synapse/synapse/handlers/appservice.py
Andrew Morgan 52c43d91f5 Add a new ephemeral AS handler for to_device message edus
Here we add the ability for the application service ephemeral message
processor to handle new events on the "to_device" stream.

We keep track of a stream id (token) per application service, and every
time a new to-device message comes in, for each appservice we pull the
messages between the last-recorded and current stream id and check
whether any of the messages are for a user in that appservice's user
namespace.

get_new_messages is implemented in the next commit.
2021-11-10 16:10:41 +00:00

718 lines
29 KiB
Python

# Copyright 2015, 2016 OpenMarket 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
from typing import (
TYPE_CHECKING,
Collection,
Dict,
Iterable,
List,
Optional,
Tuple,
Union,
)
from prometheus_client import Counter
from twisted.internet import defer
import synapse
from synapse.api.constants import EventTypes
from synapse.appservice import ApplicationService
from synapse.events import EventBase
from synapse.handlers.presence import format_user_presence_state
from synapse.logging.context import make_deferred_yieldable, run_in_background
from synapse.metrics import (
event_processing_loop_counter,
event_processing_loop_room_count,
)
from synapse.metrics.background_process_metrics import (
run_as_background_process,
wrap_as_background_process,
)
from synapse.storage.databases.main.directory import RoomAliasMapping
from synapse.types import JsonDict, RoomAlias, RoomStreamToken, UserID
from synapse.util.async_helpers import Linearizer
from synapse.util.metrics import Measure
if TYPE_CHECKING:
from synapse.server import HomeServer
logger = logging.getLogger(__name__)
MAX_TO_DEVICE_MESSAGES_PER_AS_TRANSACTION = 100
events_processed_counter = Counter("synapse_handlers_appservice_events_processed", "")
class ApplicationServicesHandler:
def __init__(self, hs: "HomeServer"):
self.store = hs.get_datastore()
self.is_mine_id = hs.is_mine_id
self.appservice_api = hs.get_application_service_api()
self.scheduler = hs.get_application_service_scheduler()
self.started_scheduler = False
self.clock = hs.get_clock()
self.notify_appservices = hs.config.appservice.notify_appservices
self.event_sources = hs.get_event_sources()
self.msc2409_to_device_messages_enabled = (
hs.config.experimental.msc2409_to_device_messages_enabled
)
self.current_max = 0
self.is_processing = False
self._ephemeral_events_linearizer = Linearizer(
name="appservice_ephemeral_events"
)
def notify_interested_services(self, max_token: RoomStreamToken) -> None:
"""Notifies (pushes) all application services interested in this event.
Pushing is done asynchronously, so this method won't block for any
prolonged length of time.
"""
# We just use the minimum stream ordering and ignore the vector clock
# component. This is safe to do as long as we *always* ignore the vector
# clock components.
current_id = max_token.stream
services = self.store.get_app_services()
if not services or not self.notify_appservices:
return
self.current_max = max(self.current_max, current_id)
if self.is_processing:
return
# We only start a new background process if necessary rather than
# optimistically (to cut down on overhead).
self._notify_interested_services(max_token)
@wrap_as_background_process("notify_interested_services")
async def _notify_interested_services(self, max_token: RoomStreamToken) -> None:
with Measure(self.clock, "notify_interested_services"):
self.is_processing = True
try:
limit = 100
upper_bound = -1
while upper_bound < self.current_max:
(
upper_bound,
events,
) = await self.store.get_new_events_for_appservice(
self.current_max, limit
)
events_by_room: Dict[str, List[EventBase]] = {}
for event in events:
events_by_room.setdefault(event.room_id, []).append(event)
async def handle_event(event: EventBase) -> None:
# Gather interested services
services = await self._get_services_for_event(event)
if len(services) == 0:
