diff options
Diffstat (limited to 'synapse/rest')
| -rw-r--r-- | synapse/rest/admin/__init__.py | 2 | ||||
| -rw-r--r-- | synapse/rest/admin/devices.py | 2 | ||||
| -rw-r--r-- | synapse/rest/admin/media.py | 33 | ||||
| -rw-r--r-- | synapse/rest/admin/rooms.py | 141 | ||||
| -rw-r--r-- | synapse/rest/admin/users.py | 50 | ||||
| -rw-r--r-- | synapse/rest/client/relations.py | 11 | ||||
| -rw-r--r-- | synapse/rest/client/room.py | 2 | ||||
| -rw-r--r-- | synapse/rest/client/room_batch.py | 42 | ||||
| -rw-r--r-- | synapse/rest/client/versions.py | 1 | ||||
| -rw-r--r-- | synapse/rest/media/v1/filepath.py | 26 | ||||
| -rw-r--r-- | synapse/rest/media/v1/media_repository.py | 12 | ||||
| -rw-r--r-- | synapse/rest/media/v1/oembed.py | 15 | ||||
| -rw-r--r-- | synapse/rest/media/v1/preview_url_resource.py | 113 | ||||
| -rw-r--r-- | synapse/rest/media/v1/upload_resource.py | 2 | ||||
| -rw-r--r-- | synapse/rest/well_known.py | 47 |
15 files changed, 270 insertions, 229 deletions
diff --git a/synapse/rest/admin/__init__.py b/synapse/rest/admin/__init__.py index e1506deb2b..70514e814f 100644 --- a/synapse/rest/admin/__init__.py +++ b/synapse/rest/admin/__init__.py @@ -42,7 +42,6 @@ from synapse.rest.admin.registration_tokens import ( RegistrationTokenRestServlet, ) from synapse.rest.admin.rooms import ( - DeleteRoomRestServlet, ForwardExtremitiesRestServlet, JoinRoomAliasServlet, ListRoomRestServlet, @@ -221,7 +220,6 @@ def register_servlets(hs: "HomeServer", http_server: HttpServer) -> None: RoomStateRestServlet(hs).register(http_server) RoomRestServlet(hs).register(http_server) RoomMembersRestServlet(hs).register(http_server) - DeleteRoomRestServlet(hs).register(http_server) JoinRoomAliasServlet(hs).register(http_server) VersionServlet(hs).register(http_server) UserAdminServlet(hs).register(http_server) diff --git a/synapse/rest/admin/devices.py b/synapse/rest/admin/devices.py index a6fa03c90f..80fbf32f17 100644 --- a/synapse/rest/admin/devices.py +++ b/synapse/rest/admin/devices.py @@ -110,7 +110,7 @@ class DevicesRestServlet(RestServlet): def __init__(self, hs: "HomeServer"): """ Args: - hs (synapse.server.HomeServer): server + hs: server """ self.hs = hs self.auth = hs.get_auth() diff --git a/synapse/rest/admin/media.py b/synapse/rest/admin/media.py index 8ce443049e..30a687d234 100644 --- a/synapse/rest/admin/media.py +++ b/synapse/rest/admin/media.py @@ -40,7 +40,7 @@ class QuarantineMediaInRoom(RestServlet): """ PATTERNS = [ - *admin_patterns("/room/(?P<room_id>[^/]+)/media/quarantine"), + *admin_patterns("/room/(?P<room_id>[^/]+)/media/quarantine$"), # This path kept around for legacy reasons *admin_patterns("/quarantine_media/(?P<room_id>[^/]+)"), ] @@ -70,7 +70,7 @@ class QuarantineMediaByUser(RestServlet): this server. """ - PATTERNS = admin_patterns("/user/(?P<user_id>[^/]+)/media/quarantine") + PATTERNS = admin_patterns("/user/(?P<user_id>[^/]+)/media/quarantine$") def __init__(self, hs: "HomeServer"): self.store = hs.get_datastore() @@ -199,7 +199,7 @@ class UnprotectMediaByID(RestServlet): class ListMediaInRoom(RestServlet): """Lists all of the media in a given room.""" - PATTERNS = admin_patterns("/room/(?P<room_id>[^/]+)/media") + PATTERNS = admin_patterns("/room/(?P<room_id>[^/]+)/media$") def __init__(self, hs: "HomeServer"): self.store = hs.get_datastore() @@ -219,7 +219,7 @@ class ListMediaInRoom(RestServlet): class PurgeMediaCacheRestServlet(RestServlet): - PATTERNS = admin_patterns("/purge_media_cache") + PATTERNS = admin_patterns("/purge_media_cache$") def __init__(self, hs: "HomeServer"): self.media_repository = hs.get_media_repository() @@ -231,6 +231,20 @@ class PurgeMediaCacheRestServlet(RestServlet): before_ts = parse_integer(request, "before_ts", required=True) logger.info("before_ts: %r", before_ts) + if before_ts < 0: + raise SynapseError( + 400, + "Query parameter before_ts must be a positive integer.", + errcode=Codes.INVALID_PARAM, + ) + elif before_ts < 30000000000: # Dec 1970 in milliseconds, Aug 2920 in seconds + raise SynapseError( + 400, + "Query parameter before_ts you provided is from the year 1970. " + + "Double check that you are providing a timestamp