app.py

import gradio as gr
import os
from langchain.chains import ConversationalRetrievalChain
from langchain.text_splitter import CharacterTextSplitter
from langchain_community.document_loaders import PyPDFLoader, Docx2txtLoader, TextLoader
from langchain_community.vectorstores import Chroma
from langchain_openai import ChatOpenAI, OpenAIEmbeddings
from dotenv import load_dotenv

# Load environment variables
load_dotenv()
os.environ["OPENAI_API_KEY"] = os.getenv("OPENAI_API_KEY")

# Validate OpenAI API Key
api_key = os.getenv('OPENAI_API_KEY')
if not api_key:
    raise ValueError("Please set the 'OPENAI_API_KEY' environment variable")

# OpenAI API key
openai_api_key = api_key

# Transform chat history for LangChain format
def transform_history_for_langchain(history):
    return [(chat[0], chat[1]) for chat in history if chat[0]]

# Transform chat history for OpenAI format
def transform_history_for_openai(history):
    new_history = []
    for chat in history:
        if chat[0]:
            new_history.append({"role": "user", "content": chat[0]})
        if chat[1]:
            new_history.append({"role": "assistant", "content": chat[1]})
    return new_history

# Load and process documents function
def load_and_process_documents(folder_path):
    documents = []
    for file in os.listdir(folder_path):
        file_path = os.path.join(folder_path, file)
        if file.endswith(".pdf"):
            loader = PyPDFLoader(file_path)
            documents.extend(loader.load())
        elif file.endswith('.docx') or file.endswith('.doc'):
            loader = Docx2txtLoader(file_path)
            documents.extend(loader.load())
        elif file.endswith('.txt'):
            loader = TextLoader(file_path)
            documents.extend(loader.load())

    text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=10)
    documents = text_splitter.split_documents(documents)

    vectordb = Chroma.from_documents(
        documents,
        embedding=OpenAIEmbeddings(),
        persist_directory="./tmp"
    )
    return vectordb

# Initialize vector database as a global variable
if 'vectordb' not in globals():
    vectordb = load_and_process_documents("./")

# Define query handling function for RAG
def handle_query(user_message, temperature, chat_history):
    try:
        if not user_message:
            return chat_history  # Return unchanged chat history

        # Use LangChain's ConversationalRetrievalChain to handle the query
        preface = """
Instruction: Answer in Traditional Chinese, within 200 characters.這是AI論壇,只回答AI相關問題
If the question is unrelated to the documents, respond with: 此事無可奉告,話說這件事須請教海虔王...
"""
        query = f"{preface} Query content: {user_message}"

        # Extract previous answers as context, converting them to LangChain format
        previous_answers = transform_history_for_langchain(chat_history)

        pdf_qa = ConversationalRetrievalChain.from_llm(
            ChatOpenAI(temperature=temperature, model_name='gpt-4'),
            retriever=vectordb.as_retriever(search_kwargs={'k': 6}),
            return_source_documents=True,
            verbose=False
        )

        # Invoke the model to handle the query
        result = pdf_qa.invoke({"question": query, "chat_history": previous_answers})

        # Ensure 'answer' is present in the result
        if "answer" not in result:
            return chat_history + [("System", "Sorry, an error occurred.")]

        # Update the AI response in chat history
        chat_history[-1] = (user_message, result["answer"])  # Update the last record, pairing user input with AI response

        return chat_history

    except Exception as e:
        return chat_history + [("System", f"An error occurred: {str(e)}")]

# Create a custom chat interface using Gradio Blocks API
with gr.Blocks() as demo:
    gr.Markdown("<h1 style='text-align: center;'>AI Assistant for AI Forum</h1>")

    chatbot = gr.Chatbot()
    state = gr.State([])

    with gr.Row():
        with gr.Column(scale=0.85):
            txt = gr.Textbox(show_label=False, placeholder="Please enter your question...")
        with gr.Column(scale=0.15, min_width=0):
            submit_btn = gr.Button("Ask")

    # Immediately show user input without response part, and clear input box
    def user_input(user_message, history):
        history.append((user_message, ""))  # Show user message, response part as empty string
        return history, "", history  # Return cleared input box and updated chat history

    # Handle AI response, update response part
    def bot_response(history):
        user_message = history[-1][0]  # Get the latest user input
        history = handle_query(user_message, 0.7, history)  # Call the query handler
        return history, history  # Return updated chat history

    # First show user message, then handle AI response, clear input box
    submit_btn.click(user_input, [txt, state], [chatbot, txt, state], queue=False).then(
        bot_response, state, [chatbot, state]
    )

    # Support pressing "Enter" to submit question, immediately show user input, clear input box
    txt.submit(user_input, [txt, state], [chatbot, txt, state], queue=False).then(
        bot_response, state, [chatbot, state]
    )

# Launch Gradio app
demo.launch()