eval/respace/tools/complement_bottom.py

import os
import json
from pathlib import Path

def integrate_bounds_fields(scenes_dir: str, grouped_layouts_dir: str, backup: bool = True):
    """
    将scenes目录下的bounds_top和bounds_bottom字段整合到grouped_layouts目录下的同名文件中
    
    Args:
        scenes_dir: dataset-ssr3dfront/scenes目录路径
        grouped_layouts_dir: grouped_layouts目录路径
        backup: 是否在修改前创建备份
    """
    
    scenes_path = Path(scenes_dir)
    grouped_path = Path(grouped_layouts_dir)
    
    if not scenes_path.exists():
        print(f"Scenes目录不存在: {scenes_path}")
        return
        
    if not grouped_path.exists():
        print(f"Grouped layouts目录不存在: {grouped_path}")
        return
    
    # 创建备份目录（如果需要）
    if backup:
        backup_dir = grouped_path.parent / "grouped_layouts_backup"
        backup_dir.mkdir(exist_ok=True)
        print(f"备份目录: {backup_dir}")
    
    # 统计信息
    total_files = 0
    processed_files = 0
    missing_scenes = 0
    errors = 0
    
    print("开始处理文件...")
    
    # 遍历grouped_layouts目录中的所有JSON文件
    for grouped_file in grouped_path.glob("*.json"):
        total_files += 1
        
        # 查找对应的scene文件
        scene_file = scenes_path / grouped_file.name
        
        if not scene_file.exists():
            missing_scenes += 1
            print(f"警告: 未找到对应的scene文件: {scene_file.name}")
            continue
        
        try:
            # 读取grouped layout文件
            with open(grouped_file, 'r', encoding='utf-8') as f:
                grouped_data = json.load(f)
            
            # 读取scene文件
            with open(scene_file, 'r', encoding='utf-8') as f:
                scene_data = json.load(f)
            
            # 检查scene文件是否包含所需字段
            if 'bounds_top' not in scene_data or 'bounds_bottom' not in scene_data:
                print(f"警告: Scene文件缺少bounds字段: {scene_file.name}")
                continue
            
            # 创建备份（如果需要）
            if backup:
                backup_file = backup_dir / grouped_file.name
                with open(backup_file, 'w', encoding='utf-8') as f:
                    json.dump(grouped_data, f, ensure_ascii=False, indent=2)
            
            # 添加bounds字段到grouped data
            grouped_data['bounds_top'] = scene_data['bounds_top']
            grouped_data['bounds_bottom'] = scene_data['bounds_bottom']
            
            # 写回grouped layout文件
            with open(grouped_file, 'w', encoding='utf-8') as f:
                json.dump(grouped_data, f, ensure_ascii=False, indent=2)
            
            processed_files += 1
            
            if processed_files % 100 == 0:
                print(f"已处理: {processed_files}/{total_files} 文件")
                
        except Exception as e:
            errors += 1
            print(f"处理文件时出错 {grouped_file.name}: {e}")
    
    # 输出统计信息
    print("\n=== 处理完成 ===")
    print(f"总文件数: {total_files}")
    print(f"成功处理: {processed_files}")
    print(f"缺少对应scene文件: {missing_scenes}")
    print(f"处理错误: {errors}")
    
    if backup:
        print(f"原始文件已备份到: {backup_dir}")

def verify_integration(grouped_layouts_dir: str, sample_size: int = 10):
    """
    验证整合结果，检查部分文件是否正确添加了bounds字段
    
