NaxsiRules.__validate_genericstr() - Code Metrics - Inspection of "fix zone_name" - nbs-system/spike - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Branch — master (1c2da9)

by jvo

created 2016-03-17 11:57 UTC

NaxsiRules.__validate_genericstr() B

↳ Parent: NaxsiRules

Complexity

Conditions

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	7
CRAP Score	8.6047

Metric	Value
cc	6
dl	0
loc	12
ccs	7
cts	12
cp	0.5833
crap	8.6047
rs	8

from time import strftime, localtime
import logging


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 explanation(self):
        """ Return a string explainign a rule """
        assoc = {'ARGS': 'argument', 'BODY': 'body', 'URL': 'url', 'HEADER': 'header'}
        expl = 'The rule number <strong>%d</strong> is ' % self.sid
        if self.negative:
            expl += '<strong>not</strong> '
        expl += 'setting the '
        scores = []
        for score in self.score.split(','):
            scores.append('<strong>{0}</strong> to <strong>{1}</strong> '.format(*score.split(':', 3)))
        expl += ', '.join(scores) + 'when it '
        if self.detection.startswith('str:'):
            expl += 'finds the string <strong>{}</strong> '.format(self.detection[4:])
        else:
            expl += 'matches the regexp <strong>{}</strong> '.format(self.detection[3:])

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

                zone_name = "?"
                for tmpzone in assoc:
                    if tmpzone in zone:
                        zone_name = assoc[tmpzone]
                if "$URL" in zone:
                    expl += "on the URL {} '{}' ".format("matching regex" if zone == "$URL_X" else "",
                                                          arg)
                else:
                    expl += "in the var with name {} '{}' of {} ".format("matching regex" if zone.endswith("_X") else "",

                                                                  arg, zone_name)
            else:
                zones.append('the <strong>{0}</strong>'.format(assoc[mz]))
        return expl

    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:") or p_str.startswith("rx:"):
            if assign is True:
                self.detection = p_str
        else:
            return self.__fail("detection {} is neither rx: or str:".format(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()
        locs = p_str.split('|')
        for loc in locs:
            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(":")
            # check it is a valid keyword
            if keyword not in self.sub_mz:
                return self.__fail("'{0}' no a known sub-part of mz : {1}".format(keyword, self.sub_mz))
            mz_state.add(keyword)
            # verify 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 $*_X ({})".format(', '.join(mz_state)))
            # just a gentle reminder
            if arg and arg.islower() is False:
                self.warnings.append("{0} in {1} is not lowercase. naxsi is case-insensitive".format(arg, loc))
            # the rule targets an actual zone
            if keyword not in ["$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
        items = list(iter(lexer.get_token, ''))
        return items

