format

Author: Bessouat40

examples/hybrid_search_example.py

@@ -34,9 +34,13 @@
         Settings.CHROMA,
         persist_directory=persist_directory,
         collection_name=collection_name,
-        search_type=Settings.SEARCH_SEMANTIC,   # default — can be omitted
+        search_type=Settings.SEARCH_SEMANTIC,  # default — can be omitted
+    )
+    .with_llm(
+        Settings.OLLAMA,
+        model_name=model_name,
+        system_prompt=Settings.DEFAULT_SYSTEM_PROMPT,
     )
-    .with_llm(Settings.OLLAMA, model_name=model_name, system_prompt=Settings.DEFAULT_SYSTEM_PROMPT)
     .build_rag(k=5)
 )
 rag_semantic.vector_store.ingest(data_path=data_path)
@@ -54,7 +58,11 @@
         collection_name=collection_name + "_bm25",
         search_type=Settings.SEARCH_BM25,
     )
-    .with_llm(Settings.OLLAMA, model_name=model_name, system_prompt=Settings.DEFAULT_SYSTEM_PROMPT)
+    .with_llm(
+        Settings.OLLAMA,
+        model_name=model_name,
+        system_prompt=Settings.DEFAULT_SYSTEM_PROMPT,
+    )
     .build_rag(k=5)
 )
 rag_bm25.vector_store.ingest(data_path=data_path)
@@ -71,9 +79,13 @@
         persist_directory=persist_directory,
         collection_name=collection_name + "_hybrid",
         search_type=Settings.SEARCH_HYBRID,
-        alpha=0.5,   # weight of semantic vs BM25 in the RRF merge (0=BM25 only, 1=semantic only)
+        alpha=0.5,  # weight of semantic vs BM25 in the RRF merge (0=BM25 only, 1=semantic only)
+    )
+    .with_llm(
+        Settings.OLLAMA,
+        model_name=model_name,
+        system_prompt=Settings.DEFAULT_SYSTEM_PROMPT,
     )
-    .with_llm(Settings.OLLAMA, model_name=model_name, system_prompt=Settings.DEFAULT_SYSTEM_PROMPT)
     .build_rag(k=5)
 )
 rag_hybrid.vector_store.ingest(data_path=data_path)
@@ -93,7 +105,7 @@
     database=Settings.CHROMA,
     persist_directory=persist_directory,
     collection_name=collection_name + "_api",
-    search_type=Settings.SEARCH_HYBRID,   # <-- hybrid mode
+    search_type=Settings.SEARCH_HYBRID,  # <-- hybrid mode
     hybrid_alpha=0.5,
 )
 

examples/langfuse_example.py

@@ -80,6 +80,5 @@
     print(f"A: {response}")
 
 print(
-    "\nAll traces have been sent to Langfuse. "
-    "Open your dashboard to inspect them."
+    "\nAll traces have been sent to Langfuse. " "Open your dashboard to inspect them."
 )

examples/qdrant_example.py

@@ -23,7 +23,9 @@
 
 # --- Configuration -----------------------------------------------------------
 
-persist_directory = "./qdrantDb"       # local mode — change to None + set host/port for remote
+persist_directory = (
+    "./qdrantDb"  # local mode — change to None + set host/port for remote
+)
 # host = "localhost"                   # remote mode
 # port = 6333                          # remote mode (default)
 

tests/__init__.py

@@ -3,6 +3,7 @@
 even though langgraph==1.0.5 does not expose the expected symbols.
 Loaded automatically by both unittest and pytest before any test module.
 """
+
 import os
 import sys
 from unittest.mock import MagicMock

tests/tests_api/test_router.py

@@ -32,7 +32,9 @@ def test_health(self):
 
     def test_generate(self):
         self.pipeline.generate.return_value = "Paris is the capital of France."
-        response = self.client.post("/generate", json={"question": "What is the capital of France?"})
+        response = self.client.post(
+            "/generate", json={"question": "What is the capital of France?"}
+        )
         self.assertEqual(response.status_code, 200)
         self.assertEqual(response.json(), {"answer": "Paris is the capital of France."})
         self.pipeline.generate.assert_called_once_with("What is the capital of France?")
@@ -56,7 +58,9 @@ def test_ingest_missing_params(self):
         response = self.client.post("/ingest", json={})
         self.assertEqual(response.status_code, 400)
         detail = response.json()["detail"]
-        self.assertTrue(any(k in detail for k in ("data_path", "file_paths", "github_url")))
+        self.assertTrue(
+            any(k in detail for k in ("data_path", "file_paths", "github_url"))
+        )
 
