Register media servlets via regex. (#16419)

This converts the media servlet URLs in the same way as
(most) of the rest of Synapse. This will give more flexibility
in the versions each endpoint exists under.

9 files changed, 103 insertions, 118 deletions

diff --git a/synapse/http/server.py b/synapse/http/server.py
index 3bbf91298e..1e4e56f36b 100644
--- a/synapse/http/server.py
+++ b/synapse/http/server.py
@@ -266,7 +266,7 @@ class HttpServer(Protocol):
     def register_paths(
         self,
         method: str,
-        path_patterns: Iterable[Pattern],
+        path_patterns: Iterable[Pattern[str]],
         callback: ServletCallback,
         servlet_classname: str,
     ) -> None:
diff --git a/synapse/media/_base.py b/synapse/media/_base.py
index 80c448de2b..d103b43449 100644
--- a/synapse/media/_base.py
+++ b/synapse/media/_base.py
@@ -26,11 +26,11 @@ from twisted.internet.interfaces import IConsumer
 from twisted.protocols.basic import FileSender
 from twisted.web.server import Request
 
-from synapse.api.errors import Codes, SynapseError, cs_error
+from synapse.api.errors import Codes, cs_error
 from synapse.http.server import finish_request, respond_with_json
 from synapse.http.site import SynapseRequest
 from synapse.logging.context import make_deferred_yieldable
-from synapse.util.stringutils import is_ascii, parse_and_validate_server_name
+from synapse.util.stringutils import is_ascii
 
 logger = logging.getLogger(__name__)
 
@@ -84,52 +84,12 @@ INLINE_CONTENT_TYPES = [
 ]
 
 
-def parse_media_id(request: Request) -> Tuple[str, str, Optional[str]]:
-    """Parses the server name, media ID and optional file name from the request URI
-
-    Also performs some rough validation on the server name.
-
-    Args:
-        request: The `Request`.
-
-    Returns:
-        A tuple containing the parsed server name, media ID and optional file name.
-
-    Raises:
-        SynapseError(404): if parsing or validation fail for any reason
-    """
-    try:
-        # The type on postpath seems incorrect in Twisted 21.2.0.
-        postpath: List[bytes] = request.postpath  # type: ignore
-        assert postpath
-
-        # This allows users to append e.g. /test.png to the URL. Useful for
-        # clients that parse the URL to see content type.
-        server_name_bytes, media_id_bytes = postpath[:2]
-        server_name = server_name_bytes.decode("utf-8")
-        media_id = media_id_bytes.decode("utf8")
-
-        # Validate the server name, raising if invalid
-        parse_and_validate_server_name(server_name)
-
-        file_name = None
-        if len(postpath) > 2:
-            try:
-                file_name = urllib.parse.unquote(postpath[-1].decode("utf-8"))
-            except UnicodeDecodeError:
-                pass
-        return server_name, media_id, file_name
-    except Exception:
-        raise SynapseError(
-            404, "Invalid media id token %r" % (request.postpath,), Codes.UNKNOWN
-        )
-
-
 def respond_404(request: SynapseRequest) -> None:
+    assert request.path is not None
     respond_with_json(
         request,
         404,
-        cs_error("Not found %r" % (request.postpath,), code=Codes.NOT_FOUND),
+        cs_error("Not found '%s'" % (request.path.decode(),), code=Codes.NOT_FOUND),
         send_cors=True,
     )
 
diff --git a/synapse/media/media_repository.py b/synapse/media/media_repository.py
index 1b7b014f9a..d11c2ff4ee 100644
--- a/synapse/media/media_repository.py
+++ b/synapse/media/media_repository.py
@@ -48,6 +48,7 @@ from synapse.media.filepath import MediaFilePaths
 from synapse.media.media_storage import MediaStorage
 from synapse.media.storage_provider import StorageProviderWrapper
 from synapse.media.thumbnailer import Thumbnailer, ThumbnailError
+from synapse.media.url_previewer import UrlPreviewer
 from synapse.metrics.background_process_metrics import run_as_background_process
 from synapse.types import UserID
 from synapse.util.async_helpers import Linearizer
@@ -114,7 +115,7 @@ class MediaRepository:
             )
             storage_providers.append(provider)
 
-        self.media_storage = MediaStorage(
+        self.media_storage: MediaStorage = MediaStorage(
             self.hs, self.primary_base_path, self.filepaths, storage_providers
         )
 
@@ -142,6 +143,13 @@ class MediaRepository:
                 MEDIA_RETENTION_CHECK_PERIOD_MS,
             )
 
