Add packaging support files and assets

imgui.ini 是运行时生成的 ImGui 配置文件，
不应纳入版本控制。
已在 .gitignore 中配置忽略。

.gitignore

@@ -58,3 +58,6 @@ Resources/models/Women_1.glb
 /venv/
 /engine/
 /panda3d_imgui-1.1.0-py3-none-any.whl
+imgui.ini
+dist
+build

AGENTS.md

@@ -1,219 +1,120 @@
-# EG 项目概览与开发指南
+# 元泰 EG 项目开发指南
 
-## 项目简介
+元泰 EG 是基于 Panda3D 的 3D 编辑器和游戏引擎。
-
-EG 是一个基于 Panda3D 引擎开发的 3D 编辑器和游戏引擎，集成了高级渲染管线、VR 支持、物理模拟、脚本系统等功能。该项目主要用于创建 3D 场景、游戏开发和交互式应用程序。
-
-## 核心技术栈
-
-- **渲染引擎**: Panda3D 1.10.15 + RenderPipeline (延迟渲染、PBR材质)
-- **GUI框架**: PyQt5 + imgui_bundle (用于编辑器界面)
-- **VR支持**: OpenVR 2.2.0
-- **脚本系统**: Python 3.10.12
-- **其他依赖**: PyQt5-WebEngine, assimp, pillow 等
-
-## 项目架构
-
-### 核心模块
-
-- **main.py** - 应用程序入口点，初始化所有系统
-- **Start_Run.py** - 启动脚本，设置环境变量和路径
-- **core/** - 核心功能模块
-  - `world.py` - 世界管理器，处理场景渲染和更新
-  - `selection.py` - 选择系统，处理对象选择
-  - `event_handler.py` - 事件处理系统
-  - `tool_manager.py` - 工具管理器
-  - `script_system.py` - 脚本系统
-  - `patrol_system.py` - 巡逻系统
-  - `Command_System.py` - 命令系统
-  - `terrain_manager.py` - 地形管理
-  - `vr_manager.py` - VR 功能管理
-  - `collision_manager.py` - 碰撞检测管理
-  - `resource_manager.py` - 资源管理
-  - `assembly_interaction.py` - 装配交互
-  - `maintenance_gui.py` - 维护界面
-  - `render_pipeline_utils.py` - 渲染管线工具
-  - `InfoPanelManager.py` - 信息面板管理
-  - `CustomMouseController.py` - 自定义鼠标控制器
-
-### GUI 系统
-
-- **gui/gui_manager.py** - GUI管理器，处理2D/3D界面元素
-- **ui/icon_manager.py** - 图标管理器
-
-### 场景管理
-
-- **scene/scene_manager.py** - 场景管理器，处理模型导入、场景树构建
-- **scene/util.py** - 场景工具函数
-
-### 项目管理
-
-- **project/project_manager.py** - 项目管理器，处理项目创建、保存、加载
-
-### 脚本系统
-
-- **scripts/** - 包含各种预定义脚本
-  - `MoverScript.py` - 移动脚本
-  - `RotatorScript.py` - 旋转脚本
-  - `ScalerScript.py` - 缩放脚本
-  - `ColorChangerScript.py` - 颜色变化脚本
-  - `FollowerScript.py` - 跟随脚本
-  - `BouncerScript.py` - 弹跳脚本
-  - `ComboAnimatorScript.py` - 组合动画脚本
-
-### VR 系统
-
-- **vr_actions/** - VR动作配置
-  - `actions.json` - VR动作定义
-  - `bindings_*.json` - 不同VR设备的绑定配置
-
-### 资源管理
-
-- **Resources/** - 资源目录
-  - `models/` - 3D模型
-  - `textures/` - 纹理贴图
-  - `materials/` - 材质文件
-  - `animations/` - 动画文件
-  - `icons/` - 图标资源
-
-### 渲染管线
-
-- **RenderPipelineFile/** - 高级渲染管线
-  - `rpcore/` - 渲染管线核心
-  - `rpplugins/` - 渲染插件
-  - `effects/` - 后处理效果
-  - `config/` - 渲染配置
-
-## 启动和运行
-
-### 环境要求
-
-- Python 3.10.12
-- Panda3D 1.10.15
-- PyQt5 5.15.9
-- OpenVR 2.2.0 (VR功能)
-
-### 运行方式
-
-1. **直接运行主程序**:
-   ```bash
-   python Start_Run.py
-   ```
-
-2. **带项目路径运行**:
-   ```bash
-   python Start_Run.py /path/to/project
-   ```
-
-3. **从main.py运行**:
-   ```bash
-   python main.py
-   ```
-
-### 配置文件
-
-- **config/vr_settings.json** - VR渲染配置
-- **imgui.ini** - ImGui界面配置
-- **.gitignore** - Git忽略文件配置
-
-## 开发约定
-
-### 代码风格
-
-- 使用UTF-8编码
-- 遵循PEP 8代码规范
-- 类名使用驼峰命名法
-- 函数和变量使用下划线命名法
-- 文件头部包含模块说明和导入信息
-
-### 脚本开发
-
-- 所有用户脚本应继承 `ScriptBase` 类
-- 脚本文件放在 `scripts/` 目录下
-- 使用 `ScriptManager` 管理脚本生命周期
-- 脚本API通过 `world` 对象提供
-
-### 插件开发
-
-- 插件系统支持动态加载
-- 插件配置文件使用JSON格式
-- 插件应实现标准接口
-
-### VR开发
-
-- VR动作配置在 `vr_actions/actions.json` 中定义
-- 支持多种VR设备（Vive、Oculus、Index）
-- VR渲染配置在 `config/vr_settings.json` 中
-
-## 构建和部署
-
-### 依赖安装
-
-```bash
-pip install -r requirements/requirements.txt
-```
-
-### 测试
-
-项目包含多个测试脚本：
-- `test_quick_script.py` - 快速测试脚本
-- `TestMover.py` - 移动测试
-- `TestRotator.py` - 旋转测试
-- `TestScaler.py` - 缩放测试
-
-### 项目文件
-
-- 项目配置使用JSON格式
-- 项目文件包含场景、资源、脚本等信息
-- 使用 `ProjectManager` 管理项目生命周期
-
-## 常见问题
-
-### VR相关问题
-
-1. 确保VR设备已正确连接
-2. 检查OpenVR运行时是否安装
-3. 验证VR动作配置是否正确
-
-### 渲染问题
-
-1. 检查显卡驱动是否最新
-2. 确认RenderPipeline配置正确
-3. 验证材质和纹理路径
-
-### 性能优化
-
-1. 使用合适的LOD设置
-2. 优化场景复杂度
-3. 调整渲染质量设置
-
-## 扩展开发
-
-### 添加新脚本
-
-1. 在 `scripts/` 目录创建新脚本文件
-2. 继承 `ScriptBase` 类
-3. 实现必要的方法
-4. 通过 `ScriptManager` 注册脚本
-
-### 添加新工具
-
-1. 在 `core/tool_manager.py` 中注册新工具
-2. 实现工具逻辑
-3. 添加GUI界面元素
-
-### 添加新渲染效果
-
-1. 在 `RenderPipelineFile/rpplugins/` 目录创建插件
-2. 实现渲染逻辑
-3. 添加配置选项
-
-## 联系和支持
-
-- 项目Git仓库: http://10.0.0.99:4000/Rowland/EG.git
-- 当前分支: imgui
-- 最新提交: 移除qt依赖 (33e62bd1e4c2c8d3aac15e045b419edb8992d7ff)
 
