app.py

import gradio as gr
import os
import time
import requests

from langchain.document_loaders import PyPDFLoader
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 "Loading..."

def pdf_changes(pdf_doc, open_ai_key):
    if openai_key is not None:
        os.environ['OPENAI_API_KEY'] = open_ai_key
        loader = PyPDFLoader(pdf_doc.name)
        documents = loader.load()
        text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
        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=0.5), 
            retriever=retriever, 
            return_source_documents=False)
        return "✅ Ready: Upload PDF"
    else:
        return "Please input correct OpenAI API key"

def pdf_url(url, open_ai_key):
    destination = 'url.pdf'  # download
    response = requests.get(url)

    if response.status_code == 200:
        with open(destination, 'wb') as file:
            file.write(response.content)
        print(f"File downloaded to {destination}")
    else:
        print(f"Failed to download the file. Status code: {response.status_code}")

    if openai_key is not None:
        os.environ['OPENAI_API_KEY'] = open_ai_key
        loader = PyPDFLoader("url.pdf")
        documents = loader.load()
        text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
        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=0.5), 
            retriever=retriever, 
            return_source_documents=False)
        return "✅ Ready: Upload from URL"
    else:
        return "Please input correct OpenAI API key"

def pdf_example(open_ai_key):
    if openai_key is not None:
        os.environ['OPENAI_API_KEY'] = open_ai_key
        loader = PyPDFLoader("sample.pdf")
        documents = loader.load()
        text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
        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=0.5), 
            retriever=retriever, 
            return_source_documents=False)
        return "✅ Ready: Load example PDF"
    else:
        return "Please input correct OpenAI API key"


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
    #print(chat_history)
    query = question
    result = qa({"question": query, "chat_history": chat_history})
    #print(result)
    return result["answer"]

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

title = """
<div style="text-align: center;max-width: 700px;">
    <h1>language-document-extractor-QA</h1>
    <p style="text-align: left;">Instruction: <br />
    1. Input your Open API key <br />
    2. There are 3 options: <br />
    2.1 Upload PDF file and click [Upload PDF]  <br />
    2.2 Input PDF url and click [Upload from URL]  <br />
    2.3 Click [Load example PDF] to use example  <br />
    3. When status is ready, you can ask question about the pdf. <br /> 
    </p>
</div>
"""
version = """
<div style="text-align: center;max-width: 700px;">
    <p style="text-align: left;">
    version: 1.01
    </p>
</div>
"""

with gr.Blocks(css=css) as demo:
    with gr.Column(elem_id="col-container"):
        gr.HTML(title)
        
        with gr.Column():
            openai_key = gr.Textbox(label="You OpenAI API key", type="password")
            pdf_doc = gr.File(label="Load a pdf", file_types=['.pdf'], type="file")
            url = gr.Textbox(label='Enter PDF URL here', placeholder="https://huggingface.co/spaces/jingwora/language-PDF-extractor-QA/resolve/main/sample.pdf")
            with gr.Row():
                load_pdf = gr.Button("Upload PDF")
                load_url = gr.Button("Upload from URL", )
                load_example = gr.Button("Load example PDF")   

        status = gr.Textbox(label="Status", placeholder="", interactive=False)
        chatbot = gr.Chatbot([], elem_id="chatbot").style(height=350)
        question = gr.Textbox(label="Question", placeholder="Type your question and hit Enter ")
        submit_btn = gr.Button("Send Message")
        gr.HTML(version)

    load_pdf.click(loading_pdf, None, status, queue=False)    
    load_pdf.click(pdf_changes, inputs=[pdf_doc, openai_key], outputs=[status], queue=False)

    load_example.click(loading_pdf, None, status, queue=False)    
    load_example.click(pdf_example, inputs=[openai_key], outputs=[status], queue=False)
    
    load_url.click(loading_pdf, None, status, queue=False)
    load_url.click(pdf_url, inputs=[url, openai_key], outputs=[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.queue(concurrency_count=5, max_size=20).launch(debug=True)