     def test_ingest_github(self):
         self.vector_store.ingest.return_value = None
@@ -66,7 +70,10 @@ def test_ingest_github(self):
             MockScrapper.return_value = instance
             response = self.client.post(
                 "/ingest",
-                json={"github_url": "https://github.com/example/repo", "github_branch": "main"},
+                json={
+                    "github_url": "https://github.com/example/repo",
+                    "github_branch": "main",
+                },
             )
         self.assertEqual(response.status_code, 200)
         instance.set_repositories.assert_called_once()
@@ -88,7 +95,9 @@ def test_ingest_file_paths(self):
 
     def test_ingest_file_paths_missing_file(self):
         with patch("os.path.isfile", return_value=False):
-            response = self.client.post("/ingest", json={"file_paths": ["/nonexistent/file.pdf"]})
+            response = self.client.post(
+                "/ingest", json={"file_paths": ["/nonexistent/file.pdf"]}
+            )
         self.assertEqual(response.status_code, 500)
         self.assertIn("not found", response.json()["detail"].lower())
 
@@ -116,7 +125,10 @@ def test_ingest_upload_no_files(self):
     # ── /collections ──────────────────────────────────────────────────────────
 
     def test_collections(self):
-        self.vector_store.get_available_collections.return_value = ["default", "project_x"]
+        self.vector_store.get_available_collections.return_value = [
+            "default",
+            "project_x",
+        ]
         response = self.client.get("/collections")
         self.assertEqual(response.status_code, 200)
         self.assertEqual(response.json(), {"collections": ["default", "project_x"]})

tests/tests_api/test_server_config.py

@@ -74,7 +74,11 @@ def test_to_vector_store_config_returns_correct_type(self):
         self.assertEqual(vs_config.collection_name, cfg.collection)
 
     def test_to_vector_store_config_chroma_host(self):
-        env = {**_clean_env(), "RAGLIGHT_CHROMA_HOST": "chromadb", "RAGLIGHT_CHROMA_PORT": "8000"}
+        env = {
+            **_clean_env(),
+            "RAGLIGHT_CHROMA_HOST": "chromadb",
+            "RAGLIGHT_CHROMA_PORT": "8000",
+        }
         with patch.dict(os.environ, env, clear=True):
             cfg = ServerConfig()
         vs_config = cfg.to_vector_store_config()

tests/tests_embeddings/test_gemini_embeddings.py

@@ -4,58 +4,64 @@
 from ..test_config import TestsConfig
 from google.genai import types
 
+
 class TestGeminiEmbeddings(unittest.TestCase):
-    
+
     @patch("raglight.embeddings.gemini_embeddings.Client")
     def test_model_load(self, MockClient: MagicMock):
         """
         Test that the client is instantiated with the correct parameters.
         Note: The code uses Client(...) instead of configure().
         """
         model = GeminiEmbeddingsModel(TestsConfig.GEMINI_EMBEDDING_MODEL)
-        
+
         self.assertTrue(MockClient.called)
-        
-        self.assertIsNotNone(model.model, "Model (genai client instance) should be loaded.")
+
+        self.assertIsNotNone(
+            model.model, "Model (genai client instance) should be loaded."
+        )
 
     @patch("raglight.embeddings.gemini_embeddings.Client")
     def test_embed_documents(self, MockClient: MagicMock):
         """Test document embedding with the correct task_type."""
         mock_client_instance = MockClient.return_value
-        
-        mock_client_instance.embed_content.return_value = {'embedding': [[0.1, 0.2], [0.3, 0.4]]}
-        
+
+        mock_client_instance.embed_content.return_value = {
+            "embedding": [[0.1, 0.2], [0.3, 0.4]]
+        }
+
         model = GeminiEmbeddingsModel(TestsConfig.GEMINI_EMBEDDING_MODEL)
         texts = ["doc1", "doc2"]
-        
+
         result = model.embed_documents(texts)
-        
+
         self.assertEqual(len(result), 2)
-        
+
         mock_client_instance.embed_content.assert_called_with(
             model=TestsConfig.GEMINI_EMBEDDING_MODEL,
             content=texts,
-            task_type="retrieval_document"
+            task_type="retrieval_document",
         )
 