 ---
 
-*该文档由iFlow CLI自动生成，最后更新时间: 2026年1月28日*
+## 🚀 快速开始
+
+### 环境要求
+
+- **Python 3.11.x**（必须）
+- Windows 10/11 64位
+- Visual Studio Build Tools（用于构建）
+
+### 安装步骤
+
+```bash
+# 1. 检查 Python 版本（必须是 3.11）
+python --version
+
+# 2. 验证版本
+python build_scripts/check_python_version.py
+
+# 3. 安装依赖
+pip install -r requirements/requirements-minimal.txt
+
+# 4. 运行程序
+python Start_Run.py
+```
+
+---
+
+## 📦 构建安装程序
+
+### Windows
+
+```powershell
+# 在 VS Dev Shell 中执行
+.\build_scripts\build_windows.ps1 -Version "1.0.0"
+```
+
+详细步骤请参考 `build_scripts/BUILD_GUIDE.md`
+
+---
+
+## 🗂️ 项目结构
+
+```
+元泰 EG/
+├── AGENTS.md                    # 本文件
+├── main.py                      # 主程序入口
+├── Start_Run.py                 # 启动脚本
+├── requirements/                # 依赖配置
+│   ├── requirements-minimal.txt # 最小依赖（推荐）
+│   └── requirements.txt         # 完整依赖（旧）
+├── build_scripts/               # 构建脚本和文档
+│   ├── build_windows.ps1        # Windows 构建脚本
+│   ├── build_linux.sh           # Linux 构建脚本
+│   ├── BUILD_GUIDE.md           # 构建指南 ⭐
+│   ├── BUILD_README.md          # 构建说明
+│   ├── INSTALLER_GUIDE.md       # 完整指南
+│   ├── QUICK_REFERENCE.md       # 快速参考
+│   ├── ICON_GUIDE.md            # 图标准备
+│   ├── PYTHON_VERSION_REQUIREMENT.md  # Python版本要求
+│   └── PACKAGING_CHECKLIST.md   # 打包检查清单
+├── core/                        # 核心模块
+├── RenderPipelineFile/          # 渲染管线
+├── Resources/                   # 资源文件
+└── ...
+```
+
+---
+
+## 🛠️ 技术栈
+
+- **Python**: 3.11.x（必须）
+- **渲染引擎**: Panda3D 1.10.16
+- **UI框架**: imgui_bundle + p3dimgui
+- **VR支持**: OpenVR 2.2.0
+- **图像处理**: Pillow
+
+### 已移除的依赖
+
+为减少体积，已移除以下未使用的依赖：
+
+- ❌ PyQt5 (~50MB)
+- ❌ PySide6 (~100MB)
+- ❌ 相关 Qt 配套库 (~20MB)
+
+**总计节省 ~170MB**
+
+---
+
+## 📚 重要文档
+
+| 文档 | 位置 | 用途 |
+|------|------|------|
+| 构建指南 | `build_scripts/BUILD_GUIDE.md` | 从零开始构建 ⭐ |
+| Python版本要求 | `build_scripts/PYTHON_VERSION_REQUIREMENT.md` | Python 3.11 安装指南 |
+| 快速参考 | `build_scripts/QUICK_REFERENCE.md` | 命令速查 |
+| 打包检查清单 | `build_scripts/PACKAGING_CHECKLIST.md` | 构建前检查 |
+
+---
+
+## ⚠️ 重要提示
+
+1. **Python 版本**: 必须使用 3.11，构建脚本会检查
+2. **构建环境**: Windows 必须在 VS Dev Shell 中构建
+3. **清理旧依赖**: 如果之前有安装 PyQt5/PySide6，建议卸载以节省空间
+
+---
+
+## 🔗 相关链接
+
+- Python 3.11: https://www.python.org/downloads/release/python-3119/
+- Visual Studio Build Tools: https://visualstudio.microsoft.com/visual-cpp-build-tools/
+
+---
+
+**最后更新**: 2024年

EG

@@ -1 +0,0 @@
-Subproject commit 69e2bda47e9713705ad5c45a08b6fc643a2b51f6

RenderPipelineFile/config/plugins.yaml

@@ -13,6 +13,7 @@ enabled:
     - scattering
     - skin_shading
     - sky_ao
+    - selection_outline
     - smaa
     - ssr
     # - clouds

RenderPipelineFile/config/stages.yaml

@@ -63,5 +63,6 @@ global_stage_order:
 
   # Finishing stages, do not insert anything below
   - UpscaleStage
+  - SelectionOutlineStage
   - FinalStage
   - UpdatePreviousPipesStage

RenderPipelineFile/rpcore/native/__init__.py

@@ -40,7 +40,22 @@ from rpcore.rpobject import RPObject
 NATIVE_CXX_LOADED = False
 
 # Read the configuration from the flag-file
+# Fix for Nuitka/PyInstaller: try multiple methods to find the flag file
+import os
 current_path = dirname(realpath(__file__))
+
+# Nuitka standalone mode fix: if __file__ is empty or invalid, use sys.argv[0]
+if not current_path or not isfile(join(current_path, "use_cxx.flag")):
+    # Try relative to executable
+    exe_dir = dirname(realpath(sys.argv[0]))
+    current_path = join(exe_dir, "RenderPipelineFile", "rpcore", "native")
+    
+    # Try one more fallback: check environment variable
+    if not isfile(join(current_path, "use_cxx.flag")):
+        rp_path = os.environ.get("RENDER_PIPELINE_PATH", "")
+        if rp_path:
+            current_path = join(rp_path, "rpcore", "native")
+
 cxx_flag_path = join(current_path, "use_cxx.flag")
 if not isfile(cxx_flag_path):
     RPObject.global_error("CORE", "Could not find cxx flag, please run the setup.py!")

RenderPipelineFile/rpplugins/selection_outline/__init__.py

@@ -0,0 +1,2 @@
+"""Selection outline plugin package."""
+

RenderPipelineFile/rpplugins/selection_outline/config.yaml

@@ -0,0 +1,3 @@
+settings:
+daytime_settings:
+

RenderPipelineFile/rpplugins/selection_outline/plugin.py

@@ -0,0 +1,15 @@
+from rpcore.pluginbase.base_plugin import BasePlugin
+
+from .selection_outline_stage import SelectionOutlineStage
+
+
+class Plugin(BasePlugin):
+
+    name = "Selection Outline"
+    author = "EG Team"
+    description = "Adds Unity-style selected-object outline as a post-process stage."
+    version = "1.0"
+
+    def on_stage_setup(self):
+        self.stage = self.create_stage(SelectionOutlineStage)
+

RenderPipelineFile/rpplugins/selection_outline/selection_outline_stage.py

@@ -0,0 +1,68 @@
+from panda3d.core import PNMImage, SamplerState, Texture, Vec4
+
+from rpcore.render_stage import RenderStage
+
+
+class SelectionOutlineStage(RenderStage):
+
+    required_pipes = ["ShadedScene"]
+
+    def __init__(self, pipeline):
+        RenderStage.__init__(self, pipeline)
+        self._enabled = False
+        self._outline_color = Vec4(1.0, 0.55, 0.0, 1.0)
+        self._outline_width = 2.0
+        self._fill_alpha = 0.0
+        self._mask_tex = self._make_default_mask()
+
+    @property
+    def produced_pipes(self):
+        return {"ShadedScene": self.target.color_tex}
+
+    def create(self):
+        self.target = self.create_target("SelectionOutline")
+        self.target.add_color_attachment(bits=16)
+        self.target.prepare_buffer()
+        self._apply_inputs()
+
+    def reload_shaders(self):
+        self.target.shader = self.load_plugin_shader("selection_outline.frag.glsl")
+        self._apply_inputs()
+
+    def set_mask_texture(self, mask_texture):
+        self._mask_tex = mask_texture if mask_texture else self._make_default_mask()
+        self._apply_inputs()
+
+    def set_enabled_outline(self, enabled):
+        self._enabled = bool(enabled)
+        self._apply_inputs()
+
+    def set_outline_style(self, color=None, width_px=None, fill_alpha=None):
+        if color is not None:
+            self._outline_color = Vec4(color)
+        if width_px is not None:
+            self._outline_width = max(0.0, float(width_px))
+        if fill_alpha is not None:
+            self._fill_alpha = max(0.0, min(1.0, float(fill_alpha)))
+        self._apply_inputs()
+
+    def _apply_inputs(self):
+        enabled_val = 1.0 if self._enabled else 0.0
+        self.target.set_shader_input("SelectionMaskTex", self._mask_tex)
+        self.target.set_shader_input("SelectionOutlineEnabled", enabled_val)
+        self.target.set_shader_input("SelectionOutlineColor", self._outline_color)
+        self.target.set_shader_input("SelectionOutlineWidth", self._outline_width)
+        self.target.set_shader_input("SelectionFillAlpha", self._fill_alpha)
+
+    def _make_default_mask(self):
+        image = PNMImage(1, 1, 4)
+        image.fill(0.0, 0.0, 0.0)
+        image.alpha_fill(0.0)
+        texture = Texture("selection_outline_default_mask")
+        texture.load(image)
+        texture.set_minfilter(SamplerState.FT_nearest)
+        texture.set_magfilter(SamplerState.FT_nearest)
+        texture.set_wrap_u(SamplerState.WM_clamp)
+        texture.set_wrap_v(SamplerState.WM_clamp)
+        return texture
+

RenderPipelineFile/rpplugins/selection_outline/shader/selection_outline.frag.glsl

