Move Spam Checker callbacks to a dedicated file (#15453)

15 files changed, 84 insertions, 63 deletions

diff --git a/synapse/app/_base.py b/synapse/app/_base.py
index f7b866978c..954402e4d2 100644
--- a/synapse/app/_base.py
+++ b/synapse/app/_base.py
@@ -64,7 +64,6 @@ from synapse.config.homeserver import HomeServerConfig
 from synapse.config.server import ListenerConfig, ManholeConfig, TCPListenerConfig
 from synapse.crypto import context_factory
 from synapse.events.presence_router import load_legacy_presence_router
-from synapse.events.spamcheck import load_legacy_spam_checkers
 from synapse.events.third_party_rules import load_legacy_third_party_event_rules
 from synapse.handlers.auth import load_legacy_password_auth_providers
 from synapse.http.site import SynapseSite
@@ -73,6 +72,7 @@ from synapse.logging.opentracing import init_tracer
 from synapse.metrics import install_gc_manager, register_threadpool
 from synapse.metrics.background_process_metrics import wrap_as_background_process
 from synapse.metrics.jemalloc import setup_jemalloc_stats
+from synapse.module_api.callbacks.spamchecker_callbacks import load_legacy_spam_checkers
 from synapse.types import ISynapseReactor
 from synapse.util import SYNAPSE_VERSION
 from synapse.util.caches.lrucache import setup_expire_lru_cache_entries
diff --git a/synapse/federation/federation_base.py b/synapse/federation/federation_base.py
index 29fae716f5..3df975958d 100644
--- a/synapse/federation/federation_base.py
+++ b/synapse/federation/federation_base.py
@@ -51,7 +51,7 @@ class FederationBase:
 
         self.server_name = hs.hostname
         self.keyring = hs.get_keyring()
-        self.spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
         self.store = hs.get_datastores().main
         self._clock = hs.get_clock()
         self._storage_controllers = hs.get_storage_controllers()
@@ -137,9 +137,9 @@ class FederationBase:
                     )
             return redacted_event
 
-        spam_check = await self.spam_checker.check_event_for_spam(pdu)
+        spam_check = await self._spam_checker_module_callbacks.check_event_for_spam(pdu)
 
-        if spam_check != self.spam_checker.NOT_SPAM:
+        if spam_check != self._spam_checker_module_callbacks.NOT_SPAM:
             logger.warning("Event contains spam, soft-failing %s", pdu.event_id)
             log_kv(
                 {
diff --git a/synapse/federation/federation_server.py b/synapse/federation/federation_server.py
index 64e99292ec..d7740eb3b4 100644
--- a/synapse/federation/federation_server.py
+++ b/synapse/federation/federation_server.py
@@ -130,7 +130,7 @@ class FederationServer(FederationBase):
         super().__init__(hs)
 
         self.handler = hs.get_federation_handler()
-        self._spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
         self._federation_event_handler = hs.get_federation_event_handler()
         self.state = hs.get_state_handler()
         self._event_auth_handler = hs.get_event_auth_handler()
@@ -1129,7 +1129,7 @@ class FederationServer(FederationBase):
             logger.warning("event id %s: %s", pdu.event_id, e)
             raise FederationError("ERROR", 403, str(e), affected=pdu.event_id)
 
-        if await self._spam_checker.should_drop_federated_event(pdu):
+        if await self._spam_checker_module_callbacks.should_drop_federated_event(pdu):
             logger.warning(
                 "Unstaged federated event contains spam, dropping %s", pdu.event_id
             )
@@ -1174,7 +1174,9 @@ class FederationServer(FederationBase):
 
             origin, event = next
 
-            if await self._spam_checker.should_drop_federated_event(event):
+            if await self._spam_checker_module_callbacks.should_drop_federated_event(
+                event
+            ):
                 logger.warning(
                     "Staged federated event contains spam, dropping %s",
                     event.event_id,
diff --git a/synapse/handlers/directory.py b/synapse/handlers/directory.py
index 1fb23cc9bf..5e8316e2e5 100644
--- a/synapse/handlers/directory.py
+++ b/synapse/handlers/directory.py
@@ -60,7 +60,7 @@ class DirectoryHandler:
             "directory", self.on_directory_query
         )
 
-        self.spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
 
     async def _create_association(
         self,
@@ -145,10 +145,12 @@ class DirectoryHandler:
                         403, "You must be in the room to create an alias for it"
                     )
 
