app.py

import streamlit as st
import pandas as pd
import plotly.express as px
import requests
from io import StringIO
from datetime import datetime

# Page configuration
st.set_page_config(
    page_title="Terror Finance & Maritime Watch",
    page_icon="🛡️",
    layout="wide",
    initial_sidebar_state="collapsed"
)

# Custom CSS for styling
st.markdown("""
<style>
    .main {
        padding: 0rem 1rem;
    }
    .stTabs [data-baseweb="tab-list"] {
        gap: 24px;
    }
    .stTabs [data-baseweb="tab"] {
        padding-left: 20px;
        padding-right: 20px;
        background-color: transparent;
        border: none;
        color: #666;
        font-weight: 500;
    }
    .stTabs [aria-selected="true"] {
        background-color: transparent;
        color: #E53E3E;
        border-bottom: 2px solid #E53E3E;
    }

    /* Metric Cards */
    .metric-card {
        background: white;
        padding: 24px;
        border-radius: 12px;
        border: 1px solid #E2E8F0;
        text-align: center;
        box-shadow: 0 1px 3px rgba(0,0,0,0.1);
    }
    .metric-number {
        font-size: 2.5rem;
        font-weight: 700;
        margin: 8px 0;
        color: #1A202C !important;
    }
    .metric-label {
        color: #4A5568 !important;
        font-size: 0.875rem;
        text-transform: uppercase;
        letter-spacing: 0.5px;
    }

    /* Entity Cards */
    .entity-card {
        background: white;
        padding: 24px;
        border-radius: 12px;
        border: 1px solid #E2E8F0;
        margin-bottom: 16px;
        box-shadow: 0 1px 3px rgba(0,0,0,0.1);
    }
    .entity-card h3,
    .entity-card p,
    .entity-card span,
    .entity-card div {
        color: #1A202C !important;
    }

    /* Tags */
    .tag {
        display: inline-block;
        padding: 4px 12px;
        background-color: #EBF8FF;
        color: #2B6CB0;
        border-radius: 16px;
        font-size: 0.875rem;
        margin-right: 8px;
        margin-bottom: 8px;
    }
    .tag-danger {
        background-color: #FED7D7;
        color: #C53030;
    }
    .tag-warning {
        background-color: #FEFCBF;
        color: #975A16;
    }

    /* Headers */
    h1 {
        color: #1A202C;
        font-weight: 700;
    }
    h2 {
        color: #2D3748;
        font-weight: 600;
        margin-top: 2rem;
        margin-bottom: 1rem;
    }

    /* Status indicators */
    .status-active {
        color: #48BB78;
        font-weight: bold;
    }
    .status-inactive {
        color: #E53E3E;
        font-weight: bold;
    }

    /* Pariente AI Branding */
    .pariente-ai {
        color: #4A5568;
        font-size: 0.875rem;
        font-style: italic;
        text-align: right;
    }
    .pariente-ai-footer {
        background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
        -webkit-background-clip: text;
        -webkit-text-fill-color: transparent;
        font-weight: bold;
    }
</style>
""", unsafe_allow_html=True)

# Function to extract IMO from Information field
def extract_imo_from_info(info_text):
    """Extract IMO number from Information field"""
    import re
    if not info_text or pd.isna(info_text):
        return "N/A"
    
    # Look for IMO pattern: IMO- followed by 7 digits
    imo_match = re.search(r'IMO-?\s*(\d{7})', str(info_text), re.IGNORECASE)
    if imo_match:
        return imo_match.group(1)
    return "N/A"

def extract_mmsi_from_info(info_text):
    """Extract MMSI from Information field"""
    import re
    if not info_text or pd.isna(info_text):
        return "N/A"
    
    # Look for MMSI pattern
    mmsi_match = re.search(r'MMSI-?\s*(\d{9})', str(info_text), re.IGNORECASE)
    if mmsi_match:
        return mmsi_match.group(1)
    return "N/A"

def extract_call_sign_from_info(info_text):
    """Extract Call Sign from Information field"""
    import re
    if not info_text or pd.isna(info_text):
        return "N/A"
    
