Basic implementation of backup media store

synapse/config/repository.py

@@ -70,7 +70,19 @@ def read_config(self, config):
         self.max_upload_size = self.parse_size(config["max_upload_size"])
         self.max_image_pixels = self.parse_size(config["max_image_pixels"])
         self.max_spider_size = self.parse_size(config["max_spider_size"])
+
         self.media_store_path = self.ensure_directory(config["media_store_path"])
+
+        self.backup_media_store_path = config.get("backup_media_store_path")
+        if self.backup_media_store_path:
+            self.backup_media_store_path = self.ensure_directory(
+                self.backup_media_store_path
+            )
+
+        self.synchronous_backup_media_store = config.get(
+            "synchronous_backup_media_store", False
+        )
+
         self.uploads_path = self.ensure_directory(config["uploads_path"])
         self.dynamic_thumbnails = config["dynamic_thumbnails"]
         self.thumbnail_requirements = parse_thumbnail_requirements(
@@ -115,6 +127,14 @@ def default_config(self, **kwargs):
         # Directory where uploaded images and attachments are stored.
         media_store_path: "%(media_store)s"
 
+        # A secondary directory where uploaded images and attachments are
+        # stored as a backup.
+        # backup_media_store_path: "%(media_store)s"
+
+        # Whether to wait for successful write to backup media store before
+        # returning successfully.
+        # synchronous_backup_media_store: false
+
         # Directory where in-progress uploads are stored.
         uploads_path: "%(uploads_path)s"
 

synapse/rest/media/v1/filepath.py

@@ -15,80 +15,111 @@
 
 import os
 import re
+import functools
 
 NEW_FORMAT_ID_RE = re.compile(r"^\d\d\d\d-\d\d-\d\d")
 
 
+def _wrap_in_base_path(func):
+    """Takes a function that returns a relative path and turns it into an
+    absolute path based on the location of the primary media store
+    """
+    @functools.wraps(func)
+    def _wrapped(self, *args, **kwargs):
+        path = func(self, *args, **kwargs)
+        return os.path.join(self.base_path, path)
+
+    return _wrapped
+
+
 class MediaFilePaths(object):
+    """Describes where files are stored on disk.
+
+    Most of the functions have a `*_rel` variant which returns a file path that
+    is relative to the base media store path. This is mainly used when we want
+    to write to the backup media store (when one is configured)
+    """
 
-    def __init__(self, base_path):
-        self.base_path = base_path
+    def __init__(self, primary_base_path):
+        self.base_path = primary_base_path
 
-    def default_thumbnail(self, default_top_level, default_sub_type, width,
-                          height, content_type, method):
+    def default_thumbnail_rel(self, default_top_level, default_sub_type, width,
+                              height, content_type, method):
         top_level_type, sub_type = content_type.split("/")
         file_name = "%i-%i-%s-%s-%s" % (
             width, height, top_level_type, sub_type, method
         )
         return os.path.join(
-            self.base_path, "default_thumbnails", default_top_level,
+            "default_thumbnails", default_top_level,
             default_sub_type, file_name
         )
 
-    def local_media_filepath(self, media_id):
+    default_thumbnail = _wrap_in_base_path(default_thumbnail_rel)
+
+    def local_media_filepath_rel(self, media_id):
         return os.path.join(
-            self.base_path, "local_content",
+            "local_content",
             media_id[0:2], media_id[2:4], media_id[4:]
         )
 
-    def local_media_thumbnail(self, media_id, width, height, content_type,
-                              method):
+    local_media_filepath = _wrap_in_base_path(local_media_filepath_rel)
+
+    def local_media_thumbnail_rel(self, media_id, width, height, content_type,
+                                  method):
         top_level_type, sub_type = content_type.split("/")
         file_name = "%i-%i-%s-%s-%s" % (
             width, height, top_level_type, sub_type, method
         )
         return os.path.join(
-            self.base_path, "local_thumbnails",
+            "local_thumbnails",
             media_id[0:2], media_id[2:4], media_id[4:],
             file_name
         )
 
-    def remote_media_filepath(self, server_name, file_id):
+    local_media_thumbnail = _wrap_in_base_path(local_media_thumbnail_rel)
+
+    def remote_media_filepath_rel(self, server_name, file_id):
         return os.path.join(
-            self.base_path, "remote_content", server_name,
+            "remote_content", server_name,
             file_id[0:2], file_id[2:4], file_id[4:]
         )
 
-    def remote_media_thumbnail(self, server_name, file_id, width, height,
-                               content_type, method):
+    remote_media_filepath = _wrap_in_base_path(remote_media_filepath_rel)
+
+    def remote_media_thumbnail_rel(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)
         return os.path.join(
-            self.base_path, "remote_thumbnail", server_name,
+            "remote_thumbnail", server_name,
             file_id[0:2], file_id[2:4], file_id[4:],
             file_name
         )
 
+    remote_media_thumbnail = _wrap_in_base_path(remote_media_thumbnail_rel)
+
     def remote_media_thumbnail_dir(self, server_name, file_id):
         return os.path.join(
             self.base_path, "remote_thumbnail", server_name,
             file_id[0:2], file_id[2:4], file_id[4:],
         )
 
-    def url_cache_filepath(self, media_id):
+    def url_cache_filepath_rel(self, media_id):
         if NEW_FORMAT_ID_RE.match(media_id):
             # Media id is of the form <DATE><RANDOM_STRING>
             # E.g.: 2017-09-28-fsdRDt24DS234dsf
             return os.path.join(
-                self.base_path, "url_cache",
+                "url_cache",
                 media_id[:10], media_id[11:]
             )
         else:
             return os.path.join(
-                self.base_path, "url_cache",
+                "url_cache",
                 media_id[0:2], media_id[2:4], media_id[4:],
             )
 
+    url_cache_filepath = _wrap_in_base_path(url_cache_filepath_rel)
+
     def url_cache_filepath_dirs_to_delete(self, media_id):
         "The dirs to try and remove if we delete the media_id file"
         if NEW_FORMAT_ID_RE.match(media_id):
@@ -110,8 +141,8 @@ def url_cache_filepath_dirs_to_delete(self, media_id):
                 ),
             ]
 
-    def url_cache_thumbnail(self, media_id, width, height, content_type,
-                            method):
+    def url_cache_thumbnail_rel(self, media_id, width, height, content_type,
+                                method):
         # Media id is of the form <DATE><RANDOM_STRING>
         # E.g.: 2017-09-28-fsdRDt24DS234dsf
 
@@ -122,17 +153,19 @@ def url_cache_thumbnail(self, media_id, width, height, content_type,
 
         if NEW_FORMAT_ID_RE.match(media_id):
             return os.path.join(
-                self.base_path, "url_cache_thumbnails",
+                "url_cache_thumbnails",
                 media_id[:10], media_id[11:],
                 file_name
             )
         else:
             return os.path.join(
-                self.base_path, "url_cache_thumbnails",
+                "url_cache_thumbnails",
                 media_id[0:2], media_id[2:4], media_id[4:],
                 file_name
             )
 
+    url_cache_thumbnail = _wrap_in_base_path(url_cache_thumbnail_rel)
+
     def url_cache_thumbnail_directory(self, media_id):
         # Media id is of the form <DATE><RANDOM_STRING>
         # E.g.: 2017-09-28-fsdRDt24DS234dsf