Support MSC3916 by adding a federation /thumbnail endpoint and authenticated `_matrix/client/v1/media/thumbnail` endpoint (#17388)

[MSC3916](https://github.com/matrix-org/matrix-spec-proposals/pull/3916)
added the endpoints `_matrix/federation/v1/media/thumbnail` and the
authenticated `_matrix/client/v1/media/thumbnail`.

This PR implements those endpoints, along with stabilizing
`_matrix/client/v1/media/config` and
`_matrix/client/v1/media/preview_url`.

Complement tests are at
https://github.com/matrix-org/complement/pull/728

8 files changed, 178 insertions, 47 deletions

diff --git a/synapse/config/experimental.py b/synapse/config/experimental.py
index 1b72727b75..c21b7eb37e 100644
--- a/synapse/config/experimental.py
+++ b/synapse/config/experimental.py
@@ -437,10 +437,6 @@ class ExperimentalConfig(Config):
             "msc3823_account_suspension", False
         )
 
-        self.msc3916_authenticated_media_enabled = experimental.get(
-            "msc3916_authenticated_media_enabled", False
-        )
-
         # MSC4151: Report room API (Client-Server API)
         self.msc4151_enabled: bool = experimental.get("msc4151_enabled", False)
 
diff --git a/synapse/federation/transport/server/__init__.py b/synapse/federation/transport/server/__init__.py
index c44e5daa47..5f997040d0 100644
--- a/synapse/federation/transport/server/__init__.py
+++ b/synapse/federation/transport/server/__init__.py
@@ -33,6 +33,7 @@ from synapse.federation.transport.server.federation import (
     FEDERATION_SERVLET_CLASSES,
     FederationAccountStatusServlet,
     FederationMediaDownloadServlet,
+    FederationMediaThumbnailServlet,
     FederationUnstableClientKeysClaimServlet,
 )
 from synapse.http.server import HttpServer, JsonResource
@@ -316,7 +317,10 @@ def register_servlets(
             ):
                 continue
 
-            if servletclass == FederationMediaDownloadServlet:
+            if (
+                servletclass == FederationMediaDownloadServlet
+                or servletclass == FederationMediaThumbnailServlet
+            ):
                 if not hs.config.server.enable_media_repo:
                     continue
 
diff --git a/synapse/federation/transport/server/_base.py b/synapse/federation/transport/server/_base.py
index e124481474..9094201da0 100644
--- a/synapse/federation/transport/server/_base.py
+++ b/synapse/federation/transport/server/_base.py
@@ -363,6 +363,8 @@ class BaseFederationServlet:
                             if (
                                 func.__self__.__class__.__name__  # type: ignore
                                 == "FederationMediaDownloadServlet"
+                                or func.__self__.__class__.__name__  # type: ignore
+                                == "FederationMediaThumbnailServlet"
                             ):
                                 response = await func(
                                     origin, content, request, *args, **kwargs
@@ -375,6 +377,8 @@ class BaseFederationServlet:
                         if (
                             func.__self__.__class__.__name__  # type: ignore
                             == "FederationMediaDownloadServlet"
+                            or func.__self__.__class__.__name__  # type: ignore
+                            == "FederationMediaThumbnailServlet"
                         ):
                             response = await func(
                                 origin, content, request, *args, **kwargs
diff --git a/synapse/federation/transport/server/federation.py b/synapse/federation/transport/server/federation.py
index ec957768d4..b075a86f68 100644
--- a/synapse/federation/transport/server/federation.py
+++ b/synapse/federation/transport/server/federation.py
@@ -46,11 +46,13 @@ from synapse.http.servlet import (
     parse_boolean_from_args,
     parse_integer,
     parse_integer_from_args,
+    parse_string,
     parse_string_from_args,
     parse_strings_from_args,
 )
 from synapse.http.site import SynapseRequest
 from synapse.media._base import DEFAULT_MAX_TIMEOUT_MS, MAXIMUM_ALLOWED_MAX_TIMEOUT_MS
+from synapse.media.thumbnailer import ThumbnailProvider
 from synapse.types import JsonDict
 from synapse.util import SYNAPSE_VERSION
 from synapse.util.ratelimitutils import FederationRateLimiter
@@ -826,6 +828,59 @@ class FederationMediaDownloadServlet(BaseFederationServerServlet):
         )
 