return # no services need notifying
# Do we know this user exists? If not, poke the user
# query API for all services which match that user regex.
# This needs to block as these user queries need to be
# made BEFORE pushing the event.
await self._check_user_exists(event.sender)
if event.type == EventTypes.Member:
await self._check_user_exists(event.state_key)
if not self.started_scheduler:
async def start_scheduler() -> None:
try:
await self.scheduler.start()
except Exception:
logger.error("Application Services Failure")
run_as_background_process("as_scheduler", start_scheduler)
self.started_scheduler = True
# Fork off pushes to these services
for service in services:
self.scheduler.submit_event_for_as(service, event)
now = self.clock.time_msec()
ts = await self.store.get_received_ts(event.event_id)
assert ts is not None
synapse.metrics.event_processing_lag_by_event.labels(
"appservice_sender"
).observe((now - ts) / 1000)
async def handle_room_events(events: Iterable[EventBase]) -> None:
for event in events:
await handle_event(event)
await make_deferred_yieldable(
defer.gatherResults(
[
run_in_background(handle_room_events, evs)
for evs in events_by_room.values()
],
consumeErrors=True,
)
)
await self.store.set_appservice_last_pos(upper_bound)
synapse.metrics.event_processing_positions.labels(
"appservice_sender"
).set(upper_bound)
events_processed_counter.inc(len(events))
event_processing_loop_room_count.labels("appservice_sender").inc(
len(events_by_room)
)
event_processing_loop_counter.labels("appservice_sender").inc()
if events:
now = self.clock.time_msec()
ts = await self.store.get_received_ts(events[-1].event_id)
assert ts is not None
synapse.metrics.event_processing_lag.labels(
"appservice_sender"
).set(now - ts)
synapse.metrics.event_processing_last_ts.labels(
"appservice_sender"
).set(ts)
finally:
self.is_processing = False
def notify_interested_services_ephemeral(
self,
stream_key: str,
new_token: Union[int, RoomStreamToken],
users: Collection[Union[str, UserID]],
) -> None:
"""
This is called by the notifier in the background when an ephemeral event is handled
by the homeserver.
This will determine which appservices are interested in the event, and submit them.
Args:
stream_key: The stream the event came from.
`stream_key` can be "typing_key", "receipt_key", "presence_key" or
"to_device_key". Any other value for `stream_key` will cause this function
to return early.
Ephemeral events will only be pushed to appservices that have opted into
receiving them by setting `push_ephemeral` to true in their registration
file. Note that while MSC2409 is experimental, this option is called
`de.sorunome.msc2409.push_ephemeral`.
Appservices will only receive ephemeral events that fall within their
registered user and room namespaces.
new_token: The stream token of the event.
users: The users that should be informed of the new event, if any.
"""
if not self.notify_appservices:
return
# Assert that new_token is an integer (and not a RoomStreamToken).
# All of the supported streams that this function handles use an
# integer to track progress (rather than a RoomStreamToken - a
# vector clock implementation) as they don't support multiple
# stream writers.
#
# As a result, we simply assert that new_token is an integer.
# If we do end up needing to pass a RoomStreamToken down here
# in the future, using RoomStreamToken.stream (the minimum stream
# position) to convert to an ascending integer value should work.
# Additional context: https://github.com/matrix-org/synapse/pull/11137
assert isinstance(new_token, int)
# Ignore to-device messages if the feature flag is not enabled
if (
stream_key == "to_device_key"
and not self.msc2409_to_device_messages_enabled
):
return
# Check whether there are any appservices which have registered to receive
# ephemeral events.
#
# Note that whether these events are actually relevant to these appservices
# is decided later on.
services = [
service
for service in self.store.get_app_services()
if service.supports_ephemeral
]
if not services:
# Bail out early if none of the target appservices have explicitly registered
# to receive these ephemeral events.
return
# We only start a new background process if necessary rather than
# optimistically (to cut down on overhead).
self._notify_interested_services_ephemeral(
services, stream_key, new_token, users
)
@wrap_as_background_process("notify_interested_services_ephemeral")
async def _notify_interested_services_ephemeral(
self,
services: List[ApplicationService],
stream_key: str,
new_token: int,
users: Collection[Union[str, UserID]],
) -> None:
logger.debug("Checking interested services for %s", stream_key)
with Measure(self.clock, "notify_interested_services_ephemeral"):
for service in services:
if stream_key == "typing_key":
