Uniformize spam-checker API, part 5: expand other spam-checker callbacks to return `Tuple[Codes, dict]` (#13044)

Signed-off-by: David Teller <davidt@element.io>
Co-authored-by: Brendan Abolivier <babolivier@matrix.org>

8 files changed, 176 insertions, 50 deletions

diff --git a/synapse/api/errors.py b/synapse/api/errors.py
index cc7b785472..1c74e131f2 100644
--- a/synapse/api/errors.py
+++ b/synapse/api/errors.py
@@ -297,8 +297,14 @@ class AuthError(SynapseError):
     other poorly-defined times.
     """
 
-    def __init__(self, code: int, msg: str, errcode: str = Codes.FORBIDDEN):
-        super().__init__(code, msg, errcode)
+    def __init__(
+        self,
+        code: int,
+        msg: str,
+        errcode: str = Codes.FORBIDDEN,
+        additional_fields: Optional[dict] = None,
+    ):
+        super().__init__(code, msg, errcode, additional_fields)
 
 
 class InvalidClientCredentialsError(SynapseError):
diff --git a/synapse/events/spamcheck.py b/synapse/events/spamcheck.py
index 32712d2042..4a3bfb38f1 100644
--- a/synapse/events/spamcheck.py
+++ b/synapse/events/spamcheck.py
@@ -21,7 +21,6 @@ from typing import (
     Awaitable,
     Callable,
     Collection,
-    Dict,
     List,
     Optional,
     Tuple,
@@ -32,10 +31,11 @@ from typing import (
 from typing_extensions import Literal
 
 import synapse
+from synapse.api.errors import Codes
 from synapse.rest.media.v1._base import FileInfo
 from synapse.rest.media.v1.media_storage import ReadableFileWrapper
 from synapse.spam_checker_api import RegistrationBehaviour
-from synapse.types import RoomAlias, UserProfile
+from synapse.types import JsonDict, RoomAlias, UserProfile
 from synapse.util.async_helpers import delay_cancellation, maybe_awaitable
 from synapse.util.metrics import Measure
 
@@ -50,12 +50,12 @@ CHECK_EVENT_FOR_SPAM_CALLBACK = Callable[
     Awaitable[
         Union[
             str,
-            "synapse.api.errors.Codes",
+            Codes,
             # Highly experimental, not officially part of the spamchecker API, may
             # disappear without warning depending on the results of ongoing
             # experiments.
             # Use this to return additional information as part of an error.
-            Tuple["synapse.api.errors.Codes", Dict],
+            Tuple[Codes, JsonDict],
             # Deprecated
             bool,
         ]
@@ -70,7 +70,12 @@ USER_MAY_JOIN_ROOM_CALLBACK = Callable[
     Awaitable[
         Union[
             Literal["NOT_SPAM"],
-            "synapse.api.errors.Codes",
+            Codes,
+            # Highly experimental, not officially part of the spamchecker API, may
+            # disappear without warning depending on the results of ongoing
+            # experiments.
+            # Use this to return additional information as part of an error.
+            Tuple[Codes, JsonDict],
             # Deprecated
             bool,
         ]
@@ -81,7 +86,12 @@ USER_MAY_INVITE_CALLBACK = Callable[
     Awaitable[
         Union[
             Literal["NOT_SPAM"],
-            "synapse.api.errors.Codes",
+            Codes,
+            # Highly experimental, not officially part of the spamchecker API, may
+            # disappear without warning depending on the results of ongoing
+            # experiments.
+            # Use this to return additional information as part of an error.
+            Tuple[Codes, JsonDict],
             # Deprecated
             bool,
         ]
@@ -92,7 +102,12 @@ USER_MAY_SEND_3PID_INVITE_CALLBACK = Callable[
     Awaitable[
         Union[
             Literal["NOT_SPAM"],
-            "synapse.api.errors.Codes",
+            Codes,
+            # Highly experimental, not officially part of the spamchecker API, may
+            # disappear without warning depending on the results of ongoing
+            # experiments.
+            # Use this to return additional information as part of an error.
+            Tuple[Codes, JsonDict],
             # Deprecated
             bool,
         ]
@@ -103,7 +118,12 @@ USER_MAY_CREATE_ROOM_CALLBACK = Callable[
     Awaitable[
         Union[
             Literal["NOT_SPAM"],
-            "synapse.api.errors.Codes",
+            Codes,
+            # Highly experimental, not officially part of the spamchecker API, may
+            # disappear without warning depending on the results of ongoing
+            # experiments.
+            # Use this to return additional information as part of an error.
+            Tuple[Codes, JsonDict],
             # Deprecated
             bool,
         ]
@@ -114,7 +134,12 @@ USER_MAY_CREATE_ROOM_ALIAS_CALLBACK = Callable[
     Awaitable[
         Union[
             Literal["NOT_SPAM"],
-            "synapse.api.errors.Codes",
+            Codes,
+            # Highly experimental, not officially part of the spamchecker API, may
+            # disappear without warning depending on the results of ongoing
+            # experiments.
+            # Use this to return additional information as part of an error.
+            Tuple[Codes, JsonDict],
             # Deprecated
             bool,
         ]
@@ -125,7 +150,12 @@ USER_MAY_PUBLISH_ROOM_CALLBACK = Callable[
     Awaitable[
         Union[
             Literal["NOT_SPAM"],
-            "synapse.api.errors.Codes",
+            Codes,
+            # Highly experimental, not officially part of the spamchecker API, may
+            # disappear without warning depending on the results of ongoing
+            # experiments.
+            # Use this to return additional information as part of an error.
+            Tuple[Codes, JsonDict],
             # Deprecated
             bool,
         ]
@@ -154,7 +184,12 @@ CHECK_MEDIA_FILE_FOR_SPAM_CALLBACK = Callable[
     Awaitable[
         Union[
             Literal["NOT_SPAM"],
-            "synapse.api.errors.Codes",
+            Codes,
+            # Highly experimental, not officially part of the spamchecker API, may
+            # disappear without warning depending on the results of ongoing
+            # experiments.
+            # Use this to return additional information as part of an error.
+            Tuple[Codes, JsonDict],
             # Deprecated
             bool,
         ]
@@ -345,7 +380,7 @@ class SpamChecker:
 
