app.py

'''Copyright 2024 Ashok Kumar

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.'''

import os
import requests
import json
import pandas as pd
import geopandas as gpd
import tzlocal
import pytz
from PIL import Image
import contextily as ctx
from datetime import datetime
from geopy.exc import GeocoderTimedOut
from geopy.geocoders import Nominatim
import folium
from folium import plugins
import streamlit as st
import streamlit_folium as st_folium
from data import flight_data
from huggingface_hub import InferenceClient
import branca.colormap as cm
from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_similarity
from difflib import get_close_matches
import warnings
warnings.filterwarnings('ignore')
import time

# Cache the airport data to avoid reloading it every time
@st.cache_data(ttl=3600)  # Cache for 1 hour
def load_airport_data():
    data_url = "https://raw.githubusercontent.com/ashok2216-A/ashok_airport-data/main/data/airports.dat"
    column_names = ["Airport ID", "Name", "City", "Country", "IATA/FAA", "ICAO", "Latitude", "Longitude",
                    "Altitude", "Timezone", "DST", "Tz database time zone", "Type", "Source"]
    return pd.read_csv(data_url, header=None, names=column_names)

# Cache geocoding results
@st.cache_data(ttl=3600)
def get_location(country):
    geolocator = Nominatim(user_agent="flight_tracker")
    return geolocator.geocode(country)

# Cache flight data fetching
@st.cache_data(ttl=60)  # Cache for 1 minute
def fetch_flight_data(lat_min, lat_max, lon_min, lon_max):
    try:
        # OpenSky Network API endpoint
        url = "https://opensky-network.org/api/states/all"
        
        # Parameters for the request
        params = {
            'lamin': lat_min,
            'lamax': lat_max,
            'lomin': lon_min,
            'lomax': lon_max
        }
        
        # Make the request with a timeout
        response = requests.get(url, params=params, timeout=10)
        
        # Check if the request was successful
        response.raise_for_status()
        
        # Parse the JSON response
        data = response.json()
        
        # Check if we got valid data
        if not data or 'states' not in data:
            st.warning("No flight data available for the selected area.")
            return {'states': [], 'time': 0}
            
        return data
        
    except requests.exceptions.RequestException as e:
        st.error(f"Error fetching flight data: {str(e)}")
        return {'states': [], 'time': 0}
    except json.JSONDecodeError as e:
        st.error(f"Error parsing flight data: {str(e)}")
        return {'states': [], 'time': 0}
    except Exception as e:
        st.error(f"Unexpected error: {str(e)}")
        return {'states': [], 'time': 0}

# Hugging Face model configuration
HF_API_URL = "https://api-inference.huggingface.co/models/google/flan-t5-large"
HF_TOKEN = os.getenv("HF_TOKEN")
headers = {"Authorization": f"Bearer {HF_TOKEN}"}

def query_llm(prompt):
    try:
        payload = {
            "inputs": prompt,
            "parameters": {
                "max_length": 250,
                "temperature": 0.1,
                "top_p": 0.95,
                "do_sample": False
            }
        }
        
        response = requests.post(HF_API_URL, headers=headers, json=payload)
        response.raise_for_status()
        return response.json()[0]['generated_text']
    except requests.exceptions.HTTPError as e:
        if e.response.status_code == 403:
            st.warning("Language model access is currently restricted. Using direct flight data display instead.")
        else:
            st.error(f"Error querying language model: {str(e)}")
        return None
    except Exception as e:
        st.error(f"Error querying language model: {str(e)}")
        return None

def create_flight_embeddings(geo_df):
    """Create embeddings for flight data to enable semantic search"""
    try:
        model = SentenceTransformer('all-MiniLM-L6-v2')
        
        # Create text representations of flight data
        flight_texts = []
        for _, row in geo_df.iterrows():
            text = f"Flight {row['callsign']} from {row['origin_country']} "
            text += f"at altitude {row['baro_altitude']}m, speed {row['velocity']}m/s, "
            text += f"heading {row['true_track']}°"
            flight_texts.append(text)
        
