NaxsiRules.fullstr() - Code Metrics - Inspection of "Add regexp validation for rules" - nbs-system/spike - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( 7424d3...b047e1 )

by -

created 2016-03-17 13:40 UTC

NaxsiRules.fullstr() A

↳ Parent: NaxsiRules

Complexity

Conditions

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	4
CRAP Score	1

Metric	Value
cc	1
dl	0
loc	4
ccs	4
cts	4
cp	1
crap	1
rs	10

from time import strftime, localtime

from spike.model import db
from shlex import shlex


class NaxsiRules(db.Model):
    __bind_key__ = 'rules'
    __tablename__ = 'naxsi_rules'

    id = db.Column(db.Integer, primary_key=True)
    msg = db.Column(db.String(), nullable=False)
    detection = db.Column(db.String(1024), nullable=False)
    mz = db.Column(db.String(1024), nullable=False)
    score = db.Column(db.String(1024), nullable=False)
    sid = db.Column(db.Integer, nullable=False, unique=True)
    ruleset = db.Column(db.String(1024), nullable=False)
    rmks = db.Column(db.Text, nullable=True, server_default="")
    active = db.Column(db.Integer, nullable=False, server_default="1")
    negative = db.Column(db.Integer, nullable=False, server_default='0')
    timestamp = db.Column(db.Integer, nullable=False)

    mr_kw = ["MainRule", "BasicRule", "main_rule", "basic_rule"]
    static_mz = {"$ARGS_VAR", "$BODY_VAR", "$URL", "$HEADERS_VAR"}
    full_zones = {"ARGS", "BODY", "URL", "HEADERS", "FILE_EXT", "RAW_BODY"}
    rx_mz = {"$ARGS_VAR_X", "$BODY_VAR_X", "$URL_X", "$HEADERS_VAR_X"}
    sub_mz = list(static_mz) + list(full_zones) + list(rx_mz)

    def __init__(self, msg="", detection="", mz="", score="", sid=42000, ruleset="", rmks="", active=0, negative=0,
                 timestamp=0):
        self.msg = msg
        self.detection = detection
        self.mz = mz
        self.score = score
        self.sid = sid
        self.ruleset = ruleset
        self.rmks = rmks
        self.active = active
        self.negative = 1 if negative == 'checked' else 0
        self.timestamp = timestamp
        self.warnings = []
        self.error = []

    def fullstr(self):
        rdate = strftime("%F - %H:%M", localtime(float(str(self.timestamp))))
        rmks = "# ".join(self.rmks.strip().split("\n"))
        return "#\n# sid: {0} | date: {1}\n#\n# {2}\n#\n{3}".format(self.sid, rdate, rmks, self.__str__())

    def __str__(self):
        negate = 'negative' if self.negative == 1 else ''
        return 'MainRule {} "{}" "msg:{}" "mz:{}" "s:{}" id:{} ;'.format(
            negate, self.detection, self.msg, self.mz, self.score, self.sid)

    def explain(self):
        """ Return a string explaining the rule

         :return str: A textual explanation of the rule
         """
        translation = {'ARGS': 'argument', 'BODY': 'body', 'URL': 'url', 'HEADER': 'header'}
        explanation = 'The rule number <strong>%d</strong> is ' % self.sid
        if self.negative:
            explanation += '<strong>not</strong> '
        explanation += 'setting the '

        scores = []
        for score in self.score.split(','):
            scores.append('<strong>{0}</strong> to <strong>{1}</strong> '.format(*score.split(':', 1)))
        explanation += ', '.join(scores) + 'when it '
        if self.detection.startswith('str:'):
            explanation += 'finds the string <strong>{}</strong> '.format(self.detection[4:])
        else:
            explanation += 'matches the regexp <strong>{}</strong> '.format(self.detection[3:])

        zones = []
        for mz in self.mz.split('|'):
            if mz.startswith('$'):
                current_zone, arg = mz.split(":", 1)
                zone_name = "?"