 
+class FederationMediaThumbnailServlet(BaseFederationServerServlet):
+    """
+    Implementation of new federation media `/thumbnail` endpoint outlined in MSC3916. Returns
+    a multipart/mixed response consisting of a JSON object and the requested media
+    item. This endpoint only returns local media.
+    """
+
+    PATH = "/media/thumbnail/(?P<media_id>[^/]*)"
+    RATELIMIT = True
+
+    def __init__(
+        self,
+        hs: "HomeServer",
+        ratelimiter: FederationRateLimiter,
+        authenticator: Authenticator,
+        server_name: str,
+    ):
+        super().__init__(hs, authenticator, ratelimiter, server_name)
+        self.media_repo = self.hs.get_media_repository()
+        self.dynamic_thumbnails = hs.config.media.dynamic_thumbnails
+        self.thumbnail_provider = ThumbnailProvider(
+            hs, self.media_repo, self.media_repo.media_storage
+        )
+
+    async def on_GET(
+        self,
+        origin: Optional[str],
+        content: Literal[None],
+        request: SynapseRequest,
+        media_id: str,
+    ) -> None:
+
+        width = parse_integer(request, "width", required=True)
+        height = parse_integer(request, "height", required=True)
+        method = parse_string(request, "method", "scale")
+        # TODO Parse the Accept header to get an prioritised list of thumbnail types.
+        m_type = "image/png"
+        max_timeout_ms = parse_integer(
+            request, "timeout_ms", default=DEFAULT_MAX_TIMEOUT_MS
+        )
+        max_timeout_ms = min(max_timeout_ms, MAXIMUM_ALLOWED_MAX_TIMEOUT_MS)
+
+        if self.dynamic_thumbnails:
+            await self.thumbnail_provider.select_or_generate_local_thumbnail(
+                request, media_id, width, height, method, m_type, max_timeout_ms, True
+            )
+        else:
+            await self.thumbnail_provider.respond_local_thumbnail(
+                request, media_id, width, height, method, m_type, max_timeout_ms, True
+            )
+        self.media_repo.mark_recently_accessed(None, media_id)
+
+
 FEDERATION_SERVLET_CLASSES: Tuple[Type[BaseFederationServlet], ...] = (
     FederationSendServlet,
     FederationEventServlet,
@@ -858,4 +913,5 @@ FEDERATION_SERVLET_CLASSES: Tuple[Type[BaseFederationServlet], ...] = (
     FederationMakeKnockServlet,
     FederationAccountStatusServlet,
     FederationMediaDownloadServlet,
+    FederationMediaThumbnailServlet,
 )
diff --git a/synapse/media/media_repository.py b/synapse/media/media_repository.py
index 542642b900..87c929eb20 100644
--- a/synapse/media/media_repository.py
+++ b/synapse/media/media_repository.py
@@ -542,7 +542,12 @@ class MediaRepository:
             respond_404(request)
 
     async def get_remote_media_info(
-        self, server_name: str, media_id: str, max_timeout_ms: int, ip_address: str
+        self,
+        server_name: str,
+        media_id: str,
+        max_timeout_ms: int,
+        ip_address: str,
+        use_federation: bool,
     ) -> RemoteMedia:
         """Gets the media info associated with the remote file, downloading
         if necessary.
@@ -553,6 +558,8 @@ class MediaRepository:
             max_timeout_ms: the maximum number of milliseconds to wait for the
                 media to be uploaded.
             ip_address: IP address of the requester
+            use_federation: if a download is necessary, whether to request the remote file
+                over the federation `/download` endpoint
 
         Returns:
             The media info of the file
@@ -573,7 +580,7 @@ class MediaRepository:
                 max_timeout_ms,
                 self.download_ratelimiter,
                 ip_address,
-                False,
+                use_federation,
             )
 
