cbrkit 0.28.0__tar.gz → 0.28.2__tar.gz

{cbrkit-0.28.0 → cbrkit-0.28.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: cbrkit
-Version: 0.28.0
+Version: 0.28.2
 Summary: Customizable Case-Based Reasoning (CBR) toolkit for Python with a built-in API and CLI
 Keywords: cbr,case-based reasoning,api,similarity,nlp,retrieval,cli,tool,library
 Author: Mirko Lenz

{cbrkit-0.28.0 → cbrkit-0.28.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "cbrkit"
-version = "0.28.0"
+version = "0.28.2"
 description = "Customizable Case-Based Reasoning (CBR) toolkit for Python with a built-in API and CLI"
 authors = [{ name = "Mirko Lenz", email = "mirko@mirkolenz.com" }]
 readme = "README.md"

{cbrkit-0.28.0 → cbrkit-0.28.2}/src/cbrkit/sim/embed.py

@@ -191,18 +191,19 @@ class cache(BatchConversionFunc[str, NumpyArray]):
 
             self.path.parent.mkdir(parents=True, exist_ok=True)
 
-            with self.connect() as connection:
-                connection.execute(f"""
+            with self.connect() as con:
+                con.execute(f"""
                     CREATE TABLE IF NOT EXISTS "{self.table}" (
                         id INTEGER PRIMARY KEY AUTOINCREMENT,
                         text TEXT NOT NULL UNIQUE,
                         vector BLOB NOT NULL
                     )
                 """)
+                con.commit()
 
-                cursor = connection.execute(f'SELECT text, vector FROM "{self.table}"')
+                cur = con.execute(f'SELECT text, vector FROM "{self.table}"')
 
-                for text, vector_blob in cursor:
+                for text, vector_blob in cur:
                     self.store[text] = np.frombuffer(vector_blob, dtype=np.float64)
 
     @contextmanager
@@ -210,12 +211,12 @@ class cache(BatchConversionFunc[str, NumpyArray]):
         if self.path is None:
             raise ValueError("Path must be set to use the cache")
 
-        connection = sqlite3.connect(self.path)
+        con = sqlite3.connect(self.path, autocommit=False)
 
         try:
-            yield connection
+            yield con
         finally:
-            connection.close()
+            con.close()
 
     @override
     def __call__(self, texts: Sequence[str]) -> Sequence[NumpyArray]:
@@ -230,13 +231,17 @@ class cache(BatchConversionFunc[str, NumpyArray]):
                     self.store[text] = vector
 
                 if self.path is not None:
-                    with self.connect() as connection:
-                        for text, vector in zip(new_texts, new_vectors, strict=True):
-                            vector_blob = vector.astype(np.float64).tobytes()
-                            connection.execute(
-                                f'INSERT OR IGNORE INTO "{self.table}" (text, vector) VALUES (?, ?)',
-                                (text, vector_blob),
-                            )
+                    with self.connect() as con:
+                        con.executemany(
+                            f'INSERT OR IGNORE INTO "{self.table}" (text, vector) VALUES(?, ?)',
+                            [
+                                (text, vector.astype(np.float64).tobytes())
+                                for text, vector in zip(
+                                    new_texts, new_vectors, strict=True
+                                )
+                            ],
+                        )
+                        con.commit()
 
         return [self.store[text] for text in texts]
 

{cbrkit-0.28.0 → cbrkit-0.28.2}/src/cbrkit/system.py

@@ -17,18 +17,17 @@ __all__ = [
 
 
 @dataclass(slots=True, frozen=True)
-class System[K: str | int, V: BaseModel, S: Float]:
+class System[K: str | int, V: BaseModel, S: Float, P: str]:
     casebase: cbrkit.typing.Casebase[K, V]
-    model: type[V]
     retriever_pipelines: Mapping[
-        str, MaybeSequence[cbrkit.typing.RetrieverFunc[K, V, S]]
+        P, MaybeSequence[cbrkit.typing.RetrieverFunc[K, V, S]]
     ] = field(default_factory=dict)
-    reuser_pipelines: Mapping[str, MaybeSequence[cbrkit.typing.ReuserFunc[K, V, S]]] = (
+    reuser_pipelines: Mapping[P, MaybeSequence[cbrkit.typing.ReuserFunc[K, V, S]]] = (
         field(default_factory=dict)
     )
 
     def get_retriever_pipeline(
-        self, name: str, limit: int | None
+        self, name: P, limit: int | None
     ) -> Sequence[cbrkit.typing.RetrieverFunc[K, V, S]]:
         retrievers = produce_sequence(self.retriever_pipelines[name])
 
@@ -43,7 +42,7 @@ class System[K: str | int, V: BaseModel, S: Float]:
     def retrieve(
         self,
         query: V,
-        retriever_pipeline: str,
+        retriever_pipeline: P,
         limit: int | None = None,
     ) -> cbrkit.retrieval.QueryResultStep[K, V, S]:
         return cbrkit.retrieval.apply_query(
@@ -55,7 +54,7 @@ class System[K: str | int, V: BaseModel, S: Float]:
     def reuse(
         self,
         query: V,
-        reuser_pipeline: str,
+        reuser_pipeline: P,
     ) -> cbrkit.retrieval.QueryResultStep[K, V, S]:
         return cbrkit.reuse.apply_query(
             self.casebase,
@@ -66,8 +65,8 @@ class System[K: str | int, V: BaseModel, S: Float]:
     def cycle(
         self,
         query: V,
-        retriever_pipeline: str,
-        reuser_pipeline: str,
+        retriever_pipeline: P,
+        reuser_pipeline: P,
         limit: int | None = None,
     ) -> cbrkit.retrieval.QueryResultStep[K, V, S]:
         return cbrkit.cycle.apply_query(
@@ -117,9 +116,7 @@ class System[K: str | int, V: BaseModel, S: Float]:
 with cbrkit.helpers.optional_dependencies():
     from fastapi import FastAPI
 
-    def to_fastapi(system: System) -> FastAPI:
-        app = FastAPI()
-
+    def to_fastapi(system: System, app: FastAPI) -> FastAPI:
         for value in system.tools:
             app.post(f"/tool/{value.__name__}")(value)
 
@@ -135,9 +132,7 @@ with cbrkit.helpers.optional_dependencies():
 with cbrkit.helpers.optional_dependencies():
     from fastmcp import FastMCP
 
-    def to_fastmcp(system: System) -> FastMCP[Any]:
-        app = FastMCP()
-
+    def to_fastmcp[T](system: System, app: FastMCP[T]) -> FastMCP[T]:
         for value in system.tools:
             app.tool(value)
 
@@ -153,5 +148,5 @@ with cbrkit.helpers.optional_dependencies():
 with cbrkit.helpers.optional_dependencies():
     from pydantic_ai.toolsets import FunctionToolset
 
-    def to_pydantic_ai(system: System) -> FunctionToolset[Any]:
+    def to_pydantic_ai(system: System) -> FunctionToolset[None]:
         return FunctionToolset(system.tools)