Support MSC3916 by adding unstable media endpoints to _matrix/client (#17213)

[MSC3916](https://github.com/matrix-org/matrix-spec-proposals/blob/rav/authentication-for-media/proposals/3916-authentication-for-media.md)
adds new media endpoints under `_matrix/client`. This PR adds the
`/preview_url`, `/config`, and `/thumbnail` endpoints. `/download` will
be added in a follow-up PR once the work for the federation `/download`
endpoint is complete (see
https://github.com/element-hq/synapse/pull/17172).

Should be reviewable commit-by-commit.

changelog.d/17213.feature

@@ -0,0 +1 @@
+Support MSC3916 by adding unstable media endpoints to `_matrix/client` (#17213).

synapse/config/experimental.py

@@ -439,3 +439,7 @@ class ExperimentalConfig(Config):
         self.msc4115_membership_on_events = experimental.get(
             "msc4115_membership_on_events", False
         )
+
+        self.msc3916_authenticated_media_enabled = experimental.get(
+            "msc3916_authenticated_media_enabled", False
+        )

synapse/media/thumbnailer.py

@@ -22,11 +22,27 @@
 import logging
 from io import BytesIO
 from types import TracebackType
-from typing import Optional, Tuple, Type
+from typing import TYPE_CHECKING, List, Optional, Tuple, Type
 
 from PIL import Image
 
+from synapse.api.errors import Codes, SynapseError, cs_error
+from synapse.config.repository import THUMBNAIL_SUPPORTED_MEDIA_FORMAT_MAP
+from synapse.http.server import respond_with_json
+from synapse.http.site import SynapseRequest
 from synapse.logging.opentracing import trace
+from synapse.media._base import (
+    FileInfo,
+    ThumbnailInfo,
+    respond_404,
+    respond_with_file,
+    respond_with_responder,
+)
+from synapse.media.media_storage import MediaStorage
+
+if TYPE_CHECKING:
+    from synapse.media.media_repository import MediaRepository
+    from synapse.server import HomeServer
 
 logger = logging.getLogger(__name__)
 