    # Look for Call Sign pattern
    call_match = re.search(r'Call Sign-?\s*([A-Z0-9]+)', str(info_text), re.IGNORECASE)
    if call_match:
        return call_match.group(1)
    return "N/A"

def extract_owner_from_info(info_text):
    """Extract owner information from Information field"""
    import re
    if not info_text or pd.isna(info_text):
        return "N/A"
    
    # Look for Registered owner or Commercial manager
    owner_match = re.search(r'Registered owner-?\s*([^,]+)', str(info_text), re.IGNORECASE)
    if owner_match:
        return owner_match.group(1).strip()
    
    manager_match = re.search(r'Commercial manager-?\s*([^,]+)', str(info_text), re.IGNORECASE)
    if manager_match:
        return manager_match.group(1).strip()
    
    return "N/A"
def safe_get(row, key, default="N/A"):
    try:
        # Special handling for Name field
        if key == 'Name':
            # Try to get Name column first
            if 'Name' in row.index and pd.notna(row['Name']) and str(row['Name']).strip():
                return str(row['Name']).strip()
            
            # Fallback: combine Last_Name and First_Name
            last_name = row.get('Last_Name', '') if 'Last_Name' in row.index else ''
            first_name = row.get('First_Name', '') if 'First_Name' in row.index else ''
            
            # Clean up the names
            last_name = str(last_name).strip() if pd.notna(last_name) else ''
            first_name = str(first_name).strip() if pd.notna(first_name) else ''
            
            # Combine names
            if last_name and first_name:
                return f"{last_name} {first_name}"
            elif first_name:
                return first_name
            elif last_name:
                return last_name
            else:
                dc_id = row.get('DC_ID', 'Unknown')
                return f"Entity {dc_id}"
        
        # Try exact key first
        if key in row.index:
            val = row[key]
        else:
            # Try alternative column names for common fields
            alternatives = {
                'DC_ID': ['ID', 'Entity_ID'],
                'Countries': ['Country', 'Location'],
                'Companies': ['Company', 'Business'],
                'Phone#': ['Phone', 'Telephone', 'Contact_Phone'],
                'Linked To': ['Organization', 'Terror_Organization', 'Group'],
                'IMO': ['IMO_Number', 'IMO_No', 'International_Maritime_Organization'],
                'Flag': ['Flag_State', 'Flag_Country'],
                'DWT': ['Deadweight', 'Dead_Weight_Tonnage'],
                'Built_Year': ['Year_Built', 'Construction_Year', 'DOB'],
                'Status': ['Vessel_Status', 'Ship_Status', 'AIS_Status'],
                'Insurance': ['Insurer', 'Insurance_Company']
            }
            
            val = None
            if key in alternatives:
                for alt_key in alternatives[key]:
                    if alt_key in row.index:
                        val = row[alt_key]
                        break
            
            if val is None:
                return default
        
        # Handle None, NaN, empty strings
        if pd.isna(val) or str(val).strip() == '' or str(val).lower() in ['nan', 'none', 'null']:
            return default
            
        return str(val).strip()
    except (KeyError, TypeError, AttributeError):
        return default

# Load clean datasets from Hugging Face
@st.cache_data
def load_clean_data():
    """Load pre-processed clean datasets"""
    base_url = "https://huggingface.co/spaces/Malaji71/list/resolve/main/"
    
    datasets = {}
    files_to_load = {
        'individuals': 'individuals_clean.csv',
        'companies': 'companies_clean.csv', 
        'vessels': 'vessels_clean.csv'
    }
    
    for category, filename in files_to_load.items():
        try:
            url = base_url + filename
            response = requests.get(url)
            response.raise_for_status()
            
