Merge branch 'main' into feat/add-cognee

davidsbatista · web-flow · commit 3a4a82931cbe · 2026-03-18T16:58:52.000+01:00
diff --git a/.github/workflows/CI_docusaurus_sync.yml b/.github/workflows/CI_docusaurus_sync.yml
@@ -117,5 +117,4 @@ jobs:
           add-paths: |
             docs-website
           body: |
-            This PR syncs the Core Integrations API reference (${{ env.INTEGRATION_NAME }}) on Docusaurus. Just approve and merge it.
-          reviewers: "${{ github.actor }}"
+            This PR syncs the Core Integrations API reference (${{ env.INTEGRATION_NAME }}) on Docusaurus. Just approve and merge it.
diff --git a/integrations/azure_doc_intelligence/pyproject.toml b/integrations/azure_doc_intelligence/pyproject.toml
@@ -90,6 +90,7 @@ line-length = 120
 [tool.ruff.lint]
 select = [
     "A",
+    "ANN",
     "ARG",
     "B",
     "C",
@@ -115,6 +116,7 @@ select = [
     "YTT",
 ]
 ignore = [
+    "ANN401",  # Allow Any - used legitimately for dynamic types and SDK boundaries
     # Allow non-abstract empty methods in abstract base classes
     "B027",
     # Allow function calls in argument defaults (common Haystack pattern for Secret.from_env_var)
@@ -145,7 +147,7 @@ ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252"]
+"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]
diff --git a/integrations/azure_doc_intelligence/src/haystack_integrations/components/converters/azure_doc_intelligence/converter.py b/integrations/azure_doc_intelligence/src/haystack_integrations/components/converters/azure_doc_intelligence/converter.py
@@ -67,7 +67,7 @@ def __init__(
         api_key: Secret = Secret.from_env_var("AZURE_DI_API_KEY"),
         model_id: str = "prebuilt-document",
         store_full_path: bool = False,
-    ):
+    ) -> None:
         """
         Creates an AzureDocumentIntelligenceConverter component.
 
@@ -93,7 +93,7 @@ def __init__(
         self.store_full_path = store_full_path
         self.client: DocumentIntelligenceClient | None = None
 
-    def warm_up(self):
+    def warm_up(self) -> None:
         """
         Initializes the Azure Document Intelligence client.
         """
diff --git a/integrations/cohere/pyproject.toml b/integrations/cohere/pyproject.toml
@@ -84,6 +84,7 @@ line-length = 120
 [tool.ruff.lint]
 select = [
   "A",
+  "ANN",
   "ARG",
   "B",
   "C",
@@ -124,6 +125,8 @@ ignore = [
   # Misc
   "B008",
   "S101",
+  # Allow `Any` - used legitimately for dynamic types and SDK boundaries
+  "ANN401",
 ]
 
 [tool.ruff.lint.isort]
@@ -133,8 +136,8 @@ known-first-party = ["haystack_integrations"]
 ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
-# Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252"]
+# Tests can use magic values, assertions, relative imports, and don't need type annotations
+"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]
diff --git a/integrations/cohere/src/haystack_integrations/components/embedders/cohere/document_embedder.py b/integrations/cohere/src/haystack_integrations/components/embedders/cohere/document_embedder.py
@@ -49,7 +49,7 @@ def __init__(
         meta_fields_to_embed: list[str] | None = None,
         embedding_separator: str = "\n",
         embedding_type: EmbeddingTypes | None = None,
-    ):
+    ) -> None:
         """
         :param api_key: the Cohere API key.
         :param model: the name of the model to use. Supported Models are:
diff --git a/integrations/cohere/src/haystack_integrations/components/embedders/cohere/embedding_types.py b/integrations/cohere/src/haystack_integrations/components/embedders/cohere/embedding_types.py
@@ -21,7 +21,7 @@ class EmbeddingTypes(Enum):
     BINARY = "binary"
     UBINARY = "ubinary"
 
