rag.core.index
==============

.. py:module:: rag.core.index

.. autoapi-nested-parse::

   Chroma persistence helpers for the papers collection.



Exceptions
----------

.. autoapisummary::

   rag.core.index.IndexMissingError


Functions
---------

.. autoapisummary::

   rag.core.index.open_chroma_collection
   rag.core.index.load_vector_index


Module Contents
---------------

.. py:exception:: IndexMissingError

   Bases: :py:obj:`FileNotFoundError`


   Raised when the Chroma collection has no documents yet.


.. py:function:: open_chroma_collection(*, config: rag.core.config.RagConfig) -> Any

   Open (or create) the persistent Chroma collection.

   :param config: Paths and collection name.
   :type config: RagConfig

   :returns: Open collection handle.
   :rtype: chromadb.Collection


.. py:function:: load_vector_index(*, config: rag.core.config.RagConfig, embed_model: llama_index.core.embeddings.BaseEmbedding, require_nonempty: bool = True) -> llama_index.core.VectorStoreIndex

   Load a ``VectorStoreIndex`` backed by the persistent Chroma store.

   :param config: Index configuration.
   :type config: RagConfig
   :param embed_model: Embedding model used at query time (must match ingest).
   :type embed_model: BaseEmbedding
   :param require_nonempty: When True, raise if the collection has zero documents.
   :type require_nonempty: bool

   :returns: LlamaIndex vector index over Chroma.
   :rtype: VectorStoreIndex

   :raises IndexMissingError: If ``require_nonempty`` and the collection is empty.