in milliseconds.", + errcode=Codes.INVALID_PARAM, + ) + ret = await self.media_repository.delete_old_remote_media(before_ts) return 200, ret @@ -271,7 +285,7 @@ class DeleteMediaByDateSize(RestServlet): timestamp and size. """ - PATTERNS = admin_patterns("/media/(?P<server_name>[^/]+)/delete") + PATTERNS = admin_patterns("/media/(?P<server_name>[^/]+)/delete$") def __init__(self, hs: "HomeServer"): self.store = hs.get_datastore() @@ -291,7 +305,14 @@ class DeleteMediaByDateSize(RestServlet): if before_ts < 0: raise SynapseError( 400, - "Query parameter before_ts must be a string representing a positive integer.", + "Query parameter before_ts must be a positive integer.", + errcode=Codes.INVALID_PARAM, + ) + elif before_ts < 30000000000: # Dec 1970 in milliseconds, Aug 2920 in seconds + raise SynapseError( + 400, + "Query parameter before_ts you provided is from the year 1970. " + + "Double check that you are providing a timestamp in milliseconds.", errcode=Codes.INVALID_PARAM, ) if size_gt < 0: diff --git a/synapse/rest/admin/rooms.py b/synapse/rest/admin/rooms.py index a4823ca6e7..05c5b4bf0c 100644 --- a/synapse/rest/admin/rooms.py +++ b/synapse/rest/admin/rooms.py @@ -46,41 +46,6 @@ if TYPE_CHECKING: logger = logging.getLogger(__name__) -class DeleteRoomRestServlet(RestServlet): - """Delete a room from server. - - It is a combination and improvement of shutdown and purge room. - - Shuts down a room by removing all local users from the room. - Blocking all future invites and joins to the room is optional. - - If desired any local aliases will be repointed to a new room - created by `new_room_user_id` and kicked users will be auto- - joined to the new room. - - If 'purge' is true, it will remove all traces of a room from the database. - """ - - PATTERNS = admin_patterns("/rooms/(?P<room_id>[^/]+)/delete$") - - def __init__(self, hs: "HomeServer"): - self.hs = hs - self.auth = hs.get_auth() - self.room_shutdown_handler = hs.get_room_shutdown_handler() - self.pagination_handler = hs.get_pagination_handler() - - async def on_POST( - self, request: SynapseRequest, room_id: str - ) -> Tuple[int, JsonDict]: - return await _delete_room( - request, - room_id, - self.auth, - self.room_shutdown_handler, - self.pagination_handler, - ) - - class ListRoomRestServlet(RestServlet): """ List all rooms that are known to the homeserver. Results are returned @@ -218,7 +183,7 @@ class RoomRestServlet(RestServlet): async def on_DELETE( self, request: SynapseRequest, room_id: str ) -> Tuple[int, JsonDict]: - return await _delete_room( + return await self._delete_room( request, room_id, self.auth, @@ -226,6 +191,58 @@ class RoomRestServlet(RestServlet): self.pagination_handler, ) + async def _delete_room( + self, + request: SynapseRequest, + room_id: str, + auth: "Auth", + room_shutdown_handler: "RoomShutdownHandler", + pagination_handler: "PaginationHandler", + ) -> Tuple[int, JsonDict]: + requester = await auth.get_user_by_req(request) + await assert_user_is_admin(auth, requester.user) + + content = parse_json_object_from_request(request) + + block = content.get("block", False) + if not isinstance(block, bool): + raise SynapseError( + HTTPStatus.BAD_REQUEST, + "Param 'block' must be a boolean, if given", + Codes.BAD_JSON, + ) + + purge = content.get("purge", True) + if not isinstance(purge, bool): + raise SynapseError( + HTTPStatus.BAD_REQUEST, + "Param 'purge' must be a boolean, if given", + Codes.BAD_JSON, + ) + + force_purge = content.get("force_purge", False) + if not isinstance(force_purge, bool): + raise SynapseError( + HTTPStatus.BAD_REQUEST, + "Param 'force_purge' must be a boolean, if given", + Codes.BAD_JSON, + ) + + ret = await room_shutdown_handler.shutdown_room( + room_id=room_id, + new_room_user_id=content.get("new_room_user_id"), + new_room_name=content.get("room_name"), + message=content.get("message"), + requester_user_id=requester.user.to_string(), + block=block, + ) + + # Purge room + if purge: + await pagination_handler.purge_room(room_id, force=force_purge) + + return 200, ret + class RoomMembersRestServlet(RestServlet): """ @@ -617,55 +634,3 @@ class RoomEventContextServlet(RestServlet): ) return 