        # Generate embeddings
        embeddings = model.encode(flight_texts)
        return embeddings, flight_texts
    except Exception as e:
        st.warning(f"Could not create embeddings: {str(e)}")
        return None, None

def find_similar_flights(identifier, geo_df, embeddings, flight_texts, threshold=0.7):
    """Find similar flights using semantic search"""
    try:
        model = SentenceTransformer('all-MiniLM-L6-v2')
        
        # Create query embedding
        query_embedding = model.encode([identifier])
        
        # Calculate similarities
        similarities = cosine_similarity(query_embedding, embeddings)[0]
        
        # Find similar flights
        similar_indices = [i for i, sim in enumerate(similarities) if sim > threshold]
        if similar_indices:
            return geo_df.iloc[similar_indices]
        return None
    except Exception as e:
        st.warning(f"Error in semantic search: {str(e)}")
        return None

def query_flight_data(geo_df, question):
    # Preprocess the question to extract key information
    question = question.lower().strip()
    
    # Common flight information queries and their corresponding columns
    query_mappings = {
        'callsign': ['callsign'],
        'altitude': ['baro_altitude', 'geo_altitude'],
        'speed': ['velocity'],
        'direction': ['true_track'],
        'country': ['origin_country'],
        'squawk': ['squawk'],
        'icao': ['icao24'],
        'vertical': ['vertical_rate'],
        'ground': ['on_ground'],
        'position': ['latitude', 'longitude'],
        'time': ['time_position', 'last_contact']
    }
    
    # Extract the identifier (usually callsign or icao) from the question
    identifier = None
    if 'for' in question:
        identifier = question.split('for')[-1].strip()
    elif 'of' in question:
        identifier = question.split('of')[-1].strip()
    elif 'about' in question:
        identifier = question.split('about')[-1].strip()
    
    if not identifier:
        return "Please specify a flight identifier (callsign or ICAO code) in your question."
    
    # Clean and normalize the identifier
    identifier = identifier.strip().upper()
    
    # Try to find the flight by callsign or icao (case-insensitive)
    flight_data = None
    
    # First try exact match
    if identifier in geo_df['callsign'].str.upper().values:
        flight_data = geo_df[geo_df['callsign'].str.upper() == identifier]
    elif identifier in geo_df['icao24'].str.upper().values:
        flight_data = geo_df[geo_df['icao24'].str.upper() == identifier]
    
    # If no exact match, try partial match
    if flight_data is None or flight_data.empty:
        # Try matching without spaces or special characters
        clean_identifier = ''.join(filter(str.isalnum, identifier))
        if not geo_df['callsign'].empty:
            clean_callsigns = geo_df['callsign'].fillna('').apply(lambda x: ''.join(filter(str.isalnum, str(x).upper())))
            matches = clean_callsigns == clean_identifier
            if matches.any():
                flight_data = geo_df[matches]
    
    # If still no match, try fuzzy matching
    if flight_data is None or flight_data.empty:
        try:
            all_callsigns = geo_df['callsign'].fillna('').str.upper().unique()
            close_matches = get_close_matches(identifier, all_callsigns, n=1, cutoff=0.8)
            if close_matches:
                flight_data = geo_df[geo_df['callsign'].str.upper() == close_matches[0]]
        except:
            pass
    
    # If still no match, try semantic search using RAG
    if flight_data is None or flight_data.empty:
        try:
            # Create embeddings for all flights
            embeddings, flight_texts = create_flight_embeddings(geo_df)
            if embeddings is not None:
                # Try to find similar flights
                similar_flights = find_similar_flights(identifier, geo_df, embeddings, flight_texts)
                if similar_flights is not None and not similar_flights.empty:
                    flight_data = similar_flights
                    st.info(f"Found similar flight(s) to {identifier}")
        except Exception as e:
            st.warning(f"Semantic search failed: {str(e)}")
    
    if flight_data is None or flight_data.empty:
        # If still no match, show available flights
        available_flights = geo_df['callsign'].dropna().unique()
        if len(available_flights) > 0:
            return f"Could not find flight {identifier}. Available flights: {', '.join(available_flights[:10])}..."
        return f"Could not find flight {identifier}. No flights currently available in the selected area."
    