    Args:
        grouped_layouts_dir: grouped_layouts目录路径
        sample_size: 检查的样本文件数量
    """
    
    grouped_path = Path(grouped_layouts_dir)
    json_files = list(grouped_path.glob("*.json"))
    
    if not json_files:
        print("未找到JSON文件")
        return
    
    # 随机选择样本文件进行检查
    import random
    sample_files = random.sample(json_files, min(sample_size, len(json_files)))
    
    print(f"\n=== 验证结果 (检查 {len(sample_files)} 个样本文件) ===")
    
    success_count = 0
    for file_path in sample_files:
        try:
            with open(file_path, 'r', encoding='utf-8') as f:
                data = json.load(f)
            
            has_bounds_top = 'bounds_top' in data
            has_bounds_bottom = 'bounds_bottom' in data
            
            if has_bounds_top and has_bounds_bottom:
                success_count += 1
                print(f"✓ {file_path.name}: 包含bounds字段")
            else:
                missing_fields = []
                if not has_bounds_top:
                    missing_fields.append('bounds_top')
                if not has_bounds_bottom:
                    missing_fields.append('bounds_bottom')
                print(f"✗ {file_path.name}: 缺少字段 {missing_fields}")
                
        except Exception as e:
            print(f"✗ {file_path.name}: 读取错误 - {e}")
    
    print(f"\n验证成功率: {success_count}/{len(sample_files)} ({success_count/len(sample_files)*100:.1f}%)")

def show_file_structure_comparison(scenes_dir: str, grouped_layouts_dir: str, filename: str):
    """
    显示指定文件在两个目录中的结构对比
    
    Args:
        scenes_dir: scenes目录路径
        grouped_layouts_dir: grouped_layouts目录路径
        filename: 要对比的文件名
    """
    
    scenes_file = Path(scenes_dir) / filename
    grouped_file = Path(grouped_layouts_dir) / filename
    
    print(f"\n=== 文件结构对比: {filename} ===")
    
    # 显示scene文件结构
    if scenes_file.exists():
        try:
            with open(scenes_file, 'r', encoding='utf-8') as f:
                scene_data = json.load(f)
            print(f"\nScene文件字段: {list(scene_data.keys())}")
            if 'bounds_top' in scene_data:
                print(f"bounds_top: {scene_data['bounds_top'][:2]}...")  # 显示前2个元素
            if 'bounds_bottom' in scene_data:
                print(f"bounds_bottom: {scene_data['bounds_bottom'][:2]}...")
        except Exception as e:
            print(f"读取scene文件错误: {e}")
    else:
        print("Scene文件不存在")
    
    # 显示grouped layout文件结构
    if grouped_file.exists():
        try:
            with open(grouped_file, 'r', encoding='utf-8') as f:
                grouped_data = json.load(f)
            print(f"\nGrouped layout文件字段: {list(grouped_data.keys())}")
            if 'bounds_top' in grouped_data:
                print("✓ 包含bounds_top字段")
            else:
                print("✗ 缺少bounds_top字段")
            if 'bounds_bottom' in grouped_data:
                print("✓ 包含bounds_bottom字段")
            else:
                print("✗ 缺少bounds_bottom字段")
        except Exception as e:
            print(f"读取grouped layout文件错误: {e}")
    else:
        print("Grouped layout文件不存在")

if __name__ == "__main__":
    # 设置目录路径
    SCENES_DIR = "/home/v-meiszhang/amlt-project/respace/dataset-ssr3dfront/scenes"
    GROUPED_LAYOUTS_DIR = "/home/v-meiszhang/amlt-project/respace/grouped_layouts_v2"
    
    # 示例文件名进行结构对比
    EXAMPLE_FILE = "0a8d471a-2587-458a-9214-586e003e9cf9-c944563d-1e2a-4ed0-9457-5a8f43c9f17c.json"
    
    print("步骤1: 显示文件结构对比")
    show_file_structure_comparison(SCENES_DIR, GROUPED_LAYOUTS_DIR, EXAMPLE_FILE)
    
    print("\n" + "="*50)
    print("步骤2: 开始整合bounds字段")
    
    # 执行整合
    integrate_bounds_fields(SCENES_DIR, GROUPED_LAYOUTS_DIR, backup=False)
    
    print("\n" + "="*50)
    print("步骤3: 验证整合结果")
    
    # 验证结果
    verify_integration(GROUPED_LAYOUTS_DIR, sample_size=5)
    
    print("\n" + "="*50)
    print("步骤4: 显示整合后的文件结构")
    
    # 再次显示结构对比以确认更改
    show_file_structure_comparison(SCENES_DIR, GROUPED_LAYOUTS_DIR, EXAMPLE_FILE)