New upstream version 1.21.1

7 files changed, 154 insertions, 31 deletions

diff --git a/synapse/rest/media/v1/filepath.py b/synapse/rest/media/v1/filepath.py
index d2826374..7447eeae 100644
--- a/synapse/rest/media/v1/filepath.py
+++ b/synapse/rest/media/v1/filepath.py
@@ -80,7 +80,7 @@ class MediaFilePaths:
         self, server_name, file_id, width, height, content_type, method
     ):
         top_level_type, sub_type = content_type.split("/")
-        file_name = "%i-%i-%s-%s" % (width, height, top_level_type, sub_type)
+        file_name = "%i-%i-%s-%s-%s" % (width, height, top_level_type, sub_type, method)
         return os.path.join(
             "remote_thumbnail",
             server_name,
@@ -92,6 +92,23 @@ class MediaFilePaths:
 
     remote_media_thumbnail = _wrap_in_base_path(remote_media_thumbnail_rel)
 
+    # Legacy path that was used to store thumbnails previously.
+    # Should be removed after some time, when most of the thumbnails are stored
+    # using the new path.
+    def remote_media_thumbnail_rel_legacy(
+        self, server_name, file_id, width, height, content_type
+    ):
+        top_level_type, sub_type = content_type.split("/")
+        file_name = "%i-%i-%s-%s" % (width, height, top_level_type, sub_type)
+        return os.path.join(
+            "remote_thumbnail",
+            server_name,
+            file_id[0:2],
+            file_id[2:4],
+            file_id[4:],
+            file_name,
+        )
+
     def remote_media_thumbnail_dir(self, server_name, file_id):
         return os.path.join(
             self.base_path,
diff --git a/synapse/rest/media/v1/media_repository.py b/synapse/rest/media/v1/media_repository.py
index 9a1b7779..e1192b47 100644
--- a/synapse/rest/media/v1/media_repository.py
+++ b/synapse/rest/media/v1/media_repository.py
@@ -53,7 +53,7 @@ from .media_storage import MediaStorage
 from .preview_url_resource import PreviewUrlResource
 from .storage_provider import StorageProviderWrapper
 from .thumbnail_resource import ThumbnailResource
-from .thumbnailer import Thumbnailer
+from .thumbnailer import Thumbnailer, ThumbnailError
 from .upload_resource import UploadResource
 
 logger = logging.getLogger(__name__)
@@ -139,7 +139,7 @@ class MediaRepository:
     async def create_content(
         self,
         media_type: str,
-        upload_name: str,
+        upload_name: Optional[str],
         content: IO,
         content_length: int,
         auth_user: str,
@@ -147,8 +147,8 @@ class MediaRepository:
         """Store uploaded content for a local user and return the mxc URL
 
         Args:
-            media_type: The content type of the file
-            upload_name: The name of the file
+            media_type: The content type of the file.
+            upload_name: The name of the file, if provided.
             content: A file like object that is the content to store
             content_length: The length of the content
             auth_user: The user_id of the uploader
@@ -156,6 +156,7 @@ class MediaRepository:
         Returns:
             The mxc url of the stored content
         """
+
         media_id = random_string(24)
 
         file_info = FileInfo(server_name=None, file_id=media_id)
@@ -460,13 +461,30 @@ class MediaRepository:
         return t_byte_source
 
     async def generate_local_exact_thumbnail(
-        self, media_id, t_width, t_height, t_method, t_type, url_cache
-    ):
+        self,
+        media_id: str,
+        t_width: int,
+        t_height: int,
+        t_method: str,
+        t_type: str,
+        url_cache: str,
+    ) -> Optional[str]:
         input_path = await self.media_storage.ensure_media_is_in_local_cache(
             FileInfo(None, media_id, url_cache=url_cache)
         )
 
-        thumbnailer = Thumbnailer(input_path)
+        try:
+            thumbnailer = Thumbnailer(input_path)
+        except ThumbnailError as e:
+            logger.warning(
+                "Unable to generate a thumbnail for local media %s using a method of %s and type of %s: %s",
+                media_id,
+                t_method,
+                t_type,
+                e,
+            )
+            return None
+
         t_byte_source = await defer_to_thread(
             self.hs.get_reactor(),
             self._generate_thumbnail,
@@ -506,14 +524,36 @@ class MediaRepository:
 
             return output_path
 
+        # Could not generate thumbnail.
+        return None
+
     async def generate_remote_exact_thumbnail(
-        self, server_name, file_id, media_id, t_width, t_height, t_method, t_type
-    ):
+        self,
+        server_name: str,
+        file_id: str,
+        media_id: str,
+        t_width: int,
+        t_height: int,
+        t_method: str,
+        t_type: str,
+    ) -> Optional[str]:
         input_path = await self.media_storage.ensure_media_is_in_local_cache(
             FileInfo(server_name, file_id, url_cache=False)
         )
 
