Feature/sha256

src/matchbox/common/hash.py

@@ -14,7 +14,7 @@
 T = TypeVar("T")
 HashableItem = TypeVar("HashableItem", bytes, bool, str, int, float, bytearray)
 
-HASH_FUNC = hashlib.sha1
+HASH_FUNC = hashlib.sha256
 
 
 def dataset_to_hashlist(dataset: Source, model_hash: bytes) -> list[dict[str, Any]]:

src/matchbox/server/postgresql/utils/insert.py

@@ -113,7 +113,7 @@ def insert_model(
     Args:
         model: Name of the new model
         left: Name of the left parent model
-        right: Name of the left parent model. Same as left in a link job
+        right: Name of the right parent model. Same as left in a link job
         description: Model description
         engine: SQLAlchemy engine instance
 
@@ -139,7 +139,8 @@ def insert_model(
         session.flush()
 
         def _create_closure_entries(parent_model: Models) -> None:
-            """Create closure entries for the new model."""
+            """Create closure entries for the new model, i.e. mappings between
+            nodes and any of their direct or indirect parents"""
             session.add(
                 ModelsFrom(
                     parent=parent_model.hash,

test/client/test_linkers.py

@@ -37,7 +37,6 @@ def test_linkers(
         1. That the input data is as expected
         2. That the data is linked correctly
         3. That the linked probabilities are inserted correctly
-        4. That the correct number of clusters are resolved
         4. That the correct number of clusters are resolved and inserted correctly
     """
     # i. Ensure database is ready, collect fixtures, perform any special linker cleaning