#!/usr/bin/env python3
"""Validate the curated Brazilian institutions YAML file."""

import yaml
from collections import Counter

# Read the curated file
with open('data/instances/brazilian_institutions_curated.yaml', 'r') as f:
    institutions = yaml.safe_load(f)

print(f"✓ Valid YAML with {len(institutions)} institutions\n")

# Analyze by type
types = Counter(inst['institution_type'] for inst in institutions)
print("Institutions by Type:")
for inst_type, count in sorted(types.items()):
    print(f"  {inst_type}: {count}")

print(f"\nTotal: {sum(types.values())}")

# Count features
features = {
    'alternative_names': 0,
    'wikidata_ids': 0,
    'digital_platforms': 0,
    'collections': 0,
    'change_history': 0,
}

total_platforms = 0
total_collections = 0
total_events = 0
confidence_scores = []

for inst in institutions:
    if inst.get('alternative_names'):
        features['alternative_names'] += 1
    
    if inst.get('identifiers'):
        for id_obj in inst['identifiers']:
            if id_obj['identifier_scheme'] == 'Wikidata':
                features['wikidata_ids'] += 1
                break
    
    if inst.get('digital_platforms'):
        features['digital_platforms'] += 1
        total_platforms += len(inst['digital_platforms'])
    
    if inst.get('collections'):
        features['collections'] += 1
        total_collections += len(inst['collections'])
    
    if inst.get('change_history'):
        features['change_history'] += 1
        total_events += len(inst['change_history'])
    
    if inst.get('provenance', {}).get('confidence_score'):
        confidence_scores.append(inst['provenance']['confidence_score'])

print("\nFeature Coverage:")
for feature, count in features.items():
    pct = (count / len(institutions)) * 100
    print(f"  {feature}: {count}/{len(institutions)} ({pct:.0f}%)")

print(f"\nAggregate Counts:")
print(f"  Total digital platforms: {total_platforms}")
print(f"  Total collections: {total_collections}")
print(f"  Total change events: {total_events}")

if confidence_scores:
    print(f"\nConfidence Scores:")
    print(f"  Min: {min(confidence_scores):.2f}")
    print(f"  Max: {max(confidence_scores):.2f}")
    print(f"  Average: {sum(confidence_scores)/len(confidence_scores):.2f}")

# List all institutions
print("\n" + "="*60)
print("All Curated Institutions:")
print("="*60)
for i, inst in enumerate(institutions, 1):
    print(f"\n{i}. {inst['name']}")
    print(f"   Type: {inst['institution_type']}")
    if inst.get('identifiers'):
        wikidata = next((id_obj for id_obj in inst['identifiers'] 
                        if id_obj['identifier_scheme'] == 'Wikidata'), None)
        if wikidata:
            print(f"   Wikidata: {wikidata['identifier_value']}")
    if inst.get('digital_platforms'):
        print(f"   Platforms: {len(inst['digital_platforms'])}")
    if inst.get('collections'):
        print(f"   Collections: {len(inst['collections'])}")
    confidence = inst.get('provenance', {}).get('confidence_score')
    if confidence:
        print(f"   Confidence: {confidence:.2f}")