-            spam_check = await self.spam_checker.user_may_create_room_alias(
-                user_id, room_alias
+            spam_check = (
+                await self._spam_checker_module_callbacks.user_may_create_room_alias(
+                    user_id, room_alias
+                )
             )
-            if spam_check != self.spam_checker.NOT_SPAM:
+            if spam_check != self._spam_checker_module_callbacks.NOT_SPAM:
                 raise AuthError(
                     403,
                     "This user is not permitted to create this alias",
@@ -444,7 +446,9 @@ class DirectoryHandler:
         """
         user_id = requester.user.to_string()
 
-        spam_check = await self.spam_checker.user_may_publish_room(user_id, room_id)
+        spam_check = await self._spam_checker_module_callbacks.user_may_publish_room(
+            user_id, room_id
+        )
         if spam_check != NOT_SPAM:
             raise AuthError(
                 403,
diff --git a/synapse/handlers/federation.py b/synapse/handlers/federation.py
index 65461a0787..d1a88cc604 100644
--- a/synapse/handlers/federation.py
+++ b/synapse/handlers/federation.py
@@ -141,7 +141,7 @@ class FederationHandler:
         self.server_name = hs.hostname
         self.keyring = hs.get_keyring()
         self.is_mine_id = hs.is_mine_id
-        self.spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
         self.event_creation_handler = hs.get_event_creation_handler()
         self.event_builder_factory = hs.get_event_builder_factory()
         self._event_auth_handler = hs.get_event_auth_handler()
@@ -1042,7 +1042,7 @@ class FederationHandler:
         if self.hs.config.server.block_non_admin_invites:
             raise SynapseError(403, "This server does not accept room invites")
 
-        spam_check = await self.spam_checker.user_may_invite(
+        spam_check = await self._spam_checker_module_callbacks.user_may_invite(
             event.sender, event.state_key, event.room_id
         )
         if spam_check != NOT_SPAM:
diff --git a/synapse/handlers/message.py b/synapse/handlers/message.py
index a17fe3bf53..2e964ed37e 100644
--- a/synapse/handlers/message.py
+++ b/synapse/handlers/message.py
@@ -508,7 +508,7 @@ class EventCreationHandler:
 
         self._bulk_push_rule_evaluator = hs.get_bulk_push_rule_evaluator()
 
-        self.spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
         self.third_party_event_rules: "ThirdPartyEventRules" = (
             self.hs.get_third_party_event_rules()
         )
@@ -1035,8 +1035,12 @@ class EventCreationHandler:
                     event.sender,
                 )
 
-                spam_check_result = await self.spam_checker.check_event_for_spam(event)
-                if spam_check_result != self.spam_checker.NOT_SPAM:
+                spam_check_result = (
+                    await self._spam_checker_module_callbacks.check_event_for_spam(
+                        event
+                    )
+                )
+                if spam_check_result != self._spam_checker_module_callbacks.NOT_SPAM:
                     if isinstance(spam_check_result, tuple):
                         try:
                             [code, dict] = spam_check_result
diff --git a/synapse/handlers/register.py b/synapse/handlers/register.py
index c8bf2439af..61c4b833bd 100644
--- a/synapse/handlers/register.py
+++ b/synapse/handlers/register.py
@@ -110,7 +110,7 @@ class RegistrationHandler:
         self._server_notices_mxid = hs.config.servernotices.server_notices_mxid
         self._server_name = hs.hostname
 
-        self.spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
 
         if hs.config.worker.worker_app:
             self._register_client = ReplicationRegisterServlet.make_client(hs)
@@ -259,7 +259,7 @@ class RegistrationHandler:
 
         await self.check_registration_ratelimit(address)
 
-        result = await self.spam_checker.check_registration_for_spam(
+        result = await self._spam_checker_module_callbacks.check_registration_for_spam(
             threepid,
             localpart,
             user_agent_ips or [],
diff --git a/synapse/handlers/room.py b/synapse/handlers/room.py
index 2d69cabf43..efd9612d90 100644
--- a/synapse/handlers/room.py
+++ b/synapse/handlers/room.py
@@ -106,7 +106,7 @@ class RoomCreationHandler:
         self.auth_blocking = hs.get_auth_blocking()
         self.clock = hs.get_clock()
         self.hs = hs
-        self.spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
         self.event_creation_handler = hs.get_event_creation_handler()
         self.room_member_handler = hs.get_room_member_handler()
         self._event_auth_handler = hs.get_event_auth_handler()
@@ -449,7 +449,9 @@ class RoomCreationHandler:
         """
         user_id = requester.user.to_string()
 
