summary refs log tree commit diff
path: root/synapse/replication/tcp/streams/_base.py
blob: a8d568b14a86ff564939a411e39df61088122ce4 (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
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
# -*- coding: utf-8 -*-
# Copyright 2017 Vector Creations Ltd
# Copyright 2019 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 itertools
import logging
from collections import namedtuple
from typing import Any, List, Optional

import attr

logger = logging.getLogger(__name__)


MAX_EVENTS_BEHIND = 10000

BackfillStreamRow = namedtuple(
    "BackfillStreamRow",
    (
        "event_id",  # str
        "room_id",  # str
        "type",  # str
        "state_key",  # str, optional
        "redacts",  # str, optional
        "relates_to",  # str, optional
    ),
)
PresenceStreamRow = namedtuple(
    "PresenceStreamRow",
    (
        "user_id",  # str
        "state",  # str
        "last_active_ts",  # int
        "last_federation_update_ts",  # int
        "last_user_sync_ts",  # int
        "status_msg",  # str
        "currently_active",  # bool
    ),
)
TypingStreamRow = namedtuple(
    "TypingStreamRow", ("room_id", "user_ids")  # str  # list(str)
)
ReceiptsStreamRow = namedtuple(
    "ReceiptsStreamRow",
    (
        "room_id",  # str
        "receipt_type",  # str
        "user_id",  # str
        "event_id",  # str
        "data",  # dict
    ),
)
PushRulesStreamRow = namedtuple("PushRulesStreamRow", ("user_id",))  # str
PushersStreamRow = namedtuple(
    "PushersStreamRow",
    ("user_id", "app_id", "pushkey", "deleted"),  # str  # str  # str  # bool
)


@attr.s
class CachesStreamRow:
    """Stream to inform workers they should invalidate their cache.

    Attributes:
        cache_func: Name of the cached function.
        keys: The entry in the cache to invalidate. If None then will
            invalidate all.
        invalidation_ts: Timestamp of when the invalidation took place.
    """

    cache_func = attr.ib(type=str)
    keys = attr.ib(type=Optional[List[Any]])
    invalidation_ts = attr.ib(type=int)


PublicRoomsStreamRow = namedtuple(
    "PublicRoomsStreamRow",
    (
        "room_id",  # str
        "visibility",  # str
        "appservice_id",  # str, optional
        "network_id",  # str, optional
    ),
)
DeviceListsStreamRow = namedtuple(
    "DeviceListsStreamRow", ("user_id", "destination")  # str  # str
)
ToDeviceStreamRow = namedtuple("ToDeviceStreamRow", ("entity",))  # str
TagAccountDataStreamRow = namedtuple(
    "TagAccountDataStreamRow", ("user_id", "room_id", "data")  # str  # str  # dict
)
AccountDataStreamRow = namedtuple(
    "AccountDataStream", ("user_id", "room_id", "data_type")  # str  # str  # str
)
GroupsStreamRow = namedtuple(
    "GroupsStreamRow",
    ("group_id", "user_id", "type", "content"),  # str  # str  # str  # dict
)
UserSignatureStreamRow = namedtuple("UserSignatureStreamRow", ("user_id"))  # str


class Stream(object):
    """Base class for the streams.

    Provides a `get_updates()` function that returns new updates since the last
    time it was called up until the point `advance_current_token` was called.
    """

    NAME = None  # type: str  # The name of the stream
    # The type of the row. Used by the default impl of parse_row.
    ROW_TYPE = None  # type: Any
    _LIMITED = True  # Whether the update function takes a limit

    @classmethod
    def parse_row(cls, row):
        """Parse a row received over replication

        By default, assumes that the row data is an array object and passes its contents
        to the constructor of the ROW_TYPE for this stream.