     async def check_event_for_spam(
         self, event: "synapse.events.EventBase"
-    ) -> Union[Tuple["synapse.api.errors.Codes", Dict], str]:
+    ) -> Union[Tuple[Codes, JsonDict], str]:
         """Checks if a given event is considered "spammy" by this server.
 
         If the server considers an event spammy, then it will be rejected if
@@ -376,7 +411,16 @@ class SpamChecker:
                 elif res is True:
                     # This spam-checker rejects the event with deprecated
                     # return value `True`
-                    return (synapse.api.errors.Codes.FORBIDDEN, {})
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
+                elif (
+                    isinstance(res, tuple)
+                    and len(res) == 2
+                    and isinstance(res[0], synapse.api.errors.Codes)
+                    and isinstance(res[1], dict)
+                ):
+                    return res
+                elif isinstance(res, synapse.api.errors.Codes):
+                    return res, {}
                 elif not isinstance(res, str):
                     # mypy complains that we can't reach this code because of the
                     # return type in CHECK_EVENT_FOR_SPAM_CALLBACK, but we don't know
@@ -422,7 +466,7 @@ class SpamChecker:
 
     async def user_may_join_room(
         self, user_id: str, room_id: str, is_invited: bool
-    ) -> Union["synapse.api.errors.Codes", Literal["NOT_SPAM"]]:
+    ) -> Union[Tuple[Codes, JsonDict], Literal["NOT_SPAM"]]:
         """Checks if a given users is allowed to join a room.
         Not called when a user creates a room.
 
@@ -432,7 +476,7 @@ class SpamChecker:
             is_invited: Whether the user is invited into the room
 
         Returns:
-            NOT_SPAM if the operation is permitted, Codes otherwise.
+            NOT_SPAM if the operation is permitted, [Codes, Dict] otherwise.
         """
         for callback in self._user_may_join_room_callbacks:
             with Measure(
@@ -443,21 +487,28 @@ class SpamChecker:
                 if res is True or res is self.NOT_SPAM:
                     continue
                 elif res is False:
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
                 elif isinstance(res, synapse.api.errors.Codes):
+                    return res, {}
+                elif (
+                    isinstance(res, tuple)
+                    and len(res) == 2
+                    and isinstance(res[0], synapse.api.errors.Codes)
+                    and isinstance(res[1], dict)
+                ):
                     return res
                 else:
                     logger.warning(
                         "Module returned invalid value, rejecting join as spam"
                     )
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
 
