import streamlit as st
import pandas as pd
import numpy as np
from sklearn.neighbors import KNeighborsRegressor
from geopy.distance import geodesic
import googlemaps
from geopy.exc import GeocoderTimedOut

# Function to calculate distance in meters between two coordinates
def calculate_distance(lat1, lon1, lat2, lon2):
    coords_1 = (lat1, lon1)
    coords_2 = (lat2, lon2)
    return geodesic(coords_1, coords_2).meters

# Function to apply KNN and return Vunit values
def knn_predict(df, target_column, features_columns, k=5):
    # Separate features and target variable
    X = df[features_columns]
    y = df[target_column]

    # Create KNN regressor
    knn = KNeighborsRegressor(n_neighbors=k)

    # Fit the model
    knn.fit(X, y)

    # Use the model to predict Vunit for the filtered_data
    predictions = knn.predict(df[features_columns])

    return predictions

# Set wide mode
st.set_page_config(layout="wide")

# Set dark theme
st.markdown(
    """
    <style>
        body {
            color: white;
            background-color: #1e1e1e;
        }
        .st-df-header, .st-df-body, .st-df-caption {
            color: #f8f9fa;  /* Bootstrap table header text color */
        }
        .st-eb {
            background-color: #343a40;  /* Streamlit exception box background color */
        }
    </style>
    """,
    unsafe_allow_html=True
)

# Create a DataFrame with sample data
data = pd.read_excel('ven_ter_fim_PEDÓ.xlsx')

# Initialize variables to avoid NameError
selected_coords = 'Custom'
radius_visible = True
custom_address_initial = 'Av. Senador Alberto Pasqualini, 177 - Centro, Lajeado - RS, 95900-034'  # Initial custom address
custom_lat = data['latitude'].mean()
custom_lon = data['longitude'].mean()
radius_in_meters = 1500
filtered_data = data  # Initialize with the entire dataset

# Find the maximum distance between coordinates
max_distance = 0
for index, row in data.iterrows():
    distance = calculate_distance(row['latitude'], row['longitude'], data['latitude'].mean(), data['longitude'].mean())
    if distance > max_distance:
        max_distance = distance

# Calculate a zoom level based on the radius
if radius_visible:
    radius_zoom_mapping = {
        500: 15,
        1000: 14,
        2000: 13,
        4000: 12,
        5000: 11
    }

    # Find the closest radius in the mapping
    closest_radius = min(radius_zoom_mapping.keys(), key=lambda x: abs(x - radius_in_meters))

    # Set the zoom level based on the mapping
    zoom_level = radius_zoom_mapping[closest_radius]

else:
    # Calculate a zoom level based on the maximum distance
    zoom_level = round(15 - np.log10(max_distance))

# Create a sidebar for controls
with st.sidebar:
    st.title('avalia.se')

    selected_coords = st.selectbox('Selecione Coordenadas', ['Random', 'Custom'])

    if selected_coords == 'Custom':
        custom_address = st.text_input('Informe o endereço', custom_address_initial)
        radius_visible = True  # Show radius slider for custom coordinates
        # No need to initialize max_distance_all here
    else:
        custom_address = "Lajeado, Rio Grande do Sul, Brazil"  # Default address
        radius_visible = False  # Hide radius slider for random coordinates
        max_distance_all = 0  # Initialize max_distance_all here

        max_distance_all = 0  # Initialize max_distance_all here

    # Geocode the custom address using the Google Maps API
    gmaps = googlemaps.Client(key='AIzaSyDoJ6C7NE2CHqFcaHTnhreOfgJeTk4uSH0')  # Replace with your API key

    try:
        location = gmaps.geocode(custom_address)[0]['geometry']['location']
        custom_lat, custom_lon = location['lat'], location['lng']
    except (IndexError, GeocoderTimedOut):
        st.error("Erro: Não foi possível geocodificar o endereço fornecido. Por favor, verifique e tente novamente.")

    # Slider for setting the zoom level
    if selected_coords == 'Custom':
        # The zoom level slider is no longer needed if radius is used for zoom adjustment
        st.markdown("O nível de zoom é ajustado automaticamente com base no raio.")
    else:
        for index, row in data.iterrows():
            distance_all = calculate_distance(row['latitude'], row['longitude'], data['latitude'].mean(), data['longitude'].mean())
            if distance_all > max_distance_all:
                max_distance_all = distance_all

        # Calculate a zoom level based on the maximum distance of the entire dataset
        zoom_level_all = round(15 - np.log10(max_distance_all))

        # Slider for setting the zoom level based on the entire dataset
        zoom_level = st.slider('Nível de zoom', min_value=1, max_value=15, value=zoom_level_all)

    # Conditionally render the radius slider
    if radius_visible:
        radius_in_meters = st.slider('Selecione raio (em metros)', min_value=100, max_value=5000, value=1000)

# Filter data based on the radius
if selected_coords == 'Custom':
    filtered_data = data[data.apply(lambda x: calculate_distance(x['latitude'], x['longitude'], custom_lat, custom_lon), axis=1) <= radius_in_meters]
    filtered_data = filtered_data.dropna()  # Drop rows with NaN values

# Add a custom CSS class to the map container
st.markdown(f"""<style>
.map {{
  width: 100%;
  height: 100vh;
}}
</style>""", unsafe_allow_html=True)

# Check if KNN should be applied
if selected_coords == 'Custom' and radius_visible:
    # Apply KNN and get predicted Vunit values
    predicted_vunit = knn_predict(filtered_data, 'Vunit', ['latitude', 'longitude', 'Area'])  # Update with your features
    # Add predicted Vunit values to filtered_data
    filtered_data['Predicted_Vunit'] = predicted_vunit

# Display the map and filtered_data
with st.container():
    if selected_coords == 'Custom':
        st.map(filtered_data, zoom=zoom_level, use_container_width=True)
    elif selected_coords == 'Random':
        st.map(data, zoom=zoom_level, use_container_width=True)

# Display the predicted Vunit values if applicable
if 'Predicted_Vunit' in filtered_data.columns:
    st.write("Valores (R$/m²) previstos com algoritmo KNN:")
    st.write(filtered_data[['latitude', 'longitude', 'Vunit', 'Predicted_Vunit']])