#!/usr/bin/env python3
"""
从附表补全 FtSPL5-24 实体
"""

import json


def main():
    print("=" * 70)
    print("Adding Missing FtSPL5-24 Genes")
    print("=" * 70)
    
    # 加载基因列表
    with open('ftspl_gene_list.json') as f:
        gene_list = json.load(f)
    
    # 加载当前结果
    with open('evaluation/auto_results/pmid31286919_final_v2.json') as f:
        current = json.load(f)
    
    # 当前已有的实体ID
    existing_ids = {e['id'] for e in current['entities']}
    
    print(f"\nCurrent entities: {len(existing_ids)}")
    print(f"Gene list: {len(gene_list)}")
    
    # 添加缺失的基因实体
    added = 0
    for gene in gene_list:
        symbol = gene['symbol']
        if symbol not in existing_ids:
            entity = {
                "id": symbol,
                "name": symbol,
                "type": "gene",
                "species": "Fagopyrum tataricum",
                "primary_id": gene['locus_id'],
                "protein_length": gene['protein_length'],
                "attributes": {"domain": gene.get('domain', '')}
            }
            current['entities'].append(entity)
            existing_ids.add(symbol)
            added += 1
    
    print(f"\nAdded {added} new gene entities")
    
    # 不添加 member_of 关系（Gold 中没有这些具体关系）
    print(f"Note: Not adding member_of relations for all genes (not in Gold Standard)")
    
    # 保存
    print(f"\nFinal stats:")
    print(f"  Total entities: {len(current['entities'])}")
    print(f"  Total relations: {len(current['relations'])}")
    
    output_path = 'evaluation/auto_results/pmid31286919_complete.json'
    with open(output_path, 'w') as f:
        json.dump(current, f, indent=2)
    
    print(f"\n✓ Saved to: {output_path}")
    
    # 评估
    print("\n" + "=" * 70)
    print("Evaluation")
    print("=" * 70)
    import subprocess
    subprocess.run([
        'python3', 'evaluation/compare_extraction.py',
        '--gold', 'evaluation/gold_standard/pmid31286919_layer2_full.json',
        '--pred', output_path
    ])


if __name__ == "__main__":
    main()