+        if hs.config.media.url_preview_enabled:
+            self.url_previewer: Optional[UrlPreviewer] = UrlPreviewer(
+                hs, self, self.media_storage
+            )
+        else:
+            self.url_previewer = None
+
     def _start_update_recently_accessed(self) -> Deferred:
         return run_as_background_process(
             "update_recently_accessed_media", self._update_recently_accessed
diff --git a/synapse/rest/media/config_resource.py b/synapse/rest/media/config_resource.py
index a95804d327..dbf5133c72 100644
--- a/synapse/rest/media/config_resource.py
+++ b/synapse/rest/media/config_resource.py
@@ -14,17 +14,19 @@
 # limitations under the License.
 #
 
+import re
 from typing import TYPE_CHECKING
 
-from synapse.http.server import DirectServeJsonResource, respond_with_json
+from synapse.http.server import respond_with_json
+from synapse.http.servlet import RestServlet
 from synapse.http.site import SynapseRequest
 
 if TYPE_CHECKING:
     from synapse.server import HomeServer
 
 
-class MediaConfigResource(DirectServeJsonResource):
-    isLeaf = True
+class MediaConfigResource(RestServlet):
+    PATTERNS = [re.compile("/_matrix/media/(r0|v3|v1)/config$")]
 
     def __init__(self, hs: "HomeServer"):
         super().__init__()
@@ -33,9 +35,6 @@ class MediaConfigResource(DirectServeJsonResource):
         self.auth = hs.get_auth()
         self.limits_dict = {"m.upload.size": config.media.max_upload_size}
 
-    async def _async_render_GET(self, request: SynapseRequest) -> None:
+    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)
-
-    async def _async_render_OPTIONS(self, request: SynapseRequest) -> None:
-        respond_with_json(request, 200, {}, send_cors=True)
diff --git a/synapse/rest/media/download_resource.py b/synapse/rest/media/download_resource.py
index 3c618ef60a..65b9ff52fa 100644
--- a/synapse/rest/media/download_resource.py
+++ b/synapse/rest/media/download_resource.py
@@ -13,16 +13,14 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import logging
-from typing import TYPE_CHECKING
+import re
+from typing import TYPE_CHECKING, Optional
 
-from synapse.http.server import (
-    DirectServeJsonResource,
-    set_corp_headers,
-    set_cors_headers,
-)
-from synapse.http.servlet import parse_boolean
+from synapse.http.server import set_corp_headers, set_cors_headers
+from synapse.http.servlet import RestServlet, parse_boolean
 from synapse.http.site import SynapseRequest
-from synapse.media._base import parse_media_id, respond_404
+from synapse.media._base import respond_404
+from synapse.util.stringutils import parse_and_validate_server_name
 
 if TYPE_CHECKING:
     from synapse.media.media_repository import MediaRepository
@@ -31,15 +29,28 @@ if TYPE_CHECKING:
 logger = logging.getLogger(__name__)
 
 
-class DownloadResource(DirectServeJsonResource):
-    isLeaf = True
+class DownloadResource(RestServlet):
+    PATTERNS = [
+        re.compile(
+            "/_matrix/media/(r0|v3|v1)/download/(?P<server_name>[^/]*)/(?P<media_id>[^/]*)(/(?P<file_name>[^/]*))?$"
+        )
+    ]
 
     def __init__(self, hs: "HomeServer", media_repo: "MediaRepository"):
         super().__init__()
         self.media_repo = media_repo
         self._is_mine_server_name = hs.is_mine_server_name
 
-    async def _async_render_GET(self, request: SynapseRequest) -> None:
+    async def on_GET(
+        self,
+        request: SynapseRequest,
+        server_name: str,
+        media_id: str,
+        file_name: Optional[str] = None,
+    ) -> None:
+        # Validate the server name, raising if invalid
+        parse_and_validate_server_name(server_name)
+
         set_cors_headers(request)
         set_corp_headers(request)
         request.setHeader(
@@ -58,9 +69,8 @@ class DownloadResource(DirectServeJsonResource):
             b"Referrer-Policy",
             b"no-referrer",
         )
-        server_name, media_id, name = parse_media_id(request)
         if self._is_mine_server_name(server_name):
-            await self.media_repo.get_local_media(request, media_id, name)
+            await self.media_repo.get_local_media(request, media_id, file_name)
         else:
             allow_remote = parse_boolean(request, "allow_remote", default=True)
             if not allow_remote:
@@ -72,4 +82,6 @@ class DownloadResource(DirectServeJsonResource):
                 respond_404(request)
                 return
 
-            await self.media_repo.get_remote_media(request, server_name, media_id, name)
+            await self.media_repo.get_remote_media(
+                request, server_name, media_id, file_name
+            )
diff --git a/synapse/rest/media/media_repository_resource.py b/synapse/rest/media/media_repository_resource.py
index 5ebaa3b032..2089bb1029 100644
--- a/synapse/rest/media/media_repository_resource.py
+++ b/synapse/rest/media/media_repository_resource.py
@@ -15,7 +15,7 @@
 from typing import TYPE_CHECKING
 
 from synapse.config._base import ConfigError
-from synapse.http.server import UnrecognizedRequestResource
+from synapse.http.server import HttpServer, JsonResource
 
 from .config_resource import MediaConfigResource
 from .download_resource import DownloadResource
@@ -27,7 +27,7 @@ if TYPE_CHECKING:
     from synapse.server import HomeServer
 
 
-class MediaRepositoryResource(UnrecognizedRequestResource):
+class MediaRepositoryResource(JsonResource):
     """File uploading and downloading.
 