         # Ensure we actually use the responder so that it releases resources
diff --git a/synapse/media/thumbnailer.py b/synapse/media/thumbnailer.py
index f8a9560784..413a720e40 100644
--- a/synapse/media/thumbnailer.py
+++ b/synapse/media/thumbnailer.py
@@ -36,9 +36,11 @@ from synapse.media._base import (
     ThumbnailInfo,
     respond_404,
     respond_with_file,
+    respond_with_multipart_responder,
     respond_with_responder,
 )
-from synapse.media.media_storage import MediaStorage
+from synapse.media.media_storage import FileResponder, MediaStorage
+from synapse.storage.databases.main.media_repository import LocalMedia
 
 if TYPE_CHECKING:
     from synapse.media.media_repository import MediaRepository
@@ -271,6 +273,7 @@ class ThumbnailProvider:
         method: str,
         m_type: str,
         max_timeout_ms: int,
+        for_federation: bool,
     ) -> None:
         media_info = await self.media_repo.get_local_media_info(
             request, media_id, max_timeout_ms
@@ -290,6 +293,8 @@ class ThumbnailProvider:
             media_id,
             url_cache=bool(media_info.url_cache),
             server_name=None,
+            for_federation=for_federation,
+            media_info=media_info,
         )
 
     async def select_or_generate_local_thumbnail(
@@ -301,6 +306,7 @@ class ThumbnailProvider:
         desired_method: str,
         desired_type: str,
         max_timeout_ms: int,
+        for_federation: bool,
     ) -> None:
         media_info = await self.media_repo.get_local_media_info(
             request, media_id, max_timeout_ms
@@ -326,10 +332,16 @@ class ThumbnailProvider:
 
                 responder = await self.media_storage.fetch_media(file_info)
                 if responder:
-                    await respond_with_responder(
-                        request, responder, info.type, info.length
-                    )
-                    return
+                    if for_federation:
+                        await respond_with_multipart_responder(
+                            self.hs.get_clock(), request, responder, media_info
+                        )
+                        return
+                    else:
+                        await respond_with_responder(
+                            request, responder, info.type, info.length
+                        )
+                        return
 
         logger.debug("We don't have a thumbnail of that size. Generating")
 
@@ -344,7 +356,15 @@ class ThumbnailProvider:
         )
 
         if file_path:
-            await respond_with_file(request, desired_type, file_path)
+            if for_federation:
+                await respond_with_multipart_responder(
+                    self.hs.get_clock(),
+                    request,
+                    FileResponder(open(file_path, "rb")),
+                    media_info,
+                )
+            else:
+                await respond_with_file(request, desired_type, file_path)
         else:
             logger.warning("Failed to generate thumbnail")
             raise SynapseError(400, "Failed to generate thumbnail.")
@@ -360,9 +380,10 @@ class ThumbnailProvider:
         desired_type: str,
         max_timeout_ms: int,
         ip_address: str,
+        use_federation: bool,
     ) -> None:
         media_info = await self.media_repo.get_remote_media_info(
-            server_name, media_id, max_timeout_ms, ip_address
+            server_name, media_id, max_timeout_ms, ip_address, use_federation
         )
         if not media_info:
             respond_404(request)
@@ -424,12 +445,13 @@ class ThumbnailProvider:
         m_type: str,
         max_timeout_ms: int,
         ip_address: str,
+        use_federation: bool,
     ) -> None:
         # TODO: Don't download the whole remote file
         # We should proxy the thumbnail from the remote server instead of
         # downloading the remote file and generating our own thumbnails.
         media_info = await self.media_repo.get_remote_media_info(
-            server_name, media_id, max_timeout_ms, ip_address
+            server_name, media_id, max_timeout_ms, ip_address, use_federation
         )
         if not media_info:
             return
@@ -448,6 +470,7 @@ class ThumbnailProvider:
             media_info.filesystem_id,
             url_cache=False,
             server_name=server_name,
+            for_federation=False,
         )
 