            # Load CSV from response content
            df = pd.read_csv(StringIO(response.text))
            
            # Clean up any remaining data issues
            for col in df.columns:
                if df[col].dtype == 'object':
                    df[col] = df[col].fillna('').astype(str)
            
            datasets[category] = df
            
        except Exception as e:
            st.error(f"Error loading {filename}: {str(e)}")
            # Create empty dataframe as fallback
            datasets[category] = pd.DataFrame()
    
    return datasets

# Create fallback data if CSV loading fails
def create_sample_data():
    """Create sample data if CSV files are not available"""
    individuals = pd.DataFrame({
        'Name': ['Sample Individual 1', 'Sample Individual 2'],
        'DC_ID': [1, 2],
        'Position': ['Financier', 'Operator'],
        'Countries': ['Lebanon', 'Syria'],
        'Email': ['sample1@email.com', 'sample2@email.com'],
        'Phone#': ['+961-xxx-xxxx', '+963-xxx-xxxx'],
        'Companies': ['Sample Corp', 'Test LLC'],
        'Linked To': ['Hamas', 'Hezbollah'],
        'Auto_Category': ['Individual', 'Individual']
    })
    
    companies = pd.DataFrame({
        'Name': ['Sample Company 1', 'Sample Company 2'],
        'DC_ID': [101, 102],
        'Sub_Category': ['Shell Company', 'Front Company'],
        'Countries': ['Panama', 'Cyprus'],
        'Owner': ['Unknown', 'Sample Person'],
        'Key_Individuals': ['Person A', 'Person B'],
        'Linked To': ['Hamas', 'Hezbollah'],
        'Auto_Category': ['Company', 'Company']
    })
    
    vessels = pd.DataFrame({
        'Name': ['Sample Vessel 1', 'Sample Vessel 2'],
        'DC_ID': [201, 202],
        'IMO': ['1234567', '7654321'],
        'Flag': ['Panama', 'Liberia'],
        'Owner': ['Sample Maritime LLC', 'Ocean Holdings'],
        'Status': ['AIS Off', 'Active'],
        'Insurance': ['Unknown', 'Lloyd\'s'],
        'Auto_Category': ['Vessel', 'Vessel']
    })
    
    return {
        'individuals': individuals,
        'companies': companies,
        'vessels': vessels
    }

# Header
col1, col2, col3 = st.columns([6, 1, 1])
with col1:
    st.markdown("# 🛡️ Terror Finance & Maritime Watch")
    st.markdown("*Powered by Pariente AI - Advanced Intelligence Analytics*")
with col2:
    theme_toggle = st.checkbox("🌙", key="theme")
with col3:
    st.markdown('<div class="pariente-ai">🤖 <span class="pariente-ai-footer">Pariente AI</span><br><em>Intelligence Platform</em></div>', unsafe_allow_html=True)

# Load data
try:
    data = load_clean_data()
    # Check if data was loaded successfully
    if all(len(df) == 0 for df in data.values()):
        st.warning("⚠️ Using sample data - CSV files not found. Please upload the processed CSV files.")
        data = create_sample_data()
except Exception as e:
    st.error(f"Error loading data: {str(e)}")
    data = create_sample_data()

# Navigation tabs
tab1, tab2, tab3, tab4, tab5 = st.tabs(["👥 Individuals", "🏢 Companies", "🚢 Vessels", "📊 Summary", "📋 Data Reports"])

# Summary Tab
with tab4:
    st.markdown("## Key Statistics")
    col1, col2, col3, col4 = st.columns(4)

    with col1:
        st.markdown(f"""
        <div class="metric-card">
            <div class="metric-number" style="color: #E53E3E;">{len(data['individuals'])}</div>
            <div class="metric-label">Individuals Tracked</div>
        </div>
        """, unsafe_allow_html=True)
    
    with col2:
        st.markdown(f"""
        <div class="metric-card">
            <div class="metric-number" style="color: #3182CE;">{len(data['companies'])}</div>
            <div class="metric-label">Companies Monitored</div>
        </div>
        """, unsafe_allow_html=True)
    
    with col3:
        st.markdown(f"""
        <div class="metric-card">
            <div class="metric-number" style="color: #F6AD55;">{len(data['vessels'])}</div>
            <div class="metric-label">Vessels Documented</div>
        </div>
        """, unsafe_allow_html=True)
    
    with col4:
        # Calculate unique countries safely
        try:
            all_countries = []
            for df in data.values():
                if 'Countries' in df.columns:
                    countries = df['Countries'].dropna().astype(str)
                    for country_list in countries:
                        if ',' in country_list:
                            all_countries.extend([c.strip() for c in country_list.split(',') if c.strip()])
                        else:
                            all_countries.append(country_list.strip())
            unique_countries = len(set([c for c in all_countries if c and c.lower() != 'nan']))
        except:
            unique_countries = 0
            
        st.markdown(f"""
        <div class="metric-card">
            <div class="metric-number" style="color: #48BB78;">{unique_countries}</div>
            <div class="metric-label">Countries Involved</div>
        </div>
        """, unsafe_allow_html=True)