    # Prepare flight data for display
    flight_info = {}
    for col in flight_data.columns:
        if col in flight_data.columns:
            value = flight_data[col].iloc[0]
            if pd.notna(value):
                if col == 'baro_altitude' or col == 'geo_altitude':
                    flight_info[col] = f"{value} meters"
                elif col == 'velocity':
                    flight_info[col] = f"{value} m/s"
                elif col == 'true_track':
                    flight_info[col] = f"{value} degrees"
                elif col == 'vertical_rate':
                    flight_info[col] = f"{value} m/s"
                elif col == 'latitude':
                    flight_info[col] = f"{value}° N"
                elif col == 'longitude':
                    flight_info[col] = f"{value}° E"
    else:
                    flight_info[col] = str(value)
    
    if not flight_info:
        return f"No information available for flight {identifier}."
    
    # Try to get LLM response, but fall back to direct display if it fails
    try:
        # Create a prompt for the LLM
        prompt = f"""Answer this question about flight {identifier}: {question}

Available flight data:
{json.dumps(flight_info, indent=2)}

Provide a clear and concise answer focusing on the specific information requested."""
        
        llm_response = query_llm(prompt)
        if llm_response:
            return llm_response
    except:
        pass
    
    # Fallback to direct data display
    response = f"Flight Information for {identifier}:\n"
    for key, value in flight_info.items():
        response += f"- {key.replace('_', ' ').title()}: {value}\n"
    return response

@st.cache_data(ttl=60)  # Cache for 1 minute
def get_traffic_gdf(lat_min, lat_max, lon_min, lon_max, local_time_zone, _loc, flight_info):
    # Get cached flight data
    json_dict = fetch_flight_data(lat_min, lat_max, lon_min, lon_max)
    
    if not json_dict or not json_dict.get('states'):
        st.warning("No flight data available for the selected area.")
        return None

    try:
        # Define columns for the DataFrame
        columns = ["icao24","callsign","origin_country","time_position","last_contact","longitude","latitude",
                "baro_altitude","on_ground","velocity","true_track","vertical_rate","sensors","geo_altitude",
                "squawk","spi","position_source"]
        
        unix_timestamp = int(json_dict["time"])
        local_timezone = pytz.timezone(local_time_zone)
        local_time = datetime.fromtimestamp(unix_timestamp, local_timezone).strftime('%Y-%m-%d %H:%M:%S')
        
        # Optimize DataFrame creation
        state_df = pd.DataFrame(json_dict["states"], columns=columns)
        state_df['time'] = local_time
        
        # Create GeoDataFrame more efficiently
        gdf = gpd.GeoDataFrame(
            state_df,
            geometry=gpd.points_from_xy(state_df.longitude, state_df.latitude),
            crs="EPSG:4326"
        )
        
        # Display information
        st.title("Live Flight Tracker")
        st.subheader('Flight Details', divider='rainbow')
        st.write('Location: {0}'.format(_loc))
        st.write('Current Local Time: {0}-{1}:'.format(local_time, local_time_zone))
        st.write("Minimum_latitude is {0} and Maximum_latitude is {1}".format(lat_min, lat_max))
        st.write("Minimum_longitude is {0} and Maximum_longitude is {1}".format(lon_min, lon_max))
        st.write('Number of Visible Flights: {}'.format(len(json_dict['states'])))
        st.write('Plotting the flight: {}'.format(flight_info))
        st.subheader('Map Visualization', divider='rainbow')
        st.write('****Click ":orange[Update Map]" Button to Refresh the Map****')
        return gdf
        
    except Exception as e:
        st.error(f"Error processing flight data: {str(e)}")
        return None

def flight_tracking(flight_view_level, country, local_time_zone, flight_info, airport, color):
    # Get cached location data
    loc = get_location(country)
    if not loc:
        st.error("Could not find location. Please try a different country name.")
        return
    
    loc_box = loc[1]
    extend_left =+12*flight_view_level
    extend_right =+10*flight_view_level
    extend_top =+10*flight_view_level
    extend_bottom =+ 18*flight_view_level
    lat_min, lat_max = (loc_box[0] - extend_left), loc_box[0]+extend_right
    lon_min, lon_max = (loc_box[1] - extend_bottom), loc_box[1]+extend_top
    
    columns = ["icao24","callsign","origin_country","time_position","last_contact","longitude","latitude",
            "baro_altitude","on_ground","velocity","true_track","vertical_rate","sensors","geo_altitude",
            "squawk","spi","position_source",]
    