-    def __str__(self):
+    def __str__(self) -> str:
         return self.value
 
     @staticmethod
diff --git a/integrations/cohere/src/haystack_integrations/components/embedders/cohere/text_embedder.py b/integrations/cohere/src/haystack_integrations/components/embedders/cohere/text_embedder.py
@@ -41,7 +41,7 @@ def __init__(
         truncate: str = "END",
         timeout: float = 120.0,
         embedding_type: EmbeddingTypes | None = None,
-    ):
+    ) -> None:
         """
         :param api_key: the Cohere API key.
         :param model: the name of the model to use. Supported Models are:
diff --git a/integrations/cohere/src/haystack_integrations/components/generators/cohere/chat/chat_generator.py b/integrations/cohere/src/haystack_integrations/components/generators/cohere/chat/chat_generator.py
@@ -491,7 +491,7 @@ def __init__(
         *,
         timeout: float | None = None,
         max_retries: int | None = None,
-    ):
+    ) -> None:
         """
         Initialize the CohereChatGenerator instance.
 
diff --git a/integrations/cohere/src/haystack_integrations/components/generators/cohere/generator.py b/integrations/cohere/src/haystack_integrations/components/generators/cohere/generator.py
@@ -37,7 +37,7 @@ def __init__(
         streaming_callback: Callable | None = None,
         api_base_url: str | None = None,
         **kwargs: Any,
-    ):
+    ) -> None:
         """
         Instantiates a `CohereGenerator` component.
 
diff --git a/integrations/cohere/src/haystack_integrations/components/rankers/cohere/ranker.py b/integrations/cohere/src/haystack_integrations/components/rankers/cohere/ranker.py
@@ -41,7 +41,7 @@ def __init__(
         meta_fields_to_embed: list[str] | None = None,
         meta_data_separator: str = "\n",
         max_tokens_per_doc: int = 4096,
-    ):
+    ) -> None:
         """
         Creates an instance of the 'CohereRanker'.
 
diff --git a/integrations/deepeval/pyproject.toml b/integrations/deepeval/pyproject.toml
@@ -82,6 +82,7 @@ line-length = 120
 [tool.ruff.lint]
 select = [
   "A",
+  "ANN",
   "ARG",
   "B",
   "C",
@@ -110,6 +111,7 @@ select = [
 ignore = [
   # Allow non-abstract empty methods in abstract base classes
   "B027",
+  "ANN401",  # Allow Any - used legitimately for dynamic types and SDK boundaries
   # Allow boolean positional values in function calls, like `dict.get(... True)`
   "FBT003",
   # Ignore checks for possible passwords
@@ -139,7 +141,7 @@ ban-relative-imports = "all"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252"]
+"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]
diff --git a/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/evaluator.py b/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/evaluator.py
@@ -54,7 +54,7 @@ def __init__(
         self,
         metric: str | DeepEvalMetric,
         metric_params: dict[str, Any] | None = None,
-    ):
+    ) -> None:
         """
         Construct a new DeepEval evaluator.
 
@@ -144,7 +144,7 @@ def from_dict(cls, data: dict[str, Any]) -> "DeepEvalEvaluator":
     def _invoke_deepeval(test_cases: list[LLMTestCase], metric: BaseMetric) -> EvaluationResult:
         return evaluate(test_cases=test_cases, metrics=[metric])
 
-    def _init_backend(self):
+    def _init_backend(self) -> None:
         """
         Initialize the DeepEval backend.
         """
diff --git a/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/metrics.py b/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/metrics.py
@@ -49,7 +49,7 @@ class DeepEvalMetric(Enum):
     #: Inputs - `questions: List[str], contexts: List[List[str]], responses: List[str]`
     CONTEXTUAL_RELEVANCE = "contextual_relevance"
 
-    def __str__(self):
+    def __str__(self) -> str:
         return self.value
 
     @classmethod
@@ -87,7 +87,13 @@ class MetricResult:
     score: float | None = None
     explanation: str | None = None
 