     Uploads are POSTed to a resource which returns a token which is used to GET
@@ -70,6 +70,11 @@ class MediaRepositoryResource(UnrecognizedRequestResource):
     width and height are close to the requested size and the aspect matches
     the requested size. The client should scale the image if it needs to fit
     within a given rectangle.
+
+    This gets mounted at various points under /_matrix/media, including:
+       * /_matrix/media/r0
+       * /_matrix/media/v1
+       * /_matrix/media/v3
     """
 
     def __init__(self, hs: "HomeServer"):
@@ -77,17 +82,23 @@ class MediaRepositoryResource(UnrecognizedRequestResource):
         if not hs.config.media.can_load_media_repo:
             raise ConfigError("Synapse is not configured to use a media repo.")
 
-        super().__init__()
+        JsonResource.__init__(self, hs, canonical_json=False)
+        self.register_servlets(self, hs)
+
+    @staticmethod
+    def register_servlets(http_server: HttpServer, hs: "HomeServer") -> None:
         media_repo = hs.get_media_repository()
 
-        self.putChild(b"upload", UploadResource(hs, media_repo))
-        self.putChild(b"download", DownloadResource(hs, media_repo))
-        self.putChild(
-            b"thumbnail", ThumbnailResource(hs, media_repo, media_repo.media_storage)
+        # Note that many of these should not exist as v1 endpoints, but empirically
+        # a lot of traffic still goes to them.
+
+        UploadResource(hs, media_repo).register(http_server)
+        DownloadResource(hs, media_repo).register(http_server)
+        ThumbnailResource(hs, media_repo, media_repo.media_storage).register(
+            http_server
         )
         if hs.config.media.url_preview_enabled:
-            self.putChild(
-                b"preview_url",
-                PreviewUrlResource(hs, media_repo, media_repo.media_storage),
+            PreviewUrlResource(hs, media_repo, media_repo.media_storage).register(
+                http_server
             )
-        self.putChild(b"config", MediaConfigResource(hs))
+        MediaConfigResource(hs).register(http_server)
diff --git a/synapse/rest/media/preview_url_resource.py b/synapse/rest/media/preview_url_resource.py
index 58513c4be4..c8acb65dca 100644
--- a/synapse/rest/media/preview_url_resource.py
+++ b/synapse/rest/media/preview_url_resource.py
@@ -13,24 +13,20 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import re
 from typing import TYPE_CHECKING
 
-from synapse.http.server import (
-    DirectServeJsonResource,
-    respond_with_json,
-    respond_with_json_bytes,
-)
-from synapse.http.servlet import parse_integer, parse_string
+from synapse.http.server import respond_with_json_bytes
+from synapse.http.servlet import RestServlet, parse_integer, parse_string
 from synapse.http.site import SynapseRequest
 from synapse.media.media_storage import MediaStorage
-from synapse.media.url_previewer import UrlPreviewer
 
 if TYPE_CHECKING:
     from synapse.media.media_repository import MediaRepository
     from synapse.server import HomeServer
 
 
-class PreviewUrlResource(DirectServeJsonResource):
+class PreviewUrlResource(RestServlet):
     """
     The `GET /_matrix/media/r0/preview_url` endpoint provides a generic preview API
     for URLs which outputs Open Graph (https://ogp.me/) responses (with some Matrix
@@ -48,7 +44,7 @@ class PreviewUrlResource(DirectServeJsonResource):
       * Matrix cannot be used to distribute the metadata between homeservers.
     """
 
-    isLeaf = True
+    PATTERNS = [re.compile("/_matrix/media/(r0|v3|v1)/preview_url$")]
 
     def __init__(
         self,
@@ -62,14 +58,10 @@ class PreviewUrlResource(DirectServeJsonResource):
         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
 
-        self._url_previewer = UrlPreviewer(hs, media_repo, media_storage)
-
-    async def _async_render_OPTIONS(self, request: SynapseRequest) -> None:
-        request.setHeader(b"Allow", b"OPTIONS, GET")
-        respond_with_json(request, 200, {}, send_cors=True)
-
-    async def _async_render_GET(self, request: SynapseRequest) -> None:
+    async def on_GET(self, request: SynapseRequest) -> None:
         # XXX: if get_user_by_req fails, what should we do in an async render?
         requester = await self.auth.get_user_by_req(request)
         url = parse_string(request, "url", required=True)
@@ -77,5 +69,5 @@ class PreviewUrlResource(DirectServeJsonResource):
         if ts is None:
             ts = self.clock.time_msec()
 