                for translated_name in translation:  # translate zone names
                    if translated_name in current_zone:
                        zone_name = translation[translated_name]

                if "$URL" in current_zone:
                    regexp = "matching regex" if current_zone == "$URL_X" else ""
                    explanation += "on the URL {} '{}' ".format(regexp, arg)
                else:
                    regexp = "matching regex" if current_zone.endswith("_X") else ""
                    explanation += "in the var with name {} '{}' of {} ".format(regexp, arg, zone_name)
            else:
                zones.append('the <strong>{0}</strong>'.format(translation[mz]))
        return explanation

    def validate(self):
        self.__validate_matchzone(self.mz)
        self.__validate_id(self.sid)
        self.__validate_detection(self.detection)

        if not self.msg:
            self.warnings.append("Rule has no 'msg:'.")
        if not self.score:
            self.error.append("Rule has no score.")

    def __fail(self, msg):
        self.error.append(msg)
        return False

    # Bellow are parsers for specific parts of a rule

    def __validate_detection(self, p_str, label="", assign=False):
        p_str = label + p_str
        if not p_str.islower():
            self.warnings.append("detection {} is not lower-case. naxsi is case-insensitive".format(p_str))

        if p_str.startswith("str:"):  # no further validation on strings
            pass
        elif p_str.startswith("rx:"):
            try:  # try to validate the regex with PCRE's python bindings
                import pcre
                try:  # if we can't compile the regex, it's likely invalid
                    pcre.compile(p_str[3:])
                except pcre.PCREError:
                    return self.__fail("{} is not a valid regex:".format(p_str))
            except ImportError:  # python-pcre is an optional dependency
try:
    raises_exception()
except:  # Could be removed
    pass
                pass
        else:
            return self.__fail("detection {} is neither rx: or str:".format(p_str))

        if assign is True:
            self.detection = p_str
        return True

    def __validate_genericstr(self, p_str, label="", assign=False):
        if assign is False:
            return True
        if label == "s:":
            self.score = p_str
        elif label == "msg:":
            self.msg = p_str
        elif label == "negative":
            self.negative = 1
        elif label != "":
            return self.__fail("Unknown fragment {}".format(label+p_str))
        return True

    def __validate_matchzone(self, p_str, label="", assign=False):
        has_zone = False
        mz_state = set()
        for loc in p_str.split('|'):
            keyword, arg = loc, None
            if loc.startswith("$"):
                if loc.find(":") == -1:
                    return self.__fail("Missing 2nd part after ':' in {0}".format(loc))
                keyword, arg = loc.split(":")

            if keyword not in self.sub_mz:  # check if `keyword` is a valid keyword
                return self.__fail("'{0}' no a known sub-part of mz : {1}".format(keyword, self.sub_mz))

            mz_state.add(keyword)

            # verify that the rule doesn't attempt to target REGEX and STATIC _VAR/URL at the same time
            if len(self.rx_mz & mz_state) and len(self.static_mz & mz_state):
                return self.__fail("You can't mix static $* with regex $*_X ({})".format(', '.join(mz_state)))

            if arg and not arg.islower():  # just a gentle reminder
                self.warnings.append("{0} in {1} is not lowercase. naxsi is case-insensitive".format(arg, loc))

            # the rule targets an actual zone
            if keyword not in ["$URL", "$URL_X"] and keyword in (self.rx_mz | self.full_zones | self.static_mz):

                has_zone = True

        if has_zone is False:
            return self.__fail("The rule/whitelist doesn't target any zone.")

        if assign is True:
            self.mz = p_str

        return True

    def __validate_id(self, p_str, label="", assign=False):
        try:
            num = int(p_str)
            if num < 10000:
                self.warnings.append("rule IDs below 10k are reserved ({0})".format(num))
        except ValueError:
            self.error.append("id:{0} is not numeric".format(p_str))
            return False
        if assign is True:
            self.sid = num
        return True

    @staticmethod
    def splitter(s):

        lexer = shlex(s)
        lexer.whitespace_split = True
        return list(iter(lexer.get_token, ''))

    def parse_rule(self, full_str):
        """
        Parse and validate a full naxsi rule
        :param full_str: raw rule
        :return: [True|False, dict]
        """
        self.warnings = []
        self.error = []

        func_map = {"id:": self.__validate_id, "str:": self.__validate_detection,
                    "rx:": self.__validate_detection, "msg:": self.__validate_genericstr,
                    "mz:": self.__validate_matchzone, "negative": self.__validate_genericstr,
                    "s:": self.__validate_genericstr}
        ret = False
        split = self.splitter(full_str)  # parse string
        intersection = set(split).intersection(set(self.mr_kw))

        if not intersection:
            return self.__fail("No mainrule/basicrule keyword.")
        elif len(intersection) > 1:
            return self.__fail("Multiple mainrule/basicrule keywords.")