-    def to_dict(self):
+    def to_dict(self) -> dict[str, Any]:
+        """
+        Convert the metric result to a dictionary.
+
+        :returns:
+            A dictionary with the metric result fields.
+        """
         return dataclasses.asdict(self)
 
 
@@ -129,6 +135,23 @@ def new(
         *,
         init_parameters: Mapping[str, type] | None = None,
     ) -> "MetricDescriptor":
+        """
+        Create a new metric descriptor, inferring input parameters from the converter signature.
+
+        :param metric:
+            The metric enum value.
+        :param backend:
+            The DeepEval metric class to instantiate.
+        :param input_converter:
+            Callable that converts Haystack inputs to DeepEval test cases.
+        :param output_converter:
+            Callable that converts DeepEval results to `MetricResult` objects.
+            If ``None``, the default output converter is used.
+        :param init_parameters:
+            Optional mapping of parameter names to types accepted by the backend metric's constructor.
+        :returns:
+            A new `MetricDescriptor` instance.
+        """
         input_converter_signature = inspect.signature(input_converter)
         input_parameters = {}
         for name, param in input_converter_signature.parameters.items():
@@ -158,7 +181,7 @@ class InputConverters:
     """
 
     @staticmethod
-    def _validate_input_elements(**kwargs):
+    def _validate_input_elements(**kwargs: Any) -> None:
         for k, collection in kwargs.items():
             if not isinstance(collection, list):
                 msg = (
@@ -177,6 +200,18 @@ def _validate_input_elements(**kwargs):
 
     @staticmethod
     def validate_input_parameters(metric: DeepEvalMetric, expected: dict[str, Any], received: dict[str, Any]) -> None:
+        """
+        Validate that all expected input parameters are present in the received inputs.
+
+        :param metric:
+            The metric being evaluated, used for error messages.
+        :param expected:
+            Dictionary of expected parameter names to their types.
+        :param received:
+            Dictionary of received parameter names to their values.
+        :raises ValueError:
+            If a required parameter is missing from ``received``.
+        """
         for param, _ in expected.items():
             if param not in received:
                 msg = f"DeepEval evaluator expected input parameter '{param}' for metric '{metric}'"
@@ -186,6 +221,18 @@ def validate_input_parameters(metric: DeepEvalMetric, expected: dict[str, Any],
     def question_context_response(
         questions: list[str], contexts: list[list[str]], responses: list[str]
     ) -> Iterable[LLMTestCase]:
+        """
+        Convert question, context, and response inputs to DeepEval test cases.
+
+        :param questions:
+            List of input questions.
+        :param contexts:
+            List of retrieval context lists, one per question.
+        :param responses:
+            List of model responses, one per question.
+        :returns:
+            An iterable of `LLMTestCase` objects.
+        """
         InputConverters._validate_input_elements(questions=questions, contexts=contexts, responses=responses)
         for q, c, r in zip(questions, contexts, responses, strict=True):  # type: ignore
             test_case = LLMTestCase(input=q, actual_output=r, retrieval_context=c)
@@ -195,6 +242,20 @@ def question_context_response(
     def question_context_response_ground_truth(
         questions: list[str], contexts: list[list[str]], responses: list[str], ground_truths: list[str]
     ) -> Iterable[LLMTestCase]:
+        """
+        Convert question, context, response, and ground truth inputs to DeepEval test cases.
+
+        :param questions:
+            List of input questions.
+        :param contexts:
+            List of retrieval context lists, one per question.
+        :param responses:
+            List of model responses, one per question.
+        :param ground_truths:
+            List of expected (ground truth) responses, one per question.
+        :returns:
+            An iterable of `LLMTestCase` objects.
+        """
         InputConverters._validate_input_elements(questions=questions, contexts=contexts, responses=responses)
         for q, c, r, gt in zip(questions, contexts, responses, ground_truths, strict=True):  # type: ignore
             test_case = LLMTestCase(input=q, actual_output=r, retrieval_context=c, expected_output=gt)
@@ -212,6 +273,15 @@ class OutputConverters:
     def default(
         metric: DeepEvalMetric,
     ) -> Callable[[TestResult], list[MetricResult]]:
+        """
+        Return the default output converter for a given metric.
+
+        :param metric:
+            The metric for which to create the converter.
+        :returns:
+            A callable that converts a `TestResult` to a list of `MetricResult` objects.
+        """
+
         def inner(output: TestResult, metric: DeepEvalMetric) -> list[MetricResult]:
             metric_name = str(metric)
             assert output.metrics_data
diff --git a/integrations/elasticsearch/pyproject.toml b/integrations/elasticsearch/pyproject.toml
@@ -85,6 +85,7 @@ line-length = 120
 [tool.ruff.lint]
 select = [
   "A",
+  "ANN",
   "ARG",
   "B",
   "C",
@@ -125,6 +126,8 @@ ignore = [
   "PLR0912",
   "PLR0913",
   "PLR0915",
+  # Allow `Any` - used legitimately for dynamic types and SDK boundaries
+  "ANN401",
 ]
 