    # Get cached airport data
    airport_df = load_airport_data()
    airport_locations = airport_df[["Name", "City", "Country", "IATA/FAA", "Latitude", "Longitude"]]
    airport_country_loc = airport_locations[airport_locations['Country'] == str(loc)]
    airport_country_loc = airport_country_loc[(airport_country_loc['Country'] == str(loc)) & 
                                            (airport_country_loc['Latitude'] >= lat_min) &
                                            (airport_country_loc['Latitude'] <= lat_max) & 
                                            (airport_country_loc['Longitude'] >= lon_min) &
                            (airport_country_loc['Longitude'] <= lon_max)]
    
    # Get traffic data
    geo_df = get_traffic_gdf(lat_min, lat_max, lon_min, lon_max, local_time_zone, loc, flight_info)
    if geo_df is None:
        return
    
    # Create a base map
    m = folium.Map(
        location=[loc_box[0], loc_box[1]],
        zoom_start=6,
        tiles='CartoDB dark_matter',
        show_grid=False
    )
    
    # Create colormap
    if color == "rainbow":
        colormap = cm.LinearColormap(
            colors=['red', 'yellow', 'green', 'blue', 'purple'],
            vmin=geo_df[flight_info].min(),
            vmax=geo_df[flight_info].max()
        )
    elif color == "ice":
        colormap = cm.LinearColormap(
            colors=['white', 'lightblue', 'blue'],
            vmin=geo_df[flight_info].min(),
            vmax=geo_df[flight_info].max()
        )
    else:  # hot
        colormap = cm.LinearColormap(
            colors=['yellow', 'orange', 'red'],
            vmin=geo_df[flight_info].min(),
            vmax=geo_df[flight_info].max()
        )
    
    # Pre-compute icon HTML template
    icon_template = """
    <div style="transform: rotate({rotation_angle}deg);">
        <svg width="24" height="24" viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg">
            <path d="M21 16v-2l-8-5V3.5c0-.83-.67-1.5-1.5-1.5S10 2.67 10 3.5V9l-8 5v2l8-2.5V19l-2 1.5V22l3.5-1 3.5 1v-1.5L13 19v-5.5l8 2.5z" fill="{color_hex}"/>
        </svg>
    </div>
    """
    
    # Pre-compute tooltip template
    tooltip_template = """
    <div style="font-size: 12px; font-family: Arial, sans-serif; max-width: 300px;">
        <div style="font-weight: bold; font-size: 14px; margin-bottom: 5px; color: #2c3e50;">
            Flight: {callsign}
        </div>
        <div style="display: grid; grid-template-columns: 1fr 1fr; gap: 5px;">
            {rows}
        </div>
    </div>
    """
    
    # Add flight markers
    for idx, row in geo_df.iterrows():
        if pd.notna(row['latitude']) and pd.notna(row['longitude']):
            # Get color based on flight_info value
            value = row[flight_info] if pd.notna(row[flight_info]) else geo_df[flight_info].min()
            color_hex = colormap(value)
            
            # Create custom flight icon with rotation
            rotation_angle = row['true_track'] if pd.notna(row['true_track']) else 0
            icon_html = icon_template.format(rotation_angle=rotation_angle, color_hex=color_hex)
            
            # Create tooltip rows
            tooltip_rows = []
            for col in columns:
                val = row[col] if pd.notna(row[col]) else 'N/A'
                if col in ['baro_altitude', 'geo_altitude']:
                    val = f"{val} m"
                elif col == 'velocity':
                    val = f"{val} m/s"
                elif col == 'true_track':
                    val = f"{val}°"
                tooltip_rows.append(f'<div style="font-weight: bold;">{col}:</div><div>{val}</div>')
            