# Note that we don't persist the token (via set_type_stream_id_for_appservice)
# for typing_key due to performance reasons and due to their highly
# ephemeral nature.
#
# Instead we simply grab the latest typing updates in _handle_typing
# and, if they apply to this application service, send it off.
events = await self._handle_typing(service, new_token)
if events:
self.scheduler.submit_ephemeral_events_for_as(service, events)
continue
# Since we read/update the stream position for this AS/stream
with (
await self._ephemeral_events_linearizer.queue(
(service.id, stream_key)
)
):
if stream_key == "receipt_key":
events = await self._handle_receipts(service, new_token)
if events:
self.scheduler.submit_ephemeral_events_for_as(
service, events
)
# Persist the latest handled stream token for this appservice
await self.store.set_type_stream_id_for_appservice(
service, "read_receipt", new_token
)
elif stream_key == "presence_key":
events = await self._handle_presence(service, users, new_token)
if events:
self.scheduler.submit_ephemeral_events_for_as(
service, events
)
# Persist the latest handled stream token for this appservice
await self.store.set_type_stream_id_for_appservice(
service, "presence", new_token
)
elif (
stream_key == "to_device_key"
and new_token is not None
and self.msc2409_to_device_messages_enabled
):
# Retrieve an iterable of to-device message events, as well as the
# maximum stream token of the messages we were able to retrieve.
events, max_stream_token = await self._handle_to_device(
service, new_token, users
)
if events:
self.scheduler.submit_ephemeral_events_for_as(
service, events
)
# TODO: If max_stream_token != new_token, schedule another transaction immediately,
# instead of waiting for another to-device to be sent?
# https://github.com/matrix-org/synapse/issues/11150#issuecomment-960726449
# Persist the latest handled stream token for this appservice
await self.store.set_type_stream_id_for_appservice(
service, "to_device", max_stream_token
)
async def _handle_typing(
self, service: ApplicationService, new_token: int
) -> List[JsonDict]:
"""
Return the typing events since the given stream token that the given application
service should receive.
First fetch all typing events between the given typing stream token (non-inclusive)
and the latest typing event stream token (inclusive). Then return only those typing
events that the given application service may be interested in.
Args:
service: The application service to check for which events it should receive.
new_token: A typing event stream token.
Returns:
A list of JSON dictionaries containing data derived from the typing events that
should be sent to the given application service.
"""
typing_source = self.event_sources.sources.typing
# Get the typing events from just before current
typing, _ = await typing_source.get_new_events_as(
service=service,
# For performance reasons, we don't persist the previous
# token in the DB and instead fetch the latest typing event
# for appservices.
# TODO: It'd likely be more efficient to simply fetch the
# typing event with the given 'new_token' stream token and
# check if the given service was interested, rather than
# iterating over all typing events and only grabbing the
# latest few.
from_key=new_token - 1,
)
return typing
async def _handle_receipts(
self, service: ApplicationService, new_token: Optional[int]
) -> List[JsonDict]:
"""
Return the latest read receipts that the given application service should receive.
First fetch all read receipts between the last receipt stream token that this
application service should have previously received (non-inclusive) and the
latest read receipt stream token (inclusive). Then from that set, return only
those read receipts that the given application service may be interested in.
Args:
service: The application service to check for which events it should receive.
new_token: A receipts event stream token. Purely used to double-check that the
from_token we pull from the database isn't greater than or equal to this
token. Prevents accidentally duplicating work.
Returns:
A list of JSON dictionaries containing data derived from the read receipts that
should be sent to the given application service.
"""
from_key = await self.store.get_type_stream_id_for_appservice(
service, "read_receipt"
)
if new_token is not None and new_token <= from_key:
logger.debug(
"Rejecting token lower than or equal to stored: %s" % (new_token,)
)
return []
receipts_source = self.event_sources.sources.receipt
receipts, _ = await receipts_source.get_new_events_as(
service=service, from_key=from_key
)
return receipts
async def _handle_presence(
self,
service: ApplicationService,
users: Collection[Union[str, UserID]],
new_token: Optional[int],
) -> List[JsonDict]:
"""
Return the latest presence updates that the given application service should receive.