-        og = await self._url_previewer.preview(url, requester.user, ts)
+        og = await self.url_previewer.preview(url, requester.user, ts)
         respond_with_json_bytes(request, 200, og, send_cors=True)
diff --git a/synapse/rest/media/thumbnail_resource.py b/synapse/rest/media/thumbnail_resource.py
index 661e604b85..f9cd773f77 100644
--- a/synapse/rest/media/thumbnail_resource.py
+++ b/synapse/rest/media/thumbnail_resource.py
@@ -13,29 +13,24 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-
 import logging
+import re
 from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
 
 from synapse.api.errors import Codes, SynapseError, cs_error
 from synapse.config.repository import THUMBNAIL_SUPPORTED_MEDIA_FORMAT_MAP
-from synapse.http.server import (
-    DirectServeJsonResource,
-    respond_with_json,
-    set_corp_headers,
-    set_cors_headers,
-)
-from synapse.http.servlet import parse_integer, parse_string
+from synapse.http.server import respond_with_json, 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 (
     FileInfo,
     ThumbnailInfo,
-    parse_media_id,
     respond_404,
     respond_with_file,
     respond_with_responder,
 )
 from synapse.media.media_storage import MediaStorage
+from synapse.util.stringutils import parse_and_validate_server_name
 
 if TYPE_CHECKING:
     from synapse.media.media_repository import MediaRepository
@@ -44,8 +39,12 @@ if TYPE_CHECKING:
 logger = logging.getLogger(__name__)
 
 
-class ThumbnailResource(DirectServeJsonResource):
-    isLeaf = True
+class ThumbnailResource(RestServlet):
+    PATTERNS = [
+        re.compile(
+            "/_matrix/media/(r0|v3|v1)/thumbnail/(?P<server_name>[^/]*)/(?P<media_id>[^/]*)$"
+        )
+    ]
 
     def __init__(
         self,
@@ -60,12 +59,17 @@ class ThumbnailResource(DirectServeJsonResource):
         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
 
-    async def _async_render_GET(self, request: SynapseRequest) -> None:
+    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)
+
         set_cors_headers(request)
         set_corp_headers(request)
-        server_name, media_id, _ = parse_media_id(request)
         width = parse_integer(request, "width", required=True)
         height = parse_integer(request, "height", required=True)
         method = parse_string(request, "method", "scale")
@@ -418,13 +422,14 @@ class ThumbnailResource(DirectServeJsonResource):
             # `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 (%r). 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.)"
+                    "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.postpath,
+                        request.path.decode(),
                         ", ".join(THUMBNAIL_SUPPORTED_MEDIA_FORMAT_MAP.keys()),
                     ),
                     code=Codes.UNKNOWN,
diff --git a/synapse/rest/media/upload_resource.py b/synapse/rest/media/upload_resource.py
index 043e8d6077..949326d85d 100644
--- a/synapse/rest/media/upload_resource.py
+++ b/synapse/rest/media/upload_resource.py
@@ -14,11 +14,12 @@
 # limitations under the License.
 
 import logging
+import re
 from typing import IO, TYPE_CHECKING, Dict, List, Optional
 
 from synapse.api.errors import Codes, SynapseError
-from synapse.http.server import DirectServeJsonResource, respond_with_json
-from synapse.http.servlet import parse_bytes_from_args
+from synapse.http.server import respond_with_json
+from synapse.http.servlet import RestServlet, parse_bytes_from_args
 from synapse.http.site import SynapseRequest
 from synapse.media.media_storage import SpamMediaException
 
@@ -29,8 +30,8 @@ if TYPE_CHECKING:
 logger = logging.getLogger(__name__)
 
 
-class UploadResource(DirectServeJsonResource):
-    isLeaf = True
+class UploadResource(RestServlet):
+    PATTERNS = [re.compile("/_matrix/media/(r0|v3|v1)/upload")]
 
     def __init__(self, hs: "HomeServer", media_repo: "MediaRepository"):
         super().__init__()
@@ -43,10 +44,7 @@ class UploadResource(DirectServeJsonResource):
         self.max_upload_size = hs.config.media.max_upload_size
         self.clock = hs.get_clock()
 
-    async def _async_render_OPTIONS(self, request: SynapseRequest) -> None:
-        respond_with_json(request, 200, {}, send_cors=True)
-
-    async def _async_render_POST(self, request: SynapseRequest) -> None:
+    async def on_POST(self, request: SynapseRequest) -> None:
         requester = await self.auth.get_user_by_req(request)
         raw_content_length = request.getHeader("Content-Length")
         if raw_content_length is None: