Add `is_dm` filtering to Sliding Sync `/sync` (#17277)

Based on [MSC3575](https://github.com/matrix-org/matrix-spec-proposals/pull/3575): Sliding Sync

5 files changed, 416 insertions, 7 deletions

diff --git a/changelog.d/17277.feature b/changelog.d/17277.feature
new file mode 100644
index 0000000000..5c16342c11
--- /dev/null
+++ b/changelog.d/17277.feature
@@ -0,0 +1 @@
+Add `is_dm` filtering to experimental [MSC3575](https://github.com/matrix-org/matrix-spec-proposals/pull/3575) Sliding Sync `/sync` endpoint.
diff --git a/synapse/handlers/sliding_sync.py b/synapse/handlers/sliding_sync.py
index de4f33abb8..78fb66d6e2 100644
--- a/synapse/handlers/sliding_sync.py
+++ b/synapse/handlers/sliding_sync.py
@@ -22,7 +22,7 @@ from typing import TYPE_CHECKING, AbstractSet, Dict, List, Optional
 
 from immutabledict import immutabledict
 
-from synapse.api.constants import Membership
+from synapse.api.constants import AccountDataTypes, Membership
 from synapse.events import EventBase
 from synapse.types import Requester, RoomStreamToken, StreamToken, UserID
 from synapse.types.handlers import OperationType, SlidingSyncConfig, SlidingSyncResult
@@ -69,9 +69,19 @@ class SlidingSyncHandler:
         from_token: Optional[StreamToken] = None,
         timeout_ms: int = 0,
     ) -> SlidingSyncResult:
-        """Get the sync for a client if we have new data for it now. Otherwise
+        """
+        Get the sync for a client if we have new data for it now. Otherwise
         wait for new data to arrive on the server. If the timeout expires, then
         return an empty sync result.
+
+        Args:
+            requester: The user making the request
+            sync_config: Sync configuration
+            from_token: The point in the stream to sync from. Token of the end of the
+                previous batch. May be `None` if this is the initial sync request.
+            timeout_ms: The time in milliseconds to wait for new data to arrive. If 0,
+                we will immediately but there might not be any new data so we just return an
+                empty response.
         """
         # If the user is not part of the mau group, then check that limits have
         # not been exceeded (if not part of the group by this point, almost certain
@@ -143,6 +153,14 @@ class SlidingSyncHandler:
         """
         Generates the response body of a Sliding Sync result, represented as a
         `SlidingSyncResult`.
+
+        We fetch data according to the token range (> `from_token` and <= `to_token`).
+
+        Args:
+            sync_config: Sync configuration
+            to_token: The point in the stream to sync up to.
+            from_token: The point in the stream to sync from. Token of the end of the
+                previous batch. May be `None` if this is the initial sync request.
         """
         user_id = sync_config.user.to_string()
         app_service = self.store.get_app_service_by_user_id(user_id)
@@ -163,11 +181,12 @@ class SlidingSyncHandler:
         lists: Dict[str, SlidingSyncResult.SlidingWindowList] = {}
         if sync_config.lists:
             for list_key, list_config in sync_config.lists.items():
-                # TODO: Apply filters
-                #
-                # TODO: Exclude partially stated rooms unless the `required_state` has
-                # `["m.room.member", "$LAZY"]`
+                # Apply filters
                 filtered_room_ids = room_id_set
+                if list_config.filters is not None:
+                    filtered_room_ids = await self.filter_rooms(
+                        sync_config.user, room_id_set, list_config.filters, to_token
+                    )
                 # TODO: Apply sorts
                 sorted_room_ids = sorted(filtered_room_ids)
 
@@ -217,6 +236,12 @@ class SlidingSyncHandler:
           `forgotten` flag to the `room_memberships` table in Synapse. There isn't a way
           to tell when a room was forgotten at the moment so we can't factor it into the
           from/to range.
+
+
+        Args:
+            user: User to fetch rooms for
+            to_token: The token to fetch rooms up to.
+            from_token: The point in the stream to sync from.
         """
         user_id = user.to_string()
 