         # No spam-checker has rejected the request, let it pass.
         return self.NOT_SPAM
 
     async def user_may_invite(
         self, inviter_userid: str, invitee_userid: str, room_id: str
-    ) -> Union["synapse.api.errors.Codes", Literal["NOT_SPAM"]]:
+    ) -> Union[Tuple[Codes, dict], Literal["NOT_SPAM"]]:
         """Checks if a given user may send an invite
 
         Args:
@@ -479,21 +530,28 @@ class SpamChecker:
                 if res is True or res is self.NOT_SPAM:
                     continue
                 elif res is False:
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
                 elif isinstance(res, synapse.api.errors.Codes):
+                    return res, {}
+                elif (
+                    isinstance(res, tuple)
+                    and len(res) == 2
+                    and isinstance(res[0], synapse.api.errors.Codes)
+                    and isinstance(res[1], dict)
+                ):
                     return res
                 else:
                     logger.warning(
                         "Module returned invalid value, rejecting invite as spam"
                     )
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
 
         # No spam-checker has rejected the request, let it pass.
         return self.NOT_SPAM
 
     async def user_may_send_3pid_invite(
         self, inviter_userid: str, medium: str, address: str, room_id: str
-    ) -> Union["synapse.api.errors.Codes", Literal["NOT_SPAM"]]:
+    ) -> Union[Tuple[Codes, dict], Literal["NOT_SPAM"]]:
         """Checks if a given user may invite a given threepid into the room
 
         Note that if the threepid is already associated with a Matrix user ID, Synapse
@@ -519,20 +577,27 @@ class SpamChecker:
                 if res is True or res is self.NOT_SPAM:
                     continue
                 elif res is False:
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
                 elif isinstance(res, synapse.api.errors.Codes):
+                    return res, {}
+                elif (
+                    isinstance(res, tuple)
+                    and len(res) == 2
+                    and isinstance(res[0], synapse.api.errors.Codes)
+                    and isinstance(res[1], dict)
+                ):
                     return res
                 else:
                     logger.warning(
                         "Module returned invalid value, rejecting 3pid invite as spam"
                     )
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
 
         return self.NOT_SPAM
 
     async def user_may_create_room(
         self, userid: str
-    ) -> Union["synapse.api.errors.Codes", Literal["NOT_SPAM"]]:
+    ) -> Union[Tuple[Codes, dict], Literal["NOT_SPAM"]]:
         """Checks if a given user may create a room
 
         Args:
@@ -546,20 +611,27 @@ class SpamChecker:
                 if res is True or res is self.NOT_SPAM:
                     continue
                 elif res is False:
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
                 elif isinstance(res, synapse.api.errors.Codes):
+                    return res, {}
+                elif (
+                    isinstance(res, tuple)
+                    and len(res) == 2
+                    and isinstance(res[0], synapse.api.errors.Codes)
+                    and isinstance(res[1], dict)
+                ):
                     return res
                 else:
                     logger.warning(
                         "Module returned invalid value, rejecting room creation as spam"
                     )
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
 
         return self.NOT_SPAM
 
     async def user_may_create_room_alias(
         self, userid: str, room_alias: RoomAlias
-    ) -> Union["synapse.api.errors.Codes", Literal["NOT_SPAM"]]:
+    ) -> Union[Tuple[Codes, dict], Literal["NOT_SPAM"]]:
         """Checks if a given user may create a room alias
 
         Args:
@@ -575,20 +647,27 @@ class SpamChecker:
                 if res is True or res is self.NOT_SPAM:
                     continue
                 elif res is False:
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
                 elif isinstance(res, synapse.api.errors.Codes):
+                    return res, {}
+                elif (
+                    isinstance(res, tuple)
+                    and len(res) == 2
+                    and isinstance(res[0], synapse.api.errors.Codes)
+                    and isinstance(res[1], dict)
+                ):
                     return res
                 else:
                     logger.warning(
                         "Module returned invalid value, rejecting room create as spam"
                     )
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
 
         return self.NOT_SPAM
 
     async def user_may_publish_room(
         self, userid: str, room_id: str
-    ) -> Union["synapse.api.errors.Codes", Literal["NOT_SPAM"]]:
+    ) -> Union[Tuple[Codes, dict], Literal["NOT_SPAM"]]:
         """Checks if a given user may publish a room to the directory
 