@@ -231,3 +247,471 @@ class Thumbnailer:
     def __del__(self) -> None:
         # Make sure we actually do close the image, rather than leak data.
         self.close()
+
+
+class ThumbnailProvider:
+    def __init__(
+        self,
+        hs: "HomeServer",
+        media_repo: "MediaRepository",
+        media_storage: MediaStorage,
+    ):
+        self.hs = hs
+        self.media_repo = media_repo
+        self.media_storage = media_storage
+        self.store = hs.get_datastores().main
+        self.dynamic_thumbnails = hs.config.media.dynamic_thumbnails
+
+    async def respond_local_thumbnail(
+        self,
+        request: SynapseRequest,
+        media_id: str,
+        width: int,
+        height: int,
+        method: str,
+        m_type: str,
+        max_timeout_ms: int,
+    ) -> None:
+        media_info = await self.media_repo.get_local_media_info(
+            request, media_id, max_timeout_ms
+        )
+        if not media_info:
+            return
+
+        thumbnail_infos = await self.store.get_local_media_thumbnails(media_id)
+        await self._select_and_respond_with_thumbnail(
+            request,
+            width,
+            height,
+            method,
+            m_type,
+            thumbnail_infos,
+            media_id,
+            media_id,
+            url_cache=bool(media_info.url_cache),
+            server_name=None,
+        )
+
+    async def select_or_generate_local_thumbnail(
+        self,
+        request: SynapseRequest,
+        media_id: str,
+        desired_width: int,
+        desired_height: int,
+        desired_method: str,
+        desired_type: str,
+        max_timeout_ms: int,
+    ) -> None:
+        media_info = await self.media_repo.get_local_media_info(
+            request, media_id, max_timeout_ms
+        )
+
+        if not media_info:
+            return
+
+        thumbnail_infos = await self.store.get_local_media_thumbnails(media_id)
+        for info in thumbnail_infos:
+            t_w = info.width == desired_width
+            t_h = info.height == desired_height
+            t_method = info.method == desired_method
+            t_type = info.type == desired_type
+
+            if t_w and t_h and t_method and t_type:
+                file_info = FileInfo(
+                    server_name=None,
+                    file_id=media_id,
+                    url_cache=bool(media_info.url_cache),
+                    thumbnail=info,
+                )
+
+                responder = await self.media_storage.fetch_media(file_info)
+                if responder:
+                    await respond_with_responder(
+                        request, responder, info.type, info.length
+                    )
+                    return
+
+        logger.debug("We don't have a thumbnail of that size. Generating")
+
+        # Okay, so we generate one.
+        file_path = await self.media_repo.generate_local_exact_thumbnail(
+            media_id,
+            desired_width,
+            desired_height,
+            desired_method,
+            desired_type,
+            url_cache=bool(media_info.url_cache),
+        )
+
+        if file_path:
+            await respond_with_file(request, desired_type, file_path)
+        else:
+            logger.warning("Failed to generate thumbnail")
+            raise SynapseError(400, "Failed to generate thumbnail.")
+
+    async def select_or_generate_remote_thumbnail(
+        self,
+        request: SynapseRequest,
+        server_name: str,
+        media_id: str,
+        desired_width: int,
+        desired_height: int,
+        desired_method: str,
+        desired_type: str,
+        max_timeout_ms: int,
+    ) -> None:
+        media_info = await self.media_repo.get_remote_media_info(
+            server_name, media_id, max_timeout_ms
+        )
+        if not media_info:
+            respond_404(request)
+            return
+
+        thumbnail_infos = await self.store.get_remote_media_thumbnails(
+            server_name, media_id
+        )
+
+        file_id = media_info.filesystem_id
+
+        for info in thumbnail_infos:
+            t_w = info.width == desired_width
+            t_h = info.height == desired_height
+            t_method = info.method == desired_method
+            t_type = info.type == desired_type
+
+            if t_w and t_h and t_method and t_type:
+                file_info = FileInfo(
+                    server_name=server_name,
+                    file_id=file_id,
+                    thumbnail=info,
+                )
+
+                responder = await self.media_storage.fetch_media(file_info)
+                if responder:
+                    await respond_with_responder(
+                        request, responder, info.type, info.length
+                    )
+                    return
+
+        logger.debug("We don't have a thumbnail of that size. Generating")
+
+        # Okay, so we generate one.
+        file_path = await self.media_repo.generate_remote_exact_thumbnail(
+            server_name,
+            file_id,
+            media_id,
+            desired_width,
+            desired_height,
+            desired_method,
+            desired_type,
+        )
+
+        if file_path:
+            await respond_with_file(request, desired_type, file_path)
+        else:
+            logger.warning("Failed to generate thumbnail")
+            raise SynapseError(400, "Failed to generate thumbnail.")
+
+    async def respond_remote_thumbnail(
+        self,
+        request: SynapseRequest,
+        server_name: str,
+        media_id: str,
+        width: int,
+        height: int,
+        method: str,
+        m_type: str,
+        max_timeout_ms: int,
+    ) -> 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
+        )
+        if not media_info:
+            return
+
+        thumbnail_infos = await self.store.get_remote_media_thumbnails(
+            server_name, media_id
+        )
+        await self._select_and_respond_with_thumbnail(
+            request,
+            width,
+            height,
+            method,
+            m_type,
+            thumbnail_infos,
+            media_id,
+            media_info.filesystem_id,
+            url_cache=False,
+            server_name=server_name,
+        )
+
+    async def _select_and_respond_with_thumbnail(
+        self,
+        request: SynapseRequest,
+        desired_width: int,
+        desired_height: int,
+        desired_method: str,
+        desired_type: str,
+        thumbnail_infos: List[ThumbnailInfo],
+        media_id: str,
+        file_id: str,
+        url_cache: bool,
+        server_name: Optional[str] = None,
+    ) -> None:
+        """
+        Respond to a request with an appropriate thumbnail from the previously generated thumbnails.
+
+        Args:
+            request: The incoming request.
+            desired_width: The desired width, the returned thumbnail may be larger than this.
+            desired_height: The desired height, the returned thumbnail may be larger than this.
+            desired_method: The desired method used to generate the thumbnail.
+            desired_type: The desired content-type of the thumbnail.
+            thumbnail_infos: A list of thumbnail info of candidate thumbnails.
+            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.
+        """
+        logger.debug(
+            "_select_and_respond_with_thumbnail: media_id=%s desired=%sx%s (%s) thumbnail_infos=%s",
+            media_id,
+            desired_width,
+            desired_height,
+            desired_method,
+            thumbnail_infos,
+        )
+
+        # If `dynamic_thumbnails` is enabled, we expect Synapse to go down a
+        # different code path to handle it.
+        assert not self.dynamic_thumbnails
+
+        if thumbnail_infos:
+            file_info = self._select_thumbnail(
+                desired_width,
+                desired_height,
+                desired_method,
+                desired_type,
+                thumbnail_infos,
+                file_id,
+                url_cache,
+                server_name,
+            )
+            if not file_info:
+                logger.info("Couldn't find a thumbnail matching the desired inputs")
+                respond_404(request)
+                return
+
+            # The thumbnail property must exist.
+            assert file_info.thumbnail is not None