    # Charts
    st.markdown("## Data Visualization")
    col1, col2, col3 = st.columns(3)
    
    with col1:
        st.markdown("### Organizations")
        try:
            if len(data['individuals']) > 0 and 'Linked To' in data['individuals'].columns:
                org_data = data['individuals']['Linked To'].value_counts().reset_index()
                org_data.columns = ['Organization', 'Count']
                if not org_data.empty:
                    fig = px.pie(org_data, names='Organization', values='Count')
                    fig.update_layout(showlegend=True, height=300)
                    st.plotly_chart(fig, use_container_width=True)
                else:
                    st.write("No organization data available")
            else:
                st.write("No organization data available")
        except Exception as e:
            st.write("Error creating organization chart")
    
    with col2:
        st.markdown("### Entity Types")
        type_data = pd.DataFrame({
            'Type': ['Individuals', 'Companies', 'Vessels'],
            'Count': [len(data['individuals']), len(data['companies']), len(data['vessels'])]
        })
        fig = px.bar(type_data, x='Type', y='Count', color_discrete_sequence=['#3182CE'])
        fig.update_layout(showlegend=False, height=300)
        st.plotly_chart(fig, use_container_width=True)
    
    with col3:
        st.markdown("### Top Countries")
        try:
            if unique_countries > 0:
                country_counts = {}
                for df in data.values():
                    if 'Countries' in df.columns:
                        countries = df['Countries'].dropna().astype(str)
                        for country_list in countries:
                            if ',' in country_list:
                                for country in country_list.split(','):
                                    country = country.strip()
                                    if country and country.lower() != 'nan':
                                        country_counts[country] = country_counts.get(country, 0) + 1
                            else:
                                country = country_list.strip()
                                if country and country.lower() != 'nan':
                                    country_counts[country] = country_counts.get(country, 0) + 1
                
                if country_counts:
                    country_data = pd.DataFrame(list(country_counts.items()), columns=['Country', 'Count'])
                    country_data = country_data.sort_values('Count', ascending=False).head(10)
                    fig = px.pie(country_data, names='Country', values='Count')
                    fig.update_layout(showlegend=True, height=300)
                    st.plotly_chart(fig, use_container_width=True)
                else:
                    st.write("No country data available")
            else:
                st.write("No country data available")
        except Exception as e:
            st.write("Error creating country chart")

# Individuals Tab
with tab1:
    st.markdown("## Individuals Database")
    
    # Export and filters
    col1, col2 = st.columns([6, 1])
    with col2:
        if st.button("📥 Export", key="export_individuals"):
            csv_data = data['individuals'].to_csv(index=False)
            st.download_button(
                label="Download CSV",
                data=csv_data,
                file_name=f"individuals_{datetime.now().strftime('%Y%m%d')}.csv",
                mime="text/csv"
            )
    
    # Filters
    col1, col2, col3 = st.columns([3, 2, 2])
    
    with col1:
        name_filter = st.text_input("Search by name", placeholder="Enter name...", key="search_individuals")
    
    with col2:
        country_options = ["All Countries"]
        if len(data['individuals']) > 0 and 'Countries' in data['individuals'].columns:
            all_countries = set()
            for country_list in data['individuals']['Countries'].dropna().astype(str):
                if ',' in country_list:
                    all_countries.update([c.strip() for c in country_list.split(',') if c.strip() and c.lower() != 'nan'])
                else:
                    if country_list.strip() and country_list.lower() != 'nan':
                        all_countries.add(country_list.strip())
            country_options += sorted(list(all_countries))
        country_filter = st.selectbox("Country", country_options, key="country_individuals")
    
    with col3:
        org_options = ["All Organizations"]
        if len(data['individuals']) > 0 and 'Linked To' in data['individuals'].columns:
            orgs = data['individuals']['Linked To'].dropna().astype(str)
            org_options += sorted(orgs.unique().tolist())
        org_filter = st.selectbox("Organization", org_options, key="org_individuals")