     async def _select_and_respond_with_thumbnail(
@@ -461,7 +484,9 @@ class ThumbnailProvider:
         media_id: str,
         file_id: str,
         url_cache: bool,
+        for_federation: bool,
         server_name: Optional[str] = None,
+        media_info: Optional[LocalMedia] = None,
     ) -> None:
         """
         Respond to a request with an appropriate thumbnail from the previously generated thumbnails.
@@ -476,6 +501,8 @@ class ThumbnailProvider:
             file_id: The ID of the media that a thumbnail is being requested for.
             url_cache: True if this is from a URL cache.
             server_name: The server name, if this is a remote thumbnail.
+            for_federation: whether the request is from the federation /thumbnail request
+            media_info: metadata about the media being requested.
         """
         logger.debug(
             "_select_and_respond_with_thumbnail: media_id=%s desired=%sx%s (%s) thumbnail_infos=%s",
@@ -511,13 +538,20 @@ class ThumbnailProvider:
 
             responder = await self.media_storage.fetch_media(file_info)
             if responder:
-                await respond_with_responder(
-                    request,
-                    responder,
-                    file_info.thumbnail.type,
-                    file_info.thumbnail.length,
-                )
-                return
+                if for_federation:
+                    assert media_info is not None
+                    await respond_with_multipart_responder(
+                        self.hs.get_clock(), request, responder, media_info
+                    )
+                    return
+                else:
+                    await respond_with_responder(
+                        request,
+                        responder,
+                        file_info.thumbnail.type,
+                        file_info.thumbnail.length,
+                    )
+                    return
 
             # If we can't find the thumbnail we regenerate it. This can happen
             # if e.g. we've deleted the thumbnails but still have the original
@@ -558,12 +592,18 @@ class ThumbnailProvider:
                 )
 
             responder = await self.media_storage.fetch_media(file_info)
-            await respond_with_responder(
-                request,
-                responder,
-                file_info.thumbnail.type,
-                file_info.thumbnail.length,
-            )
+            if for_federation:
+                assert media_info is not None
+                await respond_with_multipart_responder(
+                    self.hs.get_clock(), request, responder, media_info
+                )
+            else:
+                await respond_with_responder(
+                    request,
+                    responder,
+                    file_info.thumbnail.type,
+                    file_info.thumbnail.length,
+                )
         else:
             # This might be because:
             # 1. We can't create thumbnails for the given media (corrupted or
diff --git a/synapse/rest/client/media.py b/synapse/rest/client/media.py
index c0ae5dd66f..c30e3022de 100644
--- a/synapse/rest/client/media.py
+++ b/synapse/rest/client/media.py
@@ -47,7 +47,7 @@ from synapse.util.stringutils import parse_and_validate_server_name
 logger = logging.getLogger(__name__)
 
 
-class UnstablePreviewURLServlet(RestServlet):
+class PreviewURLServlet(RestServlet):
     """
     Same as `GET /_matrix/media/r0/preview_url`, this endpoint provides a generic preview API
     for URLs which outputs Open Graph (https://ogp.me/) responses (with some Matrix
@@ -65,9 +65,7 @@ class UnstablePreviewURLServlet(RestServlet):
       * Matrix cannot be used to distribute the metadata between homeservers.
     """
 
-    PATTERNS = [
-        re.compile(r"^/_matrix/client/unstable/org.matrix.msc3916/media/preview_url$")
-    ]
+    PATTERNS = [re.compile(r"^/_matrix/client/v1/media/preview_url$")]
 
     def __init__(
         self,
@@ -95,10 +93,8 @@ class UnstablePreviewURLServlet(RestServlet):
         respond_with_json_bytes(request, 200, og, send_cors=True)
 
 
-class UnstableMediaConfigResource(RestServlet):
-    PATTERNS = [
-        re.compile(r"^/_matrix/client/unstable/org.matrix.msc3916/media/config$")
-    ]
+class MediaConfigResource(RestServlet):
+    PATTERNS = [re.compile(r"^/_matrix/client/v1/media/config$")]
 
     def __init__(self, hs: "HomeServer"):
         super().__init__()
@@ -112,10 +108,10 @@ class UnstableMediaConfigResource(RestServlet):
         respond_with_json(request, 200, self.limits_dict, send_cors=True)
 
