src/pages/02_Passport_Power_Explorer.py

import streamlit as st
import pandas as pd
import numpy as np
import re
import os

# --- Page Configuration ---
st.set_page_config(
    page_title="Passport Power Explorer",
    page_icon="✈️",
    layout="wide"
)


@st.cache_data
def load_and_process_data(filepath):
    """
    Loads visa data from a CSV, cleans it, and extracts structured information.
    Caches the result for performance.
    """
    try:
        df = pd.read_csv(filepath)
    except FileNotFoundError:
        st.error(f"Error: The file was not found at '{filepath}'. Please make sure 'asean_visa_avaibility_playwright.csv' is in the project's src/data directory.")
        return pd.DataFrame()

    def extract_days(text):
        if not isinstance(text, str):
            return np.nan
        matches = re.findall(r'(\d+)\s*days', text, re.IGNORECASE)
        if matches:
            return int(matches[0])
        return np.nan

    df['duration_days'] = df['text'].apply(extract_days)

    # 2. Categorize Visa Type
    def categorize_visa(text):
        if not isinstance(text, str):
            return "Unknown"
        text_lower = text.lower()
        if 'visa-free' in text_lower:
            return "✅ Visa-Free"
        if 'visa on arrival' in text_lower:
            return "🛬 Visa on Arrival"
        if 'evisa' in text_lower or 'eta' in text_lower:
            return "📧 eVisa / eTA"
        if 'visa required' in text_lower:
            return "❌ Visa Required"
        return "📄 Other / Pre-enrollment"

    df['visa_category'] = df['text'].apply(categorize_visa)
    
    # Clean up column names for display
    df.rename(columns={
        'text': 'Requirement',
        'pass': 'Passport',
        'dest': 'Destination',
        'source': 'Source Country',
        'duration_days': 'Duration (Days)',
        'visa_category': 'Visa Category'
    }, inplace=True)
    
    return df

# --- Main Application UI ---

st.title("✈️ Passport Power & Visa Explorer")
st.markdown("Analyze inbound and outbound visa requirements for Southeast Asian countries based on the provided dataset.")

# Load the data
# Adjust the path to be relative to the script's location if needed
# Assuming the script is in src/pages and the CSV is in the root
# The path should be ../../visa_avaibility_playwright_20251015.csv
file_path = "src/data/asean_visa_avaibility_playwright.csv"
visa_df = load_and_process_data(file_path)

if not visa_df.empty:
    
    # --- Sidebar Filters ---
    st.sidebar.header("🔍 Filters")

    # 1. Select Focus Country
    focus_countries = ['Indonesia', 'Vietnam', 'Philippines', 'Thailand']
    selected_country = st.sidebar.selectbox(
        "Select your focus country:",
        options=focus_countries
    )

    # 2. Select Travel Direction (Inbound/Outbound)
    travel_direction = st.sidebar.radio(
        "Select travel direction:",
        options=['Outbound', 'Inbound'],
        horizontal=True
    )

    # --- Filtering Logic ---
    if travel_direction == 'Outbound':
        filtered_df = visa_df[visa_df['Source Country'] == selected_country].copy()
        st.header(f"🛂 Outbound Travel for {selected_country} Passport Holders")
    else: # Inbound
        filtered_df = visa_df[visa_df['Destination'] == selected_country].copy()
        st.header(f"🛬 Inbound Travel to {selected_country}")

    # --- Dynamic Filters Based on Data ---
    
    # 3. Filter by Visa Category
    visa_categories = sorted(filtered_df['Visa Category'].unique())
    selected_categories = st.sidebar.multiselect(
        "Filter by Visa Category:",
        options=visa_categories,
        default=visa_categories # Default to all selected
    )

    # 4. Filter by Duration
    min_duration, max_duration = 0, int(filtered_df['Duration (Days)'].max())
    
    duration_range = st.sidebar.slider(
        "Filter by Stay Duration (Days):",
        min_value=min_duration,
        max_value=max_duration,
        value=(min_duration, max_duration) # Default to full range
    )

    # Apply category and duration filters
    filtered_df = filtered_df[filtered_df['Visa Category'].isin(selected_categories)]

    # Handle slider filtering, including NaNs
    min_select, max_select = duration_range
    # Keep rows where duration is within range OR where duration is NaN (not specified)
    filtered_df = filtered_df[
        (filtered_df['Duration (Days)'].between(min_select, max_select)) | 
        (filtered_df['Duration (Days)'].isnull())
    ]
    
    # --- Display Results ---
    
    st.markdown("---")
    
    # Display summary metrics
    col1, col2, col3 = st.columns(3)
    col1.metric("Total Countries Found", f"{len(filtered_df)}")
    
    visa_free_count = len(filtered_df[filtered_df['Visa Category'] == '✅ Visa-Free'])
    col2.metric("Visa-Free Destinations", f"{visa_free_count}")
    
    visa_required_count = len(filtered_df[filtered_df['Visa Category'] == '❌ Visa Required'])
    col3.metric("Visa Required", f"{visa_required_count}")

    st.markdown("### Visa Requirements Data")
    
    # Define columns to display
    display_cols_outbound = ['Destination', 'Visa Category', 'Requirement', 'Duration (Days)']
    display_cols_inbound = ['Source Country', 'Passport', 'Visa Category', 'Requirement', 'Duration (Days)']
    
    display_columns = display_cols_outbound if travel_direction == 'Outbound' else display_cols_inbound
    
    st.dataframe(
        filtered_df[display_columns].reset_index(drop=True),
        use_container_width=True,
        hide_index=True
    )
else:
    st.warning("Could not load visa data to display.")