    # Apply filters
    filtered_individuals = data['individuals'].copy()
    
    if name_filter:
        filtered_individuals = filtered_individuals[
            filtered_individuals['Name'].str.contains(name_filter, case=False, na=False)
        ]
    
    if country_filter != "All Countries":
        filtered_individuals = filtered_individuals[
            filtered_individuals['Countries'].str.contains(country_filter, case=False, na=False)
        ]
    
    if org_filter != "All Organizations":
        filtered_individuals = filtered_individuals[
            filtered_individuals['Linked To'] == org_filter
        ]

    # Display individuals
    for _, person in filtered_individuals.iterrows():
        org_value = safe_get(person, 'Linked To')
        org_color = "tag-danger" if "Hamas" in org_value else "tag-warning"
        
        st.markdown(f"""
        <div class="entity-card">
            <h3>{safe_get(person, 'Name')}</h3>
            <p><strong>ID:</strong> {safe_get(person, 'DC_ID')}</p>
            <p><strong>Position:</strong> {safe_get(person, 'Position')}</p>
            <p><strong>Countries:</strong> {safe_get(person, 'Countries')}</p>
            <p><strong>Email:</strong> {safe_get(person, 'Email')}</p>
            <p><strong>Phone:</strong> {safe_get(person, 'Phone#')}</p>
            <p><strong>Companies:</strong> {safe_get(person, 'Companies')}</p>
            <span class="tag {org_color}">{org_value}</span>
        </div>
        """, unsafe_allow_html=True)

# Companies Tab
with tab2:
    st.markdown("## Companies Database")
    
    # Export
    col1, col2 = st.columns([6, 1])
    with col2:
        if st.button("📥 Export", key="export_companies"):
            csv_data = data['companies'].to_csv(index=False)
            st.download_button(
                label="Download CSV",
                data=csv_data,
                file_name=f"companies_{datetime.now().strftime('%Y%m%d')}.csv",
                mime="text/csv"
            )
    
    # Search
    name_filter = st.text_input("Search companies", placeholder="Enter company name...", key="search_companies")
    
    # Apply filters
    filtered_companies = data['companies'].copy()
    if name_filter:
        filtered_companies = filtered_companies[
            filtered_companies['Name'].str.contains(name_filter, case=False, na=False)
        ]

    # Display companies
    for _, company in filtered_companies.iterrows():
        org_value = safe_get(company, 'Linked To')
        org_color = "tag-danger" if "Hamas" in org_value else "tag-warning"
        
        st.markdown(f"""
        <div class="entity-card">
            <h3>{safe_get(company, 'Name')}</h3>
            <p><strong>ID:</strong> {safe_get(company, 'DC_ID')}</p>
            <p><strong>Type:</strong> {safe_get(company, 'Sub_Category')}</p>
            <p><strong>Countries:</strong> {safe_get(company, 'Countries')}</p>
            <p><strong>Owner:</strong> {safe_get(company, 'Owner')}</p>
            <p><strong>Key Individuals:</strong> {safe_get(company, 'Key_Individuals')}</p>
            <span class="tag {org_color}">{org_value}</span>
        </div>
        """, unsafe_allow_html=True)

# Vessels Tab
with tab3:
    st.markdown("## Maritime Vessels Database")
    
    # Export
    col1, col2 = st.columns([6, 1])
    with col2:
        if st.button("📥 Export", key="export_vessels"):
            csv_data = data['vessels'].to_csv(index=False)
            st.download_button(
                label="Download CSV",
                data=csv_data,
                file_name=f"vessels_{datetime.now().strftime('%Y%m%d')}.csv",
                mime="text/csv"
            )
    