-        thumbnailer = Thumbnailer(input_path)
+        try:
+            thumbnailer = Thumbnailer(input_path)
+        except ThumbnailError as e:
+            logger.warning(
+                "Unable to generate a thumbnail for remote media %s from %s using a method of %s and type of %s: %s",
+                media_id,
+                server_name,
+                t_method,
+                t_type,
+                e,
+            )
+            return None
+
         t_byte_source = await defer_to_thread(
             self.hs.get_reactor(),
             self._generate_thumbnail,
@@ -559,6 +599,9 @@ class MediaRepository:
 
             return output_path
 
+        # Could not generate thumbnail.
+        return None
+
     async def _generate_thumbnails(
         self,
         server_name: Optional[str],
@@ -590,7 +633,18 @@ class MediaRepository:
             FileInfo(server_name, file_id, url_cache=url_cache)
         )
 
-        thumbnailer = Thumbnailer(input_path)
+        try:
+            thumbnailer = Thumbnailer(input_path)
+        except ThumbnailError as e:
+            logger.warning(
+                "Unable to generate thumbnails for remote media %s from %s of type %s: %s",
+                media_id,
+                server_name,
+                media_type,
+                e,
+            )
+            return None
+
         m_width = thumbnailer.width
         m_height = thumbnailer.height
 
diff --git a/synapse/rest/media/v1/media_storage.py b/synapse/rest/media/v1/media_storage.py
index 3a352b56..a9586fb0 100644
--- a/synapse/rest/media/v1/media_storage.py
+++ b/synapse/rest/media/v1/media_storage.py
@@ -141,17 +141,34 @@ class MediaStorage:
         Returns:
             Returns a Responder if the file was found, otherwise None.
         """
+        paths = [self._file_info_to_path(file_info)]
 
-        path = self._file_info_to_path(file_info)
-        local_path = os.path.join(self.local_media_directory, path)
-        if os.path.exists(local_path):
-            return FileResponder(open(local_path, "rb"))
+        # fallback for remote thumbnails with no method in the filename
+        if file_info.thumbnail and file_info.server_name:
+            paths.append(
+                self.filepaths.remote_media_thumbnail_rel_legacy(
+                    server_name=file_info.server_name,
+                    file_id=file_info.file_id,
+                    width=file_info.thumbnail_width,
+                    height=file_info.thumbnail_height,
+                    content_type=file_info.thumbnail_type,
+                )
+            )
+
+        for path in paths:
+            local_path = os.path.join(self.local_media_directory, path)
+            if os.path.exists(local_path):
+                logger.debug("responding with local file %s", local_path)
+                return FileResponder(open(local_path, "rb"))
+            logger.debug("local file %s did not exist", local_path)
 
         for provider in self.storage_providers:
-            res = await provider.fetch(path, file_info)  # type: Any
-            if res:
-                logger.debug("Streaming %s from %s", path, provider)
-                return res
+            for path in paths:
+                res = await provider.fetch(path, file_info)  # type: Any
+                if res:
+                    logger.debug("Streaming %s from %s", path, provider)
+                    return res
+                logger.debug("%s not found on %s", path, provider)
 
         return None
 
@@ -170,6 +187,20 @@ class MediaStorage:
         if os.path.exists(local_path):
             return local_path
 
+        # Fallback for paths without method names
+        # Should be removed in the future
+        if file_info.thumbnail and file_info.server_name:
+            legacy_path = self.filepaths.remote_media_thumbnail_rel_legacy(
+                server_name=file_info.server_name,
+                file_id=file_info.file_id,
+                width=file_info.thumbnail_width,
+                height=file_info.thumbnail_height,
+                content_type=file_info.thumbnail_type,
+            )
+            legacy_local_path = os.path.join(self.local_media_directory, legacy_path)
+            if os.path.exists(legacy_local_path):
+                return legacy_local_path
+
         dirname = os.path.dirname(local_path)
         if not os.path.exists(dirname):
             os.makedirs(dirname)
diff --git a/synapse/rest/media/v1/preview_url_resource.py b/synapse/rest/media/v1/preview_url_resource.py
index cd8c2465..dce6c4d1 100644
--- a/synapse/rest/media/v1/preview_url_resource.py
+++ b/synapse/rest/media/v1/preview_url_resource.py
@@ -102,7 +102,7 @@ for endpoint, globs in _oembed_globs.items():
         _oembed_patterns[re.compile(pattern)] = endpoint
 
 
-@attr.s
+@attr.s(slots=True)
 class OEmbedResult:
     # Either HTML content or URL must be provided.
     html = attr.ib(type=Optional[str])
@@ -450,7 +450,7 @@ class PreviewUrlResource(DirectServeJsonResource):
             logger.warning("Error downloading oEmbed metadata from %s: %r", url, e)
             raise OEmbedError() from e
 
-    async def _download_url(self, url, user):
+    async def _download_url(self, url: str, user):
         # TODO: we should probably honour robots.txt... except in practice
         # we're most likely being explicitly triggered by a human rather than a
         # bot, so are we really a robot?
@@ -460,7 +460,7 @@ class PreviewUrlResource(DirectServeJsonResource):
         file_info = FileInfo(server_name=None, file_id=file_id, url_cache=True)
 