@@ -0,0 +1,59 @@
+/**
+ * Selection outline post process stage.
+ */
+
+#version 430
+
+#pragma include "render_pipeline_base.inc.glsl"
+
+uniform sampler2D ShadedScene;
+uniform sampler2D SelectionMaskTex;
+
+uniform float SelectionOutlineEnabled;
+uniform vec4 SelectionOutlineColor;
+uniform float SelectionOutlineWidth;   // pixels
+uniform float SelectionFillAlpha;      // 0..1
+
+out vec4 result;
+
+float sample_mask(vec2 uv) {
+    return textureLod(SelectionMaskTex, uv, 0).r;
+}
+
+void main() {
+    vec2 uv = get_texcoord();
+    vec3 scene_col = textureLod(ShadedScene, uv, 0).rgb;
+
+    if (SelectionOutlineEnabled < 0.5) {
+        result = vec4(scene_col, 1.0);
+        return;
+    }
+
+    vec2 texel = vec2(max(1.0, SelectionOutlineWidth)) / SCREEN_SIZE;
+    float center = sample_mask(uv);
+
+    float max_nei = 0.0;
+    max_nei = max(max_nei, sample_mask(uv + vec2( texel.x, 0.0)));
+    max_nei = max(max_nei, sample_mask(uv + vec2(-texel.x, 0.0)));
+    max_nei = max(max_nei, sample_mask(uv + vec2(0.0,  texel.y)));
+    max_nei = max(max_nei, sample_mask(uv + vec2(0.0, -texel.y)));
+    max_nei = max(max_nei, sample_mask(uv + vec2( texel.x,  texel.y)));
+    max_nei = max(max_nei, sample_mask(uv + vec2( texel.x, -texel.y)));
+    max_nei = max(max_nei, sample_mask(uv + vec2(-texel.x,  texel.y)));
+    max_nei = max(max_nei, sample_mask(uv + vec2(-texel.x, -texel.y)));
+
+    // Outer contour only.
+    float edge = clamp(max_nei - center, 0.0, 1.0);
+    float fill = center * SelectionFillAlpha;
+
+    vec3 col = scene_col;
+    if (fill > 0.0) {
+        col = mix(col, SelectionOutlineColor.rgb, fill * SelectionOutlineColor.a);
+    }
+    if (edge > 0.0) {
+        col = mix(col, SelectionOutlineColor.rgb, edge * SelectionOutlineColor.a);
+    }
+
+    result = vec4(col, 1.0);
+}
+

Start_Run.py

@@ -1,8 +1,11 @@
 import os
 import sys
+import traceback
+from datetime import datetime
 
-# 添加项目根目录到 Python 路径
+# 添加项目根目录到 Python 路径。打包后可通过环境变量覆盖，
-project_root = os.path.dirname(os.path.abspath(__file__))
+# 以适配 Windows dist / Linux AppDir 等不同目录布局。
+project_root = os.environ.get("EG_PROJECT_ROOT") or os.path.dirname(os.path.abspath(__file__))
 sys.path.insert(0, project_root)
 
 # 设置工作目录为项目根目录
@@ -20,30 +23,34 @@ sys.path.insert(0, render_pipeline_file_path)
 icons_path = os.path.join(project_root, "icons")
 sys.path.insert(0, icons_path)
 
+
+def _write_startup_log(message: str) -> None:
+    """Write startup diagnostics next to the executable/source root."""
+    log_path = os.path.join(project_root, "eg_startup.log")
+    with open(log_path, "a", encoding="utf-8") as handle:
+        handle.write(f"\n[{datetime.now().isoformat(timespec='seconds')}] {message}\n")
+
 # 现在可以导入并运行主程序
 if __name__ == "__main__":
-    args = sys.argv[1:]
+    try:
-    # args = "/home/tiger/桌面/Test1"
+        args = sys.argv[1:]
-    # args = "C:/Users/29381/Desktop/1"
+        _write_startup_log(f"Launch args: {args}")
-    print(f'Path is {args}')
+        _write_startup_log(f"Working directory: {os.getcwd()}")
-    # 将整个列表转换为字符串（包括方括号）
-    args_str = ''.join(args)
 
-    from main import MyWorld
+        # 将整个列表转换为字符串（包括方括号）
-    if args:
+        args_str = ''.join(args)
-        # print(f"[DEBUG] 启动时传入项目路径: {args_str}")
+
-        # 创建应用实例
+        from main import MyWorld
-        app = MyWorld()
+
-        # 如果传入了项目路径，尝试打开项目
+        if args:
-        if hasattr(app, 'project_manager'):
+            app = MyWorld()
-            # print(f"[DEBUG] 尝试打开项目: {args_str}")
+            if hasattr(app, 'project_manager'):
-            success = app.project_manager.openProject(args_str)
+                app.project_manager.openProject(args_str)
-            # print(f"[DEBUG] 项目打开结果: {success}")
+            app.run()
         else:
-            # print(f"[DEBUG] 项目管理器未初始化")
+            app = MyWorld()
-            pass
+            app.run()
-        app.run()
+    except Exception:
-    else:
+        _write_startup_log("Unhandled exception during startup:")
-        # print(f"[DEBUG] 无项目路径，正常启动")
+        _write_startup_log(traceback.format_exc())
-        app = MyWorld()
+        raise
-        app.run()