+
+            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 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
+            # image somewhere.
+            #
+            # Since we have an entry for the thumbnail in the DB we a) know we
+            # have have successfully generated the thumbnail in the past (so we
+            # don't need to worry about repeatedly failing to generate
+            # thumbnails), and b) have already calculated that appropriate
+            # width/height/method so we can just call the "generate exact"
+            # methods.
+
+            # First let's check that we do actually have the original image
+            # still. This will throw a 404 if we don't.
+            # TODO: We should refetch the thumbnails for remote media.
+            await self.media_storage.ensure_media_is_in_local_cache(
+                FileInfo(server_name, file_id, url_cache=url_cache)
+            )
+
+            if server_name:
+                await self.media_repo.generate_remote_exact_thumbnail(
+                    server_name,
+                    file_id=file_id,
+                    media_id=media_id,
+                    t_width=file_info.thumbnail.width,
+                    t_height=file_info.thumbnail.height,
+                    t_method=file_info.thumbnail.method,
+                    t_type=file_info.thumbnail.type,
+                )
+            else:
+                await self.media_repo.generate_local_exact_thumbnail(
+                    media_id=media_id,
+                    t_width=file_info.thumbnail.width,
+                    t_height=file_info.thumbnail.height,
+                    t_method=file_info.thumbnail.method,
+                    t_type=file_info.thumbnail.type,
+                    url_cache=url_cache,
+                )
+
+            responder = await self.media_storage.fetch_media(file_info)
+            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
+            #    unsupported file type), or
+            # 2. The thumbnailing process never ran or errored out initially
+            #    when the media was first uploaded (these bugs should be
+            #    reported and fixed).
+            # Note that we don't attempt to generate a thumbnail now because
+            # `dynamic_thumbnails` is disabled.
+            logger.info("Failed to find any generated thumbnails")
+
+            assert request.path is not None
+            respond_with_json(
+                request,
+                400,
+                cs_error(
+                    "Cannot find any thumbnails for the requested media ('%s'). This might mean the media is not a supported_media_format=(%s) or that thumbnailing failed for some other reason. (Dynamic thumbnails are disabled on this server.)"
+                    % (
+                        request.path.decode(),
+                        ", ".join(THUMBNAIL_SUPPORTED_MEDIA_FORMAT_MAP.keys()),
+                    ),
+                    code=Codes.UNKNOWN,
+                ),
+                send_cors=True,
+            )
+
+    def _select_thumbnail(
+        self,
+        desired_width: int,
+        desired_height: int,
+        desired_method: str,
+        desired_type: str,
+        thumbnail_infos: List[ThumbnailInfo],
+        file_id: str,
+        url_cache: bool,
+        server_name: Optional[str],
+    ) -> Optional[FileInfo]:
+        """
+        Choose an appropriate thumbnail from the previously generated thumbnails.
+
+        Args:
+            desired_width: The desired width, the returned thumbnail may be larger than this.
+            desired_height: The desired height, the returned thumbnail may be larger than this.
+            desired_method: The desired method used to generate the thumbnail.
+            desired_type: The desired content-type of the thumbnail.
+            thumbnail_infos: A list of thumbnail infos of candidate thumbnails.
+            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.
+
+        Returns:
+             The thumbnail which best matches the desired parameters.
+        """
+        desired_method = desired_method.lower()
+
+        # The chosen thumbnail.
+        thumbnail_info = None
+
+        d_w = desired_width
+        d_h = desired_height
+
+        if desired_method == "crop":
+            # Thumbnails that match equal or larger sizes of desired width/height.
+            crop_info_list: List[
+                Tuple[int, int, int, bool, Optional[int], ThumbnailInfo]
+            ] = []
+            # Other thumbnails.
+            crop_info_list2: List[
+                Tuple[int, int, int, bool, Optional[int], ThumbnailInfo]
+            ] = []
+            for info in thumbnail_infos:
+                # Skip thumbnails generated with different methods.
+                if info.method != "crop":
+                    continue
+
+                t_w = info.width
+                t_h = info.height
+                aspect_quality = abs(d_w * t_h - d_h * t_w)
+                min_quality = 0 if d_w <= t_w and d_h <= t_h else 1
+                size_quality = abs((d_w - t_w) * (d_h - t_h))
+                type_quality = desired_type != info.type
+                length_quality = info.length
+                if t_w >= d_w or t_h >= d_h:
+                    crop_info_list.append(
+                        (
+                            aspect_quality,
+                            min_quality,
+                            size_quality,
+                            type_quality,
+                            length_quality,
+                            info,
+                        )
+                    )
+                else:
+                    crop_info_list2.append(
+                        (
+                            aspect_quality,
+                            min_quality,
+                            size_quality,
+                            type_quality,
+                            length_quality,
+                            info,
+                        )
+                    )
+            # Pick the most appropriate thumbnail. Some values of `desired_width` and
+            # `desired_height` may result in a tie, in which case we avoid comparing on
+            # the thumbnail info and pick the thumbnail that appears earlier
+            # in the list of candidates.
+            if crop_info_list:
+                thumbnail_info = min(crop_info_list, key=lambda t: t[:-1])[-1]
+            elif crop_info_list2:
+                thumbnail_info = min(crop_info_list2, key=lambda t: t[:-1])[-1]
+        elif desired_method == "scale":
+            # Thumbnails that match equal or larger sizes of desired width/height.
+            info_list: List[Tuple[int, bool, int, ThumbnailInfo]] = []
+            # Other thumbnails.
+            info_list2: List[Tuple[int, bool, int, ThumbnailInfo]] = []
+
+            for info in thumbnail_infos:
+                # Skip thumbnails generated with different methods.
+                if info.method != "scale":
+                    continue
+
+                t_w = info.width
+                t_h = info.height
+                size_quality = abs((d_w - t_w) * (d_h - t_h))
+                type_quality = desired_type != info.type
+                length_quality = info.length
+                if t_w >= d_w or t_h >= d_h:
+                    info_list.append((size_quality, type_quality, length_quality, info))
+                else:
+                    info_list2.append(
+                        (size_quality, type_quality, length_quality, info)
+                    )
+            # Pick the most appropriate thumbnail. Some values of `desired_width` and
+            # `desired_height` may result in a tie, in which case we avoid comparing on
+            # the thumbnail info and pick the thumbnail that appears earlier
+            # in the list of candidates.
+            if info_list:
+                thumbnail_info = min(info_list, key=lambda t: t[:-1])[-1]
+            elif info_list2:
+                thumbnail_info = min(info_list2, key=lambda t: t[:-1])[-1]
+
+        if thumbnail_info:
+            return FileInfo(
+                file_id=file_id,
+                url_cache=url_cache,
+                server_name=server_name,
+                thumbnail=thumbnail_info,
+            )
+
+        # No matching thumbnail was found.
+        return None