    # Filters for vessels
    col1, col2, col3, col4 = st.columns([3, 2, 2, 2])
    
    with col1:
        name_filter = st.text_input("Search vessels", placeholder="Enter vessel name or IMO...", key="search_vessels")
    
    with col2:
        flag_options = ["All Flags"]
        if len(data['vessels']) > 0 and 'Flag' in data['vessels'].columns:
            flags = data['vessels']['Flag'].dropna().astype(str).unique()
            flag_options += sorted([f for f in flags if f and f.lower() != 'nan'])
        flag_filter = st.selectbox("Flag State", flag_options, key="flag_vessels")
    
    with col3:
        status_options = ["All Status"]
        if len(data['vessels']) > 0 and 'Status' in data['vessels'].columns:
            statuses = data['vessels']['Status'].dropna().astype(str).unique()
            status_options += sorted([s for s in statuses if s and s.lower() != 'nan'])
        status_filter = st.selectbox("Status", status_options, key="status_vessels")
    
    with col4:
        org_options = ["All Organizations"]
        if len(data['vessels']) > 0 and 'Linked To' in data['vessels'].columns:
            orgs = data['vessels']['Linked To'].dropna().astype(str).unique()
            org_options += sorted([o for o in orgs if o and o.lower() != 'nan'])
        org_filter = st.selectbox("Organization", org_options, key="org_vessels")
    
    # Apply filters
    filtered_vessels = data['vessels'].copy()
    
    if name_filter:
        # Search in Name, IMO, and other relevant fields
        mask = (
            filtered_vessels['Name'].str.contains(name_filter, case=False, na=False) |
            filtered_vessels.get('IMO', pd.Series()).astype(str).str.contains(name_filter, case=False, na=False) |
            filtered_vessels.get('Owner', pd.Series()).astype(str).str.contains(name_filter, case=False, na=False) |
            filtered_vessels.get('Information', pd.Series()).astype(str).str.contains(name_filter, case=False, na=False)
        )
        filtered_vessels = filtered_vessels[mask]
    
    if flag_filter != "All Flags":
        filtered_vessels = filtered_vessels[
            filtered_vessels['Flag'].str.contains(flag_filter, case=False, na=False)
        ]
    
    if status_filter != "All Status":
        filtered_vessels = filtered_vessels[
            filtered_vessels['Status'] == status_filter
        ]
    
    if org_filter != "All Organizations":
        filtered_vessels = filtered_vessels[
            filtered_vessels['Linked To'] == org_filter
        ]

    # Display vessels
    for _, vessel in filtered_vessels.iterrows():
        status_value = safe_get(vessel, 'Status')
        status_class = "status-inactive" if status_value == 'AIS Off' else "status-active"
        
        # Get vessel-specific information
        imo = safe_get(vessel, 'IMO')
        flag = safe_get(vessel, 'Flag')
        vessel_type = safe_get(vessel, 'Sub_Category')
        owner = safe_get(vessel, 'Owner')
        built_year = safe_get(vessel, 'Built_Year', safe_get(vessel, 'DOB'))  # Sometimes vessel age is in DOB
        dwt = safe_get(vessel, 'DWT')
        
        st.markdown(f"""
        <div class="entity-card">
            <h3>{safe_get(vessel, 'Name')}</h3>
            <p><strong>DC_ID:</strong> {safe_get(vessel, 'DC_ID')}</p>
            <p><strong>IMO Number:</strong> <span style="color: #2B6CB0; font-weight: bold;">{imo}</span></p>
            <p><strong>Vessel Type:</strong> {vessel_type}</p>
            <p><strong>Flag State:</strong> {flag}</p>
            <p><strong>Owner:</strong> {owner}</p>
            <p><strong>Built Year:</strong> {built_year}</p>
            <p><strong>DWT:</strong> {dwt}</p>
            <p><strong>Status:</strong> <span class="{status_class}">{status_value}</span></p>
            <p><strong>Insurance:</strong> {safe_get(vessel, 'Insurance')}</p>
            <p><strong>Countries:</strong> {safe_get(vessel, 'Countries')}</p>
            <p><strong>Information:</strong> {safe_get(vessel, 'Information')}</p>
        </div>
        """, unsafe_allow_html=True)

