deepset-ai · julian-risch · Mar 23, 2026 · Mar 20, 2026
@@ -88,6 +88,13 @@ select = [
   "ARG",
   "B",
   "C",
+  "D102",   # Missing docstring in public method
+  "D103",   # Missing docstring in public function
+  "D205",   # 1 blank line required between summary line and description
+  "D209",   # Closing triple quotes go to new line
+  "D213",   # summary lines must be positioned on the second physical line of the docstring
+  "D417",   # Missing argument descriptions in the docstring
+  "D419",   # Docstring is empty
   "DTZ",
   "E",
   "EM",
@@ -136,10 +143,10 @@ ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
+"tests/**/*" = ["D", "PLR2004", "S101", "TID252", "ANN"]
 # Examples can print their output
-"examples/**" = ["T201", "ANN"]
-"example/**" = ["T201", "ANN"]
+"examples/**" = ["T201", "ANN", "D"]
+"example/**" = ["T201", "ANN", "D"]
 "tests/**" = ["T201"]
 
 [tool.coverage.run]

@@ -19,6 +19,7 @@
 class LangfuseConnector:
     """
     LangfuseConnector connects Haystack LLM framework with [Langfuse](https://langfuse.com) in order to enable the
+
     tracing of operations and data flow within various components of a pipeline.
 
     To use LangfuseConnector, add it to your pipeline without connecting it to any other components.

@@ -125,6 +125,7 @@ def get_data(self) -> dict[str, Any]:
         return self._data
 
     def get_correlation_data_for_logs(self) -> dict[str, Any]:
+        """Return correlation data for log enrichment."""
         return {}
 
 
@@ -234,9 +235,11 @@ def handle(self, span: LangfuseSpan, component_type: str | None) -> None:
 
     @classmethod
     def from_dict(cls, data: dict[str, Any]) -> "SpanHandler":
+        """Deserialize a SpanHandler from a dictionary."""
         return default_from_dict(cls, data)
 
     def to_dict(self) -> dict[str, Any]:
+        """Serialize this SpanHandler to a dictionary."""
         return default_to_dict(self)
 
 
@@ -273,6 +276,7 @@ class DefaultSpanHandler(SpanHandler):
     """DefaultSpanHandler provides the default Langfuse tracing behavior for Haystack."""
 
     def create_span(self, context: SpanContext) -> LangfuseSpan:
+        """Create a Langfuse span based on the given context."""
         if self.tracer is None:
             message = (
                 "Tracer is not initialized. "
@@ -343,6 +347,7 @@ def create_span(self, context: SpanContext) -> LangfuseSpan:
             return LangfuseSpan(self.tracer.start_as_current_span(name=context.name))
 
     def handle(self, span: LangfuseSpan, component_type: str | None) -> None:
+        """Process and enrich a span after component execution."""
         # If the span is at the pipeline level, we add input and output keys to the span
         at_pipeline_level = span.get_data().get(_PIPELINE_INPUT_KEY) is not None
         if at_pipeline_level:
@@ -456,6 +461,7 @@ def __init__(
     def trace(
         self, operation_name: str, tags: dict[str, Any] | None = None, parent_span: Span | None = None
     ) -> Iterator[Span]:
+        """Create and manage a tracing span as a context manager."""
         tags = tags or {}
         span_name = tags.get(_COMPONENT_NAME_KEY, operation_name)
         component_type = tags.get(_COMPONENT_TYPE_KEY)
@@ -543,6 +549,7 @@ def trace(
                 self.flush()
 
     def flush(self) -> None:
+        """Flush all pending spans to Langfuse."""
         self._tracer.flush()
 
     def current_span(self) -> Span | None:
@@ -558,13 +565,15 @@ def current_span(self) -> Span | None:
     def get_trace_url(self) -> str:
         """
         Return the URL to the tracing data.
+
         :return: The URL to the tracing data.
         """
         return self._tracer.get_trace_url() or ""
 
     def get_trace_id(self) -> str:
         """
         Return the trace ID.
+
         :return: The trace ID.
         """
         return self._tracer.get_current_trace_id() or ""
@@ -95,6 +95,13 @@ select = [
     "ARG",
     "B",
     "C",
+  "D102",   # Missing docstring in public method
+  "D103",   # Missing docstring in public function
+  "D205",   # 1 blank line required between summary line and description
+  "D209",   # Closing triple quotes go to new line
+  "D213",   # summary lines must be positioned on the second physical line of the docstring
+  "D417",   # Missing argument descriptions in the docstring
+  "D419",   # Docstring is empty
     "DTZ",
     "E",
     "EM",
@@ -149,7 +156,7 @@ ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
+"tests/**/*" = ["D", "PLR2004", "S101", "TID252", "ANN"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]