-        spam_check = await self.spam_checker.user_may_create_room(user_id)
+        spam_check = await self._spam_checker_module_callbacks.user_may_create_room(
+            user_id
+        )
         if spam_check != NOT_SPAM:
             raise SynapseError(
                 403,
@@ -761,7 +763,9 @@ class RoomCreationHandler:
                 )
 
         if not is_requester_admin:
-            spam_check = await self.spam_checker.user_may_create_room(user_id)
+            spam_check = await self._spam_checker_module_callbacks.user_may_create_room(
+                user_id
+            )
             if spam_check != NOT_SPAM:
                 raise SynapseError(
                     403,
diff --git a/synapse/handlers/room_member.py b/synapse/handlers/room_member.py
index 1d8b0aee6f..ec317e6023 100644
--- a/synapse/handlers/room_member.py
+++ b/synapse/handlers/room_member.py
@@ -96,7 +96,7 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
         self.member_as_limiter = Linearizer(max_count=10, name="member_as_limiter")
 
         self.clock = hs.get_clock()
-        self.spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
         self.third_party_event_rules = hs.get_third_party_event_rules()
         self._server_notices_mxid = self.config.servernotices.server_notices_mxid
         self._enable_lookup = hs.config.registration.enable_3pid_lookup
@@ -806,7 +806,7 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
                     )
                     block_invite_result = (Codes.FORBIDDEN, {})
 
-                spam_check = await self.spam_checker.user_may_invite(
+                spam_check = await self._spam_checker_module_callbacks.user_may_invite(
                     requester.user.to_string(), target_id, room_id
                 )
                 if spam_check != NOT_SPAM:
@@ -940,8 +940,10 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
                 # a room then they're allowed to join it.
                 and not new_room
             ):
-                spam_check = await self.spam_checker.user_may_join_room(
-                    target.to_string(), room_id, is_invited=inviter is not None
+                spam_check = (
+                    await self._spam_checker_module_callbacks.user_may_join_room(
+                        target.to_string(), room_id, is_invited=inviter is not None
+                    )
                 )
                 if spam_check != NOT_SPAM:
                     raise SynapseError(
@@ -1550,11 +1552,13 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
             )
         else:
             # Check if the spamchecker(s) allow this invite to go through.
-            spam_check = await self.spam_checker.user_may_send_3pid_invite(
-                inviter_userid=requester.user.to_string(),
-                medium=medium,
-                address=address,
-                room_id=room_id,
+            spam_check = (
+                await self._spam_checker_module_callbacks.user_may_send_3pid_invite(
+                    inviter_userid=requester.user.to_string(),
+                    medium=medium,
+                    address=address,
+                    room_id=room_id,
+                )
             )
             if spam_check != NOT_SPAM:
                 raise SynapseError(
diff --git a/synapse/handlers/user_directory.py b/synapse/handlers/user_directory.py
index 28a92d41d6..05197edc95 100644
--- a/synapse/handlers/user_directory.py
+++ b/synapse/handlers/user_directory.py
@@ -94,7 +94,7 @@ class UserDirectoryHandler(StateDeltasHandler):
         self.is_mine_id = hs.is_mine_id
         self.update_user_directory = hs.config.worker.should_update_user_directory
         self.search_all_users = hs.config.userdirectory.user_directory_search_all_users
-        self.spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
         self._hs = hs
 
         # The current position in the current_state_delta stream
@@ -149,7 +149,9 @@ class UserDirectoryHandler(StateDeltasHandler):
         # Remove any spammy users from the results.
         non_spammy_users = []
         for user in results["results"]:
-            if not await self.spam_checker.check_username_for_spam(user):
+            if not await self._spam_checker_module_callbacks.check_username_for_spam(
+                user
+            ):
                 non_spammy_users.append(user)
         results["results"] = non_spammy_users
 
diff --git a/synapse/media/media_storage.py b/synapse/media/media_storage.py
index a7e22a91e1..a819d95407 100644
--- a/synapse/media/media_storage.py
+++ b/synapse/media/media_storage.py
@@ -36,7 +36,6 @@ from twisted.internet.defer import Deferred
 from twisted.internet.interfaces import IConsumer
 from twisted.protocols.basic import FileSender
 