@@ -439,3 +464,84 @@ class SlidingSyncHandler:
                 sync_room_id_set.add(room_id)
 
         return sync_room_id_set
+
+    async def filter_rooms(
+        self,
+        user: UserID,
+        room_id_set: AbstractSet[str],
+        filters: SlidingSyncConfig.SlidingSyncList.Filters,
+        to_token: StreamToken,
+    ) -> AbstractSet[str]:
+        """
+        Filter rooms based on the sync request.
+
+        Args:
+            user: User to filter rooms for
+            room_id_set: Set of room IDs to filter down
+            filters: Filters to apply
+            to_token: We filter based on the state of the room at this token
+        """
+        user_id = user.to_string()
+
+        # TODO: Apply filters
+        #
+        # TODO: Exclude partially stated rooms unless the `required_state` has
+        # `["m.room.member", "$LAZY"]`
+
+        filtered_room_id_set = set(room_id_set)
+
+        # Filter for Direct-Message (DM) rooms
+        if filters.is_dm is not None:
+            # We're using global account data (`m.direct`) instead of checking for
+            # `is_direct` on membership events because that property only appears for
+            # the invitee membership event (doesn't show up for the inviter). Account
+            # data is set by the client so it needs to be scrutinized.
+            #
+            # We're unable to take `to_token` into account for global account data since
+            # we only keep track of the latest account data for the user.
+            dm_map = await self.store.get_global_account_data_by_type_for_user(
+                user_id, AccountDataTypes.DIRECT
+            )
+
+            # Flatten out the map
+            dm_room_id_set = set()
+            if dm_map:
+                for room_ids in dm_map.values():
+                    # Account data should be a list of room IDs. Ignore anything else
+                    if isinstance(room_ids, list):
+                        for room_id in room_ids:
+                            if isinstance(room_id, str):
+                                dm_room_id_set.add(room_id)
+
+            if filters.is_dm:
+                # Only DM rooms please
+                filtered_room_id_set = filtered_room_id_set.intersection(dm_room_id_set)
+            else:
+                # Only non-DM rooms please
+                filtered_room_id_set = filtered_room_id_set.difference(dm_room_id_set)
+
+        if filters.spaces:
+            raise NotImplementedError()
+
+        if filters.is_encrypted:
+            raise NotImplementedError()
+
+        if filters.is_invite:
+            raise NotImplementedError()
+
+        if filters.room_types:
+            raise NotImplementedError()
+
+        if filters.not_room_types:
+            raise NotImplementedError()
+
+        if filters.room_name_like:
+            raise NotImplementedError()
+
+        if filters.tags:
+            raise NotImplementedError()
+
+        if filters.not_tags:
+            raise NotImplementedError()
+
+        return filtered_room_id_set
diff --git a/synapse/types/rest/client/__init__.py b/synapse/types/rest/client/__init__.py
index ef261518a0..ec83d0daa6 100644
--- a/synapse/types/rest/client/__init__.py
+++ b/synapse/types/rest/client/__init__.py
@@ -238,6 +238,53 @@ class SlidingSyncBody(RequestBodyModel):
         """
 