synapse/rest/client/media.py

@@ -0,0 +1,205 @@
+#
+# This file is licensed under the Affero General Public License (AGPL) version 3.
+#
+# Copyright 2020 The Matrix.org Foundation C.I.C.
+# Copyright 2015, 2016 OpenMarket Ltd
+# Copyright (C) 2024 New Vector, Ltd
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as
+# published by the Free Software Foundation, either version 3 of the
+# License, or (at your option) any later version.
+#
+# See the GNU Affero General Public License for more details:
+# <https://www.gnu.org/licenses/agpl-3.0.html>.
+#
+# Originally licensed under the Apache License, Version 2.0:
+# <http://www.apache.org/licenses/LICENSE-2.0>.
+#
+# [This file includes modifications made by New Vector Limited]
+#
+#
+
+import logging
+import re
+
+from synapse.http.server import (
+    HttpServer,
+    respond_with_json,
+    respond_with_json_bytes,
+    set_corp_headers,
+    set_cors_headers,
+)
+from synapse.http.servlet import RestServlet, parse_integer, parse_string
+from synapse.http.site import SynapseRequest
+from synapse.media._base import (
+    DEFAULT_MAX_TIMEOUT_MS,
+    MAXIMUM_ALLOWED_MAX_TIMEOUT_MS,
+    respond_404,
+)
+from synapse.media.media_repository import MediaRepository
+from synapse.media.media_storage import MediaStorage
+from synapse.media.thumbnailer import ThumbnailProvider
+from synapse.server import HomeServer
+from synapse.util.stringutils import parse_and_validate_server_name
+
+logger = logging.getLogger(__name__)
+
+
+class UnstablePreviewURLServlet(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
+    specific additions).
+
+    This does have trade-offs compared to other designs:
+
+    * Pros:
+      * Simple and flexible; can be used by any clients at any point
+    * Cons:
+      * If each homeserver provides one of these independently, all the homeservers in a
+        room may needlessly DoS the target URI
+      * The URL metadata must be stored somewhere, rather than just using Matrix
+        itself to store the media.
+      * Matrix cannot be used to distribute the metadata between homeservers.
+    """
+
+    PATTERNS = [
+        re.compile(r"^/_matrix/client/unstable/org.matrix.msc3916/media/preview_url$")
+    ]
+
+    def __init__(
+        self,
+        hs: "HomeServer",
+        media_repo: "MediaRepository",
+        media_storage: MediaStorage,
+    ):
+        super().__init__()
+
+        self.auth = hs.get_auth()
+        self.clock = hs.get_clock()
+        self.media_repo = media_repo
+        self.media_storage = media_storage
+        assert self.media_repo.url_previewer is not None
+        self.url_previewer = self.media_repo.url_previewer
+
+    async def on_GET(self, request: SynapseRequest) -> None:
+        requester = await self.auth.get_user_by_req(request)
+        url = parse_string(request, "url", required=True)
+        ts = parse_integer(request, "ts")
+        if ts is None:
+            ts = self.clock.time_msec()
+
+        og = await self.url_previewer.preview(url, requester.user, ts)
+        respond_with_json_bytes(request, 200, og, send_cors=True)
+
+
+class UnstableMediaConfigResource(RestServlet):
+    PATTERNS = [
+        re.compile(r"^/_matrix/client/unstable/org.matrix.msc3916/media/config$")
+    ]
+
+    def __init__(self, hs: "HomeServer"):
+        super().__init__()
+        config = hs.config
+        self.clock = hs.get_clock()
+        self.auth = hs.get_auth()
+        self.limits_dict = {"m.upload.size": config.media.max_upload_size}
+
+    async def on_GET(self, request: SynapseRequest) -> None:
+        await self.auth.get_user_by_req(request)
+        respond_with_json(request, 200, self.limits_dict, send_cors=True)
+
+
+class UnstableThumbnailResource(RestServlet):
+    PATTERNS = [
+        re.compile(
+            "/_matrix/client/unstable/org.matrix.msc3916/media/thumbnail/(?P<server_name>[^/]*)/(?P<media_id>[^/]*)$"
+        )
+    ]
+
+    def __init__(
+        self,
+        hs: "HomeServer",
+        media_repo: "MediaRepository",
+        media_storage: MediaStorage,
+    ):
+        super().__init__()
+
+        self.store = hs.get_datastores().main
+        self.media_repo = media_repo
+        self.media_storage = media_storage
+        self.dynamic_thumbnails = hs.config.media.dynamic_thumbnails
+        self._is_mine_server_name = hs.is_mine_server_name
+        self._server_name = hs.hostname
+        self.prevent_media_downloads_from = hs.config.media.prevent_media_downloads_from
+        self.thumbnailer = ThumbnailProvider(hs, media_repo, media_storage)
+        self.auth = hs.get_auth()
+
+    async def on_GET(
+        self, request: SynapseRequest, server_name: str, media_id: str
+    ) -> None:
+        # Validate the server name, raising if invalid
+        parse_and_validate_server_name(server_name)
+        await self.auth.get_user_by_req(request)
+
+        set_cors_headers(request)
+        set_corp_headers(request)
+        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._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
+                )
+            else:
+                await self.thumbnailer.respond_local_thumbnail(
+                    request, media_id, width, height, method, m_type, max_timeout_ms
+                )
+            self.media_repo.mark_recently_accessed(None, media_id)
+        else:
+            # Don't let users download media from configured domains, even if it
+            # is already downloaded. This is Trust & Safety tooling to make some
+            # media inaccessible to local users.
+            # See `prevent_media_downloads_from` config docs for more info.
+            if server_name in self.prevent_media_downloads_from:
+                respond_404(request)
+                return
+
+            remote_resp_function = (
+                self.thumbnailer.select_or_generate_remote_thumbnail
+                if self.dynamic_thumbnails
+                else self.thumbnailer.respond_remote_thumbnail
+            )
+            await remote_resp_function(
+                request,
+                server_name,
+                media_id,
+                width,
+                height,
+                method,
+                m_type,
+                max_timeout_ms,
+            )
+            self.media_repo.mark_recently_accessed(server_name, media_id)
+
+
+def register_servlets(hs: "HomeServer", http_server: HttpServer) -> None:
+    if hs.config.experimental.msc3916_authenticated_media_enabled:
+        media_repo = hs.get_media_repository()
+        if hs.config.media.url_preview_enabled:
+            UnstablePreviewURLServlet(
+                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
+        )

synapse/rest/media/thumbnail_resource.py

@@ -22,23 +22,18 @@
 
 import logging
 import re
-from typing import TYPE_CHECKING, List, Optional, Tuple
+from typing import TYPE_CHECKING
 
