From ace807908602cb955fc7a2cae63dc6e64bf90cc5 Mon Sep 17 00:00:00 2001
From: Erik Johnston <erik@matrix.org>
Date: Thu, 28 Sep 2017 12:52:51 +0100
Subject: [PATCH] Support new and old style media id formats

---
 synapse/rest/media/v1/filepath.py | 112 +++++++++++++++++++++---------
 1 file changed, 81 insertions(+), 31 deletions(-)

diff --git a/synapse/rest/media/v1/filepath.py b/synapse/rest/media/v1/filepath.py
index c5d43209f..d5cec1012 100644
--- a/synapse/rest/media/v1/filepath.py
+++ b/synapse/rest/media/v1/filepath.py
@@ -14,6 +14,9 @@
 # limitations under the License.
 
 import os
+import re
+
+NEW_FORMAT_ID_RE = re.compile(r"^\d\d\d\d-\d\d-\d\d")
 
 
 class MediaFilePaths(object):
@@ -73,21 +76,39 @@ class MediaFilePaths(object):
         )
 
     def url_cache_filepath(self, 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",
-            media_id[:10], media_id[11:]
-        )
+        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",
+                media_id[:10], media_id[11:]
+            )
+        else:
+            return os.path.join(
+                self.base_path, "url_cache",
+                media_id[0:2], media_id[2:4], media_id[4:],
+            )
 
     def url_cache_filepath_dirs_to_delete(self, media_id):
         "The dirs to try and remove if we delete the media_id file"
-        return [
-            os.path.join(
-                self.base_path, "url_cache",
-                media_id[:10],
-            ),
-        ]
+        if NEW_FORMAT_ID_RE.match(media_id):
+            return [
+                os.path.join(
+                    self.base_path, "url_cache",
+                    media_id[:10],
+                ),
+            ]
+        else:
+            return [
+                os.path.join(
+                    self.base_path, "url_cache",
+                    media_id[0:2], media_id[2:4],
+                ),
+                os.path.join(
+                    self.base_path, "url_cache",
+                    media_id[0:2],
+                ),
+            ]
 
     def url_cache_thumbnail(self, media_id, width, height, content_type,
                             method):
@@ -99,32 +120,61 @@ class MediaFilePaths(object):
             width, height, top_level_type, sub_type, method
         )
 
-        return os.path.join(
-            self.base_path, "url_cache_thumbnails",
-            media_id[:10], media_id[11:],
-            file_name
-        )
+        if NEW_FORMAT_ID_RE.match(media_id):
+            return os.path.join(
+                self.base_path, "url_cache_thumbnails",
+                media_id[:10], media_id[11:],
+                file_name
+            )
+        else:
+            return os.path.join(
+                self.base_path, "url_cache_thumbnails",
+                media_id[0:2], media_id[2:4], media_id[4:],
+                file_name
+            )
 
     def url_cache_thumbnail_directory(self, 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_thumbnails",
-            media_id[:10], media_id[11:],
-        )
+        if NEW_FORMAT_ID_RE.match(media_id):
+            return os.path.join(
+                self.base_path, "url_cache_thumbnails",
+                media_id[:10], media_id[11:],
+            )
+        else:
+            return os.path.join(
+                self.base_path, "url_cache_thumbnails",
+                media_id[0:2], media_id[2:4], media_id[4:],
+            )
 
     def url_cache_thumbnail_dirs_to_delete(self, media_id):
         "The dirs to try and remove if we delete the media_id thumbnails"
         # Media id is of the form <DATE><RANDOM_STRING>
         # E.g.: 2017-09-28-fsdRDt24DS234dsf
-        return [
-            os.path.join(
-                self.base_path, "url_cache_thumbnails",
-                media_id[:10], media_id[11:],
-            ),
-            os.path.join(
-                self.base_path, "url_cache_thumbnails",
-                media_id[:10],
-            ),
-        ]
+        if NEW_FORMAT_ID_RE.match(media_id):
+            return [
+                os.path.join(
+                    self.base_path, "url_cache_thumbnails",
+                    media_id[:10], media_id[11:],
+                ),
+                os.path.join(
+                    self.base_path, "url_cache_thumbnails",
+                    media_id[:10],
+                ),
+            ]
+        else:
+            return [
+                os.path.join(
+                    self.base_path, "url_cache_thumbnails",
+                    media_id[0:2], media_id[2:4], media_id[4:],
+                ),
+                os.path.join(
+                    self.base_path, "url_cache_thumbnails",
+                    media_id[0:2], media_id[2:4],
+                ),
+                os.path.join(
+                    self.base_path, "url_cache_thumbnails",
+                    media_id[0:2],
+                ),
+            ]