         Args:
@@ -603,14 +682,21 @@ class SpamChecker:
                 if res is True or res is self.NOT_SPAM:
                     continue
                 elif res is False:
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
                 elif isinstance(res, synapse.api.errors.Codes):
+                    return res, {}
+                elif (
+                    isinstance(res, tuple)
+                    and len(res) == 2
+                    and isinstance(res[0], synapse.api.errors.Codes)
+                    and isinstance(res[1], dict)
+                ):
                     return res
                 else:
                     logger.warning(
                         "Module returned invalid value, rejecting room publication as spam"
                     )
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
 
         return self.NOT_SPAM
 
@@ -678,7 +764,7 @@ class SpamChecker:
 
     async def check_media_file_for_spam(
         self, file_wrapper: ReadableFileWrapper, file_info: FileInfo
-    ) -> Union["synapse.api.errors.Codes", Literal["NOT_SPAM"]]:
+    ) -> Union[Tuple[Codes, dict], Literal["NOT_SPAM"]]:
         """Checks if a piece of newly uploaded media should be blocked.
 
         This will be called for local uploads, downloads of remote media, each
@@ -715,13 +801,20 @@ class SpamChecker:
                 if res is False or res is self.NOT_SPAM:
                     continue
                 elif res is True:
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
                 elif isinstance(res, synapse.api.errors.Codes):
+                    return res, {}
+                elif (
+                    isinstance(res, tuple)
+                    and len(res) == 2
+                    and isinstance(res[0], synapse.api.errors.Codes)
+                    and isinstance(res[1], dict)
+                ):
                     return res
                 else:
                     logger.warning(
                         "Module returned invalid value, rejecting media file as spam"
                     )
-                    return synapse.api.errors.Codes.FORBIDDEN
+                    return synapse.api.errors.Codes.FORBIDDEN, {}
 
         return self.NOT_SPAM
diff --git a/synapse/handlers/directory.py b/synapse/handlers/directory.py
index 8b0f16f965..09a7a4b238 100644
--- a/synapse/handlers/directory.py
+++ b/synapse/handlers/directory.py
@@ -149,7 +149,8 @@ class DirectoryHandler:
                 raise AuthError(
                     403,
                     "This user is not permitted to create this alias",
-                    spam_check,
+                    errcode=spam_check[0],
+                    additional_fields=spam_check[1],
                 )
 
             if not self.config.roomdirectory.is_alias_creation_allowed(
@@ -441,7 +442,8 @@ class DirectoryHandler:
             raise AuthError(
                 403,
                 "This user is not permitted to publish rooms to the room list",
-                spam_check,
+                errcode=spam_check[0],
+                additional_fields=spam_check[1],
             )
 
         if requester.is_guest:
diff --git a/synapse/handlers/federation.py b/synapse/handlers/federation.py
index e2564e9340..3b5eaf5156 100644
--- a/synapse/handlers/federation.py
+++ b/synapse/handlers/federation.py
@@ -844,7 +844,8 @@ class FederationHandler:
             raise SynapseError(
                 403,
                 "This user is not permitted to send invites to this server/user",
-                spam_check,
+                errcode=spam_check[0],
+                additional_fields=spam_check[1],
             )
 
         membership = event.content.get("membership")
diff --git a/synapse/handlers/room.py b/synapse/handlers/room.py
index 44f8084579..8dd94cbc76 100644
--- a/synapse/handlers/room.py
+++ b/synapse/handlers/room.py
@@ -440,7 +440,12 @@ class RoomCreationHandler:
 
         spam_check = await self.spam_checker.user_may_create_room(user_id)
         if spam_check != NOT_SPAM:
-            raise SynapseError(403, "You are not permitted to create rooms", spam_check)
+            raise SynapseError(
+                403,
+                "You are not permitted to create rooms",
+                errcode=spam_check[0],
+                additional_fields=spam_check[1],
+            )
 