-from synapse.api.errors import Codes, SynapseError, cs_error
-from synapse.config.repository import THUMBNAIL_SUPPORTED_MEDIA_FORMAT_MAP
-from synapse.http.server import respond_with_json, set_corp_headers, set_cors_headers
+from synapse.http.server import set_corp_headers, set_cors_headers
 from synapse.http.servlet import RestServlet, parse_integer, parse_string
 from synapse.http.site import SynapseRequest
 from synapse.media._base import (
     DEFAULT_MAX_TIMEOUT_MS,
     MAXIMUM_ALLOWED_MAX_TIMEOUT_MS,
-    FileInfo,
-    ThumbnailInfo,
     respond_404,
-    respond_with_file,
-    respond_with_responder,
 )
 from synapse.media.media_storage import MediaStorage
+from synapse.media.thumbnailer import ThumbnailProvider
 from synapse.util.stringutils import parse_and_validate_server_name
 
 if TYPE_CHECKING:
@@ -66,10 +61,11 @@ class ThumbnailResource(RestServlet):
         self.store = hs.get_datastores().main
         self.media_repo = media_repo
         self.media_storage = media_storage
-        self.dynamic_thumbnails = hs.config.media.dynamic_thumbnails
         self._is_mine_server_name = hs.is_mine_server_name
         self._server_name = hs.hostname
         self.prevent_media_downloads_from = hs.config.media.prevent_media_downloads_from
+        self.dynamic_thumbnails = hs.config.media.dynamic_thumbnails
+        self.thumbnail_provider = ThumbnailProvider(hs, media_repo, media_storage)
 
     async def on_GET(
         self, request: SynapseRequest, server_name: str, media_id: str
@@ -91,11 +87,11 @@ class ThumbnailResource(RestServlet):
 
         if self._is_mine_server_name(server_name):
             if self.dynamic_thumbnails:
-                await self._select_or_generate_local_thumbnail(
+                await self.thumbnail_provider.select_or_generate_local_thumbnail(
                     request, media_id, width, height, method, m_type, max_timeout_ms
                 )
             else:
-                await self._respond_local_thumbnail(
+                await self.thumbnail_provider.respond_local_thumbnail(
                     request, media_id, width, height, method, m_type, max_timeout_ms
                 )
             self.media_repo.mark_recently_accessed(None, media_id)
@@ -109,9 +105,9 @@ class ThumbnailResource(RestServlet):
                 return
 
