NaxsiRules

Complexity

Total Complexity

57

Size/Duplication

Total Lines	202
Duplicated Lines	0 %

Test Coverage

Coverage

88.68%

13 Methods

Rating	Name	Duplication	Size	Complexity
A	fullstr()	0	4	1
A	__init__()	0	13	2
A	__str__()	0	4	2
F	__validate_matchzone()	0	28	13
A	splitter()	0	6	1
A	__validate_genericstr()	0	4	3
A	__validate_dummy()	0	2	1
F	parse_rule()	0	49	14
A	validate()	0	9	3
A	__fail()	0	3	1
A	__validate_id()	0	11	4
B	__validate_detection()	0	8	5
C	explaination()	0	26	7

from time import strftime, localtime
import logging

The import logging seems to be unused.

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):

This line is too long as per the coding-style (129/120).

        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 explaination(self):
        """ Return a string explainign a rule """
        assoc = {'ARGS': 'request arguments', 'BODY': 'body', 'URL': 'url'}
        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:])
        expl += 'in '
        zones = []
        for mz in self.mz.split('|'):
            if mz.startswith('$'):
                if mz.lower().startswith('headers_var:cookie'):
                    zones.append('the cookies')
                else:
                    zones.append('the {0}, in the [1} field'.format(*mz.split(':')))
            else:
                zones.append('the <strong>{0}</strong>'.format(assoc[mz]))
        return expl + ', '.join(zones) + '.'

    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_dummy(self, s, assign=False):

The name s does not conform to the argument naming conventions ([a-z_][a-z0-9_]{1,30}$).

The argument s seems to be unused.

The argument assign seems to be unused.

This method could be written as a function/class method.

        return True

    def __validate_detection(self, p_str, assign=False):
        if not p_str.startswith("str:") and not p_str.startswith("rx:"):
            self.__fail("detection {} is neither rx: or str:".format(p_str))
        if not p_str.islower():
            self.warnings.append("detection {} is not lower-case. naxsi is case-insensitive".format(p_str))
        if assign is True:
            self.detection = p_str
        return True

    def __validate_genericstr(self, p_str, assign=False):

The argument assign seems to be unused.

        if p_str and not p_str.islower():
            self.warnings.append("Pattern ({0}) is not lower-case.".format(p_str))
        return True

    def __validate_matchzone(self, p_str, 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):

There is an unnecessary parenthesis after in.

                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, 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):

The name s does not conform to the argument naming conventions ([a-z_][a-z0-9_]{1,30}$).

        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_genericstr,
                    "rx:": self.__validate_genericstr, "msg:": self.__validate_dummy, "mz:": self.__validate_matchzone,
                    "negative": self.__validate_dummy, "s:": self.__validate_dummy}
        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):], 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