200, results - - -async def _delete_room( - request: SynapseRequest, - room_id: str, - auth: "Auth", - room_shutdown_handler: "RoomShutdownHandler", - pagination_handler: "PaginationHandler", -) -> Tuple[int, JsonDict]: - requester = await auth.get_user_by_req(request) - await assert_user_is_admin(auth, requester.user) - - content = parse_json_object_from_request(request) - - block = content.get("block", False) - if not isinstance(block, bool): - raise SynapseError( - HTTPStatus.BAD_REQUEST, - "Param 'block' must be a boolean, if given", - Codes.BAD_JSON, - ) - - purge = content.get("purge", True) - if not isinstance(purge, bool): - raise SynapseError( - HTTPStatus.BAD_REQUEST, - "Param 'purge' must be a boolean, if given", - Codes.BAD_JSON, - ) - - force_purge = content.get("force_purge", False) - if not isinstance(force_purge, bool): - raise SynapseError( - HTTPStatus.BAD_REQUEST, - "Param 'force_purge' must be a boolean, if given", - Codes.BAD_JSON, - ) - - ret = await room_shutdown_handler.shutdown_room( - room_id=room_id, - new_room_user_id=content.get("new_room_user_id"), - new_room_name=content.get("room_name"), - message=content.get("message"), - requester_user_id=requester.user.to_string(), - block=block, - ) - - # Purge room - if purge: - await pagination_handler.purge_room(room_id, force=force_purge) - - return 200, ret diff --git a/synapse/rest/admin/users.py b/synapse/rest/admin/users.py index f20aa65301..d14fafbbc9 100644 --- a/synapse/rest/admin/users.py +++ b/synapse/rest/admin/users.py @@ -35,6 +35,7 @@ from synapse.rest.admin._base import ( assert_user_is_admin, ) from synapse.rest.client._base import client_patterns +from synapse.storage.databases.main.registration import ExternalIDReuseException from synapse.storage.databases.main.stats import UserSortOrder from synapse.types import JsonDict, UserID @@ -228,12 +229,12 @@ class UserRestServletV2(RestServlet): if not isinstance(deactivate, bool): raise SynapseError(400, "'deactivated' parameter is not of type boolean") - # convert List[Dict[str, str]] into Set[Tuple[str, str]] + # convert List[Dict[str, str]] into List[Tuple[str, str]] if external_ids is not None: - new_external_ids = { + new_external_ids = [ (external_id["auth_provider"], external_id["external_id"]) for external_id in external_ids - } + ] # convert List[Dict[str, str]] into Set[Tuple[str, str]] if threepids is not None: @@ -275,28 +276,13 @@ class UserRestServletV2(RestServlet): ) if external_ids is not None: - # get changed external_ids (added and removed) - cur_external_ids = set( - await self.store.get_external_ids_by_user(user_id) - ) - add_external_ids = new_external_ids - cur_external_ids - del_external_ids = cur_external_ids - new_external_ids - - # remove old external_ids - for auth_provider, external_id in del_external_ids: - await self.store.remove_user_external_id( - auth_provider, - external_id, - user_id, - ) - - # add new external_ids - for auth_provider, external_id in add_external_ids: - await self.store.record_user_external_id( - auth_provider, - external_id, + try: + await self.store.replace_user_external_id( + new_external_ids, user_id, ) + except ExternalIDReuseException: + raise SynapseError(409, "External id is already in use.") if "avatar_url" in body and isinstance(body["avatar_url"], str): await self.profile_handler.set_avatar_url( @@ -340,6 +326,9 @@ class UserRestServletV2(RestServlet): target_user.to_string() ) + if "user_type" in body: + await self.store.set_user_type(target_user, user_type) + user = await self.admin_handler.get_user(target_user) assert user is not None @@ -384,12 +373,15 @@ class UserRestServletV2(RestServlet): ) if external_ids is not None: - for auth_provider, external_id in new_external_ids: - await self.store.record_user_external_id( - auth_provider, - external_id, - user_id, - ) + try: + for auth_provider, external_id in new_external_ids: + await self.store.record_user_external_id( + auth_provider, + external_id, + user_id, + ) + except ExternalIDReuseException: + raise SynapseError(409, "External id is already in use.") if "avatar_url" in body and isinstance(body["avatar_url"], str): await