Add filesystem hashtable for saved thumbnails

Signed-off-by: Alek Ratzloff <alekratz@gmail.com>

chanbans/pull.py

@@ -44,7 +44,7 @@ async def get_thumb(thumb_path: Union[str, Path], post: dict) -> Optional[bytes]
         log.info("Downloading %s", url)
         async with aiohttp.ClientSession() as session:
             async with session.get(url) as resp:
-                THUMBS_DIR.mkdir(parents=True, exist_ok=True)
+                thumb_path.parent.mkdir(parents=True, exist_ok=True)
                 content = await resp.read()
                 thumb_path.write_bytes(content)
                 return content
@@ -78,7 +78,9 @@ async def pull():
         post = defaultdict(lambda: None, posts[cols["post"]["data-pid"]])
 
         if "thumb" in post:
-            thumb_path = Path(THUMBS_DIR, f"{post['thumb']}s.jpg")
+            thumb = post['thumb']
+            thumb_name = Path(thumb[:2], f"{post['thumb']}s.jpg")
+            thumb_path = Path(THUMBS_DIR, thumb_name)
             download_jobs += [get_thumb(thumb_path, post)]
         else:
             thumb_path = ""
@@ -86,6 +88,9 @@ async def pull():
         # Try to create post in database
         try:
             with db:
+                # TODO - this would probably be faster if we select the top 25 items and did the
+                # comparison. Especially when the database gets huge.
+
                 # Check the last N bans for the given board. If the following columns are equal:
                 # * now
                 # * time