            # Create tooltip
            tooltip_html = tooltip_template.format(
                callsign=row['callsign'] if pd.notna(row['callsign']) else 'Unknown',
                rows='\n'.join(tooltip_rows)
            )
            
            # Create popup content
            popup_content = f"""
            <div style="font-size: 14px; font-family: Arial, sans-serif; max-width: 300px;">
                <div style="font-weight: bold; font-size: 16px; margin-bottom: 10px; color: #2c3e50;">
                    Flight: {row['callsign'] if pd.notna(row['callsign']) else 'Unknown'}
                </div>
                <div style="display: grid; grid-template-columns: 1fr 1fr; gap: 5px;">
                    <div style="font-weight: bold;">ICAO24:</div>
                    <div>{row['icao24'] if pd.notna(row['icao24']) else 'N/A'}</div>
                    <div style="font-weight: bold;">Origin Country:</div>
                    <div>{row['origin_country'] if pd.notna(row['origin_country']) else 'N/A'}</div>
                    <div style="font-weight: bold;">Time Position:</div>
                    <div>{row['time_position'] if pd.notna(row['time_position']) else 'N/A'}</div>
                    <div style="font-weight: bold;">Last Contact:</div>
                    <div>{row['last_contact'] if pd.notna(row['last_contact']) else 'N/A'}</div>
                    <div style="font-weight: bold;">Baro Altitude:</div>
                    <div>{row['baro_altitude'] if pd.notna(row['baro_altitude']) else 'N/A'} m</div>
                    <div style="font-weight: bold;">Geo Altitude:</div>
                    <div>{row['geo_altitude'] if pd.notna(row['geo_altitude']) else 'N/A'} m</div>
                    <div style="font-weight: bold;">Velocity:</div>
                    <div>{row['velocity'] if pd.notna(row['velocity']) else 'N/A'} m/s</div>
                    <div style="font-weight: bold;">True Track:</div>
                    <div>{row['true_track'] if pd.notna(row['true_track']) else 'N/A'}°</div>
                    <div style="font-weight: bold;">Vertical Rate:</div>
                    <div>{row['vertical_rate'] if pd.notna(row['vertical_rate']) else 'N/A'} m/s</div>
                    <div style="font-weight: bold;">Squawk:</div>
                    <div>{row['squawk'] if pd.notna(row['squawk']) else 'N/A'}</div>
                    <div style="font-weight: bold;">On Ground:</div>
                    <div>{row['on_ground'] if pd.notna(row['on_ground']) else 'N/A'}</div>
                    <div style="font-weight: bold;">SPI:</div>
                    <div>{row['spi'] if pd.notna(row['spi']) else 'N/A'}</div>
                    <div style="font-weight: bold;">Position Source:</div>
                    <div>{row['position_source'] if pd.notna(row['position_source']) else 'N/A'}</div>
                </div>
            </div>
            """
            
            # Create custom icon
            icon = folium.DivIcon(
                html=icon_html,
                icon_size=(24, 24),
                icon_anchor=(12, 12)
            )
            
            # Add marker to map
            folium.Marker(
                location=[row['latitude'], row['longitude']],
                icon=icon,
                popup=folium.Popup(popup_content, max_width=300),
                tooltip=tooltip_html
            ).add_to(m)
    
    # Add airports if selected
    if airport == 1:
        for idx, row in airport_country_loc.iterrows():
            folium.Marker(
                location=[row['Latitude'], row['Longitude']],
                icon=folium.Icon(icon='plane', prefix='fa', color='blue'),
                popup=f"<b>{row['Name']}</b><br>IATA: {row['IATA/FAA']}<br>City: {row['City']}",
                tooltip=f"Airport: {row['Name']}"
            ).add_to(m)
    
