Merge branch 'main' into encoder

Author: AlonsoGuevara

.semversioner/next-release/minor-20250916182815141332.json

@@ -0,0 +1,4 @@
+{
+  "type": "minor",
+  "description": "Add config for NLP async mode."
+}

.semversioner/next-release/minor-20250918193514400373.json

@@ -0,0 +1,4 @@
+{
+  "type": "minor",
+  "description": "add customization to vector store"
+}

.semversioner/next-release/patch-20250915234643851001.json

@@ -0,0 +1,4 @@
+{
+  "type": "patch",
+  "description": "Remove hard-coded community rate limiter."
+}

.semversioner/next-release/patch-20250918192431890892.json

@@ -0,0 +1,4 @@
+{
+  "type": "patch",
+  "description": "Fix multi-index search."
+}

.vscode/launch.json

@@ -6,21 +6,24 @@
 			"name": "Indexer",
 			"type": "debugpy",
 			"request": "launch",
-			"module": "uv",
+			"module": "graphrag",
 			"args": [
-				"poe", "index",
-				"--root", "<path_to_ragtest_root_demo>"
+				"index",
+				"--root", 
+				"<path_to_index_folder>"
 			],
+			"console": "integratedTerminal"
 		},
 		{
 			"name": "Query",
 			"type": "debugpy",
 			"request": "launch",
-			"module": "uv",
+			"module": "graphrag",
 			"args": [
-				"poe", "query",
-				"--root", "<path_to_ragtest_root_demo>",
-				"--method", "global",
+				"query",
+				"--root", 
+				"<path_to_index_folder>",
+				"--method", "basic",
 				"--query", "What are the top themes in this story",
 			]
 		},
@@ -34,6 +37,42 @@
 				"--config",
 				"<path_to_ragtest_root_demo>/settings.yaml",
 			]
-		}
+		},
+		{
+			"name": "Debug Integration Pytest",
+			"type": "debugpy",
+			"request": "launch",
+			"module": "pytest",
+			"args": [
+				"./tests/integration/vector_stores",
+				"-k", "test_azure_ai_search"
+			],
+			"console": "integratedTerminal",
+    		"justMyCode": false 
+		},
+		{
+			"name": "Debug Verbs Pytest",
+			"type": "debugpy",
+			"request": "launch",
+			"module": "pytest",
+			"args": [
+				"./tests/verbs",
+				"-k", "test_generate_text_embeddings"
+			],
+			"console": "integratedTerminal",
+    		"justMyCode": false 
+		},
+		{
+			"name": "Debug Smoke Pytest",
+			"type": "debugpy",
+			"request": "launch",
+			"module": "pytest",
+			"args": [
+				"./tests/smoke",
+				"-k", "test_fixtures"
+			],
+			"console": "integratedTerminal",
+    		"justMyCode": false 
+		},
 	]
 }

docs/get_started.md

@@ -65,11 +65,10 @@ deployment_name: <azure_model_deployment_name>
 ```
 
 #### Using Managed Auth on Azure
-To use managed auth, add an additional value to your model config and comment out or remove the api_key line:
+To use managed auth, edit the auth_type in your model config and *remove* the api_key line:
 
 ```yaml
 auth_type: azure_managed_identity # Default auth_type is is api_key
-# api_key: ${GRAPHRAG_API_KEY}
 ```
 
 You will also need to login with [az login](https://learn.microsoft.com/en-us/cli/azure/authenticate-azure-cli) and select the subscription with your endpoint.
@@ -116,4 +115,4 @@ Please refer to [Query Engine](query/overview.md) docs for detailed information
 - For more details about configuring GraphRAG, see the [configuration documentation](config/overview.md).
 - To learn more about Initialization, refer to the [Initialization documentation](config/init.md).
 - For more details about using the CLI, refer to the [CLI documentation](cli.md).
-- Check out our [visualization guide](visualization_guide.md) for a more interactive experience in debugging and exploring the knowledge graph.
+- Check out our [visualization guide](visualization_guide.md) for a more interactive experience in debugging and exploring the knowledge graph.

graphrag/cache/factory.py

@@ -97,9 +97,19 @@ def create_cosmosdb_cache(**kwargs) -> PipelineCache:
     return JsonPipelineCache(storage)
 
 
+def create_noop_cache(**_kwargs) -> PipelineCache:
+    """Create a no-op cache implementation."""
+    return NoopPipelineCache()
+
+
+def create_memory_cache(**kwargs) -> PipelineCache:
+    """Create a memory cache implementation."""
+    return InMemoryCache(**kwargs)
+
+
 # --- register built-in cache implementations ---
-CacheFactory.register(CacheType.none.value, NoopPipelineCache)
-CacheFactory.register(CacheType.memory.value, InMemoryCache)
+CacheFactory.register(CacheType.none.value, create_noop_cache)
+CacheFactory.register(CacheType.memory.value, create_memory_cache)
 CacheFactory.register(CacheType.file.value, create_file_cache)
 CacheFactory.register(CacheType.blob.value, create_blob_cache)
 CacheFactory.register(CacheType.cosmosdb.value, create_cosmosdb_cache)

graphrag/config/defaults.py

@@ -246,6 +246,7 @@ class ExtractGraphNLPDefaults:
     normalize_edge_weights: bool = True
     text_analyzer: TextAnalyzerDefaults = field(default_factory=TextAnalyzerDefaults)
     concurrent_requests: int = 25
+    async_mode: AsyncType = AsyncType.Threaded
 
 
 @dataclass
@@ -427,6 +428,7 @@ class VectorStoreDefaults:
     api_key: None = None
     audience: None = None
     database_name: None = None
+    schema: None = None
 
 
 @dataclass

graphrag/config/embeddings.py

@@ -29,14 +29,14 @@
 ]
 
 
-def create_collection_name(
+def create_index_name(
     container_name: str, embedding_name: str, validate: bool = True
 ) -> str:
     """
-    Create a collection name for the embedding store.
+    Create a index name for the embedding store.
 
     Within any given vector store, we can have multiple sets of embeddings organized into projects.
-    The `container` param is used for this partitioning, and is added as a prefix to the collection name for differentiation.
+    The `container` param is used for this partitioning, and is added as a prefix to the index name for differentiation.
 
     The embedding name is fixed, with the available list defined in graphrag.index.config.embeddings
 

graphrag/config/init_content.py

@@ -112,6 +112,7 @@
 extract_graph_nlp:
   text_analyzer:
     extractor_type: {graphrag_config_defaults.extract_graph_nlp.text_analyzer.extractor_type.value} # [regex_english, syntactic_parser, cfg]
+  async_mode: {graphrag_config_defaults.extract_graph_nlp.async_mode.value} # or asyncio
 
 cluster_graph:
   max_cluster_size: {graphrag_config_defaults.cluster_graph.max_cluster_size}