from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import Chroma
from langchain.llms import HuggingFaceHub
from langchain.chains import RetrievalQA
from knowledge_base import load_vectorstore
import os

# Load vectorstore with meal plan embeddings
db = load_vectorstore()
retriever = db.as_retriever()

# Load LLM (HuggingFace Inference API)
llm = HuggingFaceHub(
    repo_id="mistralai/Mistral-7B-Instruct-v0.1",
    model_kwargs={"temperature": 0.3, "max_new_tokens": 512}
)

# Build Retrieval QA Chain
qa_chain = RetrievalQA.from_chain_type(
    llm=llm,
    retriever=retriever,
    chain_type="stuff"
)

def get_bot_response(query):
    """
    Accepts a user query, runs through RAG chain, and returns a response.
    """
    if not query:
        return "Please ask something."

    try:
        result = qa_chain.run(query)
        return result
    except Exception as e:
        return f"Error processing query: {str(e)}"