         creation_content: JsonDict = {
             "room_version": new_room_version.identifier,
@@ -731,7 +736,10 @@ class RoomCreationHandler:
             spam_check = await self.spam_checker.user_may_create_room(user_id)
             if spam_check != NOT_SPAM:
                 raise SynapseError(
-                    403, "You are not permitted to create rooms", spam_check
+                    403,
+                    "You are not permitted to create rooms",
+                    errcode=spam_check[0],
+                    additional_fields=spam_check[1],
                 )
 
         if ratelimit:
diff --git a/synapse/handlers/room_member.py b/synapse/handlers/room_member.py
index a1d8875dd8..04c44b2ccb 100644
--- a/synapse/handlers/room_member.py
+++ b/synapse/handlers/room_member.py
@@ -685,7 +685,7 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
             if target_id == self._server_notices_mxid:
                 raise SynapseError(HTTPStatus.FORBIDDEN, "Cannot invite this user")
 
-            block_invite_code = None
+            block_invite_result = None
 
             if (
                 self._server_notices_mxid is not None
@@ -703,18 +703,21 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
                         "Blocking invite: user is not admin and non-admin "
                         "invites disabled"
                     )
-                    block_invite_code = Codes.FORBIDDEN
+                    block_invite_result = (Codes.FORBIDDEN, {})
 
                 spam_check = await self.spam_checker.user_may_invite(
                     requester.user.to_string(), target_id, room_id
                 )
                 if spam_check != NOT_SPAM:
                     logger.info("Blocking invite due to spam checker")
-                    block_invite_code = spam_check
+                    block_invite_result = spam_check
 
-            if block_invite_code is not None:
+            if block_invite_result is not None:
                 raise SynapseError(
-                    403, "Invites have been disabled on this server", block_invite_code
+                    403,
+                    "Invites have been disabled on this server",
+                    errcode=block_invite_result[0],
+                    additional_fields=block_invite_result[1],
                 )
 
         # An empty prev_events list is allowed as long as the auth_event_ids are present
@@ -828,7 +831,12 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
                     target.to_string(), room_id, is_invited=inviter is not None
                 )
                 if spam_check != NOT_SPAM:
-                    raise SynapseError(403, "Not allowed to join this room", spam_check)
+                    raise SynapseError(
+                        403,
+                        "Not allowed to join this room",
+                        errcode=spam_check[0],
+                        additional_fields=spam_check[1],
+                    )
 
             # Check if a remote join should be performed.
             remote_join, remote_room_hosts = await self._should_perform_remote_join(
@@ -1387,7 +1395,12 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
                 room_id=room_id,
             )
             if spam_check != NOT_SPAM:
-                raise SynapseError(403, "Cannot send threepid invite", spam_check)
+                raise SynapseError(
+                    403,
+                    "Cannot send threepid invite",
+                    errcode=spam_check[0],
+                    additional_fields=spam_check[1],
+                )
 
             stream_id = await self._make_and_store_3pid_invite(
                 requester,
diff --git a/synapse/module_api/__init__.py b/synapse/module_api/__init__.py
index 6191c2dc96..6d8bf54083 100644
--- a/synapse/module_api/__init__.py
+++ b/synapse/module_api/__init__.py
@@ -35,6 +35,7 @@ from typing_extensions import ParamSpec
 from twisted.internet import defer
 from twisted.web.resource import Resource
 
+from synapse.api import errors
 from synapse.api.errors import SynapseError
 from synapse.events import EventBase
 from synapse.events.presence_router import (
diff --git a/synapse/rest/media/v1/media_storage.py b/synapse/rest/media/v1/media_storage.py
index 9137417342..a5c3de192f 100644
--- a/synapse/rest/media/v1/media_storage.py
+++ b/synapse/rest/media/v1/media_storage.py
@@ -154,7 +154,9 @@ class MediaStorage:
                         # Note that we'll delete the stored media, due to the
                         # try/except below. The media also won't be stored in
                         # the DB.
-                        raise SpamMediaException(errcode=spam_check)
+                        # We currently ignore any additional field returned by
+                        # the spam-check API.
+                        raise SpamMediaException(errcode=spam_check[0])
 
                     for provider in self.storage_providers:
                         await provider.store_file(path, file_info)