self.profile_handler.set_avatar_url( diff --git a/synapse/rest/client/relations.py b/synapse/rest/client/relations.py index 0b0711c03c..58f6699073 100644 --- a/synapse/rest/client/relations.py +++ b/synapse/rest/client/relations.py @@ -128,9 +128,10 @@ class RelationSendServlet(RestServlet): content["m.relates_to"] = { "event_id": parent_id, - "key": aggregation_key, "rel_type": relation_type, } + if aggregation_key is not None: + content["m.relates_to"]["key"] = aggregation_key event_dict = { "type": event_type, @@ -232,12 +233,12 @@ class RelationPaginationServlet(RestServlet): # Similarly, we don't allow relations to be applied to relations, so we # return the original relations without any aggregations on top of them # here. - events = await self._event_serializer.serialize_events( + serialized_events = await self._event_serializer.serialize_events( events, now, bundle_aggregations=False ) return_value = pagination_chunk.to_dict() - return_value["chunk"] = events + return_value["chunk"] = serialized_events return_value["original_event"] = original_event return 200, return_value @@ -416,10 +417,10 @@ class RelationAggregationGroupPaginationServlet(RestServlet): ) now = self.clock.time_msec() - events = await self._event_serializer.serialize_events(events, now) + serialized_events = await self._event_serializer.serialize_events(events, now) return_value = result.to_dict() - return_value["chunk"] = events + return_value["chunk"] = serialized_events return 200, return_value diff --git a/synapse/rest/client/room.py b/synapse/rest/client/room.py index ed95189b6d..6a876cfa2f 100644 --- a/synapse/rest/client/room.py +++ b/synapse/rest/client/room.py @@ -914,7 +914,7 @@ class RoomTypingRestServlet(RestServlet): # If we're not on the typing writer instance we should scream if we get # requests. self._is_typing_writer = ( - hs.config.worker.writers.typing == hs.get_instance_name() + hs.get_instance_name() in hs.config.worker.writers.typing ) async def on_PUT( diff --git a/synapse/rest/client/room_batch.py b/synapse/rest/client/room_batch.py index 38ad4c2447..46f033eee2 100644 --- a/synapse/rest/client/room_batch.py +++ b/synapse/rest/client/room_batch.py @@ -32,7 +32,6 @@ from synapse.http.servlet import ( from synapse.http.site import SynapseRequest from synapse.rest.client.transactions import HttpTransactionCache from synapse.types import JsonDict -from synapse.util.stringutils import random_string if TYPE_CHECKING: from synapse.server import HomeServer @@ -132,20 +131,22 @@ class RoomBatchSendEventRestServlet(RestServlet): prev_event_ids_from_query ) + state_event_ids_at_start = [] # Create and persist all of the state events that float off on their own # before the batch. These will most likely be all of the invite/member # state events used to auth the upcoming historical messages. - state_event_ids_at_start = ( - await self.room_batch_handler.persist_state_events_at_start( - state_events_at_start=body["state_events_at_start"], - room_id=room_id, - initial_auth_event_ids=auth_event_ids, - app_service_requester=requester, + if body["state_events_at_start"]: + state_event_ids_at_start = ( + await self.room_batch_handler.persist_state_events_at_start( + state_events_at_start=body["state_events_at_start"], + room_id=room_id, + initial_auth_event_ids=auth_event_ids, + app_service_requester=requester, + ) ) - ) - # Update our ongoing auth event ID list with all of the new state we - # just created - auth_event_ids.extend(state_event_ids_at_start) + # Update our ongoing auth event ID list with all of the new state we + # just created + auth_event_ids.extend(state_event_ids_at_start) inherited_depth = await self.room_batch_handler.inherit_depth_from_prev_ids( prev_event_ids_from_query @@ -160,11 +161,6 @@ class RoomBatchSendEventRestServlet(RestServlet): base_insertion_event = None if batch_id_from_query: batch_id_to_connect_to = batch_id_from_query - # All but the first base insertion event should point at a fake - # event, which causes the HS to ask for the state at the start of - # the batch later. - fake_prev_event_id = "$" + random_string(43) - prev_event_ids = [fake_prev_event_id] # Otherwise, create