             remote_resp_function = (
-                self._select_or_generate_remote_thumbnail
+                self.thumbnail_provider.select_or_generate_remote_thumbnail
                 if self.dynamic_thumbnails
-                else self._respond_remote_thumbnail
+                else self.thumbnail_provider.respond_remote_thumbnail
             )
             await remote_resp_function(
                 request,
@@ -124,457 +120,3 @@ class ThumbnailResource(RestServlet):
                 max_timeout_ms,
             )
             self.media_repo.mark_recently_accessed(server_name, media_id)
-
-    async def _respond_local_thumbnail(
-        self,
-        request: SynapseRequest,
-        media_id: str,
-        width: int,
-        height: int,
-        method: str,
-        m_type: str,
-        max_timeout_ms: int,
-    ) -> None:
-        media_info = await self.media_repo.get_local_media_info(
-            request, media_id, max_timeout_ms
-        )
-        if not media_info:
-            return
-
-        thumbnail_infos = await self.store.get_local_media_thumbnails(media_id)
-        await self._select_and_respond_with_thumbnail(
-            request,
-            width,
-            height,
-            method,
-            m_type,
-            thumbnail_infos,
-            media_id,
-            media_id,
-            url_cache=bool(media_info.url_cache),
-            server_name=None,
-        )
-
-    async def _select_or_generate_local_thumbnail(
-        self,
-        request: SynapseRequest,
-        media_id: str,
-        desired_width: int,
-        desired_height: int,
-        desired_method: str,
-        desired_type: str,
-        max_timeout_ms: int,
-    ) -> None:
-        media_info = await self.media_repo.get_local_media_info(
-            request, media_id, max_timeout_ms
-        )
-
-        if not media_info:
-            return
-
-        thumbnail_infos = await self.store.get_local_media_thumbnails(media_id)
-        for info in thumbnail_infos:
-            t_w = info.width == desired_width
-            t_h = info.height == desired_height
-            t_method = info.method == desired_method
-            t_type = info.type == desired_type
-
-            if t_w and t_h and t_method and t_type:
-                file_info = FileInfo(
-                    server_name=None,
-                    file_id=media_id,
-                    url_cache=bool(media_info.url_cache),
-                    thumbnail=info,
-                )
-
-                responder = await self.media_storage.fetch_media(file_info)
-                if responder:
-                    await respond_with_responder(
-                        request, responder, info.type, info.length
-                    )
-                    return
-
-        logger.debug("We don't have a thumbnail of that size. Generating")
-
-        # Okay, so we generate one.
-        file_path = await self.media_repo.generate_local_exact_thumbnail(
-            media_id,
-            desired_width,
-            desired_height,
-            desired_method,
-            desired_type,
-            url_cache=bool(media_info.url_cache),
-        )
-
-        if file_path:
-            await respond_with_file(request, desired_type, file_path)
-        else:
-            logger.warning("Failed to generate thumbnail")
-            raise SynapseError(400, "Failed to generate thumbnail.")
-
-    async def _select_or_generate_remote_thumbnail(
-        self,
-        request: SynapseRequest,
-        server_name: str,
-        media_id: str,
-        desired_width: int,
-        desired_height: int,
-        desired_method: str,
-        desired_type: str,
-        max_timeout_ms: int,
-    ) -> None:
-        media_info = await self.media_repo.get_remote_media_info(
-            server_name, media_id, max_timeout_ms
-        )
-        if not media_info:
-            respond_404(request)
-            return
-
-        thumbnail_infos = await self.store.get_remote_media_thumbnails(
-            server_name, media_id
-        )
-
-        file_id = media_info.filesystem_id
-
-        for info in thumbnail_infos:
-            t_w = info.width == desired_width
-            t_h = info.height == desired_height
-            t_method = info.method == desired_method
-            t_type = info.type == desired_type
-
-            if t_w and t_h and t_method and t_type:
-                file_info = FileInfo(
-                    server_name=server_name,
-                    file_id=file_id,
-                    thumbnail=info,
-                )
-
-                responder = await self.media_storage.fetch_media(file_info)
-                if responder:
-                    await respond_with_responder(
-                        request, responder, info.type, info.length
-                    )
-                    return
-
-        logger.debug("We don't have a thumbnail of that size. Generating")
-
-        # Okay, so we generate one.
-        file_path = await self.media_repo.generate_remote_exact_thumbnail(
-            server_name,
-            file_id,
-            media_id,
-            desired_width,
-            desired_height,
-            desired_method,
-            desired_type,
-        )
-
-        if file_path:
-            await respond_with_file(request, desired_type, file_path)
-        else:
-            logger.warning("Failed to generate thumbnail")
-            raise SynapseError(400, "Failed to generate thumbnail.")
-
-    async def _respond_remote_thumbnail(
-        self,
-        request: SynapseRequest,
-        server_name: str,
-        media_id: str,
-        width: int,
-        height: int,
-        method: str,
-        m_type: str,
-        max_timeout_ms: int,
-    ) -> 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
-        )
-        if not media_info:
-            return
-
-        thumbnail_infos = await self.store.get_remote_media_thumbnails(
-            server_name, media_id
-        )
-        await self._select_and_respond_with_thumbnail(
-            request,
-            width,
-            height,
-            method,
-            m_type,
-            thumbnail_infos,
-            media_id,
-            media_info.filesystem_id,
-            url_cache=False,
-            server_name=server_name,
-        )
-
-    async def _select_and_respond_with_thumbnail(
-        self,
-        request: SynapseRequest,
-        desired_width: int,
-        desired_height: int,
-        desired_method: str,
-        desired_type: str,
-        thumbnail_infos: List[ThumbnailInfo],
-        media_id: str,
-        file_id: str,
-        url_cache: bool,
-        server_name: Optional[str] = None,
-    ) -> None:
-        """
-        Respond to a request with an appropriate thumbnail from the previously generated thumbnails.
-
-        Args:
-            request: The incoming request.
-            desired_width: The desired width, the returned thumbnail may be larger than this.
-            desired_height: The desired height, the returned thumbnail may be larger than this.
-            desired_method: The desired method used to generate the thumbnail.
-            desired_type: The desired content-type of the thumbnail.
-            thumbnail_infos: A list of thumbnail info of candidate thumbnails.
-            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.
-        """
-        logger.debug(
-            "_select_and_respond_with_thumbnail: media_id=%s desired=%sx%s (%s) thumbnail_infos=%s",
-            media_id,
-            desired_width,
-            desired_height,
-            desired_method,
-            thumbnail_infos,
-        )
-
-        # If `dynamic_thumbnails` is enabled, we expect Synapse to go down a
-        # different code path to handle it.
-        assert not self.dynamic_thumbnails
-
-        if thumbnail_infos:
-            file_info = self._select_thumbnail(
-                desired_width,
-                desired_height,
-                desired_method,
-                desired_type,
-                thumbnail_infos,
-                file_id,
-                url_cache,
-                server_name,
-            )
-            if not file_info:
-                logger.info("Couldn't find a thumbnail matching the desired inputs")
-                respond_404(request)
-                return
-
-            # The thumbnail property must exist.