     @patch("raglight.embeddings.gemini_embeddings.Client")
     def test_embed_query(self, MockClient: MagicMock):
         """Test query embedding with the correct task_type."""
         mock_client_instance = MockClient.return_value
-        mock_client_instance.embed_content.return_value = {'embedding': [0.1, 0.2]}
-        
+        mock_client_instance.embed_content.return_value = {"embedding": [0.1, 0.2]}
+
         model = GeminiEmbeddingsModel(TestsConfig.GEMINI_EMBEDDING_MODEL)
         text = "query"
-        
+
         result = model.embed_query(text)
-        
+
         self.assertEqual(len(result), 2)
-        
+
         mock_client_instance.embed_content.assert_called_with(
             model=TestsConfig.GEMINI_EMBEDDING_MODEL,
             content=text,
-            task_type="retrieval_query"
+            task_type="retrieval_query",
         )
 
+
 if __name__ == "__main__":
-    unittest.main()
+    unittest.main()

tests/tests_embeddings/test_huggingface_embeddings.py

@@ -6,14 +6,13 @@
 
 
 class TestHuggingFaceEmbeddings(unittest.TestCase):
-    
+
     @patch("raglight.embeddings.huggingface_embeddings.SentenceTransformer")
     def test_model_load(self, mock_transformer: MagicMock):
         """Test SentenceTransformer model loading."""
         mock_transformer.return_value = MagicMock()
         embeddings = HuggingfaceEmbeddingsModel(TestsConfig.HUGGINGFACE_EMBEDDINGS)
-        
-        
+
         self.assertIsNotNone(embeddings.model, "Model should be loaded successfully.")
         mock_transformer.assert_called_once_with(TestsConfig.HUGGINGFACE_EMBEDDINGS)
 
@@ -22,14 +21,14 @@ def test_embed_documents(self, mock_transformer: MagicMock):
         """Test document encoding."""
         mock_instance = mock_transformer.return_value
         mock_instance.encode.return_value = np.array([[0.1, 0.2], [0.3, 0.4]])
-        
+
         embeddings = HuggingfaceEmbeddingsModel(TestsConfig.HUGGINGFACE_EMBEDDINGS)
         result = embeddings.embed_documents(["text1", "text2"])
-        
+
         self.assertIsInstance(result, list)
         self.assertIsInstance(result[0], list)
         mock_instance.encode.assert_called_with(["text1", "text2"])
 
 
 if __name__ == "__main__":
-    unittest.main()
+    unittest.main()

tests/tests_embeddings/test_ollama_embeddings.py

@@ -5,7 +5,7 @@
 
 
 class TestOllamaEmbeddings(unittest.TestCase):
-    
+
     @patch("raglight.embeddings.ollama_embeddings.Client")
     def test_model_load(self, mock_client: MagicMock):
         """Test Ollama client initialization."""
@@ -18,40 +18,32 @@ def test_model_load(self, mock_client: MagicMock):
     def test_embed_documents(self, mock_client: MagicMock):
         """Test batch embedding with .embed() method."""
         mock_instance = mock_client.return_value
-        mock_instance.embed.return_value = {
-            "embeddings": [[0.1, 0.2], [0.3, 0.4]]
-        }
-        
+        mock_instance.embed.return_value = {"embeddings": [[0.1, 0.2], [0.3, 0.4]]}
+
         model = OllamaEmbeddingsModel(TestsConfig.OLLAMA_EMBEDDING_MODEL)
         texts = ["doc1", "doc2"]
         result = model.embed_documents(texts)
-        
+
         self.assertEqual(len(result), 2)
         mock_instance.embed.assert_called_with(
-            model=TestsConfig.OLLAMA_EMBEDDING_MODEL,
-            input=texts,
-            options=model.options
+            model=TestsConfig.OLLAMA_EMBEDDING_MODEL, input=texts, options=model.options
         )
 
     @patch("raglight.embeddings.ollama_embeddings.Client")
     def test_embed_query(self, mock_client: MagicMock):
         """Test single embedding with .embeddings() method."""
         mock_instance = mock_client.return_value
-        mock_instance.embeddings.return_value = {
-            "embedding": [0.9, 0.9]
-        }
-        
+        mock_instance.embeddings.return_value = {"embedding": [0.9, 0.9]}
+
         model = OllamaEmbeddingsModel(TestsConfig.OLLAMA_EMBEDDING_MODEL)
         text = "query text"
         result = model.embed_query(text)
-        
+
         self.assertEqual(result, [0.9, 0.9])
         mock_instance.embeddings.assert_called_with(
-            model=TestsConfig.OLLAMA_EMBEDDING_MODEL,
-            prompt=text,
-            options=model.options
+            model=TestsConfig.OLLAMA_EMBEDDING_MODEL, prompt=text, options=model.options
         )
 
 
 if __name__ == "__main__":
-    unittest.main()
+    unittest.main()