import gradio as gr
import os
import time
from langchain.document_loaders import OnlinePDFLoader
from langchain.text_splitter import CharacterTextSplitter
from langchain.llms import OpenAI
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import Chroma
from langchain.chains import ConversationalRetrievalChain

def loading_pdf():
    return "加载中...⏳"

def pdf_changes(pdf_doc, openai_api_key, chunk_size, chunk_overlap, temperature, return_source):
    if not openai_api_key:
        return "你忘记了OpenAI API密钥🗝️"
    os.environ['OPENAI_API_KEY'] = openai_api_key
    loader = OnlinePDFLoader(pdf_doc.name)
    documents = loader.load()
    text_splitter = CharacterTextSplitter(chunk_size=chunk_size, chunk_overlap=chunk_overlap)
    texts = text_splitter.split_documents(documents)
    embeddings = OpenAIEmbeddings()
    db = Chroma.from_documents(texts, embeddings)
    retriever = db.as_retriever()
    global qa 
    qa = ConversationalRetrievalChain.from_llm(
        llm=OpenAI(temperature=temperature,model="text-davinci-003",max_tokens=1000),
        retriever=retriever, 
        return_source_documents=return_source)
    return "准备就绪🚀"

def add_text(history, text):
    history = history + [(text, None)]
    return history, ""

def bot(history):
    response = infer(history[-1][0], history)
    history[-1][1] = ""
    for character in response:     
        history[-1][1] += character
        time.sleep(0.05)
        yield history
    
def infer(question, history):
    res = []
    for human, ai in history[:-1]:
        pair = (human, ai)
        res.append(pair)
    chat_history = res
    query = question
    result = qa({"question": query, "chat_history": chat_history})
    return result["answer"]

css="""
#col-container {max-width: 700px; margin-left: auto; margin-right: auto; background-color: #f0f0f0;}
"""

title = """
<div style="text-align: center;max-width: 700px;">
    <h1 style="color: #3399FF; font-family: 'Courier New', Courier, monospace;">Chat PDF[text-davinci-003]📚</h1>
    <p style="text-align: center;color: #666666; font-family: 'Courier New', Courier, monospace;">上传你的PDF，并将其加载到向量库中，<br />
    当一切准备就绪，你就可以开始提出关于pdf的问题了 🧐 <br />
    此版本使用text-davinci-003作为LLM</p>
</div>
"""

with gr.Blocks(css=css) as demo:
    with gr.Column(elem_id="col-container"):
        gr.HTML(title)
        with gr.Column():
            openai_api_key = gr.Textbox(label="你的OpenAI API密钥🔐", type="password")
            pdf_doc = gr.File(label="加载一个pdf📄", file_types=['.pdf'], type="file")
            with gr.Row():
                langchain_status = gr.Textbox(label="状态📊", placeholder="", interactive=False)
                chunk_size_slider = gr.Slider(minimum=100, maximum=2000, value=300, step=50, label='块大小📏')
                chunk_overlap_slider = gr.Slider(minimum=0, maximum=1000, value=50, step=10, label='块重叠🔀')
                temperature_slider = gr.Slider(minimum=0, maximum=1.0, value=0.5, step=0.05, label='温度🌡️')
                return_source_checkbox = gr.Checkbox(label='返回源文件📑', default=False)
                load_pdf = gr.Button("加载PDF到LangChain🔄")
        chatbot = gr.Chatbot([], elem_id="chatbot").style(height=350)
        question = gr.Textbox(label="问题❓", placeholder="输入你的问题并按回车 ")
        submit_btn = gr.Button("发送消息📨")
    load_pdf.click(loading_pdf, None, langchain_status, queue=False)    
    load_pdf.click(pdf_changes, inputs=[pdf_doc, openai_api_key, chunk_size_slider, chunk_overlap_slider, temperature_slider, return_source_checkbox], outputs=[langchain_status], queue=False)
    question.submit(add_text, [chatbot, question], [chatbot, question]).then(
    bot, chatbot, chatbot
    )
    submit_btn.click(add_text, [chatbot, question], [chatbot, question]).then(
        bot, chatbot, chatbot)

demo.launch()