-            assert file_info.thumbnail is not None
-
-            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 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
-            # image somewhere.
-            #
-            # Since we have an entry for the thumbnail in the DB we a) know we
-            # have have successfully generated the thumbnail in the past (so we
-            # don't need to worry about repeatedly failing to generate
-            # thumbnails), and b) have already calculated that appropriate
-            # width/height/method so we can just call the "generate exact"
-            # methods.
-
-            # First let's check that we do actually have the original image
-            # still. This will throw a 404 if we don't.
-            # TODO: We should refetch the thumbnails for remote media.
-            await self.media_storage.ensure_media_is_in_local_cache(
-                FileInfo(server_name, file_id, url_cache=url_cache)
-            )
-
-            if server_name:
-                await self.media_repo.generate_remote_exact_thumbnail(
-                    server_name,
-                    file_id=file_id,
-                    media_id=media_id,
-                    t_width=file_info.thumbnail.width,
-                    t_height=file_info.thumbnail.height,
-                    t_method=file_info.thumbnail.method,
-                    t_type=file_info.thumbnail.type,
-                )
-            else:
-                await self.media_repo.generate_local_exact_thumbnail(
-                    media_id=media_id,
-                    t_width=file_info.thumbnail.width,
-                    t_height=file_info.thumbnail.height,
-                    t_method=file_info.thumbnail.method,
-                    t_type=file_info.thumbnail.type,
-                    url_cache=url_cache,
-                )
-
-            responder = await self.media_storage.fetch_media(file_info)
-            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
-            #    unsupported file type), or
-            # 2. The thumbnailing process never ran or errored out initially
-            #    when the media was first uploaded (these bugs should be
-            #    reported and fixed).
-            # Note that we don't attempt to generate a thumbnail now because
-            # `dynamic_thumbnails` is disabled.
-            logger.info("Failed to find any generated thumbnails")
-
-            assert request.path is not None
-            respond_with_json(
-                request,
-                400,
-                cs_error(
-                    "Cannot find any thumbnails for the requested media ('%s'). This might mean the media is not a supported_media_format=(%s) or that thumbnailing failed for some other reason. (Dynamic thumbnails are disabled on this server.)"
-                    % (
-                        request.path.decode(),
-                        ", ".join(THUMBNAIL_SUPPORTED_MEDIA_FORMAT_MAP.keys()),
-                    ),
-                    code=Codes.UNKNOWN,
-                ),
-                send_cors=True,
-            )
-
-    def _select_thumbnail(
-        self,
-        desired_width: int,
-        desired_height: int,
-        desired_method: str,
-        desired_type: str,
-        thumbnail_infos: List[ThumbnailInfo],
-        file_id: str,
-        url_cache: bool,
-        server_name: Optional[str],
-    ) -> Optional[FileInfo]:
-        """
-        Choose an appropriate thumbnail from the previously generated thumbnails.
-
-        Args:
-            desired_width: The desired width, the returned thumbnail may be larger than this.
-            desired_height: The desired height, the returned thumbnail may be larger than this.
-            desired_method: The desired method used to generate the thumbnail.
-            desired_type: The desired content-type of the thumbnail.
-            thumbnail_infos: A list of thumbnail infos of candidate thumbnails.
-            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.
-
-        Returns:
-             The thumbnail which best matches the desired parameters.
-        """
-        desired_method = desired_method.lower()
-
-        # The chosen thumbnail.
-        thumbnail_info = None
-
-        d_w = desired_width
-        d_h = desired_height
-
-        if desired_method == "crop":
-            # Thumbnails that match equal or larger sizes of desired width/height.
-            crop_info_list: List[
-                Tuple[int, int, int, bool, Optional[int], ThumbnailInfo]
-            ] = []
-            # Other thumbnails.
-            crop_info_list2: List[
-                Tuple[int, int, int, bool, Optional[int], ThumbnailInfo]
-            ] = []
-            for info in thumbnail_infos:
-                # Skip thumbnails generated with different methods.
-                if info.method != "crop":
-                    continue
-
-                t_w = info.width
-                t_h = info.height
-                aspect_quality = abs(d_w * t_h - d_h * t_w)
-                min_quality = 0 if d_w <= t_w and d_h <= t_h else 1
-                size_quality = abs((d_w - t_w) * (d_h - t_h))
-                type_quality = desired_type != info.type
-                length_quality = info.length
-                if t_w >= d_w or t_h >= d_h:
-                    crop_info_list.append(
-                        (
-                            aspect_quality,
-                            min_quality,
-                            size_quality,
-                            type_quality,
-                            length_quality,
-                            info,
-                        )
-                    )
-                else:
-                    crop_info_list2.append(
-                        (
-                            aspect_quality,
-                            min_quality,
-                            size_quality,
-                            type_quality,
-                            length_quality,
-                            info,
-                        )
-                    )
-            # Pick the most appropriate thumbnail. Some values of `desired_width` and
-            # `desired_height` may result in a tie, in which case we avoid comparing on
-            # the thumbnail info and pick the thumbnail that appears earlier
-            # in the list of candidates.
-            if crop_info_list:
-                thumbnail_info = min(crop_info_list, key=lambda t: t[:-1])[-1]
-            elif crop_info_list2:
-                thumbnail_info = min(crop_info_list2, key=lambda t: t[:-1])[-1]
-        elif desired_method == "scale":
-            # Thumbnails that match equal or larger sizes of desired width/height.
-            info_list: List[Tuple[int, bool, int, ThumbnailInfo]] = []
-            # Other thumbnails.
-            info_list2: List[Tuple[int, bool, int, ThumbnailInfo]] = []
-
-            for info in thumbnail_infos:
-                # Skip thumbnails generated with different methods.
-                if info.method != "scale":
-                    continue
-
-                t_w = info.width
-                t_h = info.height
-                size_quality = abs((d_w - t_w) * (d_h - t_h))
-                type_quality = desired_type != info.type
-                length_quality = info.length
-                if t_w >= d_w or t_h >= d_h:
-                    info_list.append((size_quality, type_quality, length_quality, info))
-                else:
-                    info_list2.append(
-                        (size_quality, type_quality, length_quality, info)
-                    )
-            # Pick the most appropriate thumbnail. Some values of `desired_width` and
-            # `desired_height` may result in a tie, in which case we avoid comparing on
-            # the thumbnail info and pick the thumbnail that appears earlier
-            # in the list of candidates.
-            if info_list:
-                thumbnail_info = min(info_list, key=lambda t: t[:-1])[-1]
-            elif info_list2:
-                thumbnail_info = min(info_list2, key=lambda t: t[:-1])[-1]
-
-        if thumbnail_info:
-            return FileInfo(
-                file_id=file_id,
-                url_cache=url_cache,
-                server_name=server_name,
-                thumbnail=thumbnail_info,
-            )
-
-        # No matching thumbnail was found.
-        return None