        split.remove(intersection.pop())  # remove the mainrule/basicrule keyword

        if ";" in split:
            split.remove(";")

        while split:  # iterate while there is data, as handlers can defer
            for keyword in split:
                orig_kw = keyword
                keyword = keyword.strip()

                if keyword.endswith(";"):  # remove semi-colons
                    keyword = keyword[:-1]
                if keyword.startswith(('"', "'")) and (keyword[0] == keyword[-1]):  # remove (double-)quotes
                    keyword = keyword[1:-1]

                for frag_kw in func_map:
                    ret = False
                    if keyword.startswith(frag_kw):
                        # parser funcs returns True/False
                        ret = func_map[frag_kw](keyword[len(frag_kw):], label=frag_kw, assign=True)
                        if ret is True:
                            split.remove(orig_kw)
                        else:
                            return self.__fail("parsing of element '{0}' failed.".format(keyword))
                        break

                if orig_kw in split and ret is not None:  # we have an item that wasn't successfully parsed
                    return False
        return True


1	1	from time import strftime, localtime
2
3	1	from spike.model import db
4	1	from shlex import shlex
5
6
7	1	class NaxsiRules(db.Model):
8	1	__bind_key__ = 'rules'
9	1	__tablename__ = 'naxsi_rules'
10
11	1	id = db.Column(db.Integer, primary_key=True)
12	1	msg = db.Column(db.String(), nullable=False)
13	1	detection = db.Column(db.String(1024), nullable=False)
14	1	mz = db.Column(db.String(1024), nullable=False)
15	1	score = db.Column(db.String(1024), nullable=False)
16	1	sid = db.Column(db.Integer, nullable=False, unique=True)
17	1	ruleset = db.Column(db.String(1024), nullable=False)
18	1	rmks = db.Column(db.Text, nullable=True, server_default="")
19	1	active = db.Column(db.Integer, nullable=False, server_default="1")
20	1	negative = db.Column(db.Integer, nullable=False, server_default='0')
21	1	timestamp = db.Column(db.Integer, nullable=False)
22
23	1	mr_kw = ["MainRule", "BasicRule", "main_rule", "basic_rule"]
24	1	static_mz = {"$ARGS_VAR", "$BODY_VAR", "$URL", "$HEADERS_VAR"}
25	1	full_zones = {"ARGS", "BODY", "URL", "HEADERS", "FILE_EXT", "RAW_BODY"}
26	1	rx_mz = {"$ARGS_VAR_X", "$BODY_VAR_X", "$URL_X", "$HEADERS_VAR_X"}
27	1	sub_mz = list(static_mz) + list(full_zones) + list(rx_mz)
28
29	1	def __init__(self, msg="", detection="", mz="", score="", sid=42000, ruleset="", rmks="", active=0, negative=0,
30		timestamp=0):
31	1	self.msg = msg
32	1	self.detection = detection
33	1	self.mz = mz
34	1	self.score = score
35	1	self.sid = sid
36	1	self.ruleset = ruleset
37	1	self.rmks = rmks
38	1	self.active = active
39	1	self.negative = 1 if negative == 'checked' else 0
40	1	self.timestamp = timestamp
41	1	self.warnings = []
42	1	self.error = []
43
44	1	def fullstr(self):
45	1	rdate = strftime("%F - %H:%M", localtime(float(str(self.timestamp))))
46	1	rmks = "# ".join(self.rmks.strip().split("\n"))
47	1	return "#\n# sid: {0} \| date: {1}\n#\n# {2}\n#\n{3}".format(self.sid, rdate, rmks, self.__str__())
48
49	1	def __str__(self):
50	1	negate = 'negative' if self.negative == 1 else ''
51	1	return 'MainRule {} "{}" "msg:{}" "mz:{}" "s:{}" id:{} ;'.format(
52		negate, self.detection, self.msg, self.mz, self.score, self.sid)
53
54	1	def explain(self):
55		""" Return a string explaining the rule
56
57		:return str: A textual explanation of the rule
58		"""
59	1	translation = {'ARGS': 'argument', 'BODY': 'body', 'URL': 'url', 'HEADER': 'header'}
60	1	explanation = 'The rule number <strong>%d</strong> is ' % self.sid
61	1	if self.negative:
62		explanation += '<strong>not</strong> '
63	1	explanation += 'setting the '
64
65	1	scores = []
66	1	for score in self.score.split(','):
