app.py

import gradio as gr
import os
import fitz
from dotenv import load_dotenv
from langchain_community.document_loaders import UnstructuredPDFLoader
from langchain_community.vectorstores import FAISS
from langchain_community.embeddings import HuggingFaceEmbeddings
from langchain_text_splitters import CharacterTextSplitter
from langchain_groq import ChatGroq
from langchain.memory import ConversationBufferMemory
from langchain.chains import ConversationalRetrievalChain
from gtts import gTTS
import sys

try:
    import pytesseract
    from pdf2image import convert_from_path
except ImportError as e:
    print(f"Error: {e}. Please make sure all system dependencies are installed.")
    sys.exit(1)

# Rest of your imports...

# Set the Tesseract path
pytesseract.pytesseract.tesseract_cmd = '/usr/bin/tesseract'

# Test Tesseract installation
try:
    pytesseract.get_languages()
except pytesseract.TesseractNotFoundError:
    print("Error: Tesseract is not installed or not in the system PATH.")
    sys.exit(1)
    
# Load environment variables
load_dotenv()
secret_key = os.getenv("GROQ_API_KEY")

os.environ["GROQ_API_KEY"] = secret_key
embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/paraphrase-multilingual-mpnet-base-v2")
# Ensure the necessary folders exist
UPLOAD_FOLDER = 'uploads/'
AUDIO_FOLDER = 'audio/'
for folder in [UPLOAD_FOLDER, AUDIO_FOLDER]:
    if not os.path.exists(folder):
        os.makedirs(folder)


def load_pdf(file_path):
    """Load and preprocess Arabic text from a PDF file."""
    pages = convert_from_path(file_path, 500)
    documents = []
    for pageNum, imgBlob in enumerate(pages):
        text = pytesseract.image_to_string(imgBlob, lang="ara")
        documents.append(text)
    return documents

def prepare_vectorstore(data):
    text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=20, separator="\n")
    texts = data
    vectorstore = FAISS.from_texts(texts, embeddings)
    vectorstore.save_local("faiss_index")
    return vectorstore

def load_vectorstore():
    vectorstore = FAISS.load_local("faiss_index", embeddings, allow_dangerous_deserialization=True)
    return vectorstore

def create_chain(vectorstore):
    llm = ChatGroq(model="gemma2-9b-it", temperature=0)
    retriever = vectorstore.as_retriever()
    memory = ConversationBufferMemory(llm=llm, output_key="answer", memory_key="chat_history", return_messages=True)
    chain = ConversationalRetrievalChain.from_llm(
        llm=llm,
        retriever=retriever,
        memory=memory,
        verbose=False,
        chain_type="map_reduce"
    )
    return chain

def process_pdf(pdf_file):
    file_path = os.path.join(UPLOAD_FOLDER, pdf_file.name)
    with open(file_path, "wb") as f:
        f.write(pdf_file.read())
    data = load_pdf(file_path)
    vectorstore = prepare_vectorstore(data)
    return "PDF processed successfully. You can now start chatting!"

def chat(user_input, history):
    vectorstore = load_vectorstore()
    chain = create_chain(vectorstore)

    prompt = f"""
    You are an expert Arabic-language assistant specialized in analyzing and responding to queries about Arabic PDF documents. Your responses should be precise, informative, and reflect the professional tone and structure expected in formal Arabic communication. Focus on extracting and presenting relevant information from the document clearly and systematically, while avoiding colloquial or informal language.

    When responding, ensure the following:
       - Your answer directly reflects the content of the document.
       - If the requested information is not available in the document, clearly state that.
       - Keep your response concise yet comprehensive, addressing the question fully.
       - Always respond in formal Arabic, without using English.

    Question: {user_input}
    Helpful Answer:"""

    response = chain({"question": prompt})
    assistant_response = response["answer"]

    # Generate audio file
    tts = gTTS(text=assistant_response, lang='ar')
    audio_file = f"response_{len(history)}.mp3"
    tts.save(os.path.join(AUDIO_FOLDER, audio_file))

    return assistant_response, audio_file

custom_css = """
body {
    font-family: 'Noto Kufi Arabic', sans-serif;
    background: linear-gradient(135deg, #799351 0%, #A67B5B 100%);
    background-size: cover;
    background-position: center;
    background-attachment: fixed;
}

.gradio-container {
    max-width: 800px !important;
    margin: auto !important;
    background: rgba(255, 255, 255, 0.9);
    border-radius: 20px;
    box-shadow: 0 8px 32px 0 rgba(31, 38, 135, 0.37);
    backdrop-filter: blur(4px);
    border: 1px solid rgba(255, 255, 255, 0.18);
    padding: 20px;
}

h1, h2, h3 {
    color: #1A4D2E;
    font-weight: bold;
    text-align: center;
}

p {
    color: #A89F91;
}

.gradio-button {
    background-color: #5F6F65 !important;
    color: #FFFFFF !important;
}

.gradio-button:hover {
    background-color: #FFFFFF !important;
    color: #5F6F65 !important;
}

.chat-message {
    border-radius: 10px;
    padding: 10px;
    margin-bottom: 10px;
}

.chat-message.user {
    background-color: #E7F0DC;
}

.chat-message.bot {
    background-color: #F7EED3;
}

.chat-message::before {
    content: '';
    display: inline-block;
    width: 24px;
    height: 24px;
    background-size: contain;
    background-repeat: no-repeat;
    margin-right: 10px;
    vertical-align: middle;
}

.chat-message.user::before {
    content: '👤';
}

.chat-message.bot::before {
    content: '🤖';
}
"""
# Gradio interface
with gr.Blocks(css=custom_css) as demo:
    gr.Markdown("# ديمو بوت للقاء مركز حضرموت للدراسات التاريخية")
    gr.Markdown("## المنعقد السبت 14 - سبتمبر 2024")
    
    with gr.Row():
        pdf_input = gr.File(label="اختر ملف PDF للدردشة")
        process_button = gr.Button("رفع وبدء الدردشة")
    
    chat_interface = gr.ChatInterface(
        chat,
        chatbot=gr.Chatbot(height=400),
        textbox=gr.Textbox(placeholder="اكتب سؤالك هنا...", container=False),
        title="الدردشة مع البوت",
        description="اسأل أي سؤال عن محتوى الملف PDF",
        theme="soft",
        examples=["ما هو موضوع الوثيقة؟", "من هم الأشخاص المذكورون؟", "ما هي التواريخ الرئيسية المذكورة؟"],
        cache_examples=True,
        retry_btn=None,
        undo_btn="مسح آخر رسالة",
        clear_btn="مسح المحادثة",
    )

    audio_output = gr.Audio(label="الرد الصوتي")
    
    process_button.click(process_pdf, inputs=[pdf_input], outputs=[chat_interface.textbox])
    chat_interface.submit(lambda x, y: y[-1][1], inputs=[chat_interface.textbox, chat_interface.chatbot], outputs=[audio_output])

demo.launch()