[QUESTION]: Why does neo4j_graphrag, a framework for hybrid retrieval, not support tag filtering

[chenxiaopeng-com]

Question

Why does neo4j_graphrag, a framework for hybrid retrieval, not support tag filtering
What serious problems would arise if I were to modify it to support

if entity_type == EntityType.NODE:
    if search_type == SearchType.HYBRID:
        if filters:
            **raise Exception("Filters are not supported with hybrid search")**
        if ranker == HybridSearchRanker.NAIVE:
            query = _get_hybrid_query(neo4j_version_is_5_23_or_above)
        elif ranker == HybridSearchRanker.LINEAR and alpha:
            query = _get_hybrid_query_linear(
                neo4j_version_is_5_23_or_above, alpha=alpha
            )
        else:
            raise InvalidHybridSearchRankerError()
        params: dict[str, Any] = {}
    elif search_type == SearchType.VECTOR:
        if filters:
            if (
                node_label is not None
                and embedding_node_property is not None
                and embedding_dimension is not None
            ):
                query, params = _get_filtered_vector_query(
                    filters,
                    node_label,
                    embedding_node_property,
                    embedding_dimension,
                    use_parallel_runtime,
                )
            else:
                raise Exception(
                    "Vector Search with filters requires: node_label, embedding_node_property, embedding_dimension"
                )
        else:
            query, params = NODE_VECTOR_INDEX_QUERY, {}
    else:
        raise ValueError(f"Search type is not supported: {search_type}")
    fallback_return = (
        f"RETURN node {{ .*, `{embedding_node_property}`: null }} AS node, "
        "labels(node) AS nodeLabels, "
        "elementId(node) AS elementId, "
        "elementId(node) AS id, "
        "score"
    )