import streamlit as st
from transformers import pipeline
from PIL import Image
from huggingface_hub import InferenceClient
import os
from gradio_client import Client

# Set page configuration
st.set_page_config(
    page_title="DelishAI - Your Culinary Assistant",
    page_icon="🍽️",
    layout="centered",
    initial_sidebar_state="expanded",
)

# Custom CSS to improve styling and responsiveness
def local_css():
    st.markdown(
        """
        <style>
        /* Main layout */
        .main {
            background-color: #f0f2f6;
        }
        /* Title styling */
        .title h1 {
            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
            text-align: center;
            color: #ff4b4b;
            font-size: 3rem;
            margin-bottom: 20px;
        }
        /* Image styling */
        .st-image img {
            border-radius: 15px;
            margin-bottom: 20px;
            max-width: 100%;
        }
        /* Sidebar styling */
        [data-testid="stSidebar"] {
            background-color: #ff4b4b;
        }
        [data-testid="stSidebar"] .css-ng1t4o {
            color: white;
        }
        [data-testid="stSidebar"] .css-1d391kg {
            color: white;
        }
        /* File uploader styling */
        .stFileUploader {
            border: 2px dashed #ff4b4b;
            border-radius: 10px;
            padding: 20px;
            text-align: center;
            color: #ff4b4b;
            background-color: #ffffff;
            font-weight: bold;
        }
        /* File uploader hover effect */
        .stFileUploader:hover {
            background-color: #ffe5e5;
        }
        /* Button styling */
        .stButton>button {
            background-color: #ff4b4b;
            color: white;
            border: none;
            padding: 0.7rem 1.5rem;
            border-radius: 5px;
            font-size: 1.1rem;
            font-weight: bold;
            margin-top: 10px;
        }
        .stButton>button:hover {
            background-color: #e04343;
            color: white;
        }
        /* Headers styling */
        h2 {
            color: #ff4b4b;
            margin-top: 30px;
            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
        }
        h3 {
            color: #ff4b4b;
            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
        }
        /* Text styling */
        .stMarkdown p {
            font-size: 1.1rem;
        }
        /* Footer styling */
        footer {
            visibility: hidden;
        }
        /* Hide sidebar on small screens */
        @media only screen and (max-width: 600px) {
            [data-testid="stSidebar"] {
                display: none;
            }
            .main .block-container {
                padding-left: 1rem;
                padding-right: 1rem;
            }
            .title h1 {
                font-size: 2rem;
            }
            .stButton>button {
                width: 100%;
            }
        }
        /* Sample images grid */
        .sample-images {
            display: flex;
            justify-content: center;
            flex-wrap: wrap;
            gap: 10px;
        }
        .sample-images img {
            width: 150px;
            height: 150px;
            object-fit: cover;
            border-radius: 10px;
            cursor: pointer;
            border: 2px solid transparent;
        }
        .sample-images img:hover {
            border: 2px solid #ff4b4b;
        }
        </style>
        """,
        unsafe_allow_html=True
    )

local_css()

# Hugging Face API key
API_KEY = st.secrets["HF_API_KEY"]

# Initialize the Hugging Face Inference Client
client = InferenceClient(api_key=API_KEY)

# Load the image classification pipeline
@st.cache_resource
def load_image_classification_pipeline():
    """
    Load the image classification pipeline using a pretrained model.
    """
    return pipeline("image-classification", model="Shresthadev403/food-image-classification")

pipe_classification = load_image_classification_pipeline()

# Function to generate ingredients using Hugging Face Inference Client
def get_ingredients_qwen(food_name):
    """
    Generate a list of ingredients for the given food item using Qwen NLP model.
    Returns a clean, comma-separated list of ingredients.
    """
    messages = [
        {
            "role": "user",
            "content": f"List only the main ingredients for {food_name}. "
                       f"Respond in a concise, comma-separated list without any extra text or explanations."
        }
    ]
    try:
        completion = client.chat.completions.create(
            model="Qwen/Qwen2.5-Coder-32B-Instruct",
            messages=messages,
            max_tokens=50
        )
        generated_text = completion.choices[0].message["content"].strip()
        return generated_text
    except Exception as e:
        return f"Error generating ingredients: {e}"

# Main content
st.markdown('<div class="title"><h1>DelishAI - Your Culinary Assistant</h1></div>', unsafe_allow_html=True)

# Add banner image
st.image("IR_IMAGE.png", use_container_width=True)

# Sidebar for model information (hidden on small screens)
with st.sidebar:
    st.title("Model Information")
    st.write("**Image Classification Model**")
    st.write("Shresthadev403/food-image-classification")
    st.write("**LLM for Ingredients**")
    st.write("Qwen/Qwen2.5-Coder-32B-Instruct")
    st.markdown("---")
    st.markdown("<p style='text-align: center;'>Developed by Muhammad Hassan Butt.</p>", unsafe_allow_html=True)

# Sample images
st.subheader("Or try one of these sample images:")
sample_images = {
    "Burger": "sample_images/burger.jpg",
    "Pizza": "sample_images/pizza.jpg",
    "Sushi": "sample_images/sushi.jpg",
    "Salad": "sample_images/salad.jpg"
}

cols = st.columns(len(sample_images))
for idx, (name, file_path) in enumerate(sample_images.items()):
    with cols[idx]:
        if st.button(f"{name}", key=name):
            uploaded_file = file_path

# File uploader
st.subheader("Upload a food image:")
uploaded_file = st.file_uploader("", type=["jpg", "png", "jpeg"])

if 'uploaded_file' in locals() and uploaded_file is not None:
    # Display the uploaded image
    if isinstance(uploaded_file, str):
        # Sample image selected
        image = Image.open(uploaded_file)
    else:
        # User uploaded image
        image = Image.open(uploaded_file)
    st.image(image, caption="Uploaded Image", use_container_width=True)

    # Classification button
    if st.button("Classify"):
        with st.spinner("Classifying..."):
            # Make predictions
            predictions = pipe_classification(image)

            # Display only the top prediction
            top_food = predictions[0]['label']
            st.header(f"🍽️ Food: {top_food}")

            # Generate and display ingredients for the top prediction
            st.subheader("📝 Ingredients")
            try:
                ingredients = get_ingredients_qwen(top_food)
                st.write(ingredients)
            except Exception as e:
                st.error(f"Error generating ingredients: {e}")

            st.subheader("💡 Healthier Alternatives")
            try:
                client_gradio = Client("https://8a56cb969da1f9d721.gradio.live/")
                result = client_gradio.predict(
                    query=f"What's a healthy {top_food} recipe, and why is it healthy?",
                    api_name="/get_response"
                )
                st.write(result)
            except Exception as e:
                st.error(f"Unable to contact RAG: {e}")
else:
    st.info("Please select or upload an image to get started.")