import streamlit as st
from groq import Groq
from PIL import Image
import os
from dotenv import load_dotenv
import base64
import io

# Load environment variables
load_dotenv()
api_key = os.getenv("GROQ_API_KEY")
client = Groq(api_key=api_key)

# Streamlit page configuration
st.set_page_config(
    page_title="Llama OCR",
    page_icon="🦙",
    layout="wide",
    initial_sidebar_state="expanded"
)

def main_content():
    st.title("🦙 Llama OCR")
    st.markdown('<p style="margin-top: -20px;">Extract structured text from images using Llama 3.2 Vision!</p>', unsafe_allow_html=True)
    st.markdown("---")

    col1, col2 = st.columns([6, 1])
    with col2:
        if st.button("Clear 🗑️"):
            if 'ocr_result' in st.session_state:
                del st.session_state['ocr_result']
            st.rerun()

    if 'ocr_result' in st.session_state:
        st.markdown("### 🎯 **Extracted Text**")
        st.markdown(st.session_state['ocr_result'], unsafe_allow_html=True)

def sidebar_content():
    with st.sidebar:
        st.header("📥 Upload Image")
        
        if 'ocr_result' not in st.session_state:
            st.write("### Please upload an image to extract text.")
        
        uploaded_file = st.file_uploader("Choose an image...", type=['png', 'jpg', 'jpeg'])

        if uploaded_file:
            display_uploaded_image(uploaded_file)

        if uploaded_file and st.button("Extract Text 🔍") and 'ocr_result' not in st.session_state:
            with st.spinner("Processing image... Please wait."):
                process_image(uploaded_file)

        if not uploaded_file and 'ocr_result' not in st.session_state:
            st.sidebar.empty()

def display_uploaded_image(uploaded_file):
    image = Image.open(uploaded_file)
    st.image(image, caption="Uploaded Image", use_container_width=True)

def encode_image(uploaded_file):
    image = Image.open(uploaded_file)
    buffered = io.BytesIO()
    image.save(buffered, format=image.format)
    img_byte_array = buffered.getvalue()
    return base64.b64encode(img_byte_array).decode('utf-8'), image.format

def process_image(uploaded_file):
    if uploaded_file:
        base64_image, image_format = encode_image(uploaded_file)
        mime_type = f"image/{image_format.lower()}"
        base64_url = f"data:{mime_type};base64,{base64_image}"

        with st.spinner("Generating response... This may take a moment."):
            try:
                response = client.chat.completions.create(
                    model="llama-3.2-11b-vision-preview",
                    messages=[
                        {
                            "role": "user",
                            "content": [
                                {"type": "text", "text": "Analyze the text in the provided image. Extract all readable content and present it in a structured Markdown format. Use headings, lists, or code blocks as appropriate for clarity and organization."},
                                {"type": "image_url", "image_url": {"url": base64_url}},
                            ]
                        }
                    ],
                    temperature=0.2,
                    max_tokens=200,
                    top_p=0.5,
                    stream=False
                )
                message_content = response.choices[0].message.content
                st.session_state['ocr_result'] = message_content
            except Exception as e:
                st.error(f"Error during text extraction: {e}")

# Corrected execution order: process sidebar first, then main content
if __name__ == "__main__":
    sidebar_content()
    main_content()