@@ -99,6 +99,13 @@ select = [
     "ARG",
     "B",
     "C",
+  "D102",   # Missing docstring in public method
+  "D103",   # Missing docstring in public function
+  "D205",   # 1 blank line required between summary line and description
+  "D209",   # Closing triple quotes go to new line
+  "D213",   # summary lines must be positioned on the second physical line of the docstring
+  "D417",   # Missing argument descriptions in the docstring
+  "D419",   # Docstring is empty
     "DTZ",
     "E",
     "EM",
@@ -142,7 +149,7 @@ ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
+"tests/**/*" = ["D", "PLR2004", "S101", "TID252", "ANN"]
 # Examples can print their output
 "examples/**" = ["T201"]
 "tests/**" = ["T201"]

@@ -211,8 +211,11 @@ def __init__(
         model_clip_path: str | None = None,
     ) -> None:
         """
+        Initialize LlamaCppChatGenerator.
+
         :param model: The path of a quantized model for text generation, for example, "zephyr-7b-beta.Q4_0.gguf".
             If the model path is also specified in the `model_kwargs`, this parameter will be ignored.
+
         :param n_ctx: The number of tokens in the context. When set to 0, the context will be taken from the model.
         :param n_batch: Prompt processing maximum batch size.
         :param model_kwargs: Dictionary containing keyword arguments used to initialize the LLM for text generation.
@@ -274,6 +277,7 @@ def __init__(
         self._inference_lock = asyncio.Lock()
 
     def warm_up(self) -> None:
+        """Load and initialize the llama.cpp model."""
         if self._model is not None:
             return
 
@@ -462,6 +466,7 @@ def _handle_streaming_response(
     ) -> dict[str, list[ChatMessage]]:
         """
         Take streaming responses from llama.cpp, convert to Haystack StreamingChunk objects, stream them,
+
         and finally convert them to a ChatMessage.
 
         :param response_stream: The streaming responses from llama.cpp.

@@ -38,8 +38,11 @@ def __init__(
         generation_kwargs: dict[str, Any] | None = None,
     ) -> None:
         """
+        Initialize LlamaCppGenerator.
+
         :param model: The path of a quantized model for text generation, for example, "zephyr-7b-beta.Q4_0.gguf".
             If the model path is also specified in the `model_kwargs`, this parameter will be ignored.
+
         :param n_ctx: The number of tokens in the context. When set to 0, the context will be taken from the model.
         :param n_batch: Prompt processing maximum batch size.
         :param model_kwargs: Dictionary containing keyword arguments used to initialize the LLM for text generation.
@@ -69,6 +72,7 @@ def __init__(
         self.model: Llama | None = None
 
     def warm_up(self) -> None:
+        """Load and initialize the llama.cpp model."""
         if self.model is None:
             self.model = Llama(**self.model_kwargs)
 

@@ -84,6 +84,13 @@ select = [
   "ARG",
   "B",
   "C",
+  "D102",   # Missing docstring in public method
+  "D103",   # Missing docstring in public function
+  "D205",   # 1 blank line required between summary line and description
+  "D209",   # Closing triple quotes go to new line
+  "D213",   # summary lines must be positioned on the second physical line of the docstring
+  "D417",   # Missing argument descriptions in the docstring
+  "D419",   # Docstring is empty
   "DTZ",
   "E",
   "EM",
@@ -137,7 +144,7 @@ ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN", "E501", "F841"]
+"tests/**/*" = ["D", "PLR2004", "S101", "TID252", "ANN", "E501", "F841"]
 # Examples can print their output and don't need type annotations
 "examples/**/*" = ["T201", "ANN"]
 

@@ -20,6 +20,7 @@
 class LlamaStackChatGenerator(OpenAIChatGenerator):
     """
     Enables text generation using Llama Stack framework.
+
     Llama Stack Server supports multiple inference providers, including Ollama, Together,
     and vLLM and other cloud providers.
     For a complete list of inference providers, see [Llama Stack docs](https://llama-stack.readthedocs.io/en/latest/providers/inference/index.html).
@@ -70,8 +71,10 @@ def __init__(
         http_client_kwargs: dict[str, Any] | None = None,
     ) -> None:
         """
-        Creates an instance of LlamaStackChatGenerator. To use this chat generator,
-        you need to setup Llama Stack Server with an inference provider and have a model available.
+        Creates an instance of LlamaStackChatGenerator.
+
+        To use this chat generator, you need to setup Llama Stack Server with an inference provider and have a model
+        available.
 
         :param model:
             The name of the model to use for chat completion.

@@ -99,6 +99,13 @@ select = [
     "ARG",
     "B",
     "C",
+  "D102",   # Missing docstring in public method
+  "D103",   # Missing docstring in public function
+  "D205",   # 1 blank line required between summary line and description
+  "D209",   # Closing triple quotes go to new line
+  "D213",   # summary lines must be positioned on the second physical line of the docstring
+  "D417",   # Missing argument descriptions in the docstring
+  "D419",   # Docstring is empty
     "DTZ",
     "E",
     "EM",
@@ -153,8 +160,8 @@ ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
-"examples/**/*" = ["T201", "E501", "ANN"]
+"tests/**/*" = ["D", "PLR2004", "S101", "TID252", "ANN"]
+"examples/**/*" = ["T201", "E501", "ANN", "D"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]

@@ -119,8 +119,7 @@ def run_background(
         self, coro_factory: Callable[[asyncio.Event], Coroutine[Any, Any, Any]], timeout: float | None = None
     ) -> tuple[concurrent.futures.Future[Any], asyncio.Event]:
         """
-        Schedule `coro_factory` to run in the executor's event loop **without** blocking the
-        caller thread.
+        Schedule `coro_factory` to run in the executor's event loop **without** blocking the caller thread.
 