67	1	scores.append('<strong>{0}</strong> to <strong>{1}</strong> '.format(*score.split(':', 1)))
68	1	explanation += ', '.join(scores) + 'when it '
69	1	if self.detection.startswith('str:'):
70	1	explanation += 'finds the string <strong>{}</strong> '.format(self.detection[4:])
71		else:
72		explanation += 'matches the regexp <strong>{}</strong> '.format(self.detection[3:])
73
74	1	zones = []
75	1	for mz in self.mz.split('\|'):
76	1	if mz.startswith('$'):
77		current_zone, arg = mz.split(":", 1)
78		zone_name = "?"
79
80		for translated_name in translation: # translate zone names
81		if translated_name in current_zone:
82		zone_name = translation[translated_name]
83
84		if "$URL" in current_zone:
85		regexp = "matching regex" if current_zone == "$URL_X" else ""
86		explanation += "on the URL {} '{}' ".format(regexp, arg)
87		else:
88		regexp = "matching regex" if current_zone.endswith("_X") else ""
89		explanation += "in the var with name {} '{}' of {} ".format(regexp, arg, zone_name)
90		else:
91	1	zones.append('the <strong>{0}</strong>'.format(translation[mz]))
92	1	return explanation
93
94	1	def validate(self):
95	1	self.__validate_matchzone(self.mz)
96	1	self.__validate_id(self.sid)
97	1	self.__validate_detection(self.detection)
98
99	1	if not self.msg:
100		self.warnings.append("Rule has no 'msg:'.")
101	1	if not self.score:
102		self.error.append("Rule has no score.")
103
104	1	def __fail(self, msg):
105	1	self.error.append(msg)
106	1	return False
107
108		# Bellow are parsers for specific parts of a rule
109
110	1	def __validate_detection(self, p_str, label="", assign=False):
111	1	p_str = label + p_str
112	1	if not p_str.islower():
113	1	self.warnings.append("detection {} is not lower-case. naxsi is case-insensitive".format(p_str))
114
115	1	if p_str.startswith("str:"): # no further validation on strings
116	1	pass
117	1	elif p_str.startswith("rx:"):
118	1	try: # try to validate the regex with PCRE's python bindings
119	1	import pcre
120		try: # if we can't compile the regex, it's likely invalid
121		pcre.compile(p_str[3:])
122		except pcre.PCREError:
123		return self.__fail("{} is not a valid regex:".format(p_str))
124	1	except ImportError: # python-pcre is an optional dependency
		0 ignored issues – show Unused Code introduced 2016-03-17 13:42 UTC by Report Bug Copy Issue Report This except handler seems to be unused and could be removed. Except handlers which only contain `pass` and do not have an `else` clause can usually simply be removed: try: raises_exception() except: # Could be removed pass Loading history...
125	1	pass
126		else:
127	1	return self.__fail("detection {} is neither rx: or str:".format(p_str))
128
129	1	if assign is True:
130	1	self.detection = p_str
131	1	return True
132
133	1	def __validate_genericstr(self, p_str, label="", assign=False):
134	1	if assign is False:
135		return True
136	1	if label == "s:":
137	1	self.score = p_str
138	1	elif label == "msg:":
139	1	self.msg = p_str
140		elif label == "negative":
141		self.negative = 1
142		elif label != "":
143		return self.__fail("Unknown fragment {}".format(label+p_str))
144	1	return True
145
146	1	def __validate_matchzone(self, p_str, label="", assign=False):
147	1	has_zone = False
148	1	mz_state = set()
149	1	for loc in p_str.split('\|'):
150	1	keyword, arg = loc, None
151	1	if loc.startswith("$"):
152	1	if loc.find(":") == -1:
153		return self.__fail("Missing 2nd part after ':' in {0}".format(loc))
154	1	keyword, arg = loc.split(":")
155
156	1	if keyword not in self.sub_mz: # check if `keyword` is a valid keyword
157		return self.__fail("'{0}' no a known sub-part of mz : {1}".format(keyword, self.sub_mz))
158
159	1	mz_state.add(keyword)
160
161		# verify that the rule doesn't attempt to target REGEX and STATIC _VAR/URL at the same time
162	1	if len(self.rx_mz & mz_state) and len(self.static_mz & mz_state):
