langchain-qdrant[patch]: Add ruff bandit rules to linter (#31815)

- Add ruff bandit rules
- Address a few s101s
- Some formatting

libs/partners/qdrant/langchain_qdrant/qdrant.py

@@ -230,7 +230,7 @@ class QdrantVectorStore(VectorStore):
         """Get the Qdrant client instance that is being used.
 
         Returns:
-            QdrantClient: An instance of `QdrantClient`.
+            QdrantClient: An instance of ``QdrantClient``.
         """
         return self._client
 
@@ -239,10 +239,10 @@ class QdrantVectorStore(VectorStore):
         """Get the dense embeddings instance that is being used.
 
         Raises:
-            ValueError: If embeddings are `None`.
+            ValueError: If embeddings are ``None``.
 
         Returns:
-            Embeddings: An instance of `Embeddings`.
+            Embeddings: An instance of ``Embeddings``.
         """
         if self._embeddings is None:
             raise ValueError(
@@ -255,10 +255,10 @@ class QdrantVectorStore(VectorStore):
         """Get the sparse embeddings instance that is being used.
 
         Raises:
-            ValueError: If sparse embeddings are `None`.
+            ValueError: If sparse embeddings are ``None``.
 
         Returns:
-            SparseEmbeddings: An instance of `SparseEmbeddings`.
+            SparseEmbeddings: An instance of ``SparseEmbeddings``.
         """
         if self._sparse_embeddings is None:
             raise ValueError(
@@ -302,7 +302,7 @@ class QdrantVectorStore(VectorStore):
         validate_collection_config: bool = True,
         **kwargs: Any,
     ) -> QdrantVectorStore:
-        """Construct an instance of `QdrantVectorStore` from a list of texts.
+        """Construct an instance of ``QdrantVectorStore`` from a list of texts.
 
         This is a user-friendly interface that:
         1. Creates embeddings, one for each text
@@ -382,11 +382,11 @@ class QdrantVectorStore(VectorStore):
         validate_collection_config: bool = True,
         **kwargs: Any,
     ) -> QdrantVectorStore:
-        """Construct an instance of `QdrantVectorStore` from an existing collection
+        """Construct an instance of ``QdrantVectorStore`` from an existing collection
         without adding any data.
 
         Returns:
-            QdrantVectorStore: A new instance of `QdrantVectorStore`.
+            QdrantVectorStore: A new instance of ``QdrantVectorStore``.
         """
         client = QdrantClient(
             location=location,
@@ -660,7 +660,6 @@ class QdrantVectorStore(VectorStore):
         Maximal marginal relevance optimizes for similarity to query AND diversity
         among selected documents.
 
-
         Returns:
             List of Documents selected by maximal marginal relevance.
         """
@@ -911,16 +910,16 @@ class QdrantVectorStore(VectorStore):
 
     @staticmethod
     def _cosine_relevance_score_fn(distance: float) -> float:
-        """Normalize the distance to a score on a scale [0, 1]."""
+        """Normalize the distance to a score on a scale ``[0, 1]``."""
         return (distance + 1.0) / 2.0
 
     def _select_relevance_score_fn(self) -> Callable[[float], float]:
         """
-        The 'correct' relevance function
+        The "correct" relevance function may differ depending on a few things,
-        may differ depending on a few things, including:
+        including:
-        - the distance / similarity metric used by the VectorStore
+        - The distance / similarity metric used by the VectorStore
-        - the scale of your embeddings (OpenAI's are unit normed. Many others are not!)
+        - The scale of your embeddings (OpenAI's are unit normed. Many others are not!)
-        - embedding dimensionality
+        - Embedding dimensionality
         - etc.
         """
 
@@ -1039,9 +1038,10 @@ class QdrantVectorStore(VectorStore):
             dense_embeddings = self.embeddings.embed_documents(list(texts))
             sparse_embeddings = self.sparse_embeddings.embed_documents(list(texts))
 
-            assert len(dense_embeddings) == len(
+            if len(dense_embeddings) != len(sparse_embeddings):
-                sparse_embeddings
+                raise ValueError(
-            ), "Mismatched length between dense and sparse embeddings."
+                    "Mismatched length between dense and sparse embeddings."
+                )
 
             return [
                 {
@@ -1128,7 +1128,8 @@ class QdrantVectorStore(VectorStore):
                     "set `force_recreate` to `True`."
                 )
 
-        assert vector_config is not None, "VectorParams is None"
+        if vector_config is None:
+            raise ValueError("VectorParams is None")
 
         if isinstance(dense_embeddings, Embeddings):
             vector_size = len(dense_embeddings.embed_documents(["dummy_text"])[0])

libs/partners/qdrant/langchain_qdrant/vectorstores.py

@@ -105,7 +105,7 @@ class Qdrant(VectorStore):
 
         if embeddings is None and embedding_function is None:
             raise ValueError(
-                "`embeddings` value can't be None. Pass `Embeddings` instance."
+                "`embeddings` value can't be None. Pass `embeddings` instance."
             )
 
         if embeddings is not None and embedding_function is not None:
@@ -161,7 +161,7 @@ class Qdrant(VectorStore):
                 uuid-like strings.
             batch_size:
                 How many vectors upload per-request.
-                Default: 64
+                Default: ``64``
 
         Returns:
             List of ids from adding the texts into the vectorstore.
@@ -196,7 +196,7 @@ class Qdrant(VectorStore):
                 uuid-like strings.
             batch_size:
                 How many vectors upload per-request.
-                Default: 64
+                Default: ``64``
 
         Returns:
             List of ids from adding the texts into the vectorstore.
@@ -1669,7 +1669,8 @@ class Qdrant(VectorStore):
                     f"`None`. If you want to recreate the collection, set "
                     f"`force_recreate` parameter to `True`."
                 )
-            assert isinstance(current_vector_config, models.VectorParams), (
+            if not isinstance(current_vector_config, models.VectorParams):
+                raise ValueError(
                     "Expected current_vector_config to be an instance of "
                     f"models.VectorParams, but got {type(current_vector_config)}"
                 )
@@ -1829,8 +1830,8 @@ class Qdrant(VectorStore):
                     f"`None`. If you want to recreate the collection, set "
                     f"`force_recreate` parameter to `True`."
                 )
-
+            if not isinstance(current_vector_config, models.VectorParams):
-            assert isinstance(current_vector_config, models.VectorParams), (
+                raise ValueError(
                     "Expected current_vector_config to be an instance of "
                     f"models.VectorParams, but got {type(current_vector_config)}"
                 )

libs/partners/qdrant/pyproject.toml

@@ -51,7 +51,7 @@ langchain-core = { path = "../../core", editable = true }
 target-version = "py39"
 
 [tool.ruff.lint]
-select = ["E", "F", "I", "T201", "UP"]
+select = ["E", "F", "I", "T201", "UP", "S"]
 ignore = [ "UP007", ]
 
 [tool.mypy]
@@ -67,3 +67,9 @@ markers = [
     "compile: mark placeholder test used to compile integration tests without running them",
 ]
 asyncio_mode = "auto"
+
+[tool.ruff.lint.extend-per-file-ignores]
+"tests/**/*.py" = [
+    "S101", # Tests need assertions
+    "S311", # Standard pseudo-random generators are not suitable for cryptographic purposes
+]