add indices

py/core/base/__init__.py

@@ -136,7 +136,7 @@
     "generate_document_id",
     "generate_extraction_id",
     "generate_default_user_collection_id",
-    "generate_collection_id_from_name",
+    "generate_id_from_label",
     "generate_user_id",
     "increment_version",
     "EntityType",

py/core/base/api/models/__init__.py

@@ -6,6 +6,7 @@
     WrappedTokenResponse,
     WrappedUserResponse,
 )
+from shared.api.models.base import PaginatedResultsWrapper, ResultsWrapper
 from shared.api.models.ingestion.responses import (
     CreateVectorIndexResponse,
     IngestionResponse,
@@ -147,4 +148,6 @@
     "WrappedCompletionResponse",
     "WrappedRAGResponse",
     "WrappedRAGAgentResponse",
+    "PaginatedResultsWrapper",
+    "ResultsWrapper",
 ]

py/core/base/providers/database.py

@@ -564,6 +564,7 @@ async def get_chunk(self, chunk_id: UUID) -> Optional[dict[str, Any]]:
     @abstractmethod
     async def create_index(
         self,
+        name: Optional[str] = None,
         table_name: Optional[VectorTableName] = None,
         index_measure: IndexMeasure = IndexMeasure.cosine_distance,
         index_method: IndexMethod = IndexMethod.auto,
@@ -578,7 +579,7 @@ async def create_index(
 
     @abstractmethod
     async def list_indices(
-        self, table_name: Optional[VectorTableName] = None
+        self, offset: int = 0, limit: int = 10, filters: Optional[dict] = None
     ) -> list[dict]:
         pass
 
@@ -1479,9 +1480,9 @@ async def create_index(
         )
 
     async def list_indices(
-        self, table_name: Optional[VectorTableName] = None
+        self, offset: int = 0, limit: int = 10, filters: Optional[dict] = None
     ) -> list[dict]:
-        return await self.vector_handler.list_indices(table_name)
+        return await self.vector_handler.list_indices(offset, limit, filters)
 
     async def delete_index(
         self,

py/core/base/utils/__init__.py

@@ -7,12 +7,12 @@
     format_relations,
     format_search_results_for_llm,
     format_search_results_for_stream,
-    generate_collection_id_from_name,
     generate_default_prompt_id,
     generate_default_user_collection_id,
     generate_document_id,
     generate_extraction_id,
     generate_id,
+    generate_id_from_label,
     generate_user_id,
     increment_version,
     llm_cost_per_million_tokens,
@@ -35,7 +35,7 @@
     "generate_document_id",
     "generate_extraction_id",
     "generate_user_id",
-    "generate_collection_id_from_name",
+    "generate_id_from_label",
     "generate_default_prompt_id",
     "RecursiveCharacterTextSplitter",
     "TextSplitter",

py/core/main/api/v2/ingestion_router.py

@@ -572,7 +572,7 @@ async def list_vector_indices_app(
                 description=list_vector_indices_descriptions.get("table_name"),
             ),
             auth_user=Depends(self.service.providers.auth.auth_wrapper),
-        ) -> WrappedListVectorIndicesResponse:
+        ):
             indices = await self.service.providers.database.list_indices(
                 table_name=table_name
             )

py/core/main/api/v3/chunks_router.py

@@ -182,14 +182,16 @@ async def create_chunks(
             run_with_orchestration: Optional[bool] = Body(True),
             auth_user=Depends(self.providers.auth.auth_wrapper),
         ) -> ResultsWrapper[list[ChunkIngestionResponse]]:
-            f"""
+            """
             Create multiple chunks and process them through the ingestion pipeline.
 
             This endpoint allows creating multiple chunks at once, optionally associating them
             with documents and collections. The chunks will be processed asynchronously if
             run_with_orchestration is True.
 
-            Maximum of {MAX_CHUNKS_PER_REQUEST} chunks can be created in a single request.
+            Maximum of 100,000 chunks can be created in a single request.
+
+            Note, it is not yet possible to add chunks to an existing document using this endpoint.
             """
             default_document_id = generate_id()
             if len(raw_chunks) > MAX_CHUNKS_PER_REQUEST:
@@ -212,9 +214,10 @@ async def create_chunks(
                 document_id = document_id or default_document_id
                 # Convert UnprocessedChunks to RawChunks for ingestion
                 raw_chunks_for_doc = [
-                    RawChunk(
+                    UnprocessedChunk(
                         text=chunk.text if hasattr(chunk, "text") else "",
                         metadata=chunk.metadata,
+                        id=chunk.id,
                     )
                     for chunk in doc_chunks
                 ]
@@ -229,6 +232,9 @@ async def create_chunks(
                     "user": auth_user.model_dump_json(),
                 }
 
+                # TODO - Modify create_chunks so that we can add chunks to existing document
+                # TODO - Modify create_chunks so that we can add chunks to existing document
+
                 if run_with_orchestration:
                     # Run ingestion with orchestration
                     raw_message = (
@@ -384,9 +390,11 @@ async def retrieve_chunk(
 
 client = R2RClient("http://localhost:7272")
 result = client.chunks.update(
-    id="b4ac4dd6-5f27-596e-a55b-7cf242ca30aa",
-    text="Updated content",
-    metadata={"key": "new value"}
+    {
+        "id": first_chunk_id,
+        "text": "Updated content",
+        "metadata": {"key": "new value"}
+    }
 )
 """,
                     }
@@ -498,7 +506,7 @@ async def enrich_chunk(
         @self.base_endpoint
         async def delete_chunk(
             id: Json[UUID] = Path(...),
-        ) -> ResultsWrapper[ChunkResponse]:
+        ) -> ResultsWrapper[bool]:
             """
             Delete a specific chunk by ID.
 
@@ -512,7 +520,7 @@ async def delete_chunk(
                 raise R2RException(f"Chunk {id} not found", 404)
 
             await self.services["management"].delete({"$eq": {"chunk_id": id}})
-            return None
+            return True
 
         @self.router.get(
             "/chunks",

py/core/main/api/v3/collections_router.py

@@ -8,6 +8,7 @@
 
 from core.base import R2RException, RunType
 from core.base.api.models import (
+    ResultsWrapper,
     WrappedAddUserResponse,
     WrappedCollectionListResponse,
     WrappedCollectionResponse,
@@ -348,7 +349,7 @@ async def delete_collection(
                 description="The unique identifier of the collection to delete",
             ),
             auth_user=Depends(self.providers.auth.auth_wrapper),
-        ) -> WrappedDeleteResponse:
+        ) -> ResultsWrapper[bool]:
             """
             Delete an existing collection.
 
@@ -566,7 +567,7 @@ async def remove_document_from_collection(
                 description="The unique identifier of the document to remove",
             ),
             auth_user=Depends(self.providers.auth.auth_wrapper),
-        ) -> WrappedDeleteResponse:
+        ) -> ResultsWrapper[bool]:
             """
             Remove a document from a collection.
 
@@ -774,7 +775,7 @@ async def remove_user_from_collection(
                 ..., description="The unique identifier of the user to remove"
             ),
             auth_user=Depends(self.providers.auth.auth_wrapper),
-        ) -> WrappedDeleteResponse:
+        ) -> ResultsWrapper[bool]:
             """
             Remove a user from a collection.