First, filter the given users list to those that the application service is
interested in. Then retrieve the latest presence updates since the
the last-known previously received presence stream token for the given
application service. Return those presence updates.
Args:
service: The application service that ephemeral events are being sent to.
users: The users that should receive the presence update.
new_token: A presence update stream token. Purely used to double-check that the
from_token we pull from the database isn't greater than or equal to this
token. Prevents accidentally duplicating work.
Returns:
A list of json dictionaries containing data derived from the presence events
that should be sent to the given application service.
"""
events: List[JsonDict] = []
presence_source = self.event_sources.sources.presence
from_key = await self.store.get_type_stream_id_for_appservice(
service, "presence"
)
if new_token is not None and new_token <= from_key:
logger.debug(
"Rejecting token lower than or equal to stored: %s" % (new_token,)
)
return []
for user in users:
if isinstance(user, str):
user = UserID.from_string(user)
interested = await service.is_interested_in_presence(user, self.store)
if not interested:
continue
presence_events, _ = await presence_source.get_new_events(
user=user,
from_key=from_key,
)
time_now = self.clock.time_msec()
events.extend(
{
"type": "m.presence",
"sender": event.user_id,
"content": format_user_presence_state(
event, time_now, include_user_id=False
),
}
for event in presence_events
)
return events
async def _handle_to_device(
self,
service: ApplicationService,
new_token: int,
users: Collection[Union[str, UserID]],
) -> Tuple[List[JsonDict], int]:
"""
Given an application service, determine which events it should receive
from those between the last-recorded typing event stream token for this
appservice and the given stream token.
Args:
service: The application service to check for which events it should receive.
new_token: The latest to-device event stream token.
users: The users that should receive new to-device messages.
Returns:
A two-tuple containing the following:
* A list of JSON dictionaries containing data derived from the typing events
that should be sent to the given application service.
* The last-processed to-device message's stream id. If this does not equal
`new_token` then there are likely more events to send.
"""
# Get the stream token that this application service has processed up until
from_key = await self.store.get_type_stream_id_for_appservice(
service, "to_device"
)
# Filter out users that this appservice is not interested in
users_appservice_is_interested_in: List[str] = []
for user in users:
if isinstance(user, UserID):
user = user.to_string()
if service.is_interested_in_user(user):
users_appservice_is_interested_in.append(user)
if not users_appservice_is_interested_in:
# Return early if the AS was not interested in any of these users
return [], new_token
# Retrieve the to-device messages for each user
(
recipient_user_id_device_id_to_messages,
# Record the maximum stream token of the retrieved messages that
# this function was able to pull before hitting the max to-device
# message count limit.
# We return this value later, to ensure we only record the stream
# token we managed to get up to, so that the rest can be sent later.
max_stream_token,
) = await self.store.get_new_messages(
users_appservice_is_interested_in,
from_key,
new_token,
limit=MAX_TO_DEVICE_MESSAGES_PER_AS_TRANSACTION,
)
# According to MSC2409, we'll need to add 'to_user_id' and 'to_device_id' fields
# to the event JSON so that the application service will know which user/device
# combination this messages was intended for.
#
# So we mangle this dict into a flat list of to-device messages with the relevant
# user ID and device ID embedded inside each message dict.
message_payload: List[JsonDict] = []
for (
user_id,
device_id,
), messages in recipient_user_id_device_id_to_messages.items():
for message_json in messages:
# Remove 'message_id' from the to-device message, as it's an internal ID
message_json.pop("message_id", None)
message_payload.append(
{
"to_user_id": user_id,
"to_device_id": device_id,
**message_json,
}
)
return message_payload, max_stream_token
async def query_user_exists(self, user_id: str) -> bool:
"""Check if any application service knows this user_id exists.
