summary refs log tree commit diff
path: root/synapse/push/push_rule_evaluator.py
blob: 8e0d3a416d964f6ec33cec20c3e12dffc1888931 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
# -*- coding: utf-8 -*-
# Copyright 2015, 2016 OpenMarket Ltd
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import logging
import re
from typing import Pattern

from synapse.events import EventBase
from synapse.types import UserID
from synapse.util.caches import register_cache
from synapse.util.caches.lrucache import LruCache

logger = logging.getLogger(__name__)


GLOB_REGEX = re.compile(r"\\\[(\\\!|)(.*)\\\]")
IS_GLOB = re.compile(r"[\?\*\[\]]")
INEQUALITY_EXPR = re.compile("^([=<>]*)([0-9]*)$")


def _room_member_count(ev, condition, room_member_count):
    return _test_ineq_condition(condition, room_member_count)


def _sender_notification_permission(ev, condition, sender_power_level, power_levels):
    notif_level_key = condition.get("key")
    if notif_level_key is None:
        return False

    notif_levels = power_levels.get("notifications", {})
    room_notif_level = notif_levels.get(notif_level_key, 50)

    return sender_power_level >= room_notif_level


def _test_ineq_condition(condition, number):
    if "is" not in condition:
        return False
    m = INEQUALITY_EXPR.match(condition["is"])
    if not m:
        return False
    ineq = m.group(1)
    rhs = m.group(2)
    if not rhs.isdigit():
        return False
    rhs_int = int(rhs)

    if ineq == "" or ineq == "==":
        return number == rhs_int
    elif ineq == "<":
        return number < rhs_int
    elif ineq == ">":
        return number > rhs_int
    elif ineq == ">=":
        return number >= rhs_int
    elif ineq == "<=":
        return number <= rhs_int
    else:
        return False


def tweaks_for_actions(actions):
    tweaks = {}
    for a in actions:
        if not isinstance(a, dict):
            continue
        if "set_tweak" in a and "value" in a:
            tweaks[a["set_tweak"]] = a["value"]
    return tweaks


class PushRuleEvaluatorForEvent(object):
    def __init__(
        self,
        event: EventBase,
        room_member_count: int,
        sender_power_level: int,
        power_levels: dict,
    ):
        self._event = event
        self._room_member_count = room_member_count
        self._sender_power_level = sender_power_level
        self._power_levels = power_levels

        # Maps strings of e.g. 'content.body' -> event["content"]["body"]
        self._value_cache = _flatten_dict(event)

    def matches(self, condition: dict, user_id: str, display_name: str) -> bool:
        if condition["kind"] == "event_match":
            return self._event_match(condition, user_id)
        elif condition["kind"] == "contains_display_name":
            return self._contains_display_name(display_name)
        elif condition["kind"] == "room_member_count":
            return _room_member_count(self._event, condition, self._room_member_count)
        elif condition["kind"] == "sender_notification_permission":
            return _sender_notification_permission(
                self._event, condition, self._sender_power_level, self._power_levels
            )
        else:
            return True

    def _event_match(self, condition: dict, user_id: str) -> bool:
        pattern = condition.get("pattern", None)

        if not pattern:
            pattern_type = condition.get("pattern_type", None)
            if pattern_type == "user_id":
                pattern = user_id
            elif pattern_type == "user_localpart":
                pattern = UserID.from_string(user_id).localpart

        if not pattern:
            logger.warning("event_match condition with no pattern")
            return False

        # XXX: optimisation: cache our pattern regexps
        if condition["key"] == "content.body":
            body = self._event.content.get("body", None)
            if not body or not isinstance(body, str):
                return False

            return _glob_matches(pattern, body, word_boundary=True)
        else:
            haystack = self._get_value(condition["key"])
            if haystack is None:
                return False

            return _glob_matches(pattern, haystack)

    def _contains_display_name(self, display_name: str) -> bool:
        if not display_name:
            return False

        body = self._event.content.get("body", None)
        if not body or not isinstance(body, str):
            return False

        # Similar to _glob_matches, but do not treat display_name as a glob.
        r = regex_cache.get((display_name, False, True), None)
        if not r:
            r = re.escape(display_name)
            r = _re_word_boundary(r)
            r = re.compile(r, flags=re.IGNORECASE)
            regex_cache[(display_name, False, True)] = r

        return r.search(body)

    def _get_value(self, dotted_key: str) -> str:
        return self._value_cache.get(dotted_key, None)


# Caches (string, is_glob, word_boundary) -> regex for push. See _glob_matches
regex_cache = LruCache(50000)
register_cache("cache", "regex_push_cache", regex_cache)


def _glob_matches(glob: str, value: str, word_boundary: bool = False) -> bool:
    """Tests if value matches glob.

    Args:
        glob
        value: String to test against glob.
        word_boundary: Whether to match against word boundaries or entire
            string. Defaults to False.
    """

    try:
        r = regex_cache.get((glob, True, word_boundary), None)
        if not r:
            r = _glob_to_re(glob, word_boundary)
            regex_cache[(glob, True, word_boundary)] = r
        return r.search(value)
    except re.error:
        logger.warning("Failed to parse glob to regex: %r", glob)
        return False


def _glob_to_re(glob: str, word_boundary: bool) -> Pattern:
    """Generates regex for a given glob.

    Args:
        glob
        word_boundary: Whether to match against word boundaries or entire string.
    """
    if IS_GLOB.search(glob):
        r = re.escape(glob)

        r = r.replace(r"\*", ".*?")
        r = r.replace(r"\?", ".")

        # handle [abc], [a-z] and [!a-z] style ranges.
        r = GLOB_REGEX.sub(
            lambda x: (
                "[%s%s]" % (x.group(1) and "^" or "", x.group(2).replace(r"\\\-", "-"))
            ),
            r,
        )
        if word_boundary:
            r = _re_word_boundary(r)

            return re.compile(r, flags=re.IGNORECASE)
        else:
            r = "^" + r + "$"

            return re.compile(r, flags=re.IGNORECASE)
    elif word_boundary:
        r = re.escape(glob)
        r = _re_word_boundary(r)

        return re.compile(r, flags=re.IGNORECASE)
    else:
        r = "^" + re.escape(glob) + "$"
        return re.compile(r, flags=re.IGNORECASE)


def _re_word_boundary(r: str) -> str:
    """
    Adds word boundary characters to the start and end of an
    expression to require that the match occur as a whole word,
    but do so respecting the fact that strings starting or ending
    with non-word characters will change word boundaries.
    """
    # we can't use \b as it chokes on unicode. however \W seems to be okay
    # as shorthand for [^0-9A-Za-z_].
    return r"(^|\W)%s(\W|$)" % (r,)


def _flatten_dict(d, prefix=[], result=None):
    if result is None:
        result = {}
    for key, value in d.items():
        if isinstance(value, str):
            result[".".join(prefix + [key])] = value.lower()
        elif hasattr(value, "items"):
            _flatten_dict(value, prefix=(prefix + [key]), result=result)

    return result