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


1	1	from time import strftime, localtime
2	1	import logging
		0 ignored issues – show Unused Code introduced 2016-03-16 17:05 UTC by Report Bug Copy Issue Report The import `logging` seems to be unused. Loading history...
3
4	1	from spike.model import db
5	1	from shlex import shlex
6
7
8	1	class NaxsiRules(db.Model):
9	1	__bind_key__ = 'rules'
10	1	__tablename__ = 'naxsi_rules'
11
12	1	id = db.Column(db.Integer, primary_key=True)
13	1	msg = db.Column(db.String(), nullable=False)
14	1	detection = db.Column(db.String(1024), nullable=False)
15	1	mz = db.Column(db.String(1024), nullable=False)
16	1	score = db.Column(db.String(1024), nullable=False)
17	1	sid = db.Column(db.Integer, nullable=False, unique=True)
18	1	ruleset = db.Column(db.String(1024), nullable=False)
19	1	rmks = db.Column(db.Text, nullable=True, server_default="")
20	1	active = db.Column(db.Integer, nullable=False, server_default="1")
21	1	negative = db.Column(db.Integer, nullable=False, server_default='0')
22	1	timestamp = db.Column(db.Integer, nullable=False)
23
24	1	mr_kw = ["MainRule", "BasicRule", "main_rule", "basic_rule"]
25	1	static_mz = {"$ARGS_VAR", "$BODY_VAR", "$URL", "$HEADERS_VAR"}
26	1	full_zones = {"ARGS", "BODY", "URL", "HEADERS", "FILE_EXT", "RAW_BODY"}
27	1	rx_mz = {"$ARGS_VAR_X", "$BODY_VAR_X", "$URL_X", "$HEADERS_VAR_X"}
28	1	sub_mz = list(static_mz) + list(full_zones) + list(rx_mz)
29
30	1	def __init__(self, msg="", detection="", mz="", score="", sid=42000, ruleset="", rmks="", active=0, negative=0, timestamp=0):
		0 ignored issues – show Coding Style introduced 2016-03-03 14:14 UTC by Report Bug Copy Issue Report This line is too long as per the coding-style (129/120). This check looks for lines that are too long. You can specify the maximum line length. Loading history...
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 explanation(self):
55		""" Return a string explainign a rule """
56	1	assoc = {'ARGS': 'argument', 'BODY': 'body', 'URL': 'url', 'HEADER': 'header'}
57	1	expl = 'The rule number <strong>%d</strong> is ' % self.sid
58	1	if self.negative:
59		expl += '<strong>not</strong> '
60	1	expl += 'setting the '
61	1	scores = []
62	1	for score in self.score.split(','):
63	1	scores.append('<strong>{0}</strong> to <strong>{1}</strong> '.format(*score.split(':', 3)))
64	1	expl += ', '.join(scores) + 'when it '
65	1	if self.detection.startswith('str:'):
66	1	expl += 'finds the string <strong>{}</strong> '.format(self.detection[4:])
67		else:
68		expl += 'matches the regexp <strong>{}</strong> '.format(self.detection[3:])
69
70	1	zones = []
71	1	for mz in self.mz.split('\|'):
72	1	if mz.startswith('$'):
73		zone,arg = mz.split(":")
		0 ignored issues – show Coding Style introduced 2016-03-17 11:33 UTC by Report Bug Copy Issue Report Exactly one space required after comma zone,arg = mz.split(":") ^ Loading history...
74		zone_name = "?"
75		for tmpzone in assoc:
76		if tmpzone in zone:
77		zone_name = assoc[tmpzone]
78		if "$URL" in zone:
79		expl += "on the URL {} '{}' ".format("matching regex" if zone == "$URL_X" else "",
80		arg)
81		else:
82		expl += "in the var with name {} '{}' of {} ".format("matching regex" if zone.endswith("_X") else "",
		0 ignored issues – show Coding Style introduced 2016-03-17 11:33 UTC by Report Bug Copy Issue Report This line is too long as per the coding-style (121/120). This check looks for lines that are too long. You can specify the maximum line length. Loading history...
83		arg, zone_name)
84		else:
85	1	zones.append('the <strong>{0}</strong>'.format(assoc[mz]))
86	1	return expl
87
88	1	def validate(self):
89	1	self.__validate_matchzone(self.mz)
90	1	self.__validate_id(self.sid)
91	1	self.__validate_detection(self.detection)
92
93	1	if not self.msg:
94		self.warnings.append("Rule has no 'msg:'.")
95	1	if not self.score:
96		self.error.append("Rule has no score.")
97
98	1	def __fail(self, msg):
99	1	self.error.append(msg)
100	1	return False
101
102		# Bellow are parsers for specific parts of a rule
103
104	1	def __validate_detection(self, p_str, label="", assign=False):
105	1	p_str = label + p_str
106	1	if not p_str.islower():
107	1	self.warnings.append("detection {} is not lower-case. naxsi is case-insensitive".format(p_str))
108	1	if p_str.startswith("str:") or p_str.startswith("rx:"):
109	1	if assign is True:
110	1	self.detection = p_str
111		else:
112	1	return self.__fail("detection {} is neither rx: or str:".format(p_str))
113	1	return True
114
115	1	def __validate_genericstr(self, p_str, label="", assign=False):
116	1	if assign is False:
117		return True
118	1	if label == "s:":
119	1	self.score = p_str
120	1	elif label == "msg:":
121	1	self.msg = p_str
122		elif label == "negative":
123		self.negative = 1
124		elif label != "":
125		return self.__fail("Unknown fragment {}".format(label+p_str))
126	1	return True
127
