from langchain_milvus import BM25BuiltInFunction, Milvus
from config.llm import llmEmbeddings

URI = "http://10.10.10.9:19530"

knVectorstore = Milvus(
    embedding_function=llmEmbeddings,
    connection_args={"uri": URI, "token": "root:Milvus", "db_name": "bbit_ai_lab"},
    collection_name="knowledge",
    index_params={"index_type": "FLAT", "metric_type": "L2"},
    consistency_level="Strong",
    auto_id=True,

    primary_field = "id",
    text_field="text",
    vector_field="vector",
    partition_key_field = "kn_id",
    enable_dynamic_field = True,
    drop_old=False,  # set to True if seeking to drop the collection with that name if it exists
)
memVectorstore = Milvus(
    embedding_function=llmEmbeddings,
    connection_args={"uri": URI, "token": "root:Milvus", "db_name": "bbit_ai_lab"},
    collection_name="memory",
    index_params={"index_type": "FLAT", "metric_type": "L2"},
    consistency_level="Strong",
    auto_id=True,

    primary_field = "id",
    text_field="text",
    vector_field="vector",
    partition_key_field = "ai_id",
    enable_dynamic_field = True,
    drop_old=False,  # set to True if seeking to drop the collection with that name if it exists
)