163	1	return self.__fail("You can't mix static $* with regex $*_X ({})".format(', '.join(mz_state)))
164
165	1	if arg and not arg.islower(): # just a gentle reminder
166	1	self.warnings.append("{0} in {1} is not lowercase. naxsi is case-insensitive".format(arg, loc))
167
168		# the rule targets an actual zone
169	1	if keyword not in ["$URL", "$URL_X"] and keyword in (self.rx_mz \| self.full_zones \| self.static_mz):
		0 ignored issues – show Unused Code Coding Style introduced 2016-03-15 16:16 UTC by Report Bug Copy Issue Report There is an unnecessary parenthesis after `in`. Loading history...
170	1	has_zone = True
171
172	1	if has_zone is False:
173		return self.__fail("The rule/whitelist doesn't target any zone.")
174
175	1	if assign is True:
176	1	self.mz = p_str
177
178	1	return True
179
180	1	def __validate_id(self, p_str, label="", assign=False):
181	1	try:
182	1	num = int(p_str)
183	1	if num < 10000:
184	1	self.warnings.append("rule IDs below 10k are reserved ({0})".format(num))
185		except ValueError:
186		self.error.append("id:{0} is not numeric".format(p_str))
187		return False
188	1	if assign is True:
189	1	self.sid = num
190	1	return True
191
192	1	@staticmethod
193		def splitter(s):
		0 ignored issues – show Coding Style Naming introduced 2016-03-16 11:23 UTC by Report Bug Copy Issue Report The name `s` does not conform to the argument naming conventions (`[a-z_][a-z0-9_]{1,30}$`). This check looks for invalid names for a range of different identifiers. You can set regular expressions to which the identifiers must conform if the defaults do not match your requirements. If your project includes a Pylint configuration file, the settings contained in that file take precedence. To find out more about Pylint, please refer to their site. Loading history...
194	1	lexer = shlex(s)
195	1	lexer.whitespace_split = True
196	1	return list(iter(lexer.get_token, ''))
197
198	1	def parse_rule(self, full_str):
199		"""
200		Parse and validate a full naxsi rule
201		:param full_str: raw rule
202		:return: [True\|False, dict]
203		"""
204	1	self.warnings = []
205	1	self.error = []
206
207	1	func_map = {"id:": self.__validate_id, "str:": self.__validate_detection,
208		"rx:": self.__validate_detection, "msg:": self.__validate_genericstr,
209		"mz:": self.__validate_matchzone, "negative": self.__validate_genericstr,
210		"s:": self.__validate_genericstr}
211	1	ret = False
212	1	split = self.splitter(full_str) # parse string
213	1	intersection = set(split).intersection(set(self.mr_kw))
214
215	1	if not intersection:
216		return self.__fail("No mainrule/basicrule keyword.")
217	1	elif len(intersection) > 1:
218		return self.__fail("Multiple mainrule/basicrule keywords.")
219
220	1	split.remove(intersection.pop()) # remove the mainrule/basicrule keyword
221
222	1	if ";" in split:
223	1	split.remove(";")
224
225	1	while split: # iterate while there is data, as handlers can defer
226	1	for keyword in split:
227	1	orig_kw = keyword
228	1	keyword = keyword.strip()
229
230	1	if keyword.endswith(";"): # remove semi-colons
231	1	keyword = keyword[:-1]
232	1	if keyword.startswith(('"', "'")) and (keyword[0] == keyword[-1]): # remove (double-)quotes
233	1	keyword = keyword[1:-1]
234
235	1	for frag_kw in func_map:
236	1	ret = False
237	1	if keyword.startswith(frag_kw):
238		# parser funcs returns True/False
239	1	ret = func_map[frag_kw](keyword[len(frag_kw):], label=frag_kw, assign=True)
240	1	if ret is True:
241	1	split.remove(orig_kw)
242		else:
243	1	return self.__fail("parsing of element '{0}' failed.".format(keyword))
244	1	break
245
246	1	if orig_kw in split and ret is not None: # we have an item that wasn't successfully parsed
247		return False
248		return True
249

nbs-system / spike

Push — master ( 7424d3...b047e1 )

NaxsiRules.fullstr() A

Complexity

Size

Duplication

Code Coverage

Duplication Side-by-Side

Filter issues like