build_scripts/BUILD_GUIDE.md

@@ -0,0 +1,393 @@
+# 元泰 EG 构建完整指南
+
+本文档提供从零开始构建元泰 EG 安装程序的完整步骤。
+
+---
+
+## 📋 环境要求
+
+### 必须条件
+
+| 组件 | 版本 | 说明 |
+|------|------|------|
+| **Python** | **3.11.x** | 必须使用 3.11，其他版本不保证兼容 |
+| Windows | 10/11 | 64位系统 |
+| 磁盘空间 | 5GB+ | 构建需要大量临时文件 |
+| 内存 | 4GB+ | 建议 8GB |
+
+### 检查 Python 版本
+
+```powershell
+# 检查版本
+python --version
+# 必须显示: Python 3.11.x
+
+# 使用项目提供的检查脚本
+python build_scripts/check_python_version.py
+```
+
+如果版本不对，请参考 `build_scripts/PYTHON_VERSION_REQUIREMENT.md` 安装正确版本。
+
+---
+
+## 🚀 快速开始（5分钟）
+
+### 步骤 1: 安装依赖
+
+```powershell
+# 确保是 Python 3.11
+python --version
+
+# 安装最小依赖（推荐）
+python -m pip install -r requirements/requirements-minimal.txt
+```
+
+### 步骤 2: 验证安装
+
+```powershell
+# 测试程序能否运行
+python Start_Run.py
+```
+
+### 步骤 3: 构建安装程序
+
+```powershell
+# 进入 VS Dev Shell（必须有 C++ 编译器）
+# 然后执行构建脚本
+.\build_scripts\build_windows.ps1 -Version "1.0.0"
+```
+
+---
+
+## 📌 打包原则
+
+从 2026-03 的修复开始，项目采用“保守打包优先”的策略：
+
+- 优先保证发行版可运行，不先追求安装包最小化
+- 对 `RenderPipeline`、动态插件、文件路径动态导入使用显式白名单
+- 将 `RenderPipelineFile`、`Resources`、`config`、`demo` 等目录视为运行时资源，整体复制
+- 保留启动日志 `eg_startup.log`，用于定位打包后启动异常
+
+详细背景和源码整改建议请参考：
+
+- `build_scripts/PACKAGING_RISK_AUDIT.md`
+
+---
+
+## 📦 依赖说明
+
+### 最小依赖（requirements-minimal.txt）
+
+只包含程序运行必需的依赖：
+
+```
+Panda3D==1.10.16          # 3D 引擎核心
+panda3d-frame             # Panda3D 扩展
+panda3d-imgui==1.1.0      # ImGui 集成
+imgui-bundle==1.92.4      # ImGui 绑定
+openvr==2.2.0             # VR 支持
+Pillow==11.3.0            # 图像处理
+PyYAML==5.4.1             # 配置解析
+psutil==5.9.0             # 系统工具
+```
+
+### 已移除的依赖
+
+以下依赖已从项目中移除，节省约 **170MB**：
+
+| 依赖 | 节省空间 | 原因 |
+|------|---------|------|
+| PyQt5 | ~50 MB | 未使用 |
+| PySide6 | ~100 MB | 未使用 |
+| PyQt5-Qt5, PyQt5_sip 等 | ~20 MB | 配套库 |
+
+### 清理旧依赖
+
+如果之前安装过旧依赖，建议清理：
+
+```powershell
+# 卸载未使用的 Qt 库
+python -m pip uninstall -y PyQt5 PyQt5-Qt5 PyQt5_sip PySide6 PySide6_Addons PySide6_Essentials shiboken6
+
+# 清理 pip 缓存
+python -m pip cache purge
+```
+
+---
+
+## 🔧 详细构建步骤
+
+### 第一步：环境准备
+
+1. **安装 Python 3.11**
+   - 下载：https://www.python.org/downloads/release/python-3119/
+   - 安装时勾选 "Add Python to PATH"
+
+2. **安装 Visual Studio Build Tools**（Windows 必需）
+   - 下载：https://visualstudio.microsoft.com/visual-cpp-build-tools/
+   - 选择 "使用 C++ 的桌面开发"
+   - 或安装完整 Visual Studio 2022
+
+3. **验证环境**
+   ```powershell
+   # 检查 Python
+   python --version  # Python 3.11.x
+   
+   # 检查编译器（在 VS Dev Shell 中）
+   cl  # 应显示版本信息
+   ```
+
+### 第二步：安装项目依赖
+
+```powershell
+# 进入项目目录
+cd C:\path\to\EG
+
+# 安装依赖
+python -m pip install -r requirements/requirements-minimal.txt
+
+# 验证安装
+python -c "import panda3d.core; import imgui_bundle; print('✓ 依赖安装成功')"
+```
+
+### 第三步：运行测试
+
+```powershell
+# 测试程序能否正常启动
+python Start_Run.py
+```
+
+### 第四步：打开 VS Dev Shell
+
+**什么是 VS Dev Shell？**
+VS Dev Shell（Visual Studio 开发人员命令提示符）是一个包含 C++ 编译器（cl.exe）的特殊命令行窗口，Nuitka 需要它来编译 Python 代码为机器码。
+
+**如何打开 VS Dev Shell：**
+
+#### 方法 1: 通过开始菜单（推荐）
+
+1. 按 `Win` 键打开开始菜单
+2. 搜索：`Developer Command Prompt`
+   - 或中文：`开发人员命令提示符`
+3. 点击 **"Developer Command Prompt for VS 2022"**
+4. 建议右键选择 **"以管理员身份运行"**
+
+#### 方法 2: 通过 Visual Studio
+
+1. 打开 **Visual Studio 2022**
+2. 点击菜单栏的 **"工具"** → **"命令行"** → **"开发者命令提示符"**
+
+#### 方法 3: 使用 PowerShell 命令
+
+```powershell
+# 如果知道 VS 安装路径
+& "C:\Program Files\Microsoft Visual Studio\2022\Community\Common7\Tools\Launch-VsDevShell.ps1" -Arch amd64
+```
+
+**验证 VS Dev Shell 已正确打开：**
+
+```powershell
+# 在 VS Dev Shell 中执行
+cl
+
+# 应该显示类似：
+# 用于 x64 的 Microsoft (R) C/C++ 优化编译器 19.xx.xxxxx 版
+```
+
+### 第五步：构建安装程序
+
+在 **VS Dev Shell** 中执行：
+
+#### 方法 A：使用构建脚本（推荐）
+
+```powershell
+# 1. 进入项目目录
+cd C:\Users\Tellme\apps\EG
+
+# 2. 切换到 PowerShell（如果当前是 cmd）
+powershell
+
+# 3. 执行构建
+.\build_scripts\build_windows.ps1 -Version "1.0.0"
+```
+
+该脚本已包含以下关键修复：
+
+- 修正 `RenderPipeline` 在打包后可能找错 `config/pipeline.yaml` 的问题
+- 显式包含 `rpcore`、`rplibs`、`rpplugins`
+- 自动探测并尽量纳入 `plugins`、`gltf`、`playwright`、`PyQt5` 等可选包
+- 自动复制运行时资源目录
+
+#### 方法 B：手动构建
+
+```powershell
+# 在 VS Dev Shell 中
+cd C:\path\to\EG
+
+python -m nuitka `
+    --standalone `
+    --include-package=panda3d `
+    --include-package=direct `
+    --include-package=imgui_bundle `
+    --include-package=p3dimgui `
+    --include-package=openvr `
+    --include-package=RenderPipelineFile `
+    --include-package=rpcore `
+    --include-package=rplibs `
+    --include-package=rpplugins `
+    --follow-import-to=rpcore `
+    --follow-import-to=rpplugins `
+    --include-data-files="$env:APPDATA\Python\Python311\site-packages\imgui_bundle\glfw3.dll=imgui_bundle\glfw3.dll" `
+    --windows-icon-from-ico=icons/app.ico `
+    --windows-disable-console `
+    --output-dir=build/nuitka `
+    --jobs=4 `
+    Start_Run.py
+```
+
+说明：
+
+- 当前更推荐 `standalone + 资源目录复制` 的方式，而不是 `onefile`
+- `RenderPipeline`、动态插件和外部资源目录在 `onefile` 下更容易出现路径问题
+
+### 第五步：打包资源
+
+构建完成后，手动复制资源文件：
+
+```powershell
+# 创建输出目录
+$Version = "1.0.0"
+New-Item -ItemType Directory -Force -Path "dist\EG_$Version" | Out-Null
+
+# 复制可执行文件
+Copy-Item "build\nuitka\Start_Run.exe" "dist\EG_$Version\元泰.exe" -Force
+
+# 复制资源目录
+$dirs = @("config", "icons", "Resources", "tex", "templates",
+         "core", "gui", "project", "scene", "scripts",
+         "ssbo_component", "tools", "TransformGizmo", "ui",
+         "RenderPipelineFile", "vr_actions", "new")
+
+foreach ($dir in $dirs) {
+    Copy-Item $dir "dist\EG_$Version\$dir" -Recurse -Force
+}
+
+Write-Host "✓ 打包完成: dist\EG_$Version"
+```
+
+如果使用 `build_windows.ps1`，通常不需要再手动执行这一段，因为脚本已经自动复制资源。
+
+### Linux 构建说明
+
+Linux 构建脚本：
+
+```bash
+./build_scripts/build_linux.sh "1.0.0"
+```
+
+Linux 脚本已按与 Windows 相同的思路补强：
+
+- 使用保守白名单显式包含核心包
+- 自动探测并尽量纳入可选依赖
+- 整体复制运行时资源目录
+- 在 AppRun 中设置 `EG_PROJECT_ROOT`
+- 让 AppImage 内的程序从真实资源根目录启动
+
+---
+
+## 📁 构建输出
+
+构建完成后，输出目录结构：
+
+```
+dist/EG_1.0.0/
+├── 元泰.exe              # 主程序 (~120MB)
+├── config/               # 配置文件
+├── icons/                # 图标资源
+├── Resources/            # 模型、材质等资源
+├── RenderPipelineFile/   # 渲染管线
+├── ...                   # 其他数据目录
+└── (无 .py 源代码)       # 已编译为机器码
+```
+
+---
+
+## 🐛 常见问题
+
+### Q1: Python 版本错误
+
+**错误**：`必须使用 Python 3.11，当前版本是 3.13.5`
+
+**解决**：
+```powershell
+# 安装 Python 3.11
+# 参考 build_scripts/PYTHON_VERSION_REQUIREMENT.md
+```
+
+### Q2: 找不到 C++ 编译器
+
+**错误**：`未找到 Visual C++ 编译器`
+
+**解决**：
+1. 安装 Visual Studio Build Tools
+2. 在 **VS Dev Shell** 中运行构建
+
+### Q3: 缺少 DLL
+
+**错误**：`Cannot find glfw3.dll`
+
+**解决**：确保 `--include-data-files` 参数正确指向 DLL 路径：
+```powershell
+# 注意 Python 版本路径
+# Python 3.11: ...\Python311\...
+# Python 3.13: ...\Python313\...
+```
+
+### Q4: ModuleNotFoundError: rpcore
+
+**错误**：`No module named 'rpcore'`
+
+**解决**：确保使用 `--follow-import-to=rpcore`
+
+### Q5: 找不到 pipeline.yaml / rpplugins.*
+
+**错误示例**：
+
+- `Failed to load YAML file: File not found`
+- `No such file or directory: /$$rpconfig/pipeline.yaml`
+- `ModuleNotFoundError: No module named 'rpplugins.ao'`
+
+**原因**：
+
+- `RenderPipeline` 在打包后可能把基路径识别错
+- 插件通过配置文件动态加载，打包器不会稳定自动发现
+
+**解决**：
+
+- 使用仓库中的最新 `build_windows.ps1` / `build_linux.sh`
+- 确保 `RenderPipelineFile` 被整体复制
+- 确保 `rpcore`、`rplibs`、`rpplugins` 被显式纳入打包
+
+### Q6: 双击后闪退，没有控制台输出
+
+**解决**：
+
+- 查看发行目录中的 `eg_startup.log`
+- 该文件会记录启动阶段 traceback
+- 遇到打包问题时，优先保留这个日志进行排查
+
+---
+
+## 📚 相关文档
+
+| 文档 | 用途 |
+|------|------|
+| `build_scripts/README.md` | build_scripts 目录说明 |
+| `build_scripts/ICON_GUIDE.md` | 图标准备 |
+| `build_scripts/PACKAGING_RISK_AUDIT.md` | 打包风险与整改建议 |
+| `build_scripts/PYTHON_VERSION_REQUIREMENT.md` | Python 版本要求 |
+| `build_scripts/PACKAGING_CHECKLIST.md` | 打包检查清单 |
+
+---
+
+**最后更新**: 2024年

