glam/scripts/export_nde_map_json.py

#!/usr/bin/env python3
"""
Export NDE Enriched Institutions to JSON for Frontend Map

Reads the enriched YAML files and produces a lightweight JSON file
suitable for the React/Leaflet map component.
"""

import json
from pathlib import Path
from datetime import datetime, timezone
import sys

# Add project root to path for imports
project_root = Path(__file__).parent.parent
sys.path.insert(0, str(project_root))

try:
    import yaml
except ImportError:
    print("Error: PyYAML not installed. Run: pip install pyyaml")
    sys.exit(1)

# Institution type mappings
TYPE_COLORS = {
    'G': '#00bcd4',   # Gallery - cyan
    'L': '#2ecc71',   # Library - green
    'A': '#3498db',   # Archive - blue
    'M': '#e74c3c',   # Museum - red
    'O': '#f39c12',   # Official - orange
    'R': '#1abc9c',   # Research - teal
    'C': '#795548',   # Corporation - brown
    'U': '#9e9e9e',   # Unknown - gray
    'B': '#4caf50',   # Botanical - green
    'E': '#ff9800',   # Education - amber
    'S': '#9b59b6',   # Society - purple
    'F': '#95a5a6',   # Features - gray
    'I': '#673ab7',   # Intangible - deep purple
    'X': '#607d8b',   # Mixed - blue gray
    'P': '#ff5722',   # Personal - deep orange
    'H': '#607d8b',   # Holy sites - blue gray
    'D': '#34495e',   # Digital - dark gray
    'N': '#e91e63',   # NGO - pink
    'T': '#ff5722',   # Taste/smell - deep orange
}

TYPE_NAMES = {
    'G': 'Gallery',
    'L': 'Library',
    'A': 'Archive',
    'M': 'Museum',
    'O': 'Official',
    'R': 'Research',
    'C': 'Corporation',
    'U': 'Unknown',
    'B': 'Botanical',
    'E': 'Education',
    'S': 'Society',
    'F': 'Features',
    'I': 'Intangible',
    'X': 'Mixed',
    'P': 'Personal',
    'H': 'Holy sites',
    'D': 'Digital',
    'N': 'NGO',
    'T': 'Taste/smell',
}


def extract_institution_data(entry_data: dict) -> dict | None:
    """Extract the relevant data for the map from an enriched entry."""
    
    # Get original entry data
    original = entry_data.get('original_entry', {})
    enrichment = entry_data.get('wikidata_enrichment', {})
    
    # Skip if no coordinates
    coords = enrichment.get('wikidata_coordinates', {})
    if not coords or not coords.get('latitude') or not coords.get('longitude'):
        return None
    
    # Get institution type (first one if list)
    types = original.get('type', [])
    inst_type = types[0] if types else 'U'
    
    # Get name - prefer Dutch label, fall back to original name
    name = (
        enrichment.get('wikidata_label_nl') or 
        original.get('organisatie') or 
        'Unknown Institution'
    )
    
    # Get city
    city = original.get('plaatsnaam_bezoekadres', '')
    
    # Get description - prefer Dutch, fall back to English
    description = (
        enrichment.get('wikidata_description_nl') or
        enrichment.get('wikidata_description_en') or
        ''
    )
    
    # Get website
    website = enrichment.get('wikidata_official_website', '')
    
    # Get Wikidata ID
    wikidata_id = enrichment.get('wikidata_entity_id', '')
    
    return {
        'lat': coords['latitude'],
        'lon': coords['longitude'],
        'name': name,
        'city': city,
        'type': inst_type,
        'type_name': TYPE_NAMES.get(inst_type, 'Unknown'),
        'color': TYPE_COLORS.get(inst_type, '#9e9e9e'),
        'website': website,
        'wikidata_id': wikidata_id,
        'description': description[:200] + '...' if len(description) > 200 else description,
    }


def main():
    """Main export function."""
    
    # Paths
    enriched_dir = project_root / 'data' / 'nde' / 'enriched' / 'entries'
    output_dir = project_root / 'frontend' / 'public' / 'data'
    output_file = output_dir / 'nde_institutions.json'
    
    # Create output directory if needed
    output_dir.mkdir(parents=True, exist_ok=True)
    
    print(f"Reading enriched entries from: {enriched_dir}")
    
    institutions = []
    files_processed = 0
    files_with_coords = 0
    
    # Process all YAML files
    yaml_files = sorted(enriched_dir.glob('*.yaml'))
    
    for yaml_file in yaml_files:
        try:
            with open(yaml_file, 'r', encoding='utf-8') as f:
                entry_data = yaml.safe_load(f)
            
            files_processed += 1
            
            # Extract institution data
            inst_data = extract_institution_data(entry_data)
            if inst_data:
                institutions.append(inst_data)
                files_with_coords += 1
                
        except Exception as e:
            print(f"Warning: Error processing {yaml_file.name}: {e}")
            continue
    
    # Sort by name
    institutions.sort(key=lambda x: x['name'].lower())
    
    # Write JSON
    with open(output_file, 'w', encoding='utf-8') as f:
        json.dump(institutions, f, ensure_ascii=False, indent=2)
    
    print(f"\n✅ Export complete!")
    print(f"   Files processed: {files_processed}")
    print(f"   Institutions with coordinates: {files_with_coords}")
    print(f"   Output file: {output_file}")
    
    # Print type distribution
    type_counts = {}
    for inst in institutions:
        t = inst['type']
        type_counts[t] = type_counts.get(t, 0) + 1
    
    print(f"\n📊 Distribution by type:")
    for t, count in sorted(type_counts.items(), key=lambda x: -x[1]):
        print(f"   {TYPE_NAMES.get(t, t)}: {count}")


if __name__ == '__main__':
    main()