         class Filters(RequestBodyModel):
+            """
+            All fields are applied with AND operators, hence if `is_dm: True` and
+            `is_encrypted: True` then only Encrypted DM rooms will be returned. The
+            absence of fields implies no filter on that criteria: it does NOT imply
+            `False`. These fields may be expanded through use of extensions.
+
+            Attributes:
+                is_dm: Flag which only returns rooms present (or not) in the DM section
+                    of account data. If unset, both DM rooms and non-DM rooms are returned.
+                    If False, only non-DM rooms are returned. If True, only DM rooms are
+                    returned.
+                spaces: Filter the room based on the space they belong to according to
+                    `m.space.child` state events. If multiple spaces are present, a room can
+                    be part of any one of the listed spaces (OR'd). The server will inspect
+                    the `m.space.child` state events for the JOINED space room IDs given.
+                    Servers MUST NOT navigate subspaces. It is up to the client to give a
+                    complete list of spaces to navigate. Only rooms directly mentioned as
+                    `m.space.child` events in these spaces will be returned. Unknown spaces
+                    or spaces the user is not joined to will be ignored.
+                is_encrypted: Flag which only returns rooms which have an
+                    `m.room.encryption` state event. If unset, both encrypted and
+                    unencrypted rooms are returned. If `False`, only unencrypted rooms are
+                    returned. If `True`, only encrypted rooms are returned.
+                is_invite: Flag which only returns rooms the user is currently invited
+                    to. If unset, both invited and joined rooms are returned. If `False`, no
+                    invited rooms are returned. If `True`, only invited rooms are returned.
+                room_types: If specified, only rooms where the `m.room.create` event has
+                    a `type` matching one of the strings in this array will be returned. If
+                    this field is unset, all rooms are returned regardless of type. This can
+                    be used to get the initial set of spaces for an account. For rooms which
+                    do not have a room type, use `null`/`None` to include them.
+                not_room_types: Same as `room_types` but inverted. This can be used to
+                    filter out spaces from the room list. If a type is in both `room_types`
+                    and `not_room_types`, then `not_room_types` wins and they are not included
+                    in the result.
+                room_name_like: Filter the room name. Case-insensitive partial matching
+                    e.g 'foo' matches 'abFooab'. The term 'like' is inspired by SQL 'LIKE',
+                    and the text here is similar to '%foo%'.
+                tags: Filter the room based on its room tags. If multiple tags are
+                    present, a room can have any one of the listed tags (OR'd).
+                not_tags: Filter the room based on its room tags. Takes priority over
+                    `tags`. For example, a room with tags A and B with filters `tags: [A]`
+                    `not_tags: [B]` would NOT be included because `not_tags` takes priority over
+                    `tags`. This filter is useful if your rooms list does NOT include the
+                    list of favourite rooms again.
+            """
+
             is_dm: Optional[StrictBool] = None
             spaces: Optional[List[StrictStr]] = None
             is_encrypted: Optional[StrictBool] = None
diff --git a/tests/handlers/test_sliding_sync.py b/tests/handlers/test_sliding_sync.py
index 41ceb517f0..62fe1214fe 100644
--- a/tests/handlers/test_sliding_sync.py
+++ b/tests/handlers/test_sliding_sync.py
@@ -22,8 +22,9 @@ from unittest.mock import patch
 