         # If this URL can be accessed via oEmbed, use that instead.
-        url_to_download = url
+        url_to_download = url  # type: Optional[str]
         oembed_url = self._get_oembed_url(url)
         if oembed_url:
             # The result might be a new URL to download, or it might be HTML content.
@@ -520,9 +520,15 @@ class PreviewUrlResource(DirectServeJsonResource):
                 # FIXME: we should calculate a proper expiration based on the
                 # Cache-Control and Expire headers.  But for now, assume 1 hour.
                 expires = ONE_HOUR
-                etag = headers["ETag"][0] if "ETag" in headers else None
+                etag = (
+                    headers[b"ETag"][0].decode("ascii") if b"ETag" in headers else None
+                )
         else:
-            html_bytes = oembed_result.html.encode("utf-8")  # type: ignore
+            # we can only get here if we did an oembed request and have an oembed_result.html
+            assert oembed_result.html is not None
+            assert oembed_url is not None
+
+            html_bytes = oembed_result.html.encode("utf-8")
             with self.media_storage.store_into_file(file_info) as (f, fname, finish):
                 f.write(html_bytes)
                 await finish()
diff --git a/synapse/rest/media/v1/thumbnail_resource.py b/synapse/rest/media/v1/thumbnail_resource.py
index a83535b9..30421b66 100644
--- a/synapse/rest/media/v1/thumbnail_resource.py
+++ b/synapse/rest/media/v1/thumbnail_resource.py
@@ -16,6 +16,7 @@
 
 import logging
 
+from synapse.api.errors import SynapseError
 from synapse.http.server import DirectServeJsonResource, set_cors_headers
 from synapse.http.servlet import parse_integer, parse_string
 
@@ -173,7 +174,7 @@ class ThumbnailResource(DirectServeJsonResource):
             await respond_with_file(request, desired_type, file_path)
         else:
             logger.warning("Failed to generate thumbnail")
-            respond_404(request)
+            raise SynapseError(400, "Failed to generate thumbnail.")
 
     async def _select_or_generate_remote_thumbnail(
         self,
@@ -235,7 +236,7 @@ class ThumbnailResource(DirectServeJsonResource):
             await respond_with_file(request, desired_type, file_path)
         else:
             logger.warning("Failed to generate thumbnail")
-            respond_404(request)
+            raise SynapseError(400, "Failed to generate thumbnail.")
 
     async def _respond_remote_thumbnail(
         self, request, server_name, media_id, width, height, method, m_type
diff --git a/synapse/rest/media/v1/thumbnailer.py b/synapse/rest/media/v1/thumbnailer.py
index d681bf7b..32a8e4f9 100644
--- a/synapse/rest/media/v1/thumbnailer.py
+++ b/synapse/rest/media/v1/thumbnailer.py
@@ -15,7 +15,7 @@
 import logging
 from io import BytesIO
 
-from PIL import Image as Image
+from PIL import Image
 
 logger = logging.getLogger(__name__)
 
@@ -31,12 +31,22 @@ EXIF_TRANSPOSE_MAPPINGS = {
 }
 
 
+class ThumbnailError(Exception):
+    """An error occurred generating a thumbnail."""
+
+
 class Thumbnailer:
 
     FORMATS = {"image/jpeg": "JPEG", "image/png": "PNG"}
 
     def __init__(self, input_path):
-        self.image = Image.open(input_path)
+        try:
+            self.image = Image.open(input_path)
+        except OSError as e:
+            # If an error occurs opening the image, a thumbnail won't be able to
+            # be generated.
+            raise ThumbnailError from e
+
         self.width, self.height = self.image.size
         self.transpose_method = None
         try:
@@ -73,7 +83,7 @@ class Thumbnailer:
 
         Args:
             max_width: The largest possible width.
-            max_height: The larget possible height.
+            max_height: The largest possible height.
         """
 
         if max_width * self.height < max_height * self.width:
@@ -107,7 +117,7 @@ class Thumbnailer:
 
         Args:
             max_width: The largest possible width.
-            max_height: The larget possible height.
+            max_height: The largest possible height.
 
         Returns:
             BytesIO: the bytes of the encoded image ready to be written to disk
diff --git a/synapse/rest/media/v1/upload_resource.py b/synapse/rest/media/v1/upload_resource.py
index 3ebf7a68..d76f7389 100644
--- a/synapse/rest/media/v1/upload_resource.py
+++ b/synapse/rest/media/v1/upload_resource.py
@@ -63,6 +63,10 @@ class UploadResource(DirectServeJsonResource):
                     msg="Invalid UTF-8 filename parameter: %r" % (upload_name), code=400
                 )
 
+        # If the name is falsey (e.g. an empty byte string) ensure it is None.
+        else:
+            upload_name = None
+
         headers = request.requestHeaders
 
         if headers.hasHeader(b"Content-Type"):