from langchain.document_loaders import PyPDFDirectoryLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.vectorstores import FAISS
from langchain.llms import openai
from langchain.chains import ConversationalRetrievalChain
from langchain.chat_models import ChatOpenAI
from langchain.embeddings import HuggingFaceBgeEmbeddings
###########################################################################################

def get_pdf_load():
    loader=PyPDFDirectoryLoader("./data")
    document=loader.load()
    return document
########################################################################################
def get_text_split(document):
    text_splitter= RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
    texts =text_splitter.split_documents(document)
    return texts
#########################################################################################
def get_vectorstore(texts):
    #Vector and Embeddings
    DB_FAISS_PATH = 'vectore_Imstudio/faiss'
    #Vector and Embeddings
    embeddings= HuggingFaceBgeEmbeddings(model_name='Avditvs/multilingual-e5-small-distill-base-0.1', model_kwargs={'device': 'cpu'})
    db= FAISS.from_documents(texts,embeddings)
    db.save_local(DB_FAISS_PATH)
    return db
############################################################################################
def get_chain(db):
    llm=ChatOpenAI(base_url="https://bd4c-85-9-86-142.ngrok-free.app/v1", api_key="lm-studio",temperature=0.1,model="lmstudio-community/Meta-Llama-3-8B-Instruct-GGUF")
    #Build a chain
    qa_chain = ConversationalRetrievalChain.from_llm(
        llm,db.as_retriever (search_kwargs={'k':2}),return_source_documents=True)
    return  qa_chain
####################################################################################################################
def main():
    st.set_page_config(
        page_title="Chat Bot PDFs",
        page_icon=":books:",
    )
    
    st.header("Chat Bot PDFs :books:")
    user_question = st.text_input("Ask a question about your documents:")
    if st.button("Build Model"):
        with st.spinner("Waiting"):
            document=get_pdf_load()
            texts=et_text_split(document)
            db=get_vectorstore(texts)
            qa_chain=get_chain(db)
            st.write("compelete build model")
  
    if st.button("Answer"):
            with st.spinner("Answering"):
              
          
    if st.button("CLEAR"):
            with st.spinner("CLEARING"):
              st.cache_data.clear()
         
    
    #with st.sidebar:
        #if st.button("Process build model"):
 


if __name__ == "__main__": 
    main()