 [tool.ruff.lint.isort]
@@ -134,8 +137,8 @@ known-first-party = ["haystack_integrations"]
 ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
-# Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252"]
+# Tests can use magic values, assertions, relative imports, and don't need type annotations
+"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]
diff --git a/integrations/elasticsearch/src/haystack_integrations/components/retrievers/elasticsearch/bm25_retriever.py b/integrations/elasticsearch/src/haystack_integrations/components/retrievers/elasticsearch/bm25_retriever.py
@@ -52,7 +52,7 @@ def __init__(
         top_k: int = 10,
         scale_score: bool = False,
         filter_policy: str | FilterPolicy = FilterPolicy.REPLACE,
-    ):
+    ) -> None:
         """
         Initialize ElasticsearchBM25Retriever with an instance ElasticsearchDocumentStore.
 
diff --git a/integrations/elasticsearch/src/haystack_integrations/components/retrievers/elasticsearch/embedding_retriever.py b/integrations/elasticsearch/src/haystack_integrations/components/retrievers/elasticsearch/embedding_retriever.py
@@ -53,7 +53,7 @@ def __init__(
         top_k: int = 10,
         num_candidates: int | None = None,
         filter_policy: str | FilterPolicy = FilterPolicy.REPLACE,
-    ):
+    ) -> None:
         """
         Create the ElasticsearchEmbeddingRetriever component.
 
diff --git a/integrations/elasticsearch/src/haystack_integrations/components/retrievers/elasticsearch/sql_retriever.py b/integrations/elasticsearch/src/haystack_integrations/components/retrievers/elasticsearch/sql_retriever.py
@@ -42,7 +42,7 @@ def __init__(
         document_store: ElasticsearchDocumentStore,
         raise_on_failure: bool = True,
         fetch_size: int | None = None,
-    ):
+    ) -> None:
         """
         Creates the ElasticsearchSQLRetriever component.
 
diff --git a/integrations/elasticsearch/src/haystack_integrations/document_stores/elasticsearch/document_store.py b/integrations/elasticsearch/src/haystack_integrations/document_stores/elasticsearch/document_store.py
@@ -86,7 +86,7 @@ def __init__(
         api_key_id: Secret | str | None = Secret.from_env_var("ELASTIC_API_KEY_ID", strict=False),
         embedding_similarity_function: Literal["cosine", "dot_product", "l2_norm", "max_inner_product"] = "cosine",
         **kwargs: Any,
-    ):
+    ) -> None:
         """
         Creates a new ElasticsearchDocumentStore instance.
 
@@ -156,7 +156,7 @@ def __init__(
                 ],
             }
 
-    def _ensure_initialized(self):
+    def _ensure_initialized(self) -> None:
         """
         Ensures both sync and async clients are initialized and the index exists.
         """