an insertion event to act as a starting point. # # We don't always have an insertion event to start hanging more history @@ -173,8 +169,6 @@ class RoomBatchSendEventRestServlet(RestServlet): # an insertion event), in which case we just create a new insertion event # that can then get pointed to by a "marker" event later. else: - prev_event_ids = prev_event_ids_from_query - base_insertion_event_dict = ( self.room_batch_handler.create_insertion_event_dict( sender=requester.user.to_string(), @@ -182,7 +176,7 @@ class RoomBatchSendEventRestServlet(RestServlet): origin_server_ts=last_event_in_batch["origin_server_ts"], ) ) - base_insertion_event_dict["prev_events"] = prev_event_ids.copy() + base_insertion_event_dict["prev_events"] = prev_event_ids_from_query.copy() ( base_insertion_event, @@ -199,10 +193,18 @@ class RoomBatchSendEventRestServlet(RestServlet): depth=inherited_depth, ) - batch_id_to_connect_to = base_insertion_event["content"][ + batch_id_to_connect_to = base_insertion_event.content[ EventContentFields.MSC2716_NEXT_BATCH_ID ] + # Also connect the historical event chain to the end of the floating + # state chain, which causes the HS to ask for the state at the start of + # the batch later. If there is no state chain to connect to, just make + # the insertion event float itself. + prev_event_ids = [] + if len(state_event_ids_at_start): + prev_event_ids = [state_event_ids_at_start[-1]] + # Create and persist all of the historical events as well as insertion # and batch meta events to make the batch navigable in the DAG. event_ids, next_batch_id = await self.room_batch_handler.handle_batch_of_events( diff --git a/synapse/rest/client/versions.py b/synapse/rest/client/versions.py index b52a296d8f..8d888f4565 100644 --- a/synapse/rest/client/versions.py +++ b/synapse/rest/client/versions.py @@ -72,6 +72,7 @@ class VersionsRestServlet(RestServlet): "r0.4.0", "r0.5.0", "r0.6.0", + "r0.6.1", ], # as per MSC1497: "unstable_features": { diff --git a/synapse/rest/media/v1/filepath.py b/synapse/rest/media/v1/filepath.py index 08bd85f664..bec77088ee 100644 --- a/synapse/rest/media/v1/filepath.py +++ b/synapse/rest/media/v1/filepath.py @@ -16,12 +16,15 @@ import functools import os import re -from typing import Any, Callable, List +from typing import Any, Callable, List, TypeVar, cast NEW_FORMAT_ID_RE = re.compile(r"^\d\d\d\d-\d\d-\d\d") -def _wrap_in_base_path(func: Callable[..., str]) -> Callable[..., str]: +F = TypeVar("F", bound=Callable[..., str]) + + +def _wrap_in_base_path(func: F) -> F: """Takes a function that returns a relative path and turns it into an absolute path based on the location of the primary media store """ @@ -31,7 +34,7 @@ def _wrap_in_base_path(func: Callable[..., str]) -> Callable[..., str]: path = func(self, *args, **kwargs) return os.path.join(self.base_path, path) - return _wrapped + return cast(F, _wrapped) class MediaFilePaths: @@ -45,23 +48,6 @@ class MediaFilePaths: def __init__(self, primary_base_path: str): self.base_path = primary_base_path - def default_thumbnail_rel( - self, - default_top_level: str, - default_sub_type: str, - width: int, - height: int, - content_type: str, - method: str, - ) -> str: - top_level_type, sub_type = content_type.split("/") - file_name = "%i-%i-%s-%s-%s" % (width, height, top_level_type, sub_type, method) - return os.path.join( - "default_thumbnails", default_top_level, default_sub_type, file_name - ) - - default_thumbnail = _wrap_in_base_path(default_thumbnail_rel) - def local_media_filepath_rel(self, media_id: str) -> str: return os.path.join("local_content", media_id[0:2], media_id[2:4], media_id[4:]) diff --git a/synapse/rest/media/v1/media_repository.py b/synapse/rest/media/v1/media_repository.py index abd88a2d4f..244ba261bb 100644 --- a/synapse/rest/media/v1/media_repository.py +++ b/synapse/rest/media/v1/media_repository.py @@ -215,6 +215,8 @@ class MediaRepository: self.mark_recently_accessed(None, media_id) media_type = media_info["media_type"] + if not media_type: + media_type = "application/octet-stream" media_length = media_info["media_length"] upload_name = name if name else media_info["upload_name"] url_cache = media_info["url_cache"] @@ -333,6 +335,9 @@ class MediaRepository: logger.info("Media is quarantined") raise NotFoundError() + if not media_info["media_type"]: + media_info["media_type"] = "application/octet-stream" + responder = await self.media_storage.fetch_media(file_info) if responder: return responder, media_info @@ -354,6 +359,8 @@ class MediaRepository: raise e file_id = media_info["filesystem_id"] + if not media_info["media_type"]: + media_info["media_type"] = "application/octet-stream" file_info = FileInfo(server_name, file_id) # We generate thumbnails even if another process downloaded the media @@ -445,7 +452,10 @@ class MediaRepository: await finish() - media_type = headers[b"Content-Type"][0].decode("ascii") + if b"Content-Type" in headers: + media_type = headers[b"Content-Type"][0].decode("ascii") + else: + media_type = "application/octet-stream" upload_name = get_filename_from_headers(headers) time_now_ms = self.clock.time_msec() diff --git a/synapse/rest/media/v1/oembed.py b/synapse/rest/media/v1/oembed.py index 6d7e1f9064..2a59552c20 100644 --- a/synapse/rest/media/v1/oembed.py +++ b/synapse/rest/media/v1/oembed.py @@ -17,7 +17,6 @@ from typing import TYPE_CHECKING, List, Optional import attr -from synapse.http.client import SimpleHttpClient from synapse.types import JsonDict from synapse.util import json_decoder @@ -48,7 +47,7 @@ class OEmbedProvider: requesting/parsing oEmbed content. """ - def __init__(self, hs: "HomeServer", client: SimpleHttpClient): + def __init__(self, hs: "HomeServer"): self._oembed_patterns = {} for oembed_endpoint in hs.config.oembed.oembed_patterns: api_endpoint = oembed_endpoint.api_endpoint @@ -69,7 +68,6 @@ class OEmbedProvider: # Iterate through each URL pattern and point it to the endpoint. for pattern in oembed_endpoint.url_patterns: self._oembed_patterns[pattern] = api_endpoint - self._client = client def get_oembed_url(self, url: str) -> Optional[str]: """ @@ -139,10 +137,11 @@ class OEmbedProvider: # oEmbed responses *must* be UTF-8 according to the spec. oembed = json_decoder.decode(raw_body.decode("utf-8")) - # Ensure there's a version of 1.0. - oembed_version = oembed["version"] - if oembed_version != "1.0": - raise RuntimeError(f"Invalid version: {oembed_version}") + # The version is a required string field, but not always provided, + # or sometimes provided as a float. Be lenient. + oembed_version = oembed.get("version", "1.0") + if oembed_version != "1.0" and oembed_version != 1: + raise RuntimeError(f"Invalid oEmbed version: {oembed_version}") # Ensure the cache age is None or an int. cache_age = oembed.get("cache_age") @@ -191,7 +190,7 @@ class OEmbedProvider: except Exception as e: # Trap any exception and let the code follow as usual. - logger.warning(f"Error parsing oEmbed metadata from {url}: {e:r}") + logger.warning("Error parsing oEmbed metadata from %s: %r", url, e) open_graph_response = {} cache_age = None diff --git a/synapse/rest/media/v1/preview_url_resource.py b/synapse/rest/media/v1/preview_url_resource.py index 1fe0fc8aa9..8ca97b5b18 100644 --- a/synapse/rest/media/v1/preview_url_resource.py +++ b/synapse/rest/media/v1/preview_url_resource.py @@ -12,6 +12,7 @@ # 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 codecs import datetime import errno import fnmatch @@ -22,7 +23,7 @@ import re import shutil import sys import traceback -from typing import TYPE_CHECKING, Dict, Generator, Iterable, Optional, Tuple, Union +from typing import TYPE_CHECKING, Dict, Generator, Iterable, Optional, Set, Tuple, Union from urllib import parse as urlparse import attr @@ -140,7 +141,7 @@ class PreviewUrlResource(DirectServeJsonResource): self.primary_base_path = media_repo.primary_base_path self.media_storage = media_storage - self._oembed = OEmbedProvider(hs, self.client) + self._oembed = OEmbedProvider(hs) # We run the background jobs if we're the instance specified (or no # instance is specified, where we assume there is only one instance @@ -295,8 +296,7 @@ class PreviewUrlResource(DirectServeJsonResource): with open(media_info.filename, "rb") as file: body = file.read() - encoding = get_html_media_encoding(body, media_info.media_type) - tree = decode_body(body, encoding) + tree = decode_body(body, media_info.uri, media_info.media_type) if tree is not None: # Check if this HTML document points to oEmbed information and # defer to that. @@ -632,16 +632,27 @@ class PreviewUrlResource(DirectServeJsonResource): logger.debug("No media removed from url cache") -def get_html_media_encoding(body: bytes, content_type: str) -> str: +def _normalise_encoding(encoding: str) -> Optional[str]: + """Use the Python codec's name as the normalised entry.""" + try: + return codecs.lookup(encoding).name + except LookupError: + return None + + +def get_html_media_encodings(body: bytes, content_type: Optional[str]) -> Iterable[str]: """ - Get the encoding of the body based on the (presumably) HTML body or media_type. + Get potential encoding of the body based on the (presumably) HTML body or the content-type header. The precedence used for finding a character encoding is: - 1. meta tag with a charset declared. + 1. <meta> tag with a charset declared. 2. The XML document's character encoding attribute. 3. The Content-Type header. - 4. Fallback to UTF-8. + 4. Fallback to utf-8. + 5. Fallback to windows-1252. + + This roughly follows the algorithm used by BeautifulSoup's bs4.dammit.EncodingDetector. Args: body: The HTML document, as bytes. @@ -650,39 +661,55 @@ def get_html_media_encoding(body: bytes, content_type: str) -> str: Returns: The character encoding of the body, as a string. """ + # There's no point in returning an encoding more than once. + attempted_encodings: Set[str] = set() + # Limit searches to the first 1kb, since it ought to be at the top. body_start = body[:1024] - # Let's try and figure out if it has an encoding set in a meta tag. + # Check if it has an encoding set in a meta tag. match = _charset_match.search(body_start) if match: - return match.group(1).decode("ascii") + encoding = _normalise_encoding(match.group(1).decode("ascii")) + if encoding: + attempted_encodings.add(encoding) + yield encoding # TODO Support <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> - # If we didn't find a match, see if it an XML document with an encoding. + # Check if it has an XML document with an encoding. match = _xml_encoding_match.match(body_start) if match: - return match.group(1).decode("ascii") - - # If we don't find a match, we'll look at the HTTP Content-Type, and - # if that doesn't exist, we'll fall back to UTF-8. - content_match = _content_type_match.match(content_type) - if content_match: - return content_match.group(1) - - return "utf-8" + encoding = _normalise_encoding(match.group(1).decode("ascii")) + if encoding and encoding not in attempted_encodings: + attempted_encodings.add(encoding) + yield encoding + + # Check the HTTP Content-Type header for a character set. + if content_type: + content_match = _content_type_match.match(content_type) + if content_match: + encoding = _normalise_encoding(content_match.group(1)) + if encoding and encoding not in attempted_encodings: + attempted_encodings.add(encoding) + yield encoding + + # Finally, fallback to UTF-8, then windows-1252. + for fallback in ("utf-8", "cp1252"): + if fallback not in attempted_encodings: + yield fallback def decode_body( - body: bytes, request_encoding: Optional[str] = None + body: bytes, uri: str, content_type: Optional[str] = None ) -> Optional["etree.Element"]: """ This uses lxml to parse the HTML document. Args: body: The HTML document, as bytes. - request_encoding: The character encoding of the body, as a string. + uri: The URI used to download the body. + content_type: The Content-Type header. Returns: The parsed HTML body, or None if an error occurred during processed. @@ -691,32 +718,28 @@ def decode_body( if not body: return None - from lxml import etree - - # Create an HTML parser. If this fails, log and return no metadata. - try: - parser = etree.HTMLParser(recover=True, encoding=request_encoding) - except LookupError: - # blindly consider the encoding as utf-8. - parser = etree.HTMLParser(recover=True, encoding="utf-8") - except Exception as e: - logger.warning("Unable to create HTML parser: %s" % (e,)) + # The idea here