         The factory receives an :class:`asyncio.Event` that can be used to cooperatively shut
         the coroutine down. The method returns **both** the concurrent future (to observe
@@ -1292,7 +1291,8 @@ def __del__(self) -> None:
 
 
 class _MCPClientSessionManager:
-    """Runs an MCPClient connect/close inside the AsyncExecutor's event loop.
+    """
+    Runs an MCPClient connect/close inside the AsyncExecutor's event loop.
 
     Life-cycle:
       1.  Create the worker to schedule a long-running coroutine in the

@@ -113,8 +113,7 @@ def _deserialize_state_config(config: dict[str, dict[str, Any]] | None) -> dict[
 
 class MCPToolset(Toolset):
     """
-    A Toolset that connects to an MCP (Model Context Protocol) server and provides
-    access to its tools.
+    A Toolset that connects to an MCP (Model Context Protocol) server and provides access to its tools.
 
     MCPToolset dynamically discovers and loads all tools from any MCP-compliant server,
     supporting both network-based streaming connections (Streamable HTTP, SSE) and local
@@ -289,7 +288,8 @@ def __init__(
             self._warmup_called = True
 
     def warm_up(self) -> None:
-        """Connect and load tools when eager_connect is turned off.
+        """
+        Connect and load tools when eager_connect is turned off.
 
         This method is automatically called by ``ToolInvoker.warm_up()`` and ``Pipeline.warm_up()``.
         You can also call it directly before using the toolset to ensure all tool schemas

@@ -84,6 +84,13 @@ select = [
   "ARG",
   "B",
   "C",
+  "D102",   # Missing docstring in public method
+  "D103",   # Missing docstring in public function
+  "D205",   # 1 blank line required between summary line and description
+  "D209",   # Closing triple quotes go to new line
+  "D213",   # summary lines must be positioned on the second physical line of the docstring
+  "D417",   # Missing argument descriptions in the docstring
+  "D419",   # Docstring is empty
   "DTZ",
   "E",
   "EM",
@@ -133,7 +140,7 @@ ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
+"tests/**/*" = ["D", "PLR2004", "S101", "TID252", "ANN"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]

@@ -21,6 +21,7 @@
 class MetaLlamaChatGenerator(OpenAIChatGenerator):
     """
     Enables text generation using Llama generative models.
+
     For supported models, see [Llama API Docs](https://llama.developer.meta.com/docs/).
 
     Users can pass any text generation parameters valid for the Llama Chat Completion API
@@ -76,8 +77,9 @@ def __init__(
         tools: ToolsType | None = None,
     ) -> None:
         """
-        Creates an instance of LlamaChatGenerator. Unless specified otherwise in the `model`, this is for Llama's
-        `Llama-4-Scout-17B-16E-Instruct-FP8` model.
+        Creates an instance of LlamaChatGenerator.
+
+        Unless specified otherwise in the `model`, this is for Llama's `Llama-4-Scout-17B-16E-Instruct-FP8` model.
 
         :param api_key:
             The Llama API key.

@@ -87,6 +87,13 @@ select = [
   "ARG",
   "B",
   "C",
+  "D102",   # Missing docstring in public method
+  "D103",   # Missing docstring in public function
+  "D205",   # 1 blank line required between summary line and description
+  "D209",   # Closing triple quotes go to new line
+  "D213",   # summary lines must be positioned on the second physical line of the docstring
+  "D417",   # Missing argument descriptions in the docstring
+  "D419",   # Docstring is empty
   "DTZ",
   "E",
   "EM",
@@ -136,7 +143,7 @@ ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
+"tests/**/*" = ["D", "PLR2004", "S101", "TID252", "ANN"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]

@@ -26,8 +26,9 @@
 @component
 class MistralOCRDocumentConverter:
     """
-    This component extracts text from documents using Mistral's OCR API, with optional structured
-    annotations for both individual image regions (bounding boxes) and full documents.
+    Extract text from documents using Mistral's OCR API with optional structured annotations.
+
+    Supports optional structured annotations for individual image regions (bounding boxes) and full documents.
 
     Accepts document sources in various formats (str/Path for local files, ByteStream for in-memory data,
     DocumentURLChunk for document URLs, ImageURLChunk for image URLs, or FileChunk for Mistral file IDs)

@@ -12,6 +12,7 @@
 class MistralDocumentEmbedder(OpenAIDocumentEmbedder):
     """
     A component for computing Document embeddings using Mistral models.
+
     The embedding of each Document is stored in the `embedding` field of the Document.
 
     Usage example: