import streamlit as st
import numpy as np
import plotly.graph_objects as go
from sklearn.manifold import MDS
from collections import defaultdict

# Page configuration
st.set_page_config(
    page_title="Address Similarity Explorer",
    page_icon="🏠",
    layout="wide"
)

class ShingleSimilarity:
    def __init__(self, k=3):
        self.k = k
        
    def get_shingles(self, text):
        text = text.lower().strip()
        if len(text) < self.k:
            return {text}
        return {text[i:i+self.k] for i in range(len(text) - self.k + 1)}
    
    def similarity(self, text1, text2):
        shingles1 = self.get_shingles(text1)
        shingles2 = self.get_shingles(text2)
        
        intersection = len(shingles1.intersection(shingles2))
        union = len(shingles1.union(shingles2))
        
        return intersection / union if union > 0 else 0.0

def create_similarity_visualization(texts, shingle_sim):
    n = len(texts)
    similarity_matrix = np.zeros((n, n))
    
    for i in range(n):
        for j in range(n):
            similarity_matrix[i][j] = shingle_sim.similarity(texts[i], texts[j])
    
    # Create visualization
    fig = go.Figure(data=go.Heatmap(
        z=similarity_matrix,
        x=texts,
        y=texts,
        colorscale='Viridis',
        text=np.round(similarity_matrix, 4),
        texttemplate='%{text}',
        textfont={"size": 12},
    ))
    
    fig.update_layout(
        title="Address Similarity Matrix",
        height=600,
        width=800
    )
    
    return fig

def main():
    st.title("🏠 Address Similarity Analyzer")
    
    # Example addresses from the image
    example_addresses = [
        "123 North Hampton Blvd",
        "123 N Hampton Blvd",
        "123 North Hampton Boulevard",
        "123 N. Hampton Blvd",
        "123 N. Hampton Boulevard",
        "65 South Hampton Blvd"
    ]
    
    st.markdown("### Address Similarity Analysis")
    st.info("If the similarity score is over 0.4, addresses are considered to be the same location")
    
    # Initialize shingle similarity with k=3
    shingle_sim = ShingleSimilarity(k=3)
    
    # Display similarity matrix
    fig = create_similarity_visualization(example_addresses, shingle_sim)
    st.plotly_chart(fig)
    
    # Interactive comparison
    st.markdown("### Compare Two Addresses")
    col1, col2 = st.columns(2)
    
    with col1:
        addr1 = st.selectbox("Select first address:", example_addresses)
    with col2:
        addr2 = st.selectbox("Select second address:", example_addresses, index=1)
    
    if st.button("Compare Addresses"):
        similarity = shingle_sim.similarity(addr1, addr2)
        
        st.metric(
            label="Similarity Score",
            value=f"{similarity:.4f}"
        )
        
        is_same = similarity > 0.4
        status = "✅ Same Location" if is_same else "❌ Different Locations"
        color = "success" if is_same else "error"
        st.markdown(f"**Status:** :{color}[{status}]")
        
        # Show shingles comparison
        col1, col2 = st.columns(2)
        with col1:
            st.markdown(f"**Shingles for Address 1:**")
            st.write(sorted(shingle_sim.get_shingles(addr1)))
        with col2:
            st.markdown(f"**Shingles for Address 2:**")
            st.write(sorted(shingle_sim.get_shingles(addr2)))

if __name__ == "__main__":
    main()