-import synapse
 from synapse.api.errors import NotFoundError
 from synapse.logging.context import defer_to_thread, make_deferred_yieldable
 from synapse.util import Clock
@@ -74,7 +73,7 @@ class MediaStorage:
         self.local_media_directory = local_media_directory
         self.filepaths = filepaths
         self.storage_providers = storage_providers
-        self.spam_checker = hs.get_spam_checker()
+        self._spam_checker_module_callbacks = hs.get_module_api_callbacks().spam_checker
         self.clock = hs.get_clock()
 
     async def store_file(self, source: IO, file_info: FileInfo) -> str:
@@ -145,10 +144,10 @@ class MediaStorage:
                     f.flush()
                     f.close()
 
-                    spam_check = await self.spam_checker.check_media_file_for_spam(
+                    spam_check = await self._spam_checker_module_callbacks.check_media_file_for_spam(
                         ReadableFileWrapper(self.clock, fname), file_info
                     )
-                    if spam_check != synapse.module_api.NOT_SPAM:
+                    if spam_check != self._spam_checker_module_callbacks.NOT_SPAM:
                         logger.info("Blocking media due to spam checker")
                         # Note that we'll delete the stored media, due to the
                         # try/except below. The media also won't be stored in
diff --git a/synapse/module_api/__init__.py b/synapse/module_api/__init__.py
index 595c23e78d..eeafea74d1 100644
--- a/synapse/module_api/__init__.py
+++ b/synapse/module_api/__init__.py
@@ -44,20 +44,6 @@ from synapse.events.presence_router import (
     GET_USERS_FOR_STATES_CALLBACK,
     PresenceRouter,
 )
-from synapse.events.spamcheck import (
-    CHECK_EVENT_FOR_SPAM_CALLBACK,
-    CHECK_MEDIA_FILE_FOR_SPAM_CALLBACK,
-    CHECK_REGISTRATION_FOR_SPAM_CALLBACK,
-    CHECK_USERNAME_FOR_SPAM_CALLBACK,
-    SHOULD_DROP_FEDERATED_EVENT_CALLBACK,
-    USER_MAY_CREATE_ROOM_ALIAS_CALLBACK,
-    USER_MAY_CREATE_ROOM_CALLBACK,
-    USER_MAY_INVITE_CALLBACK,
-    USER_MAY_JOIN_ROOM_CALLBACK,
-    USER_MAY_PUBLISH_ROOM_CALLBACK,
-    USER_MAY_SEND_3PID_INVITE_CALLBACK,
-    SpamChecker,
-)
 from synapse.events.third_party_rules import (
     CHECK_CAN_DEACTIVATE_USER_CALLBACK,
     CHECK_CAN_SHUTDOWN_ROOM_CALLBACK,
@@ -105,6 +91,20 @@ from synapse.module_api.callbacks.account_validity_callbacks import (
     ON_LEGACY_SEND_MAIL_CALLBACK,
     ON_USER_REGISTRATION_CALLBACK,
 )
+from synapse.module_api.callbacks.spamchecker_callbacks import (
+    CHECK_EVENT_FOR_SPAM_CALLBACK,
+    CHECK_MEDIA_FILE_FOR_SPAM_CALLBACK,
+    CHECK_REGISTRATION_FOR_SPAM_CALLBACK,
+    CHECK_USERNAME_FOR_SPAM_CALLBACK,
+    SHOULD_DROP_FEDERATED_EVENT_CALLBACK,
+    USER_MAY_CREATE_ROOM_ALIAS_CALLBACK,
+    USER_MAY_CREATE_ROOM_CALLBACK,
+    USER_MAY_INVITE_CALLBACK,
+    USER_MAY_JOIN_ROOM_CALLBACK,
+    USER_MAY_PUBLISH_ROOM_CALLBACK,
+    USER_MAY_SEND_3PID_INVITE_CALLBACK,
+    SpamCheckerModuleApiCallbacks,
+)
 from synapse.rest.client.login import LoginResponse
 from synapse.storage import DataStore
 from synapse.storage.background_updates import (
@@ -147,7 +147,7 @@ are loaded into Synapse.
 """
 
