Merge pull request #5859 from matrix-org/rei/msc2197

MSC2197 Search Filters over Federation

4 files changed, 96 insertions, 16 deletions

diff --git a/changelog.d/5859.feature b/changelog.d/5859.feature
new file mode 100644
index 0000000000..52df7fc81b
--- /dev/null
+++ b/changelog.d/5859.feature
@@ -0,0 +1 @@
+Add unstable support for MSC2197 (filtered search requests over federation), in order to allow upcoming room directory query performance improvements.
diff --git a/synapse/federation/transport/client.py b/synapse/federation/transport/client.py
index 0cea0d2a10..482a101c09 100644
--- a/synapse/federation/transport/client.py
+++ b/synapse/federation/transport/client.py
@@ -327,21 +327,37 @@ class TransportLayerClient(object):
         include_all_networks=False,
         third_party_instance_id=None,
     ):
-        path = _create_v1_path("/publicRooms")
-
-        args = {"include_all_networks": "true" if include_all_networks else "false"}
-        if third_party_instance_id:
-            args["third_party_instance_id"] = (third_party_instance_id,)
-        if limit:
-            args["limit"] = [str(limit)]
-        if since_token:
-            args["since"] = [since_token]
-
-        # TODO(erikj): Actually send the search_filter across federation.
-
-        response = yield self.client.get_json(
-            destination=remote_server, path=path, args=args, ignore_backoff=True
-        )
+        if search_filter:
+            # this uses MSC2197 (Search Filtering over Federation)
+            path = _create_v1_path("/publicRooms")
+
+            data = {"include_all_networks": "true" if include_all_networks else "false"}
+            if third_party_instance_id:
+                data["third_party_instance_id"] = third_party_instance_id
+            if limit:
+                data["limit"] = str(limit)
+            if since_token:
+                data["since"] = since_token
+
+            data["filter"] = search_filter
+
+            response = yield self.client.post_json(
+                destination=remote_server, path=path, data=data, ignore_backoff=True
+            )
+        else:
+            path = _create_v1_path("/publicRooms")
+
+            args = {"include_all_networks": "true" if include_all_networks else "false"}
+            if third_party_instance_id:
+                args["third_party_instance_id"] = (third_party_instance_id,)
+            if limit:
+                args["limit"] = [str(limit)]
+            if since_token:
+                args["since"] = [since_token]
+
+            response = yield self.client.get_json(
+                destination=remote_server, path=path, args=args, ignore_backoff=True
+            )
 
         return response
 
diff --git a/synapse/federation/transport/server.py b/synapse/federation/transport/server.py
index dc53b4b170..f9930b6460 100644
--- a/synapse/federation/transport/server.py
+++ b/synapse/federation/transport/server.py
@@ -770,6 +770,42 @@ class PublicRoomList(BaseFederationServlet):
         )
         return 200, data
 
+    async def on_POST(self, origin, content, query):
+        # This implements MSC2197 (Search Filtering over Federation)
+        if not self.allow_access:
+            raise FederationDeniedError(origin)
+
+        limit = int(content.get("limit", 100))
+        since_token = content.get("since", None)
+        search_filter = content.get("filter", None)
+
+        include_all_networks = content.get("include_all_networks", False)
+        third_party_instance_id = content.get("third_party_instance_id", None)
+
+        if include_all_networks:
+            network_tuple = None
+            if third_party_instance_id is not None:
+                raise SynapseError(
+                    400, "Can't use include_all_networks with an explicit network"
+                )
+        elif third_party_instance_id is None:
+            network_tuple = ThirdPartyInstanceID(None, None)
+        else:
+            network_tuple = ThirdPartyInstanceID.from_string(third_party_instance_id)
+
+        if search_filter is None:
+            logger.warning("Nonefilter")
+
+        data = await self.handler.get_local_public_room_list(
+            limit=limit,
+            since_token=since_token,
+            search_filter=search_filter,
+            network_tuple=network_tuple,
+            from_federation=True,
+        )
+
+        return 200, data
+
 
 class FederationVersionServlet(BaseFederationServlet):
     PATH = "/version"
diff --git a/synapse/handlers/room_list.py b/synapse/handlers/room_list.py
index e9094ad02b..a7e55f00e5 100644
--- a/synapse/handlers/room_list.py
+++ b/synapse/handlers/room_list.py
@@ -25,6 +25,7 @@ from unpaddedbase64 import decode_base64, encode_base64
 from twisted.internet import defer
 
 from synapse.api.constants import EventTypes, JoinRules
+from synapse.api.errors import Codes, HttpResponseException
 from synapse.types import ThirdPartyInstanceID
 from synapse.util.async_helpers import concurrently_execute
 from synapse.util.caches.descriptors import cachedInlineCallbacks
@@ -485,7 +486,33 @@ class RoomListHandler(BaseHandler):
             return {"chunk": [], "total_room_count_estimate": 0}
 
         if search_filter:
-            # We currently don't support searching across federation, so we have
+            # Searching across federation is defined in MSC2197.
+            # However, the remote homeserver may or may not actually support it.
+            # So we first try an MSC2197 remote-filtered search, then fall back
+            # to a locally-filtered search if we must.
+
+            try:
+                res = yield self._get_remote_list_cached(
+                    server_name,
+                    limit=limit,
+                    since_token=since_token,
+                    include_all_networks=include_all_networks,
+                    third_party_instance_id=third_party_instance_id,
+                    search_filter=search_filter,
+                )
+                return res
+            except HttpResponseException as hre:
+                syn_err = hre.to_synapse_error()
+                if hre.code in (404, 405) or syn_err.errcode in (
+                    Codes.UNRECOGNIZED,
+                    Codes.NOT_FOUND,
+                ):
+                    logger.debug("Falling back to locally-filtered /publicRooms")
+                else:
+                    raise  # Not an error that should trigger a fallback.
+
+            # if we reach this point, then we fall back to the situation where
+            # we currently don't support searching across federation, so we have
             # to do it manually without pagination
             limit = None
             since_token = None