 
-class UnstableThumbnailResource(RestServlet):
+class ThumbnailResource(RestServlet):
     PATTERNS = [
         re.compile(
-            "/_matrix/client/unstable/org.matrix.msc3916/media/thumbnail/(?P<server_name>[^/]*)/(?P<media_id>[^/]*)$"
+            "/_matrix/client/v1/media/thumbnail/(?P<server_name>[^/]*)/(?P<media_id>[^/]*)$"
         )
     ]
 
@@ -159,11 +155,25 @@ class UnstableThumbnailResource(RestServlet):
         if self._is_mine_server_name(server_name):
             if self.dynamic_thumbnails:
                 await self.thumbnailer.select_or_generate_local_thumbnail(
-                    request, media_id, width, height, method, m_type, max_timeout_ms
+                    request,
+                    media_id,
+                    width,
+                    height,
+                    method,
+                    m_type,
+                    max_timeout_ms,
+                    False,
                 )
             else:
                 await self.thumbnailer.respond_local_thumbnail(
-                    request, media_id, width, height, method, m_type, max_timeout_ms
+                    request,
+                    media_id,
+                    width,
+                    height,
+                    method,
+                    m_type,
+                    max_timeout_ms,
+                    False,
                 )
             self.media_repo.mark_recently_accessed(None, media_id)
         else:
@@ -191,6 +201,7 @@ class UnstableThumbnailResource(RestServlet):
                 m_type,
                 max_timeout_ms,
                 ip_address,
+                True,
             )
             self.media_repo.mark_recently_accessed(server_name, media_id)
 
@@ -260,11 +271,9 @@ class DownloadResource(RestServlet):
 def register_servlets(hs: "HomeServer", http_server: HttpServer) -> None:
     media_repo = hs.get_media_repository()
     if hs.config.media.url_preview_enabled:
-        UnstablePreviewURLServlet(hs, media_repo, media_repo.media_storage).register(
+        PreviewURLServlet(hs, media_repo, media_repo.media_storage).register(
             http_server
         )
-    UnstableMediaConfigResource(hs).register(http_server)
-    UnstableThumbnailResource(hs, media_repo, media_repo.media_storage).register(
-        http_server
-    )
+    MediaConfigResource(hs).register(http_server)
+    ThumbnailResource(hs, media_repo, media_repo.media_storage).register(http_server)
     DownloadResource(hs, media_repo).register(http_server)
diff --git a/synapse/rest/media/thumbnail_resource.py b/synapse/rest/media/thumbnail_resource.py
index ce511c6dce..70354aa439 100644
--- a/synapse/rest/media/thumbnail_resource.py
+++ b/synapse/rest/media/thumbnail_resource.py
@@ -88,11 +88,25 @@ class ThumbnailResource(RestServlet):
         if self._is_mine_server_name(server_name):
             if self.dynamic_thumbnails:
                 await self.thumbnail_provider.select_or_generate_local_thumbnail(
-                    request, media_id, width, height, method, m_type, max_timeout_ms
+                    request,
+                    media_id,
+                    width,
+                    height,
+                    method,
+                    m_type,
+                    max_timeout_ms,
+                    False,
                 )
             else:
                 await self.thumbnail_provider.respond_local_thumbnail(
-                    request, media_id, width, height, method, m_type, max_timeout_ms
+                    request,
+                    media_id,
+                    width,
+                    height,
+                    method,
+                    m_type,
+                    max_timeout_ms,
+                    False,
                 )
             self.media_repo.mark_recently_accessed(None, media_id)
         else:
@@ -120,5 +134,6 @@ class ThumbnailResource(RestServlet):
                 m_type,
                 max_timeout_ms,
                 ip_address,
+                False,
             )
             self.media_repo.mark_recently_accessed(server_name, media_id)