import gradio as gr
import pandas as pd
import numpy as np
import folium
import sys
import os

# Add utils to path
sys.path.insert(0, os.path.join(os.path.dirname(os.path.abspath(__file__)), 'utils'))
from clean_text import clean_text
from semantic_similarity import Encoder
from main import get_recommendations

print("Loading restaurant data...")
data = pd.read_csv("data/toy_data_aggregated_embeddings.csv")
print(f"Loaded {len(data)} restaurants")

# Initialize semantic encoder
print("Loading semantic encoder model...")
try:
    encoder = Encoder()
    print("Semantic encoder loaded")
except Exception as e:
    print(f"Warning: Could not load semantic encoder: {e}")
    print("Falling back to keyword-only search")

def create_paris_map(results_df):
    """Create interactive map of Paris restaurants"""
    paris_center = [48.8566, 2.3522]
    m = folium.Map(location=paris_center, zoom_start=12, tiles='OpenStreetMap')
    
    for idx, row in results_df.iterrows():
        lat_offset = np.random.uniform(-0.05, 0.05)
        lng_offset = np.random.uniform(-0.07, 0.07)
        coords = [48.8566 + lat_offset, 2.3522 + lng_offset]
        
        rating = float(row.get('overall_rating', 0))
        color = 'green' if rating >= 4.5 else 'blue' if rating >= 4.0 else 'orange' if rating >= 3.5 else 'red'
        
        popup_html = f"""
        <div style="width:250px">
            <h4><b>{row['name']}</b></h4>
            <p>Rating: {row.get('overall_rating', 'N/A')}</p>
            <p>Reviews: {row.get('review_count', 'N/A')}</p>
            <p>Popularity Score: {row.get('pop_score', 'N/A'):.2f}</p>
        </div>
        """
        
        folium.Marker(
            location=coords,
            popup=folium.Popup(popup_html, max_width=300),
            icon=folium.Icon(color=color, icon='cutlery', prefix='fa')
        ).add_to(m)
    
    return m._repr_html_()


def search_restaurants(query_input, data_source, num_results):
    n_candidates = 2000
    query_clean = clean_text(query_input)
    restaurant_ids = get_recommendations(query_clean, n_candidates, num_results, data_source)
    
    # Subset data for recommendedations
    results = data[data["id"].isin(restaurant_ids)]
    map_html = create_paris_map(results)

    output = f"Found {len(results)} restaurants for '{query_input}'\n"
    output += f"Data Source: {data_source}\n"

    for idx, (_, row) in enumerate(results.iterrows(), 1):
        name = row.get('name', 'Unknown')
        rating = row.get('overall_rating', 'N/A')
        reviews = row.get('review_count', 'N/A')
        
        output += f"{idx}. **{name}**\n"
        output += f"   Rating: {rating} | Reviews: {reviews}\n"
        output += "\n"
        
        if 'address' in row and pd.notna(row['address']):
            addr = str(row['address'])[:100]
            output += f"   Address: {addr}\n"
        
        output += "\n"

    return output, map_html

# Create Gradio interface
with gr.Blocks(
    title="Restaurant Finder", 
    # theme=gr.themes.Soft()
    ) as app:
    gr.Markdown("""
    # Advanced Restaurant Recommendation System
    ### Search Through 5,000+ Restaurants with AI-Powered Semantic Search
    
    Find restaurants using semantic understanding and popularity ranking!
    """)
    
    with gr.Row():
        with gr.Column(scale=3):
            query_input = gr.Textbox(
                label="Search Query",
                placeholder="e.g., Italian pasta, best sushi, romantic dinner, family-friendly pizza",
                lines=2
            )
        
        with gr.Column(scale=2):
            data_source = gr.Dropdown(
                choices=["Michelin Guide", "Google", "Yelp"],
                value="Yelp",
                multiselect=True,
                label="Data Source",
                info="Select restaurant data source"
            )
    
    with gr.Row():
        
        with gr.Column(scale=1):
            num_results = gr.Slider(
                minimum=5,
                maximum=30,
                value=10,
                step=5,
                label="Results"
            )
    
    search_btn = gr.Button("Search Restaurants", variant="primary", size="lg")
    
    with gr.Row():
        with gr.Column(scale=1):
            results_output = gr.Textbox(
                label="Restaurant Results",
                lines=20,
                max_lines=30
            )
        
        with gr.Column(scale=1):
            map_output = gr.HTML(
                label="Paris Map"
            )
    
    gr.Markdown("### Try These Examples:")
    
    examples = [
        ["Italian pasta", "Yelp", 10],
        ["sushi", "Michelin Guide", 10],
        ["romantic dinner", "Google", 8],
        ["family-friendly pizza", "Yelp", 10],
        ["best seafood", "Michelin Guide", 10],
        ["cheap burger", "Google", 10]
    ]
    
    gr.Examples(
        examples=examples,
        inputs=[query_input, data_source, num_results]
    )
    
    search_btn.click(
        fn=search_restaurants,
        inputs=[query_input, data_source, num_results],
        outputs=[results_output, map_output]
    )
    
    query_input.submit(
        fn=search_restaurants,
        inputs=[query_input, data_source, num_results],
        outputs=[results_output, map_output]
    )

if __name__ == "__main__":
    print("\nStarting Advanced Restaurant Finder...")
    print(f"{len(data)} restaurants ready to search")
    print("Opening at http://127.0.0.1:7860\n")
    
    # # if run locally
    # app.launch(share=False, server_name="127.0.0.1", server_port=7860, inbrowser=True)

    # if run on HF Space
    app.launch(ssr_mode=False)