refactor: Remove uuid_from_buffer, use hashlib directly for query cache

- Remove uuid_from_buffer from hash.py (dead code)
- connection.py now uses hashlib.md5().hexdigest() directly
- Update test_hash.py to test key_hash instead

Co-Authored-By: Claude Opus 4.5 <[email protected]>

Author: dimitri-yatsenko

src/datajoint/connection.py

@@ -5,6 +5,7 @@
 
 from __future__ import annotations
 
+import hashlib
 import logging
 import pathlib
 import re
@@ -18,7 +19,6 @@
 from . import errors
 from .blob import pack, unpack
 from .dependencies import Dependencies
-from .hash import uuid_from_buffer
 from .settings import config
 from .version import __version__
 
@@ -418,7 +418,7 @@ def query(
         if use_query_cache:
             if not config[cache_key]:
                 raise errors.DataJointError(f"Provide filepath dj.config['{cache_key}'] when using query caching.")
-            hash_ = uuid_from_buffer((str(self._query_cache) + re.sub(r"`\$\w+`", "", query)).encode() + pack(args))
+            hash_ = hashlib.md5((str(self._query_cache) + re.sub(r"`\$\w+`", "", query)).encode() + pack(args)).hexdigest()
             cache_path = pathlib.Path(config[cache_key]) / str(hash_)
             try:
                 buffer = cache_path.read_bytes()

src/datajoint/hash.py

@@ -1,7 +1,6 @@
 from __future__ import annotations
 
 import hashlib
-import uuid
 from typing import Any
 
 
@@ -14,16 +13,3 @@ def key_hash(mapping: dict[str, Any]) -> str:
     for k, v in sorted(mapping.items()):
         hashed.update(str(v).encode())
     return hashed.hexdigest()
-
-
-def uuid_from_buffer(buffer: bytes = b"", *, init_string: str = "") -> uuid.UUID:
-    """
-    Compute MD5 hash of buffer data, returned as UUID.
-
-    :param buffer: bytes to hash
-    :param init_string: string to initialize the checksum (for namespacing)
-    :return: UUID based on MD5 digest
-    """
-    hashed = hashlib.md5(init_string.encode())
-    hashed.update(buffer)
-    return uuid.UUID(bytes=hashed.digest())

tests/unit/test_hash.py

@@ -1,6 +1,8 @@
 from datajoint import hash
 
 
-def test_hash():
-    assert hash.uuid_from_buffer(b"abc").hex == "900150983cd24fb0d6963f7d28e17f72"
-    assert hash.uuid_from_buffer(b"").hex == "d41d8cd98f00b204e9800998ecf8427e"
+def test_key_hash():
+    """Test that key_hash produces consistent MD5 hex digests."""
+    assert hash.key_hash({"a": 1, "b": 2}) == hash.key_hash({"b": 2, "a": 1})
+    assert hash.key_hash({"x": "hello"}) == "5d41402abc4b2a76b9719d911017c592"
+    assert hash.key_hash({}) == "d41d8cd98f00b204e9800998ecf8427e"