128	1	def __validate_matchzone(self, p_str, label="", assign=False):
		0 ignored issues – show Unused Code introduced 2016-03-17 11:33 UTC by Report Bug Copy Issue Report The argument `label` seems to be unused. Loading history...
129	1	has_zone = False
130	1	mz_state = set()
131	1	locs = p_str.split('\|')
132	1	for loc in locs:
133	1	keyword, arg = loc, None
134	1	if loc.startswith("$"):
135	1	if loc.find(":") == -1:
136		return self.__fail("Missing 2nd part after ':' in {0}".format(loc))
137	1	keyword, arg = loc.split(":")
138		# check it is a valid keyword
139	1	if keyword not in self.sub_mz:
140		return self.__fail("'{0}' no a known sub-part of mz : {1}".format(keyword, self.sub_mz))
141	1	mz_state.add(keyword)
142		# verify the rule doesn't attempt to target REGEX and STATIC _VAR/URL at the same time
143	1	if len(self.rx_mz & mz_state) and len(self.static_mz & mz_state):
144	1	return self.__fail("You can't mix static $* with regex $*_X ({})".format(', '.join(mz_state)))
145		# just a gentle reminder
146	1	if arg and arg.islower() is False:
147	1	self.warnings.append("{0} in {1} is not lowercase. naxsi is case-insensitive".format(arg, loc))
148		# the rule targets an actual zone
149	1	if keyword not in ["$URL", "$URL_X"] and keyword in (self.rx_mz \| self.full_zones \| self.static_mz):
		0 ignored issues – show Unused Code Coding Style introduced 2016-03-15 16:16 UTC by Report Bug Copy Issue Report There is an unnecessary parenthesis after `in`. Loading history...
150	1	has_zone = True
151	1	if has_zone is False:
152		return self.__fail("The rule/whitelist doesn't target any zone.")
153	1	if assign is True:
154	1	self.mz = p_str
155	1	return True
156
157	1	def __validate_id(self, p_str, label="", assign=False):
		0 ignored issues – show Unused Code introduced 2016-03-17 11:33 UTC by Report Bug Copy Issue Report The argument `label` seems to be unused. Loading history...
158	1	try:
159	1	num = int(p_str)
160	1	if num < 10000:
161	1	self.warnings.append("rule IDs below 10k are reserved ({0})".format(num))
162		except ValueError:
163		self.error.append("id:{0} is not numeric".format(p_str))
164		return False
165	1	if assign is True:
166	1	self.sid = num
167	1	return True
168
169	1	@staticmethod
170		def splitter(s):
		0 ignored issues – show Coding Style Naming introduced 2016-03-16 11:23 UTC by Report Bug Copy Issue Report The name `s` does not conform to the argument naming conventions (`[a-z_][a-z0-9_]{1,30}$`). 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...
171	1	lexer = shlex(s)
172	1	lexer.whitespace_split = True
173	1	items = list(iter(lexer.get_token, ''))
174	1	return items
175
176	1	def parse_rule(self, full_str):
177		"""
178		Parse and validate a full naxsi rule
179		:param full_str: raw rule
180		:return: [True\|False, dict]
181		"""
182	1	self.warnings = []
183	1	self.error = []
184
185	1	func_map = {"id:": self.__validate_id, "str:": self.__validate_detection,
186		"rx:": self.__validate_detection, "msg:": self.__validate_genericstr, "mz:": self.__validate_matchzone,
		0 ignored issues – show Coding Style introduced 2016-03-17 11:33 UTC by Report Bug Copy Issue Report This line is too long as per the coding-style (123/120). This check looks for lines that are too long. You can specify the maximum line length. Loading history...
187		"negative": self.__validate_genericstr, "s:": self.__validate_genericstr}
188	1	ret = False
189	1	split = self.splitter(full_str) # parse string
190	1	intersection = set(split).intersection(set(self.mr_kw))
191
192	1	if not intersection:
193		return self.__fail("No mainrule/basicrule keyword.")
194	1	elif len(intersection) > 1:
195		return self.__fail("Multiple mainrule/basicrule keywords.")
196
197	1	split.remove(intersection.pop()) # remove the mainrule/basicrule keyword
198
199	1	if ";" in split:
200	1	split.remove(";")
201
202	1	while split: # iterate while there is data, as handlers can defer
203	1	for keyword in split:
204	1	orig_kw = keyword
205	1	keyword = keyword.strip()
206
207	1	if keyword.endswith(";"): # remove semi-colons
208	1	keyword = keyword[:-1]
209	1	if keyword.startswith(('"', "'")) and (keyword[0] == keyword[-1]): # remove (double-)quotes
210	1	keyword = keyword[1:-1]
211	1	for frag_kw in func_map:
212	1	ret = False
213	1	if keyword.startswith(frag_kw):
214		# parser funcs returns True/False
215	1	ret = func_map[frag_kw](keyword[len(frag_kw):], label=frag_kw, assign=True)
216	1	if ret is True:
217	1	split.remove(orig_kw)
218		else:
219	1	return self.__fail("parsing of element '{0}' failed.".format(keyword))
220	1	break
221		# we have an item that wasn't successfully parsed
222	1	if orig_kw in split and ret is not None:
223		return False
224		return True
225

nbs-system / spike

Branch — master (1c2da9)

NaxsiRules.__validate_genericstr() B

Complexity

Size

Duplication

Code Coverage

Duplication Side-by-Side

Filter issues like