attachment_count
if self.STRIP_WHITESPACE:
predigest = re.sub(r'\s', '', predigest)
return predigest
class SuspectFilterTestCase(unittest.TestCase):
"""Test Suspectfilter"""
def setUp(self):
self.candidate = SuspectFilter(TESTDATADIR + '/headertest.regex')
def tearDown(self):
pass
def test_sf_get_args(self):
"""Test SuspectFilter files"""
suspect = Suspect('*****@*****.**',
'*****@*****.**',
TESTDATADIR + '/helloworld.eml')
suspect.tags['testtag'] = 'testvalue'
headermatches = self.candidate.get_args(suspect)
self.assertTrue('Sent to unittest domain!' in headermatches,
"To_domain not found in headercheck")
self.assertTrue(
'Envelope sender is [email protected]' in headermatches,
"Envelope Sender not matched in header chekc")
self.assertTrue('Mime Version is 1.0' in headermatches,
"Standard header Mime Version not found")
self.assertTrue('A tag match' in headermatches,
"Tag match did not work")
self.assertTrue('Globbing works' in headermatches,
"header globbing failed")
self.assertTrue('body rule works' in headermatches,
"decoded body rule failed")
self.assertTrue('full body rule works' in headermatches,
"full body failed")
self.assertTrue('mime rule works' in headermatches, "mime rule failed")
self.assertFalse(
'this should not match in a body rule' in headermatches,
'decoded body rule matched raw body')
# perl style advanced rules
self.assertTrue('perl-style /-notation works!' in headermatches,
"new rule format failed: %s" % headermatches)
self.assertTrue(
'perl-style recipient match' in headermatches,
"new rule format failed for to_domain: %s" % headermatches)
self.assertFalse('this should not match' in headermatches,
"rule flag ignorecase was not detected")
# TODO: raw body rules
def test_sf_matches(self):
"""Test SuspectFilter extended matches"""
suspect = Suspect('*****@*****.**',
'*****@*****.**',
TESTDATADIR + '/helloworld.eml')
(match, info) = self.candidate.matches(suspect, extended=True)
self.assertTrue(match, 'Match should return True')
field, matchedvalue, arg, regex = info
self.assertTrue(field == 'to_domain')
self.assertTrue(matchedvalue == 'unittests.fuglu.org')
self.assertTrue(arg == 'Sent to unittest domain!')
self.assertTrue(regex == 'unittests\.fuglu\.org')
def test_sf_get_field(self):
"""Test SuspectFilter field extract"""
suspect = Suspect('*****@*****.**',
'*****@*****.**',
TESTDATADIR + '/helloworld.eml')
# additional field tests
self.assertEqual(
self.candidate.get_field(suspect, 'clienthelo')[0], 'helo1')
self.assertEqual(
self.candidate.get_field(suspect, 'clientip')[0], '10.0.0.1')
self.assertEqual(
self.candidate.get_field(suspect, 'clienthostname')[0], 'rdns1')
def test_strip(self):
html = """foo<a href="bar">bar</a><script language="JavaScript">echo('hello world');</script>baz"""
declarationtest = """<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="de">
<head>
<title>greetings</title>
</head>
<body>
<font color="red">well met!</font>
</body>
</html>
"""
# word generated empty message
wordhtml = """<html xmlns:v=3D"urn:schemas-microsoft-com:vml"
xmlns:o=3D"urn:schemas-microsoft-com:office:office"
xmlns:w=3D"urn:schemas-microsoft-com:office:word"
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml"
xmlns=3D"http://www.w3.org/TR/REC-html40"><head><META
HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html;
charset=3Dus-ascii"><meta name=3DGenerator content=3D"Microsoft Word 15
(filtered medium)"><style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;
mso-fareast-language:EN-US;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:#0563C1;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:#954F72;
text-decoration:underline;}
span.E-MailFormatvorlage17
{mso-style-type:personal-compose;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri",sans-serif;
mso-fareast-language:EN-US;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:70.85pt 70.85pt 2.0cm 70.85pt;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DDE-CH
link=3D"#0563C1" vlink=3D"#954F72"><div class=3DWordSection1><p
class=3DMsoNormal><o:p> </o:p></p></div></body></html>"""
for use_bfs in [True, False]:
stripped = self.candidate.strip_text(html, use_bfs=use_bfs)
self.assertEqual(stripped, 'foobarbaz')
docstripped = self.candidate.strip_text(declarationtest,
use_bfs=use_bfs)
self.assertEqual(docstripped.split(),
['greetings', 'well', 'met!'])