Args:
user_id: The user to query if they exist on any AS.
Returns:
True if this user exists on at least one application service.
"""
user_query_services = self._get_services_for_user(user_id=user_id)
for user_service in user_query_services:
is_known_user = await self.appservice_api.query_user(user_service, user_id)
if is_known_user:
return True
return False
async def query_room_alias_exists(
self, room_alias: RoomAlias
) -> Optional[RoomAliasMapping]:
"""Check if an application service knows this room alias exists.
Args:
room_alias: The room alias to query.
Returns:
namedtuple: with keys "room_id" and "servers" or None if no
association can be found.
"""
room_alias_str = room_alias.to_string()
services = self.store.get_app_services()
alias_query_services = [
s for s in services if (s.is_interested_in_alias(room_alias_str))
]
for alias_service in alias_query_services:
is_known_alias = await self.appservice_api.query_alias(
alias_service, room_alias_str
)
if is_known_alias:
# the alias exists now so don't query more ASes.
return await self.store.get_association_from_room_alias(room_alias)
return None
async def query_3pe(
self, kind: str, protocol: str, fields: Dict[bytes, List[bytes]]
) -> List[JsonDict]:
services = self._get_services_for_3pn(protocol)
results = await make_deferred_yieldable(
defer.DeferredList(
[
run_in_background(
self.appservice_api.query_3pe, service, kind, protocol, fields
)
for service in services
],
consumeErrors=True,
)
)
ret = []
for (success, result) in results:
if success:
ret.extend(result)
return ret
async def get_3pe_protocols(
self, only_protocol: Optional[str] = None
) -> Dict[str, JsonDict]:
services = self.store.get_app_services()
protocols: Dict[str, List[JsonDict]] = {}
# Collect up all the individual protocol responses out of the ASes
for s in services:
for p in s.protocols:
if only_protocol is not None and p != only_protocol:
continue
if p not in protocols:
protocols[p] = []
info = await self.appservice_api.get_3pe_protocol(s, p)
if info is not None:
protocols[p].append(info)
def _merge_instances(infos: List[JsonDict]) -> JsonDict:
# Merge the 'instances' lists of multiple results, but just take
# the other fields from the first as they ought to be identical
# copy the result so as not to corrupt the cached one
combined = dict(infos[0])
combined["instances"] = list(combined["instances"])
for info in infos[1:]:
combined["instances"].extend(info["instances"])
return combined
return {
p: _merge_instances(protocols[p]) for p in protocols.keys() if protocols[p]
}
async def _get_services_for_event(
self, event: EventBase
) -> List[ApplicationService]:
"""Retrieve a list of application services interested in this event.
Args:
event: The event to check. Can be None if alias_list is not.
Returns:
A list of services interested in this event based on the service regex.
"""
services = self.store.get_app_services()
# we can't use a list comprehension here. Since python 3, list
# comprehensions use a generator internally. This means you can't yield
# inside of a list comprehension anymore.
interested_list = []
for s in services:
if await s.is_interested(event, self.store):
interested_list.append(s)
return interested_list
def _get_services_for_user(self, user_id: str) -> List[ApplicationService]:
services = self.store.get_app_services()
return [s for s in services if (s.is_interested_in_user(user_id))]
def _get_services_for_3pn(self, protocol: str) -> List[ApplicationService]:
services = self.store.get_app_services()
return [s for s in services if s.is_interested_in_protocol(protocol)]
async def _is_unknown_user(self, user_id: str) -> bool:
if not self.is_mine_id(user_id):
# we don't know if they are unknown or not since it isn't one of our
# users. We can't poke ASes.
return False
user_info = await self.store.get_user_by_id(user_id)
if user_info:
return False
# user not found; could be the AS though, so check.
services = self.store.get_app_services()
service_list = [s for s in services if s.sender == user_id]
return len(service_list) == 0
async def _check_user_exists(self, user_id: str) -> bool:
unknown_user = await self._is_unknown_user(user_id)
if unknown_user:
return await self.query_user_exists(user_id)
return True