 from twisted.test.proto_helpers import MemoryReactor
 
-from synapse.api.constants import EventTypes, JoinRules, Membership
+from synapse.api.constants import AccountDataTypes, EventTypes, JoinRules, Membership
 from synapse.api.room_versions import RoomVersions
+from synapse.handlers.sliding_sync import SlidingSyncConfig
 from synapse.rest import admin
 from synapse.rest.client import knock, login, room
 from synapse.server import HomeServer
@@ -1116,3 +1117,130 @@ class GetSyncRoomIdsForUserEventShardTestCase(BaseMultiWorkerStreamTestCase):
                 room_id3,
             },
         )
+
+
+class FilterRoomsTestCase(HomeserverTestCase):
+    """
+    Tests Sliding Sync handler `filter_rooms()` to make sure it includes/excludes rooms
+    correctly.
+    """
+
+    servlets = [
+        admin.register_servlets,
+        knock.register_servlets,
+        login.register_servlets,
+        room.register_servlets,
+    ]
+
+    def default_config(self) -> JsonDict:
+        config = super().default_config()
+        # Enable sliding sync
+        config["experimental_features"] = {"msc3575_enabled": True}
+        return config
+
+    def prepare(self, reactor: MemoryReactor, clock: Clock, hs: HomeServer) -> None:
+        self.sliding_sync_handler = self.hs.get_sliding_sync_handler()
+        self.store = self.hs.get_datastores().main
+        self.event_sources = hs.get_event_sources()
+
+    def _create_dm_room(
+        self,
+        inviter_user_id: str,
+        inviter_tok: str,
+        invitee_user_id: str,
+        invitee_tok: str,
+    ) -> str:
+        """
+        Helper to create a DM room as the "inviter" and invite the "invitee" user to the room. The
+        "invitee" user also will join the room. The `m.direct` account data will be set
+        for both users.
+        """
+
+        # Create a room and send an invite the other user
+        room_id = self.helper.create_room_as(
+            inviter_user_id,
+            is_public=False,
+            tok=inviter_tok,
+        )
+        self.helper.invite(
+            room_id,
+            src=inviter_user_id,
+            targ=invitee_user_id,
+            tok=inviter_tok,
+            extra_data={"is_direct": True},
+        )
+        # Person that was invited joins the room
+        self.helper.join(room_id, invitee_user_id, tok=invitee_tok)
+
+        # Mimic the client setting the room as a direct message in the global account
+        # data
+        self.get_success(
+            self.store.add_account_data_for_user(
+                invitee_user_id,
+                AccountDataTypes.DIRECT,
+                {inviter_user_id: [room_id]},
+            )
+        )
+        self.get_success(
+            self.store.add_account_data_for_user(
+                inviter_user_id,
+                AccountDataTypes.DIRECT,
+                {invitee_user_id: [room_id]},
+            )
+        )
+
+        return room_id
+
+    def test_filter_dm_rooms(self) -> None:
+        """
+        Test `filter.is_dm` for DM rooms
+        """
+        user1_id = self.register_user("user1", "pass")
+        user1_tok = self.login(user1_id, "pass")
+        user2_id = self.register_user("user2", "pass")
+        user2_tok = self.login(user2_id, "pass")
+
+        # Create a normal room
+        room_id = self.helper.create_room_as(
+            user1_id,
+            is_public=False,
+            tok=user1_tok,
+        )
+
+        # Create a DM room
+        dm_room_id = self._create_dm_room(
+            inviter_user_id=user1_id,
+            inviter_tok=user1_tok,
+            invitee_user_id=user2_id,
+            invitee_tok=user2_tok,
+        )
+
+        after_rooms_token = self.event_sources.get_current_token()
+
+        # Try with `is_dm=True`
+        truthy_filtered_room_ids = self.get_success(
+            self.sliding_sync_handler.filter_rooms(
+                UserID.from_string(user1_id),
+                {room_id, dm_room_id},
+                SlidingSyncConfig.SlidingSyncList.Filters(
+                    is_dm=True,
+                ),
+                after_rooms_token,
+            )
+        )
+
+        self.assertEqual(truthy_filtered_room_ids, {dm_room_id})
+
+        # Try with `is_dm=False`
+        falsy_filtered_room_ids = self.get_success(
+            self.sliding_sync_handler.filter_rooms(
+                UserID.from_string(user1_id),
+                {room_id, dm_room_id},
+                SlidingSyncConfig.SlidingSyncList.Filters(
+                    is_dm=False,
+                ),
+                after_rooms_token,
+            )
+        )
+
+        self.assertEqual(falsy_filtered_room_ids, {room_id})
diff --git a/tests/rest/client/test_sync.py b/tests/rest/client/test_sync.py
index a20a3fb40d..40870b2cfe 100644
--- a/tests/rest/client/test_sync.py
+++ b/tests/rest/client/test_sync.py
@@ -27,6 +27,7 @@ from twisted.test.proto_helpers import MemoryReactor
 
 import synapse.rest.admin
 from synapse.api.constants import (
+    AccountDataTypes,
     EventContentFields,
     EventTypes,
     ReceiptTypes,
@@ -1226,10 +1227,59 @@ class SlidingSyncTestCase(unittest.HomeserverTestCase):
         return config
 
     def prepare(self, reactor: MemoryReactor, clock: Clock, hs: HomeServer) -> None:
+        self.store = hs.get_datastores().main
         self.sync_endpoint = "/_matrix/client/unstable/org.matrix.msc3575/sync"
         self.store = hs.get_datastores().main
         self.event_sources = hs.get_event_sources()
 