tests/media/test_media_storage.py

@@ -18,6 +18,7 @@
 # [This file includes modifications made by New Vector Limited]
 #
 #
+import itertools
 import os
 import shutil
 import tempfile
@@ -46,11 +47,11 @@ from synapse.media._base import FileInfo, ThumbnailInfo
 from synapse.media.filepath import MediaFilePaths
 from synapse.media.media_storage import MediaStorage, ReadableFileWrapper
 from synapse.media.storage_provider import FileStorageProviderBackend
+from synapse.media.thumbnailer import ThumbnailProvider
 from synapse.module_api import ModuleApi
 from synapse.module_api.callbacks.spamchecker_callbacks import load_legacy_spam_checkers
 from synapse.rest import admin
-from synapse.rest.client import login
-from synapse.rest.media.thumbnail_resource import ThumbnailResource
+from synapse.rest.client import login, media
 from synapse.server import HomeServer
 from synapse.types import JsonDict, RoomAlias
 from synapse.util import Clock
@@ -153,12 +154,7 @@ class _TestImage:
     is_inline: bool = True
 
 
-@parameterized_class(
-    ("test_image",),
-    [
-        # small png
-        (
-            _TestImage(
+small_png = _TestImage(
     SMALL_PNG,
     b"image/png",
     b".png",
@@ -173,11 +169,9 @@ class _TestImage:
         b"0000001f15c4890000000d49444154789c636060606000000005"
         b"0001a5f645400000000049454e44ae426082"
     ),
-            ),
-        ),
-        # small png with transparency.
-        (
-            _TestImage(
+)
+
+small_png_with_transparency = _TestImage(
     unhexlify(
         b"89504e470d0a1a0a0000000d49484452000000010000000101000"
         b"00000376ef9240000000274524e5300010194fdae0000000a4944"
@@ -188,32 +182,25 @@ class _TestImage:
     b".png",
     # Note that we don't check the output since it varies across
     # different versions of Pillow.
-            ),
-        ),
-        # small lossless webp
-        (
-            _TestImage(
+)
+
+small_lossless_webp = _TestImage(
     unhexlify(
-                    b"524946461a000000574542505650384c0d0000002f0000001007"
-                    b"1011118888fe0700"
+        b"524946461a000000574542505650384c0d0000002f0000001007" b"1011118888fe0700"
     ),
     b"image/webp",
     b".webp",
-            ),
-        ),
-        # an empty file
-        (
-            _TestImage(
+)
+
+empty_file = _TestImage(
     b"",
     b"image/gif",
     b".gif",
     expected_found=False,
     unable_to_thumbnail=True,
-            ),
-        ),
-        # An SVG.
-        (
-            _TestImage(
+)
+
+SVG = _TestImage(
     b"""<?xml version="1.0"?>
 <!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN"
   "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
@@ -228,14 +215,27 @@ class _TestImage:
     expected_found=False,
     unable_to_thumbnail=True,
     is_inline=False,
-            ),
-        ),
-    ],
 )
+test_images = [
+    small_png,
+    small_png_with_transparency,
+    small_lossless_webp,
+    empty_file,
+    SVG,
+]
+urls = [
+    "_matrix/media/r0/thumbnail",
+    "_matrix/client/unstable/org.matrix.msc3916/media/thumbnail",
+]
+
+
+@parameterized_class(("test_image", "url"), itertools.product(test_images, urls))
 class MediaRepoTests(unittest.HomeserverTestCase):
+    servlets = [media.register_servlets]
     test_image: ClassVar[_TestImage]
     hijack_auth = True
     user_id = "@test:user"
+    url: ClassVar[str]
 
     def make_homeserver(self, reactor: MemoryReactor, clock: Clock) -> HomeServer:
         self.fetches: List[
@@ -298,6 +298,7 @@ class MediaRepoTests(unittest.HomeserverTestCase):
             "config": {"directory": self.storage_path},
         }
         config["media_storage_providers"] = [provider_config]
+        config["experimental_features"] = {"msc3916_authenticated_media_enabled": True}
 
         hs = self.setup_test_homeserver(config=config, federation_http_client=client)
 
@@ -502,7 +503,7 @@ class MediaRepoTests(unittest.HomeserverTestCase):
         params = "?width=32&height=32&method=scale"
         channel = self.make_request(
             "GET",
-            f"/_matrix/media/v3/thumbnail/{self.media_id}{params}",
+            f"/{self.url}/{self.media_id}{params}",
             shorthand=False,
             await_result=False,
         )
@@ -530,7 +531,7 @@ class MediaRepoTests(unittest.HomeserverTestCase):
 
         channel = self.make_request(
             "GET",
-            f"/_matrix/media/v3/thumbnail/{self.media_id}{params}",
+            f"/{self.url}/{self.media_id}{params}",
             shorthand=False,
             await_result=False,
         )
@@ -566,12 +567,11 @@ class MediaRepoTests(unittest.HomeserverTestCase):
         params = "?width=32&height=32&method=" + method
         channel = self.make_request(
             "GET",
-            f"/_matrix/media/r0/thumbnail/{self.media_id}{params}",
+            f"/{self.url}/{self.media_id}{params}",
             shorthand=False,
             await_result=False,
         )
         self.pump()
-
         headers = {
             b"Content-Length": [b"%d" % (len(self.test_image.data))],
             b"Content-Type": [self.test_image.content_type],
@@ -580,7 +580,6 @@ class MediaRepoTests(unittest.HomeserverTestCase):
             (self.test_image.data, (len(self.test_image.data), headers))
         )
         self.pump()
-
         if expected_found:
             self.assertEqual(channel.code, 200)
 
@@ -603,7 +602,7 @@ class MediaRepoTests(unittest.HomeserverTestCase):
                 channel.json_body,
                 {
                     "errcode": "M_UNKNOWN",
-                    "error": "Cannot find any thumbnails for the requested media ('/_matrix/media/r0/thumbnail/example.com/12345'). This might mean the media is not a supported_media_format=(image/jpeg, image/jpg, image/webp, image/gif, image/png) or that thumbnailing failed for some other reason. (Dynamic thumbnails are disabled on this server.)",
+                    "error": f"Cannot find any thumbnails for the requested media ('/{self.url}/example.com/12345'). This might mean the media is not a supported_media_format=(image/jpeg, image/jpg, image/webp, image/gif, image/png) or that thumbnailing failed for some other reason. (Dynamic thumbnails are disabled on this server.)",
                 },
             )
         else:
@@ -613,7 +612,7 @@ class MediaRepoTests(unittest.HomeserverTestCase):
                 channel.json_body,
                 {
                     "errcode": "M_NOT_FOUND",
-                    "error": "Not found '/_matrix/media/r0/thumbnail/example.com/12345'",
+                    "error": f"Not found '/{self.url}/example.com/12345'",
                 },
             )
 
@@ -625,12 +624,12 @@ class MediaRepoTests(unittest.HomeserverTestCase):
 
         content_type = self.test_image.content_type.decode()
         media_repo = self.hs.get_media_repository()
-        thumbnail_resouce = ThumbnailResource(
+        thumbnail_provider = ThumbnailProvider(
             self.hs, media_repo, media_repo.media_storage
         )
 
         self.assertIsNotNone(
-            thumbnail_resouce._select_thumbnail(
+            thumbnail_provider._select_thumbnail(
                 desired_width=desired_size,
                 desired_height=desired_size,
                 desired_method=method,