# Data Reports Tab
with tab5:
    st.markdown("## 📋 Data Analysis Reports")
    
    # Load additional analysis files
    try:
        # Load analysis JSON
        analysis_url = "https://huggingface.co/spaces/Malaji71/list/resolve/main/complete_analysis.json"
        analysis_response = requests.get(analysis_url)
        if analysis_response.status_code == 200:
            analysis_data = analysis_response.json()
            
            col1, col2 = st.columns(2)
            
            with col1:
                st.markdown("### 📊 Data Quality Metrics")
                if 'complete_analysis' in analysis_data:
                    quality_metrics = analysis_data['complete_analysis'].get('quality_metrics', {})
                    if 'completeness' in quality_metrics:
                        completeness_df = pd.DataFrame(
                            list(quality_metrics['completeness'].items()),
                            columns=['Column', 'Completeness %']
                        ).sort_values('Completeness %', ascending=False)
                        st.dataframe(completeness_df, use_container_width=True)
                    
                    if 'issues' in quality_metrics and quality_metrics['issues']:
                        st.markdown("### ⚠️ Data Issues")
                        for issue in quality_metrics['issues']:
                            st.warning(issue)
            
            with col2:
                st.markdown("### 🎯 Categorization Analysis")
                if 'complete_analysis' in analysis_data:
                    cat_analysis = analysis_data['complete_analysis'].get('categorization_analysis', {})
                    if 'content_based_categorization' in cat_analysis:
                        cat_counts = cat_analysis['content_based_categorization']
                        
                        for category, count in cat_counts.items():
                            st.metric(category.title(), count)
                
                st.markdown("### 📈 Processing Statistics")
                processing_stats = {
                    'Total Rows Processed': analysis_data.get('complete_analysis', {}).get('total_rows', 0),
                    'Total Columns Analyzed': analysis_data.get('complete_analysis', {}).get('total_columns', 0),
                    'Analysis Timestamp': analysis_data.get('complete_analysis', {}).get('timestamp', 'Unknown')
                }
                
                for stat, value in processing_stats.items():
                    st.write(f"**{stat}**: {value}")
        
        else:
            st.info("📄 Complete analysis data not available. Upload complete_analysis.json to see detailed reports.")
    
    except Exception as e:
        st.info("📄 Analysis reports will be available when you upload the JSON files.")
    
    # Load and display text reports
    try:
        # Executive Summary
        summary_url = "https://huggingface.co/spaces/Malaji71/list/resolve/main/executive_summary.txt"
        summary_response = requests.get(summary_url)
        if summary_response.status_code == 200:
            st.markdown("### 📝 Executive Summary")
            st.text(summary_response.text)
    except:
        pass
    
    try:
        # Analysis Report excerpt
        report_url = "https://huggingface.co/spaces/Malaji71/list/resolve/main/analysis_report.txt"
        report_response = requests.get(report_url)
        if report_response.status_code == 200:
            st.markdown("### 📋 Full Analysis Report")
            with st.expander("View Complete Report"):
                st.text(report_response.text)
    except:
        pass
    
    # Download section
    st.markdown("### 📥 Download Analysis Files")
    st.markdown("""
    **Available Analysis Files:**
    - `complete_analysis.json` - Complete analysis data in JSON format
    - `analysis_report.txt` - Detailed human-readable report  
    - `executive_summary.txt` - Executive summary
    - `uncertain_entities.csv` - Entities requiring manual review
    
    Upload these files to your Hugging Face Space to enable full analysis viewing.
    """)

# Footer
st.markdown("---")
col1, col2 = st.columns([3, 1])
with col1:
    st.markdown("🛡️ **Terror Finance & Maritime Watch** - Monitoring entities involved in terror financing and maritime sanctions evasion")
    st.markdown(f"📊 Data processed: {len(data['individuals'])} individuals, {len(data['companies'])} companies, {len(data['vessels'])} vessels")
    st.markdown(f"🕒 Last updated: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}")
with col2:
    st.markdown("**Powered by**")
    st.markdown("🤖 **Pariente AI**")
    st.markdown("*Advanced Intelligence Analytics*")