_get_retriever_tags() — langchain Function Reference

Architecture documentation for the _get_retriever_tags() function in qdrant.py from the langchain codebase.

Function python LangChainCore Runnables

Entity Profile

LangChainCore→ Runnables→ _get_retriever_tags() — langchain Function Reference

Dependency Diagram

graph TD
  d90db0d6_cc58_7f9b_4c05_46a2d6fe6ead["_get_retriever_tags()"]
  671b47a0_cdd3_a89d_e90f_0631a4bd67d3["QdrantVectorStore"]
  d90db0d6_cc58_7f9b_4c05_46a2d6fe6ead -->|defined in| 671b47a0_cdd3_a89d_e90f_0631a4bd67d3
  style d90db0d6_cc58_7f9b_4c05_46a2d6fe6ead fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/partners/qdrant/langchain_qdrant/qdrant.py lines 282–300

    def _get_retriever_tags(self) -> list[str]:
        """Get tags for retriever.

        Override the base class method to handle SPARSE mode where embeddings can be
        None. In SPARSE mode, embeddings is None, so we don't include embeddings class
        name in tags. In DENSE/HYBRID modes, embeddings is not None, so we include
        embeddings class name.
        """
        tags = [self.__class__.__name__]

        # Handle different retrieval modes
        if self.retrieval_mode == RetrievalMode.SPARSE:
            # SPARSE mode: no dense embeddings, so no embeddings class name in tags
            pass
        # DENSE/HYBRID modes: include embeddings class name if available
        elif self.embeddings is not None:
            tags.append(self.embeddings.__class__.__name__)

        return tags