build_scripts/ICON_GUIDE.md

@@ -0,0 +1,285 @@
+# EG 图标准备指南
+
+构建安装程序需要准备应用程序图标，本指南说明如何创建和放置图标文件。
+
+---
+
+## 📋 所需图标
+
+| 平台 | 文件名 | 格式 | 推荐尺寸 | 用途 |
+|------|--------|------|---------|------|
+| Windows | `icons/app.ico` | ICO | 256x256 (多尺寸) | 可执行文件和安装程序图标 |
+| Linux | `icons/app.png` | PNG | 256x256 | AppImage 和桌面图标 |
+
+---
+
+## 🎨 图标要求
+
+### Windows ICO 文件
+
+ICO 文件可以包含多个尺寸的图标：
+- **必须包含**: 256x256 (32位)
+- **推荐包含**: 48x48, 32x32, 16x16
+- **可选**: 128x128, 64x64
+
+**格式要求**:
+- 支持透明背景 (32位，带 Alpha 通道)
+- 大图标使用 PNG 压缩
+- 小图标使用 BMP 格式
+
+### Linux PNG 文件
+
+- **尺寸**: 256x256 像素
+- **格式**: PNG，带透明通道
+- **色彩**: RGBA (32位)
+
+---
+
+## 🔧 创建图标的方法
+
+### 方法 1: 使用现有图片转换
+
+如果你有 PNG/JPG 格式的 logo：
+
+**Windows (ICO)**:
+```bash
+# 使用 ImageMagick
+convert logo.png -define icon:auto-resize=256,128,64,48,32,16 icons/app.ico
+
+# 或在线工具
+# https://convertio.co/png-ico/
+# https://icoconvert.com/
+```
+
+**Linux (PNG)**:
+```bash
+# 直接复制或调整大小
+cp logo.png icons/app.png
+
+# 或使用 ImageMagick 调整
+convert logo.png -resize 256x256 icons/app.png
+```
+
+### 方法 2: 使用专业工具设计
+
+| 工具 | 平台 | 类型 | 链接 |
+|------|------|------|------|
+| Adobe Illustrator | Win/Mac | 专业矢量 | - |
+| Figma | Web | 免费在线 | figma.com |
+| Canva | Web | 免费在线 | canva.com |
+| GIMP | Win/Mac/Linux | 免费开源 | gimp.org |
+| Inkscape | Win/Mac/Linux | 免费矢量 | inkscape.org |
+
+### 方法 3: 使用 Python 脚本生成
+
+如果你有源图片，可以使用 Python 生成：
+
+```python
+#!/usr/bin/env python3
+"""图标生成脚本"""
+
+from PIL import Image
+import os
+
+# 源图片路径
+source_image = "source_logo.png"  # 你的源图片
+output_dir = "icons"
+
+# 确保输出目录存在
+os.makedirs(output_dir, exist_ok=True)
+
+# 打开源图片
+img = Image.open(source_image)
+
+# 转换为 RGBA 模式 (确保透明通道)
+if img.mode != 'RGBA':
+    img = img.convert('RGBA')
+
+# 生成 Windows ICO (多尺寸)
+icon_sizes = [(16, 16), (32, 32), (48, 48), (64, 64), (128, 128), (256, 256)]
+ico_images = []
+
+for size in icon_sizes:
+    resized = img.resize(size, Image.Resampling.LANCZOS)
+    ico_images.append(resized)
+
+# 保存 ICO (第一个图像作为根，其余作为子图标)
+ico_images[0].save(
+    os.path.join(output_dir, "app.ico"),
+    format='ICO',
+    sizes=[(i.width, i.height) for i in ico_images],
+    append_images=ico_images[1:]
+)
+
+print(f"✓ 生成: {output_dir}/app.ico")
+
+# 生成 Linux PNG (256x256)
+png_size = (256, 256)
+png_img = img.resize(png_size, Image.Resampling.LANCZOS)
+png_img.save(os.path.join(output_dir, "app.png"), format='PNG')
+
+print(f"✓ 生成: {output_dir}/app.png")
+
+print("\n图标生成完成！")
+```
+
+**使用方法**:
+```bash
+# 安装依赖
+pip install Pillow
+
+# 运行脚本
+python generate_icons.py
+```
+
+### 方法 4: 从项目中提取
+
+如果项目中有现有图标资源：
+
+```bash
+# 查找项目中的图标文件
+find . -name "*.ico" -o -name "*.png" | grep -i icon
+
+# 复制到正确位置
+cp 找到的图标文件 icons/app.ico
+cp 找到的图标文件 icons/app.png
+```
+
+---
+
+## 📁 目录结构
+
+放置图标后，项目结构应为：
+
+```
+EG/
+├── icons/
+│   ├── app.ico          # Windows 图标 (必需)
+│   ├── app.png          # Linux 图标 (必需)
+│   └── ...              # 其他图标资源
+├── build_scripts/
+├── ...
+```
+
+---
+
+## ✅ 验证图标
+
+### Windows
+
+```powershell
+# 检查文件存在
+Test-Path icons/app.ico
+
+# 查看 ICO 内容 (需要工具)
+# 使用 Resource Hacker: http://www.angusj.com/resourcehacker/
+```
+
+### Linux
+
+```bash
+# 检查文件存在
+ls -la icons/app.png
+
+# 查看图片信息
+file icons/app.png
+identify icons/app.png  # 需要 ImageMagick
+
+# 预览
+xdg-open icons/app.png
+```
+
+---
+
+## 🚀 测试图标
+
+构建完成后，验证图标是否正确显示：
+
+### Windows
+
+1. **可执行文件图标**:
+   ```powershell
+   # 检查 dist/EG_1.0.0/EG.exe 的图标
+   # 在资源管理器中查看，应该显示你的图标
+   ```
+
+2. **安装程序图标**:
+   ```powershell
+   # 检查 dist/EG_Setup_v1.0.0_x64.exe 的图标
+   ```
+
+### Linux
+
+```bash
+# 检查 AppImage 图标
+./dist/EG-1.0.0-x86_64.AppImage --appimage-mount
+ls squashfs-root/EG.png
+```
+
+---
+
+## 🐛 常见问题
+
+### 问题 1: ICO 文件无法识别
+
+**症状**: Windows 显示默认图标
+
+**解决**:
+- 确保 ICO 包含 256x256 尺寸
+- 使用专业工具重新生成
+- 检查文件是否损坏
+
+### 问题 2: Linux 图标不显示
+
+**症状**: AppImage 使用默认图标
+
+**解决**:
+- 确认 `icons/app.png` 存在
+- 检查文件权限: `chmod 644 icons/app.png`
+- 确认 PNG 格式正确: `file icons/app.png`
+
+### 问题 3: 图标模糊
+
+**症状**: 图标显示模糊或锯齿
+
+**解决**:
+- 使用矢量源文件生成
+- 确保包含多个尺寸
+- 使用高质量缩放算法 (LANCZOS)
+
+---
+
+## 📦 快速开始
+
+如果你没有现成的图标，可以先使用占位图标：
+
+```bash
+# Windows: 创建一个简单的 ICO
+# 使用在线工具生成，或先跳过 (Nuitka 会使用默认图标)
+
+# Linux: 创建一个简单的 PNG
+python3 -c "
+from PIL import Image
+import os
+os.makedirs('icons', exist_ok=True)
+img = Image.new('RGBA', (256, 256), (100, 150, 255, 255))
+img.save('icons/app.png')
+print('Created placeholder icon')
+"
+```
+
+**注意**: 发布前务必替换为正式图标！
+
+---
+
+## 🎨 设计建议
+
+1. **简洁明了**: 小尺寸下依然可识别
+2. **品牌一致性**: 与应用风格一致
+3. **透明背景**: 适应不同主题
+4. **测试多尺寸**: 确保 16x16 到 256x256 都清晰
+5. **考虑暗色主题**: 在深色背景下也可见
+
+---
+
+**文档版本**: 1.0