+    def _create_dm_room(
+        self,
+        inviter_user_id: str,
+        inviter_tok: str,
+        invitee_user_id: str,
+        invitee_tok: str,
+    ) -> str:
+        """
+        Helper to create a DM room as the "inviter" and invite the "invitee" user to the
+        room. The "invitee" user also will join the room. The `m.direct` account data
+        will be set for both users.
+        """
+
+        # Create a room and send an invite the other user
+        room_id = self.helper.create_room_as(
+            inviter_user_id,
+            is_public=False,
+            tok=inviter_tok,
+        )
+        self.helper.invite(
+            room_id,
+            src=inviter_user_id,
+            targ=invitee_user_id,
+            tok=inviter_tok,
+            extra_data={"is_direct": True},
+        )
+        # Person that was invited joins the room
+        self.helper.join(room_id, invitee_user_id, tok=invitee_tok)
+
+        # Mimic the client setting the room as a direct message in the global account
+        # data
+        self.get_success(
+            self.store.add_account_data_for_user(
+                invitee_user_id,
+                AccountDataTypes.DIRECT,
+                {inviter_user_id: [room_id]},
+            )
+        )
+        self.get_success(
+            self.store.add_account_data_for_user(
+                inviter_user_id,
+                AccountDataTypes.DIRECT,
+                {invitee_user_id: [room_id]},
+            )
+        )
+
+        return room_id
+
     def test_sync_list(self) -> None:
         """
         Test that room IDs show up in the Sliding Sync lists
@@ -1336,3 +1386,80 @@ class SlidingSyncTestCase(unittest.HomeserverTestCase):
         self.assertEqual(
             channel.json_body["next_pos"], future_position_token_serialized
         )
+
+    def test_filter_list(self) -> None:
+        """
+        Test that filters apply to lists
+        """
+        user1_id = self.register_user("user1", "pass")
+        user1_tok = self.login(user1_id, "pass")
+        user2_id = self.register_user("user2", "pass")
+        user2_tok = self.login(user2_id, "pass")
+
+        # Create a DM room
+        dm_room_id = self._create_dm_room(
+            inviter_user_id=user1_id,
+            inviter_tok=user1_tok,
+            invitee_user_id=user2_id,
+            invitee_tok=user2_tok,
+        )
+
+        # Create a normal room
+        room_id = self.helper.create_room_as(user1_id, tok=user1_tok, is_public=True)
+
+        # Make the Sliding Sync request
+        channel = self.make_request(
+            "POST",
+            self.sync_endpoint,
+            {
+                "lists": {
+                    "dms": {
+                        "ranges": [[0, 99]],
+                        "sort": ["by_recency"],
+                        "required_state": [],
+                        "timeline_limit": 1,
+                        "filters": {"is_dm": True},
+                    },
+                    "foo-list": {
+                        "ranges": [[0, 99]],
+                        "sort": ["by_recency"],
+                        "required_state": [],
+                        "timeline_limit": 1,
+                        "filters": {"is_dm": False},
+                    },
+                }
+            },
+            access_token=user1_tok,
+        )
+        self.assertEqual(channel.code, 200, channel.json_body)
+
+        # Make sure it has the foo-list we requested
+        self.assertListEqual(
+            list(channel.json_body["lists"].keys()),
+            ["dms", "foo-list"],
+            channel.json_body["lists"].keys(),
+        )
+
+        # Make sure the list includes the room we are joined to
+        self.assertListEqual(
+            list(channel.json_body["lists"]["dms"]["ops"]),
+            [
+                {
+                    "op": "SYNC",
+                    "range": [0, 99],
+                    "room_ids": [dm_room_id],
+                }
+            ],
+            list(channel.json_body["lists"]["dms"]),
+        )
+        self.assertListEqual(
+            list(channel.json_body["lists"]["foo-list"]["ops"]),
+            [
+                {
+                    "op": "SYNC",
+                    "range": [0, 99],
+                    "room_ids": [room_id],
+                }
+            ],
+            list(channel.json_body["lists"]["foo-list"]),
+        )