        Args:
            row: row data from the incoming RDATA command, after json decoding

        Returns:
            ROW_TYPE object for this stream
        """
        return cls.ROW_TYPE(*row)

    def __init__(self, hs):
        # The token from which we last asked for updates
        self.last_token = self.current_token()

        # The token that we will get updates up to
        self.upto_token = self.current_token()

    def advance_current_token(self):
        """Updates `upto_token` to "now", which updates up until which point
        get_updates[_since] will fetch rows till.
        """
        self.upto_token = self.current_token()

    def discard_updates_and_advance(self):
        """Called when the stream should advance but the updates would be discarded,
        e.g. when there are no currently connected workers.
        """
        self.upto_token = self.current_token()
        self.last_token = self.upto_token

    async def get_updates(self):
        """Gets all updates since the last time this function was called (or
        since the stream was constructed if it hadn't been called before),
        until the `upto_token`

        Returns:
            Deferred[Tuple[List[Tuple[int, Any]], int]:
                Resolves to a pair ``(updates, current_token)``, where ``updates`` is a
                list of ``(token, row)`` entries. ``row`` will be json-serialised and
                sent over the replication steam.
        """
        updates, current_token = await self.get_updates_since(self.last_token)
        self.last_token = current_token

        return updates, current_token

    async def get_updates_since(self, from_token):
        """Like get_updates except allows specifying from when we should
        stream updates

        Returns:
            Deferred[Tuple[List[Tuple[int, Any]], int]:
                Resolves to a pair ``(updates, current_token)``, where ``updates`` is a
                list of ``(token, row)`` entries. ``row`` will be json-serialised and
                sent over the replication steam.
        """
        if from_token in ("NOW", "now"):
            return [], self.upto_token

        current_token = self.upto_token

        from_token = int(from_token)

        if from_token == current_token:
            return [], current_token

        logger.info("get_updates_since: %s", self.__class__)
        if self._LIMITED:
            rows = await self.update_function(
                from_token, current_token, limit=MAX_EVENTS_BEHIND + 1
            )

            # never turn more than MAX_EVENTS_BEHIND + 1 into updates.
            rows = itertools.islice(rows, MAX_EVENTS_BEHIND + 1)
        else:
            rows = await self.update_function(from_token, current_token)

        updates = [(row[0], row[1:]) for row in rows]

        # check we didn't get more rows than the limit.
        # doing it like this allows the update_function to be a generator.
        if self._LIMITED and len(updates) >= MAX_EVENTS_BEHIND:
            raise Exception("stream %s has fallen behind" % (self.NAME))

        return updates, current_token

    def current_token(self):
        """Gets the current token of the underlying streams. Should be provided
        by the sub classes

        Returns:
            int
        """
        raise NotImplementedError()

    def update_function(self, from_token, current_token, limit=None):
        """Get updates between from_token and to_token. If Stream._LIMITED is
        True then limit is provided, otherwise it's not.

        Returns:
            Deferred(list(tuple)): the first entry in the tuple is the token for
                that update, and the rest of the tuple gets used to construct
                a ``ROW_TYPE`` instance
        """
        raise NotImplementedError()


class BackfillStream(Stream):
    """We fetched some old events and either we had never seen that event before
    or it went from being an outlier to not.
    """

    NAME = "backfill"
    ROW_TYPE = BackfillStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()
        self.current_token = store.get_current_backfill_token  # type: ignore
        self.update_function = store.get_all_new_backfill_event_rows  # type: ignore

        super(BackfillStream, self).__init__(hs)


class PresenceStream(Stream):
    NAME = "presence"
    _LIMITED = False
    ROW_TYPE = PresenceStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()
        presence_handler = hs.get_presence_handler()

        self.current_token = store.get_current_presence_token  # type: ignore
        self.update_function = presence_handler.get_all_presence_updates  # type: ignore

        super(PresenceStream, self).__init__(hs)


class TypingStream(Stream):
    NAME = "typing"
    _LIMITED = False
    ROW_TYPE = TypingStreamRow

    def __init__(self, hs):
        typing_handler = hs.get_typing_handler()

        self.current_token = typing_handler.get_current_token  # type: ignore
        self.update_function = typing_handler.get_all_typing_updates  # type: ignore

        super(TypingStream, self).__init__(hs)


class ReceiptsStream(Stream):
    NAME = "receipts"
    ROW_TYPE = ReceiptsStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()

        self.current_token = store.get_max_receipt_stream_id  # type: ignore
        self.update_function = store.get_all_updated_receipts  # type: ignore

        super(ReceiptsStream, self).__init__(hs)


class PushRulesStream(Stream):
    """A user has changed their push rules
    """