build_scripts/PACKAGING_CHECKLIST.md

@@ -0,0 +1,294 @@
+# EG 项目打包清单
+
+本文档详细说明安装程序中包含的内容，以及打包决策依据。
+
+---
+
+## 📦 打包内容总览
+
+### 代码部分 (由 Nuitka 编译)
+
+| 入口文件 | 处理方式 | 说明 |
+|---------|---------|------|
+| `Start_Run.py` | 编译为主程序 | 应用入口 |
+| `main.py` | 被导入编译 | 主程序逻辑 |
+| `core/` | 编译进可执行文件 | 核心模块 |
+| `gui/` | 编译进可执行文件 | GUI模块 |
+| `project/` | 编译进可执行文件 | 项目管理 |
+| `scene/` | 编译进可执行文件 | 场景管理 |
+| `ssbo_component/` | 编译进可执行文件 | SSBO组件 |
+| `tools/` | 编译进可执行文件 | 工具模块 |
+| `TransformGizmo/` | 编译进可执行文件 | 变换工具 |
+| `ui/` | 编译进可执行文件 | UI模块 |
+
+### 数据资源 (复制到输出目录)
+
+| 目录/文件 | 用途 | 必要性 |
+|----------|------|--------|
+| `config/` | 配置文件 | ✅ 必须 |
+| `icons/` | 图标资源 | ✅ 必须 |
+| `RenderPipelineFile/` | 渲染管线 | ✅ 必须 |
+| `Resources/` | 资源文件 | ✅ 必须 |
+| `scripts/` | 脚本系统 | ✅ 必须 |
+| `templates/` | 项目模板 | ✅ 必须 |
+| `tex/` | 纹理资源 | ✅ 必须 |
+| `vr_actions/` | VR配置 | ✅ 必须 |
+| `imgui.ini` | ImGui配置 | ✅ 必须 |
+
+---
+
+## ⚠️ 需要确认的问题
+
+### 问题 1: `project/` 目录
+
+**现状**: 当前配置中 **未包含** `project/` 目录
+
+**分析**:
+
+- `project/` 包含 `project_manager.py`, `webgl_packager.py`, `__init__.py`
+- `main.py` 中导入: `from project.project_manager import ProjectManager`
+
+**结论**: ⚠️ **必须包含！** 否则项目管理功能无法工作
+
+**建议操作**: 添加到 `DATA_DIRECTORIES`
+
+```python
+DATA_DIRECTORIES = [
+    # ... 现有目录 ...
+    "project",  # 添加此行
+]
+```
+
+---
+
+### 问题 2: `new/` 目录
+
+**现状**: 当前配置中 **未包含**
+
+**分析**:
+
+- 包含 `project.json` (空项目模板)
+- 可能是"新建项目"的默认模板
+
+**结论**: ✅ **建议包含** 作为默认项目模板
+
+**建议操作**: 添加到 `DATA_DIRECTORIES`
+
+---
+
+### 问题 3: `test_project/` 目录
+
+**现状**: 当前配置中 **未包含**
+
+**分析**:
+
+- 包含示例 `project.json`
+- 可用于演示功能
+
+**选项**:
+
+1. **不包含** - 减小安装包体积
+2. **包含为可选组件** - 用户可选择安装
+3. **包含** - 方便新用户学习
+
+**建议**: 选项 1 或 3，根据安装包大小决定
+
+---
+
+### 问题 4: Python 缓存文件
+
+**现状**: 已排除 `__pycache__`, `*.pyc`, `*.pyo`, `*.pyd`
+
+**分析**: ✅ 正确，这些是由 Python 运行时生成的缓存，不需要分发
+
+---
+
+### 问题 5: 开发环境文件
+
+**现状**: 已排除 `.git`, `.gitignore`, `.idea`, `.vscode`
+
+**分析**: ✅ 正确，这些是开发环境配置，用户不需要
+
+---
+
+### 问题 6: 构建脚本
+
+**现状**: `build_scripts/` 未包含
+
+**分析**: ✅ 正确，构建脚本不需要分发给最终用户
+
+---
+
+### 问题 7: 依赖文件
+
+**现状**: `requirements/` 未包含
+
+**分析**:
+
+- 最终用户不需要 `requirements.txt`（依赖已编译进可执行文件）
+- 但如果提供 SDK/开发套件，可能需要
+
+**结论**: ✅ 当前处理正确
+
+---
+
+## 🔧 修正后的配置
+
+### 更新 DATA_DIRECTORIES
+
+```python
+# ==================== 数据文件配置 ====================
+# 需要包含的目录 (相对于项目根目录)
+DATA_DIRECTORIES = [
+    # 配置和资源
+    "config",
+    "icons",
+    "Resources",
+    "tex",
+    "templates",
+    
+    # 核心模块 (这些目录中的 .py 文件会被编译，但可能包含非 .py 资源)
+    "core",              # 核心功能
+    "gui",               # GUI相关
+    "project",           # ⚠️ 新增: 项目管理 (包含非代码资源)
+    "scene",             # 场景管理
+    "scripts",           # 脚本系统
+    "ssbo_component",    # SSBO编辑器
+    "tools",             # 工具模块
+    "TransformGizmo",    # 变换Gizmo
+    "ui",                # UI模块
+    
+    # 引擎和扩展
+    "RenderPipelineFile", # 渲染管线
+    "vr_actions",         # VR配置
+    
+    # 示例和模板
+    "new",                # ⚠️ 新增: 默认项目模板
+    # "test_project",     # 可选: 示例项目
+]
+```
+
+### Nuitka 包含包确认
+
+当前配置:
+
+```python
+"include_packages": [
+    "Panda3D",
+    "direct",
+    "imgui_bundle",
+    "p3dimgui",
+    "openvr",
+    "PIL",
+    "yaml",
+    "psutil",
+],
+```
+
+**需要添加**:
+
+- 检查 `requirements.txt` 中的其他依赖是否需要显式包含
+
+---
+
+## 📊 打包体积预估
+
+| 目录 | 估算大小 | 说明 |
+|------|---------|------|
+| `RenderPipelineFile/` | ~50-100 MB | 渲染管线核心 |
+| `Resources/` | 可变 | 用户资源 |
+| `tex/` | ~10-50 MB | 默认纹理 |
+| `icons/` | ~1-5 MB | 图标 |
+| 其他代码目录 | ~5-10 MB | Python 源文件(编译前) |
+| **总计** | **~100-200 MB** | 不含用户资源 |
+
+---
+
+## ✅ 打包检查清单
+
+### 构建前请确认
+
+#### 目录配置
+
+- [x] `config/` - 配置目录 ✅
+- [x] `core/` - 核心模块 ✅  
+- [x] `gui/` - GUI模块 ✅
+- [x] `icons/` - 图标资源 ✅
+- [x] `project/` - 项目管理 ✅ **(已添加)**
+- [x] `RenderPipelineFile/` - 渲染管线 ✅
+- [x] `Resources/` - 资源文件 ✅
+- [x] `scene/` - 场景管理 ✅
+- [x] `scripts/` - 脚本系统 ✅
+- [x] `ssbo_component/` - SSBO组件 ✅
+- [x] `templates/` - 模板 ✅
+- [x] `tex/` - 纹理 ✅
+- [x] `tools/` - 工具 ✅
+- [x] `TransformGizmo/` - 变换工具 ✅
+- [x] `ui/` - UI模块 ✅
+- [x] `vr_actions/` - VR配置 ✅
+- [x] `new/` - 默认项目模板 ✅ **(已添加)**
+
+#### 文件检查
+
+- [ ] `icons/app.ico` - Windows图标 ⚠️ **需要添加**
+- [ ] `icons/app.png` - Linux图标 ⚠️ **需要添加**
+- [x] `imgui.ini` - ImGui配置 ✅
+- [x] `Start_Run.py` - 入口脚本 ✅
+- [x] `main.py` - 主程序 ✅
+
+#### 清理检查
+
+- [x] 排除了所有 `__pycache__` 和 `.pyc` 文件 ✅
+- [x] 排除了开发环境文件 (`.idea`, `.vscode`) ✅
+- [x] 排除了构建目录 (`build_scripts`, `dist`, `build`) ✅
+- [x] 排除了 `.git` 目录 ✅
+
+#### 功能验证
+
+- [ ] 测试构建成功
+- [ ] 测试安装程序能正常运行
+- [ ] 测试便携版能正常运行
+- [ ] 验证项目管理功能正常
+- [ ] 验证新建项目功能正常
+
+---
+
+## 🧪 验证方法
+
+构建完成后，验证以下内容:
+
+```powershell
+# Windows
+# 1. 检查输出目录结构
+tree /f dist\EG_1.0.0
+
+# 2. 运行程序测试基本功能
+dist\EG_1.0.0\EG.exe
+
+# 3. 检查关键文件是否存在
+Test-Path dist\EG_1.0.0\config
+Test-Path dist\EG_1.0.0\Resources
+Test-Path dist\EG_1.0.0\scripts
+```
+
+```bash
+# Linux
+# 1. 检查 AppImage 内容
+./dist/EG-1.0.0-x86_64.AppImage --appimage-mount
+
+# 2. 运行测试
+./dist/EG-1.0.0-x86_64.AppImage
+```
+
+---
+
+## 🔒 安全注意事项
+
+1. **代码保护**: Nuitka 编译后的机器码难以逆向，但资源文件(模型、纹理)是明文的
+2. **敏感配置**: 确保 `config/` 中没有包含 API 密钥等敏感信息
+3. **许可证**: 如果包含第三方资源，确保许可证允许再分发
+
+---
+
+**文档版本**: 1.0  
+**最后更新**: 2024年