is that multiple encodings are tried until one works. + # Unfortunately the result is never used and then LXML will decode the string + # again with the found encoding. + for encoding in get_html_media_encodings(body, content_type): + try: + body.decode(encoding) + except Exception: + pass + else: + break + else: + logger.warning("Unable to decode HTML body for %s", uri) return None - def _attempt_decode_body( - body_attempt: Union[bytes, str] - ) -> Optional["etree.Element"]: - # Attempt to parse the body. Returns None if the body was successfully - # parsed, but no tree was found. - return etree.fromstring(body_attempt, parser) + from lxml import etree - # Attempt to parse the body. If this fails, log and return no metadata. - try: - return _attempt_decode_body(body) - except UnicodeDecodeError: - # blindly try decoding the body as utf-8, which seems to fix - # the charset mismatches on https://google.com - return _attempt_decode_body(body.decode("utf-8", "ignore")) + # Create an HTML parser. + parser = etree.HTMLParser(recover=True, encoding=encoding) + + # Attempt to parse the body. Returns None if the body was successfully + # parsed, but no tree was found. + return etree.fromstring(body, parser) def _calc_og(tree: "etree.Element", media_uri: str) -> Dict[str, Optional[str]]: diff --git a/synapse/rest/media/v1/upload_resource.py b/synapse/rest/media/v1/upload_resource.py index 7dcb1428e4..8162094cf6 100644 --- a/synapse/rest/media/v1/upload_resource.py +++ b/synapse/rest/media/v1/upload_resource.py @@ -80,7 +80,7 @@ class UploadResource(DirectServeJsonResource): assert content_type_headers # for mypy media_type = content_type_headers[0].decode("ascii") else: - raise SynapseError(msg="Upload request missing 'Content-Type'", code=400) + media_type = "application/octet-stream" # if headers.hasHeader(b"Content-Disposition"): # disposition = headers.getRawHeaders(b"Content-Disposition")[0] diff --git a/synapse/rest/well_known.py b/synapse/rest/well_known.py index 7ac01faab4..edbf5ce5d0 100644 --- a/synapse/rest/well_known.py +++ b/synapse/rest/well_known.py @@ -21,6 +21,7 @@ from twisted.web.server import Request from synapse.http.server import set_cors_headers from synapse.types import JsonDict from synapse.util import json_encoder +from synapse.util.stringutils import parse_server_name if TYPE_CHECKING: from synapse.server import HomeServer @@ -47,8 +48,8 @@ class WellKnownBuilder: return result -class WellKnownResource(Resource): - """A Twisted web resource which renders the .well-known file""" +class ClientWellKnownResource(Resource): + """A Twisted web resource which renders the .well-known/matrix/client file""" isLeaf = 1 @@ -67,3 +68,45 @@ class WellKnownResource(Resource): logger.debug("returning: %s", r) request.setHeader(b"Content-Type", b"application/json") return json_encoder.encode(r).encode("utf-8") + + +class ServerWellKnownResource(Resource): + """Resource for .well-known/matrix/server, redirecting to port 443""" + + isLeaf = 1 + + def __init__(self, hs: "HomeServer"): + super().__init__() + self._serve_server_wellknown = hs.config.server.serve_server_wellknown + + host, port = parse_server_name(hs.config.server.server_name) + + # If we've got this far, then https://<server_name>/ must route to us, so + # we just redirect the traffic to port 443 instead of 8448. + if port is None: + port = 443 + + self._response = json_encoder.encode({"m.server": f"{host}:{port}"}).encode( + "utf-8" + ) + + def render_GET(self, request: Request) -> bytes: + if not self._serve_server_wellknown: + request.setResponseCode(404) + request.setHeader(b"Content-Type", b"text/plain") + return b"404. Is anything ever truly *well* known?\n" + + request.setHeader(b"Content-Type", b"application/json") + return self._response + + +def well_known_resource(hs: "HomeServer") -> Resource: + """Returns a Twisted web resource which handles '.well-known' requests""" + res = Resource() + matrix_resource = Resource() + res.putChild(b"matrix", matrix_resource) + + matrix_resource.putChild(b"server", ServerWellKnownResource(hs)) + matrix_resource.putChild(b"client", ClientWellKnownResource(hs)) + + return res |