from configs.model_config import *
from chains.local_doc_qa import LocalDocQA
import os
import nltk
from models.loader.args import parser
import models.shared as shared
from models.loader import LoaderCheckPoint
from models.chatglm_llm import ChatGLM

nltk.data.path = [os.path.join(os.path.dirname(__file__), "nltk_data")] + nltk.data.path

# return top-k text chunk from vector store
VECTOR_SEARCH_TOP_K = 6

# LLM input history length
LLM_HISTORY_LEN = 3

# Show reply with source text from input document
REPLY_WITH_SOURCE = True

if __name__ == "__main__":
    args = None
    args = parser.parse_args()
    args_dict = vars(args)
    shared.loaderCheckPoint = LoaderCheckPoint(args_dict)
    llm_model_ins = shared.loaderLLM()
    llm_model_ins.history_len = LLM_HISTORY_LEN

    local_doc_qa = LocalDocQA()
    local_doc_qa.init_cfg(llm_model=llm_model_ins,
                          embedding_model=EMBEDDING_MODEL,
                          embedding_device=EMBEDDING_DEVICE,
                          top_k=VECTOR_SEARCH_TOP_K)
    vs_path = None
    while not vs_path:
        filepath = input("Input your local knowledge file path 请输入本地知识文件路径：")
        vs_path, _ = local_doc_qa.init_knowledge_vector_store(filepath)
    history = []
    while True:
        query = input("Input your question 请输入问题：")
        last_print_len = 0
        for resp, history in local_doc_qa.get_knowledge_based_answer(query=query,
                                                                     vs_path=vs_path,
                                                                     chat_history=history,
                                                                     streaming=STREAMING):
            if STREAMING:
                print(resp["result"][last_print_len:], end="", flush=True)
                last_print_len = len(resp["result"])
            else:
                print(resp["result"])
        if REPLY_WITH_SOURCE:
            source_text = [f"""出处 [{inum + 1}] {os.path.split(doc.metadata['source'])[-1]}：\n\n{doc.page_content}\n\n"""
                           # f"""相关度：{doc.metadata['score']}\n\n"""
                           for inum, doc in
                           enumerate(resp["source_documents"])]
            print("\n\n" + "\n\n".join(source_text))