    NAME = "push_rules"
    ROW_TYPE = PushRulesStreamRow

    def __init__(self, hs):
        self.store = hs.get_datastore()
        super(PushRulesStream, self).__init__(hs)

    def current_token(self):
        push_rules_token, _ = self.store.get_push_rules_stream_token()
        return push_rules_token

    async def update_function(self, from_token, to_token, limit):
        rows = await self.store.get_all_push_rule_updates(from_token, to_token, limit)
        return [(row[0], row[2]) for row in rows]


class PushersStream(Stream):
    """A user has added/changed/removed a pusher
    """

    NAME = "pushers"
    ROW_TYPE = PushersStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()

        self.current_token = store.get_pushers_stream_token  # type: ignore
        self.update_function = store.get_all_updated_pushers_rows  # type: ignore

        super(PushersStream, self).__init__(hs)


class CachesStream(Stream):
    """A cache was invalidated on the master and no other stream would invalidate
    the cache on the workers
    """

    NAME = "caches"
    ROW_TYPE = CachesStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()

        self.current_token = store.get_cache_stream_token  # type: ignore
        self.update_function = store.get_all_updated_caches  # type: ignore

        super(CachesStream, self).__init__(hs)


class PublicRoomsStream(Stream):
    """The public rooms list changed
    """

    NAME = "public_rooms"
    ROW_TYPE = PublicRoomsStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()

        self.current_token = store.get_current_public_room_stream_id  # type: ignore
        self.update_function = store.get_all_new_public_rooms  # type: ignore

        super(PublicRoomsStream, self).__init__(hs)


class DeviceListsStream(Stream):
    """Someone added/changed/removed a device
    """

    NAME = "device_lists"
    _LIMITED = False
    ROW_TYPE = DeviceListsStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()

        self.current_token = store.get_device_stream_token  # type: ignore
        self.update_function = store.get_all_device_list_changes_for_remotes  # type: ignore

        super(DeviceListsStream, self).__init__(hs)


class ToDeviceStream(Stream):
    """New to_device messages for a client
    """

    NAME = "to_device"
    ROW_TYPE = ToDeviceStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()

        self.current_token = store.get_to_device_stream_token  # type: ignore
        self.update_function = store.get_all_new_device_messages  # type: ignore

        super(ToDeviceStream, self).__init__(hs)


class TagAccountDataStream(Stream):
    """Someone added/removed a tag for a room
    """

    NAME = "tag_account_data"
    ROW_TYPE = TagAccountDataStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()

        self.current_token = store.get_max_account_data_stream_id  # type: ignore
        self.update_function = store.get_all_updated_tags  # type: ignore

        super(TagAccountDataStream, self).__init__(hs)


class AccountDataStream(Stream):
    """Global or per room account data was changed
    """

    NAME = "account_data"
    ROW_TYPE = AccountDataStreamRow

    def __init__(self, hs):
        self.store = hs.get_datastore()

        self.current_token = self.store.get_max_account_data_stream_id  # type: ignore

        super(AccountDataStream, self).__init__(hs)

    async def update_function(self, from_token, to_token, limit):
        global_results, room_results = await self.store.get_all_updated_account_data(
            from_token, from_token, to_token, limit
        )

        results = list(room_results)
        results.extend(
            (stream_id, user_id, None, account_data_type)
            for stream_id, user_id, account_data_type in global_results
        )

        return results


class GroupServerStream(Stream):
    NAME = "groups"
    ROW_TYPE = GroupsStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()

        self.current_token = store.get_group_stream_token  # type: ignore
        self.update_function = store.get_all_groups_changes  # type: ignore

        super(GroupServerStream, self).__init__(hs)


class UserSignatureStream(Stream):
    """A user has signed their own device with their user-signing key
    """

    NAME = "user_signature"
    _LIMITED = False
    ROW_TYPE = UserSignatureStreamRow

    def __init__(self, hs):
        store = hs.get_datastore()

        self.current_token = store.get_device_stream_token  # type: ignore
        self.update_function = store.get_all_user_signature_changes_for_remotes  # type: ignore

        super(UserSignatureStream, self).__init__(hs)