    # Add colormap to the map
    colormap.add_to(m)
    
    # Add a layer control
    folium.LayerControl().add_to(m)
    
    # Display the map in Streamlit
    st_folium.folium_static(m, width=1200, height=600)
    return None

st.set_page_config(
    layout="wide"
)
image = Image.open('logo.png')
add_selectbox = st.sidebar.image(
    image, width=150
)
add_selectbox = st.sidebar.subheader(
    "Configure Map",divider='rainbow'
)
with st.sidebar:
    Refresh = st.button('Update Map', key=1)
    on = st.toggle('View Airports')
    if on:
        air_port = 1
        st.write(':rainbow[Nice Work Buddy!]')
        st.write('Now Airports are Visible')
    else:
        air_port=0
    view = st.slider('Increase Flight Visibility',1,6,2)
    st.write("You Selected:", view)
    cou = st.text_input('Type Country Name', 'north america')
    st.write('The current Country name is', cou)
    time = st.text_input('Type Time Zone Name (Ex: America/Toronto, Europe/Berlin)', 'Asia/Kolkata')
    st.write('The current Time Zone is', time)
    info = st.selectbox(
    'Select Flight Information',
    ('baro_altitude',
        'on_ground', 'velocity',
        'geo_altitude'))
    st.write('Plotting the data of Flight:', info)
    clr = st.radio('Pick A Color for Scatter Plot',["rainbow","ice","hot"])
    if clr == "rainbow":
        st.write('The current color is', "****:rainbow[Rainbow]****")
    elif clr == 'ice':
        st.write('The current color is', "****:blue[Ice]****")
    elif clr == 'hot':
        st.write('The current color is', "****:red[Hot]****")
    else: None
# with st.spinner('Wait!, We Requesting API Data...'):
# try:
flight_tracking(flight_view_level=view, country=cou,flight_info=info,
            local_time_zone=time, airport=air_port, color=clr)
st.subheader('Ask your Questions!', divider='rainbow')
st.write("Google's TAPAS base LLM model 🤖")
geo_df = flight_data(flight_view_level = view, country= cou, flight_info=info, local_time_zone=time, airport=1)
question = st.text_input('Type your questions here', "What is the squawk code for SWR9XD?")
result = query_flight_data(geo_df, question)
st.markdown(result)
# except TypeError:
#     st.error(':red[Error: ] Please Re-run this page.', icon="🚨")
#     st.button('Re-run', type="primary")
# st.snow()



# import streamlit as st
# from huggingface_hub import InferenceClient
# import os

# hf_token = os.getenv("HF_TOKEN")
# # Set up the Hugging Face Inference Client
# client = InferenceClient(
#     provider="together",  # Replace with the correct provider if needed
#     api_key= hf_token  # Replace with your Hugging Face API key
# )

# # Streamlit app title
# st.title("🤖 Deepseek R1 Chatbot")
# st.write("Chat with the Deepseek R1 model powered by Hugging Face Inference API.")

# # Initialize session state to store chat history
# if "messages" not in st.session_state:
#     st.session_state.messages = []

# # Display chat history
# for message in st.session_state.messages:
#     with st.chat_message(message["role"]):
#         st.markdown(message["content"])

# # User input
# if prompt := st.chat_input("What would you like to ask?"):
#     # Add user message to chat history
#     st.session_state.messages.append({"role": "user", "content": prompt})
#     with st.chat_message("user"):
#         st.markdown(prompt)

#     # Generate response from Deepseek R1 model
#     with st.spinner("Thinking..."):
#         try:
#             # Prepare the messages for the model
#             messages = [{"role": m["role"], "content": m["content"]} for m in st.session_state.messages]
            
#             # Call the Hugging Face Inference API
#             completion = client.chat.completions.create(
#                 model="deepseek-ai/DeepSeek-R1",  # Replace with the correct model name
#                 messages=messages,
#                 max_tokens=500
#             )
            
#             # Extract the model's response
#             response = completion.choices[0].message.content

#             # Add model's response to chat history
#             st.session_state.messages.append({"role": "assistant", "content": response})
#             with st.chat_message("assistant"):
#                 st.markdown(response)
        
#         except Exception as e:
#             st.error(f"An error occurred: {e}")