Merge remote-tracking branch 'origin/main' into custom_vector_store_schema

Author: Gaudy Blanco

.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."
+}

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/index/operations/summarize_communities/strategies.py

@@ -16,7 +16,6 @@
     Finding,
     StrategyConfig,
 )
-from graphrag.index.utils.rate_limiter import RateLimiter
 from graphrag.language_model.manager import ModelManager
 from graphrag.language_model.protocol.base import ChatModel
 
@@ -51,8 +50,6 @@ async def _run_extractor(
     level: int,
     args: StrategyConfig,
 ) -> CommunityReport | None:
-    # RateLimiter
-    rate_limiter = RateLimiter(rate=1, per=60)
     extractor = CommunityReportsExtractor(
         model,
         extraction_prompt=args.get("extraction_prompt", None),
@@ -63,7 +60,6 @@ async def _run_extractor(
     )
 
     try:
-        await rate_limiter.acquire()
         results = await extractor(input)
         report = results.structured_output
         if report is None:

graphrag/index/utils/rate_limiter.py

@@ -190,57 +190,58 @@ def update_context_data(
     """
     updated_context_data = {}
     for key in context_data:
+        entries = context_data[key].to_dict(orient="records")
         updated_entry = []
         if key == "reports":
             updated_entry = [
                 dict(
-                    {k: entry[k] for k in entry},
+                    entry,
                     index_name=links["community_reports"][int(entry["id"])][
                         "index_name"
                     ],
                     index_id=links["community_reports"][int(entry["id"])]["id"],
                 )
-                for entry in context_data[key]
+                for entry in entries
             ]
         if key == "entities":
             updated_entry = [
                 dict(
-                    {k: entry[k] for k in entry},
+                    entry,
                     entity=entry["entity"].split("-")[0],
                     index_name=links["entities"][int(entry["id"])]["index_name"],
                     index_id=links["entities"][int(entry["id"])]["id"],
                 )
-                for entry in context_data[key]
+                for entry in entries
             ]
         if key == "relationships":
             updated_entry = [
                 dict(
-                    {k: entry[k] for k in entry},
+                    entry,
                     source=entry["source"].split("-")[0],
                     target=entry["target"].split("-")[0],
                     index_name=links["relationships"][int(entry["id"])]["index_name"],
                     index_id=links["relationships"][int(entry["id"])]["id"],
                 )
-                for entry in context_data[key]
+                for entry in entries
             ]
         if key == "claims":
             updated_entry = [
                 dict(
-                    {k: entry[k] for k in entry},
+                    entry,
                     entity=entry["entity"].split("-")[0],
                     index_name=links["covariates"][int(entry["id"])]["index_name"],
                     index_id=links["covariates"][int(entry["id"])]["id"],
                 )
-                for entry in context_data[key]
+                for entry in entries
             ]
         if key == "sources":
             updated_entry = [
                 dict(
-                    {k: entry[k] for k in entry},
+                    entry,
                     index_name=links["text_units"][int(entry["id"])]["index_name"],
                     index_id=links["text_units"][int(entry["id"])]["id"],
                 )
-                for entry in context_data[key]
+                for entry in entries
             ]
         updated_context_data[key] = updated_entry
     return updated_context_data