wordhtmstripped = self.candidate.strip_text(wordhtml,
use_bfs=use_bfs)
self.assertEqual(wordhtmstripped.strip(), '')
class SuspectFilterTestCase(unittest.TestCase):
"""Test Header Filter"""
def setUp(self):
self.candidate = SuspectFilter(TESTDATADIR + '/headertest.regex')
def tearDown(self):
pass
def test_sf_get_args(self):
"""Test SuspectFilter files"""
suspect = Suspect('*****@*****.**',
'*****@*****.**', TESTDATADIR + '/helloworld.eml')
suspect.tags['testtag'] = 'testvalue'
headermatches = self.candidate.get_args(suspect)
self.assertTrue(
'Sent to unittest domain!' in headermatches, "To_domain not found in headercheck")
self.assertTrue('Envelope sender is [email protected]' in headermatches,
"Envelope Sender not matched in header chekc")
self.assertTrue('Mime Version is 1.0' in headermatches,
"Standard header Mime Version not found")
self.assertTrue(
'A tag match' in headermatches, "Tag match did not work")
self.assertTrue(
'Globbing works' in headermatches, "header globbing failed")
self.assertTrue(
'body rule works' in headermatches, "decoded body rule failed")
self.assertTrue(
'full body rule works' in headermatches, "full body failed")
self.assertTrue('mime rule works' in headermatches, "mime rule failed")
self.assertFalse('this should not match in a body rule' in headermatches,
'decoded body rule matched raw body')
# perl style advanced rules
self.assertTrue('perl-style /-notation works!' in headermatches,
"new rule format failed: %s" % headermatches)
self.assertTrue('perl-style recipient match' in headermatches,
"new rule format failed for to_domain: %s" % headermatches)
self.assertFalse('this should not match' in headermatches,
"rule flag ignorecase was not detected")
# TODO: raw body rules
def test_sf_matches(self):
"""Test SuspectFilter extended matches"""
suspect = Suspect('*****@*****.**',
'*****@*****.**', TESTDATADIR + '/helloworld.eml')
(match, info) = self.candidate.matches(suspect, extended=True)
self.assertTrue(match, 'Match should return True')
field, matchedvalue, arg, regex = info
self.assertTrue(field == 'to_domain')
self.assertTrue(matchedvalue == 'unittests.fuglu.org')
self.assertTrue(arg == 'Sent to unittest domain!')
self.assertTrue(regex == 'unittests\.fuglu\.org')
def test_sf_get_field(self):
"""Test SuspectFilter field extract"""
suspect = Suspect('*****@*****.**',
'*****@*****.**', TESTDATADIR + '/helloworld.eml')
# additional field tests
self.assertEqual(self.candidate.get_field(
suspect, 'clienthelo')[0], 'helo1')
self.assertEqual(self.candidate.get_field(
suspect, 'clientip')[0], '10.0.0.1')
self.assertEqual(self.candidate.get_field(
suspect, 'clienthostname')[0], 'rdns1')
def test_strip(self):
html = """foo<a href="bar">bar</a><script language="JavaScript">echo('hello world');</script>baz"""
declarationtest = """<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="de">
<head>
<title>greetings</title>
</head>
<body>
<font color="red">well met!</font>
</body>
</html>
"""
# word generated empty message
wordhtml = """<html xmlns:v=3D"urn:schemas-microsoft-com:vml"
xmlns:o=3D"urn:schemas-microsoft-com:office:office"
xmlns:w=3D"urn:schemas-microsoft-com:office:word"
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml"
xmlns=3D"http://www.w3.org/TR/REC-html40"><head><META
HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html;
charset=3Dus-ascii"><meta name=3DGenerator content=3D"Microsoft Word 15
(filtered medium)"><style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;
mso-fareast-language:EN-US;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:#0563C1;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:#954F72;
text-decoration:underline;}
span.E-MailFormatvorlage17
{mso-style-type:personal-compose;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri",sans-serif;
mso-fareast-language:EN-US;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:70.85pt 70.85pt 2.0cm 70.85pt;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DDE-CH
link=3D"#0563C1" vlink=3D"#954F72"><div class=3DWordSection1><p
class=3DMsoNormal><o:p> </o:p></p></div></body></html>"""
for use_bfs in [True, False]:
stripped = self.candidate.strip_text(html, use_bfs=use_bfs)
self.assertEqual(stripped, 'foobarbaz')
docstripped = self.candidate.strip_text(
declarationtest, use_bfs=use_bfs)
self.assertEqual(
docstripped.split(), ['greetings', 'well', 'met!'])
wordhtmstripped = self.candidate.strip_text(
wordhtml, use_bfs=use_bfs)
self.assertEqual(wordhtmstripped.strip(), '')