 PRESENCE_ALL_USERS = PresenceRouter.ALL_USERS
-NOT_SPAM = SpamChecker.NOT_SPAM
+NOT_SPAM = SpamCheckerModuleApiCallbacks.NOT_SPAM
 
 __all__ = [
     "errors",
@@ -271,7 +271,6 @@ class ModuleApi:
         self._public_room_list_manager = PublicRoomListManager(hs)
         self._account_data_manager = AccountDataManager(hs)
 
-        self._spam_checker = hs.get_spam_checker()
         self._third_party_event_rules = hs.get_third_party_event_rules()
         self._password_auth_provider = hs.get_password_auth_provider()
         self._presence_router = hs.get_presence_router()
@@ -305,7 +304,7 @@ class ModuleApi:
 
         Added in Synapse v1.37.0.
         """
-        return self._spam_checker.register_callbacks(
+        return self._callbacks.spam_checker.register_callbacks(
             check_event_for_spam=check_event_for_spam,
             should_drop_federated_event=should_drop_federated_event,
             user_may_join_room=user_may_join_room,
diff --git a/synapse/module_api/callbacks/__init__.py b/synapse/module_api/callbacks/__init__.py
index 3d977bf655..5cdb2c003a 100644
--- a/synapse/module_api/callbacks/__init__.py
+++ b/synapse/module_api/callbacks/__init__.py
@@ -12,11 +12,20 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from synapse.server import HomeServer
+
 from synapse.module_api.callbacks.account_validity_callbacks import (
     AccountValidityModuleApiCallbacks,
 )
+from synapse.module_api.callbacks.spamchecker_callbacks import (
+    SpamCheckerModuleApiCallbacks,
+)
 
 
 class ModuleApiCallbacks:
-    def __init__(self) -> None:
+    def __init__(self, hs: "HomeServer") -> None:
         self.account_validity = AccountValidityModuleApiCallbacks()
+        self.spam_checker = SpamCheckerModuleApiCallbacks(hs)
diff --git a/synapse/events/spamcheck.py b/synapse/module_api/callbacks/spamchecker_callbacks.py
index 765c15bb51..4456d1b81e 100644
--- a/synapse/events/spamcheck.py
+++ b/synapse/module_api/callbacks/spamchecker_callbacks.py
@@ -286,11 +286,10 @@ def load_legacy_spam_checkers(hs: "synapse.server.HomeServer") -> None:
         api.register_spam_checker_callbacks(**hooks)
 
 
-class SpamChecker:
+class SpamCheckerModuleApiCallbacks:
     NOT_SPAM: Literal["NOT_SPAM"] = "NOT_SPAM"
 
     def __init__(self, hs: "synapse.server.HomeServer") -> None:
-        self.hs = hs
         self.clock = hs.get_clock()
 
         self._check_event_for_spam_callbacks: List[CHECK_EVENT_FOR_SPAM_CALLBACK] = []
diff --git a/synapse/server.py b/synapse/server.py
index a191c19993..559724594b 100644
--- a/synapse/server.py
+++ b/synapse/server.py
@@ -42,7 +42,6 @@ from synapse.crypto.context_factory import RegularPolicyForHTTPS
 from synapse.crypto.keyring import Keyring
 from synapse.events.builder import EventBuilderFactory
 from synapse.events.presence_router import PresenceRouter
-from synapse.events.spamcheck import SpamChecker
 from synapse.events.third_party_rules import ThirdPartyEventRules
 from synapse.events.utils import EventClientSerializer
 from synapse.federation.federation_client import FederationClient
@@ -688,10 +687,6 @@ class HomeServer(metaclass=abc.ABCMeta):
         return StatsHandler(self)
 
     @cache_in_self
-    def get_spam_checker(self) -> SpamChecker:
-        return SpamChecker(self)
-
-    @cache_in_self
     def get_third_party_event_rules(self) -> ThirdPartyEventRules:
         return ThirdPartyEventRules(self)
 
@@ -803,7 +798,7 @@ class HomeServer(metaclass=abc.ABCMeta):
 
     @cache_in_self
     def get_module_api_callbacks(self) -> ModuleApiCallbacks:
-        return ModuleApiCallbacks()
+        return ModuleApiCallbacks(self)
 
     @cache_in_self
     def get_account_data_handler(self) -> AccountDataHandler: