let posts be found by primary key id

build_database.py

@@ -98,8 +98,8 @@ async def main_with_ctx(ctx, wanted_date):
 
                 # write to output
                 log.info("copying temp to output")
-                with output_path.open(mode="wb") as output_fd:
+                with output_path.open(mode="wb") as output_fdout:
-                    shutil.copyfileobj(temp_fd, output_fd)
+                    shutil.copyfileobj(temp_fd, output_fdout)
 
     # decompress
     for url_type, _url in urls.items():

e621_api_cloner.py

@@ -150,16 +150,33 @@ async def posts_json():
     if len(tags) != 1:
         return "unsupported query (only 1 tag)", 400
 
-    md5_tag = tags[0]
+    first_tag = tags[0]
-    if not md5_tag.startswith("md5:"):
+    tag_parts = first_tag.split(":")
-        return "unsupported query (only md5 for now)", 400
+    if len(tag_parts) != 2:
+        return "unsupported query (tag must be a:b format)", 400
 
-    md5_prefix, md5_hexencoded_hash = md5_tag.split(":")
+    type, value = tag_parts
-    assert md5_prefix == "md5"
+    if type not in ("md5", "id"):
+        return (
+            "unsupported query (tag must be a:b format, and a must be id or md5)",
+            400,
+        )
 
+    if type == "md5":
+        md5_hexencoded_hash = value
         rows = await app.db.execute_fetchall(
             "select * from posts where md5 = ?", (md5_hexencoded_hash,)
         )
+    elif type == "id":
+        try:
+            value = int(value)
+        except ValueError:
+            return "id:x x must be int", 400
+        rows = await app.db.execute_fetchall(
+            "select * from posts where id = ?", (value,)
+        )
+    else:
+        raise AssertionError("must be id or md5")
     if not rows:
         return {"posts": []}