build_scripts/PACKAGING_RISK_AUDIT.md

@@ -0,0 +1,246 @@
+# 元泰 EG 打包风险与源码整改清单
+
+本文档用于记录当前项目在 Windows/Linux 打包时容易导致运行失败、功能失效或行为不一致的风险点，并给出建议的整改方向。
+
+目标原则：
+
+- 打包优先保证可运行，不追求安装包最小化
+- 对动态导入、配置驱动加载、原生扩展依赖保持保守策略
+- 对已声明移除的依赖，源码中不应再保留残留引用
+
+---
+
+## 一、已确认的打包风险
+
+### 1. RenderPipeline 配置与插件属于运行时动态加载
+
+相关位置：
+
+- [RenderPipelineFile/config/plugins.yaml](/C:/Users/Tellme/apps/EG/RenderPipelineFile/config/plugins.yaml)
+- [core/world.py](/C:/Users/Tellme/apps/EG/core/world.py)
+- [build_scripts/build_windows.ps1](/C:/Users/Tellme/apps/EG/build_scripts/build_windows.ps1)
+
+问题说明：
+
+- `RenderPipeline` 会根据配置文件在运行时动态加载 `rpplugins.*`
+- 这类插件不是普通静态 import，打包工具无法稳定自动发现
+- 已实际出现 `ModuleNotFoundError: No module named 'rpplugins.ao'`
+
+建议：
+
+- 打包脚本显式包含 `rpcore`、`rplibs`、`rpplugins`
+- 保留整个 `RenderPipelineFile` 目录
+- 显式设置 `RenderPipeline` 的 `base_path` 和 `config_dir`
+- 后续新增插件时，同步检查打包白名单
+
+### 2. 文件路径驱动的动态模块加载
+
+相关位置：
+
+- [main.py](/C:/Users/Tellme/apps/EG/main.py)
+- [core/script_system.py](/C:/Users/Tellme/apps/EG/core/script_system.py)
+
+问题说明：
+
+- 使用 `importlib.util.spec_from_file_location()` 按文件路径加载模块
+- 例如 `demo/video_integration.py`
+- 这类模块不会被打包器通过静态分析自动发现
+
+建议：
+
+- 继续将对应目录视为运行时资源并整体复制
+- 更理想的方案是逐步改为标准包导入
+- 所有此类动态加载点都应登记到打包白名单
+
+### 3. 运行时外部工具依赖目录结构
+
+相关位置：
+
+- [core/world.py](/C:/Users/Tellme/apps/EG/core/world.py)
+- [core/tool_manager.py](/C:/Users/Tellme/apps/EG/core/tool_manager.py)
+
+问题说明：
+
+- 程序会在运行时启动 `toolkit/day_time_editor/main.py`
+- 还会启动 `toolkit/material_editor/main.py`
+- 还会启动 `toolkit/plugin_configurator/main.py`
+- 如果这些目录未被完整复制，主程序可能能启动，但局部功能会失效
+
+建议：
+
+- 将整个 `RenderPipelineFile` 视为正式运行时资源
+- 子工具路径不应依赖“开发目录结构恰好存在”
+- 后续可考虑增加启动前存在性检测和更友好的错误提示
+
+---
+
+## 二、建议项目团队整改的源码问题
+
+### 1. 残留的 PyQt5 引用需要清理
+
+相关位置：
+
+- [ui/icon_manager.py](/C:/Users/Tellme/apps/EG/ui/icon_manager.py)
+- [core/selection.py](/C:/Users/Tellme/apps/EG/core/selection.py)
+- [scene/scene_manager_convert_tiles_mixin.py](/C:/Users/Tellme/apps/EG/scene/scene_manager_convert_tiles_mixin.py)
+- [core/world.py](/C:/Users/Tellme/apps/EG/core/world.py)
+
+问题说明：
+
+- 项目文档中已经说明移除了 `PyQt5/PySide6`
+- 但源码仍保留多处 `PyQt5` 引用
+- 这会导致打包后功能行为不一致，或在特定分支下运行时报错
+
+建议：
+
+- 用当前实际 UI 栈替代残留 Qt 逻辑
+- 删除仅为兼容历史方案保留的无效 Qt 代码
+- 对必须保留的兼容逻辑，至少统一做降级处理和提示
+
+### 2. Playwright 浏览器依赖不应处于“隐式依赖”状态
+
+相关位置：
+
+- [core/imgui_webview.py](/C:/Users/Tellme/apps/EG/core/imgui_webview.py)
+
+问题说明：
+
+- 该模块依赖 `playwright.async_api`
+- 同时还依赖 Chromium 浏览器安装
+- 如果发行版没有附带这些条件，对应功能会直接不可用
+
+建议：
+
+- 明确决定是否正式支持此功能
+- 如果正式支持，则纳入依赖与打包流程
+- 如果不正式支持，则在发行版中禁用并提示，而不是运行时才报错
+
+### 3. glTF 专用加载依赖缺少明确归档
+
+相关位置：
+
+- [ui/panels/animation_tools.py](/C:/Users/Tellme/apps/EG/ui/panels/animation_tools.py)
+
+问题说明：
+
+- 这里存在运行时 `import gltf`
+- 这不是启动必经路径，因此问题可能只在用户点击特定功能时暴露
+
+建议：
+
+- 如果该功能是正式能力，则将 `gltf` 明确加入依赖清单和打包清单
+- 如果只是补充路径，应在缺模块时给出明确降级提示
+
+### 4. 平台特定依赖应显式治理
+
+相关位置：
+
+- [core/tool_manager.py](/C:/Users/Tellme/apps/EG/core/tool_manager.py)
+
+问题说明：
+
+- 这里使用 `win32gui`、`win32con`
+- 当前代码虽然做了异常兜底，但功能可能 silently fail
+
+建议：
+
+- 明确是否正式依赖 pywin32
+- 如果依赖，则加入打包环境
+- 如果不依赖，则删除窗口激活逻辑或改为可观测提示
+
+### 5. 硬编码开发机路径需要移除
+
+相关位置：
+
+- [core/world.py](/C:/Users/Tellme/apps/EG/core/world.py)
+
+问题说明：
+
+- 存在 `/home/tiger/...` 这类硬编码本地路径
+- 即使当前分支未使用，也不应保留在仓库中
+
+建议：
+
+- 清理所有开发机私有绝对路径
+- 改为项目资源路径、配置路径或用户可配置路径
+
+---
+
+## 三、建议长期执行的工程规则
+
+### 1. 建立“动态导入登记表”
+
+所有以下模式都必须登记并同步给打包脚本：
+
+- `importlib.import_module`
+- `spec_from_file_location`
+- `__import__`
+- 配置文件驱动插件加载
+- 子进程启动外部 Python 工具
+
+### 2. 建立“运行时资源目录白名单”
+
+以下内容不应依赖打包器自动分析：
+
+- 配置文件
+- shader
+- 模型
+- 字体
+- 图标
+- demo 脚本
+- 插件目录
+- 工具目录
+
+### 3. 已移除依赖必须同步清理源码
+
+如果项目文档声明：
+
+- 已移除 `PyQt5`
+- 已移除 `PySide6`
+
+则源码中不应继续保留这些引用，避免“文档说没有，代码里还在偷偷依赖”的情况。
+
+### 4. 可选功能必须有明确产品策略
+
+对以下类型的依赖，需要明确归类：
+
+- 浏览器内核
+- glTF 专用加载器
+- 平台专用 API
+- 外部命令行工具
+
+只能二选一：
+
+- 正式支持并纳入安装/打包流程
+- 默认禁用并提供明确提示
+
+不应继续保持“开发环境可用、发行版碰运气”的状态。
+
+---
+
+## 四、当前打包策略建议
+
+当前推荐策略：
+
+- 尽量显式包含相关包，不追求安装包最小化
+- 尽量整体复制运行时资源目录
+- 对启动阶段异常保留日志文件，便于定位下一层问题
+
+建议构建策略：
+
+- 显式包含 `rpcore`、`rplibs`、`rpplugins`
+- 尽量显式包含 `core`、`scene`、`project`、`ui`、`gui`、`scripts`、`ssbo_component`、`tools`、`TransformGizmo`
+- 自动探测并尽量包含 `plugins`、`gltf`、`playwright` 等可选包
+- 整体复制 `RenderPipelineFile`、`Resources`、`config`、`demo` 等目录
+
+---
+
+## 五、建议团队后续执行顺序
+
+1. 先确保打包脚本采用保守策略，优先恢复发行版可运行性
+2. 再逐步清理源码中的动态依赖、残留 Qt 引用、硬编码路径
+3. 最后再考虑做安装包体积优化，而不是反过来
+
+---
+
+最后更新：2026-03-14

build_scripts/PYTHON_VERSION_REQUIREMENT.md

@@ -0,0 +1,99 @@
+# Python 版本要求
+
+## 必须使用的版本
+
+**Python 3.11.x**
+
+```bash
+# 检查 Python 版本
+python --version
+# 应显示: Python 3.11.x
+```
+
+## 为什么不支持其他版本？
+
+| 版本 | 状态 | 原因 |
+|------|------|------|
+| Python 3.10 | ❌ 不支持 | Panda3D 1.10.16 在 3.10 上有兼容性问题 |
+| **Python 3.11** | ✅ **推荐** | 经过完整测试，所有功能正常 |
+| Python 3.12 | ⚠️ 未测试 | 可能存在兼容性问题 |
+| Python 3.13 | ⚠️ 不推荐 | 当前环境使用，但部分依赖可能有警告 |
+
+## Windows 安装 Python 3.11
+
+### 方法 1: 官网下载
+
+1. 访问: https://www.python.org/downloads/release/python-3119/
+2. 下载: `Windows installer (64-bit)`
+3. 安装时勾选:
+   - ✅ `Add Python to PATH`
+   - ✅ `Use admin privileges when installing py.exe`
+
+### 方法 2: 使用 Chocolatey
+
+```powershell
+# 安装 Chocolatey 后
+choco install python --version=3.11.9
+```
+
+### 方法 3: 使用 pyenv-win
+
+```powershell
+# 安装 pyenv-win
+pip install pyenv-win
+
+# 安装 Python 3.11
+pyenv install 3.11.9
+pyenv global 3.11.9
+```
+
+## 验证安装
+
+```powershell
+# 检查版本
+python --version
+# Python 3.11.9
+
+# 检查 pip
+python -m pip --version
+# pip 24.x from ...
+```
+
+## 安装项目依赖
+
+```powershell
+# 创建虚拟环境（推荐）
+python -m venv venv
+.\venv\Scripts\activate
+
+# 安装依赖
+python -m pip install -r requirements/requirements-minimal.txt
+```
+
+## 构建要求
+
+构建安装程序时也必须使用 Python 3.11：
+
+```powershell
+# 确保使用 Python 3.11
+python --version
+# Python 3.11.x
+
+# 然后执行构建
+.\build_scripts\build_windows.ps1
+```
+
+## 常见问题
+
+### Q: 我已经安装了 Python 3.12/3.13，需要卸载吗？
+**A**: 不需要卸载，可以同时安装多个版本。使用 `py -3.11` 或指定完整路径调用 Python 3.11。
+
+### Q: 如何切换默认 Python 版本？
+**A**: 修改系统 PATH 环境变量，将 Python 3.11 的路径放到其他版本前面。
+
+### Q: 虚拟环境中可以使用不同版本吗？
+**A**: 不可以，虚拟环境继承创建它的 Python 版本。需要用 Python 3.11 创建虚拟环境。
+
+---
+
+**最后更新**: 2024年

build_scripts/README.md

@@ -0,0 +1,143 @@
+# build_scripts 目录说明
+
+本目录包含元泰 EG 项目的所有构建相关脚本和文档。
+
+---
+
+## 📁 目录结构
+
+```
+build_scripts/
+├── 🔧 构建脚本
+│   ├── build_windows.ps1      # Windows 主构建脚本 ⭐
+│   ├── build_linux.sh         # Linux 构建脚本
+│   └── (已移除临时/重复入口脚本)
+│
+├── 🔍 检查脚本
+│   ├── check_env.ps1          # Windows 环境检查
+│   ├── check_env.sh           # Linux 环境检查
+│   ├── check_python_version.py # Python 版本检查 ⭐
+│   └── analyze_packaging.py   # 打包分析工具
+│
+└── 📖 文档
+    ├── BUILD_GUIDE.md         # 完整构建指南 ⭐⭐⭐
+    ├── ICON_GUIDE.md          # 图标准备
+    ├── PACKAGING_RISK_AUDIT.md # 打包风险与源码整改清单 ⭐
+    ├── PYTHON_VERSION_REQUIREMENT.md  # Python 3.11 要求 ⭐
+    └── PACKAGING_CHECKLIST.md # 打包检查清单
+```
+
+---
+
+## 🚀 快速开始
+
+### 第一次构建？
+
+请按顺序阅读：
+1. `BUILD_GUIDE.md` - 完整构建步骤
+2. `PYTHON_VERSION_REQUIREMENT.md` - Python 3.11 安装
+3. `ICON_GUIDE.md` - 准备应用图标（可选）
+4. `PACKAGING_RISK_AUDIT.md` - 打包风险与源码整改清单
+
+### 已经配置好环境？
+
+直接使用：
+```powershell
+# Windows
+.\build_scripts\build_windows.ps1 -Version "1.0.0"
+
+# Linux
+./build_scripts/build_linux.sh "1.0.0"
+```
+
+---
+
+## 📋 关键检查点
+
+构建前请确认：
+
+- [ ] Python 3.11.x 已安装
+- [ ] `python build_scripts/check_python_version.py` 通过
+- [ ] `pip install -r requirements/requirements-minimal.txt` 完成
+- [ ] Visual Studio Build Tools 已安装（Windows）
+- [ ] 在 VS Dev Shell 中运行构建
+
+---
+
+## 🔧 核心脚本说明
+
+### build_windows.ps1
+
+Windows 主构建脚本，功能：
+- 检查 Python 3.11
+- 检查 C++ 编译器
+- 执行保守白名单 Nuitka 编译
+- 自动探测并尽量纳入可选依赖
+- 复制运行时资源文件
+- 生成最终输出
+
+关键策略：
+- 优先保证发行版可运行，不追求最小体积
+- 显式包含 `rpcore`、`rplibs`、`rpplugins`
+- 整体复制 `RenderPipelineFile`、`Resources`、`config`、`demo` 等目录
+
+### build_linux.sh
+
+Linux 主构建脚本，功能：
+- 执行保守白名单 Nuitka 编译
+- 构建 AppDir / AppImage
+- 显式设置 `EG_PROJECT_ROOT`，保证 AppImage 中资源路径正确
+- 复制运行时资源文件
+
+用法：
+```powershell
+./build_scripts/build_linux.sh "1.0.0" --clean
+```
+
+### check_python_version.py
+
+Python 版本检查工具：
+```powershell
+python build_scripts/check_python_version.py
+```
+
+### analyze_packaging.py
+
+打包内容分析工具：
+```powershell
+python build_scripts/analyze_packaging.py
+```
+
+---
+
+## 📝 重要变更
+
+### 2024年更新
+
+- ✅ 指定 Python 3.11 为必须版本
+- ✅ 移除 PyQt5/PySide6 依赖（节省 ~170MB）
+- ✅ 简化依赖配置
+- ✅ 重新组织文档结构
+
+### 2026年更新
+
+- ✅ 清理重复/临时构建脚本与说明文档
+- ✅ Windows/Linux 构建脚本统一改为保守白名单策略
+- ✅ 新增打包风险审计文档
+
+---
+
+## ❓ 常见问题
+
+**Q: 构建失败怎么办？**  
+A: 查看 `BUILD_GUIDE.md` 的"常见问题"章节
+
+**Q: Python 版本不对？**  
+A: 阅读 `PYTHON_VERSION_REQUIREMENT.md`
+
+**Q: 缺少图标？**  
+A: 参考 `ICON_GUIDE.md` 创建或下载占位图标
+
+---
+
+**最后更新**: 2024年

build_scripts/analyze_packaging.py

@@ -0,0 +1,326 @@
+#!/usr/bin/env python3
+"""
+EG 打包分析工具
+用于分析项目结构和打包配置的完整性
+"""
+
+import os
+import sys
+from pathlib import Path
+from typing import List, Set, Tuple
+
+# 项目根目录
+PROJECT_ROOT = Path(__file__).parent.parent.absolute()
+
+# 当前的打包配置 (从 setup.py 同步)
+DATA_DIRECTORIES = [
+    # 配置和资源
+    "config",
+    "icons",
+    "Resources",
+    "tex",
+    "templates",
+    
+    # 核心模块
+    "core",
+    "gui",
+    "project",
+    "scene",
+    "scripts",
+    "ssbo_component",
+    "tools",
+    "TransformGizmo",
+    "ui",
+    
+    # 引擎和扩展
+    "RenderPipelineFile",
+    "vr_actions",
+    
+    # 项目模板
+    "new",
+]
+
+# 排除的模式
+EXCLUDE_PATTERNS = [
+    "__pycache__",
+    "*.pyc",
+    "*.pyo",
+    "*.pyd",
+    ".git",
+    ".gitignore",
+    ".idea",
+    ".vscode",
+    "*.egg-info",
+    "*.egg",
+    "dist",
+    "build",
+    "*.spec",
+]
+
+# 已知的开发/构建目录 (不应该打包)
+DEV_DIRECTORIES = [
+    ".git",
+    ".idea",
+    ".vscode",
+    "build",
+    "build_scripts",
+    "dist",
+    "test_project",  # 示例项目，可选
+    "__pycache__",
+]
+
+# 已知的文档文件 (可选打包)
+DOC_FILES = [
+    "README.md",
+    "LICENSE.txt",
+    "CHANGELOG.md",
+    "AGENTS.md",
+]
+
+
+def get_directory_size(path: Path) -> Tuple[int, int]:
+    """计算目录大小和文件数量"""
+    total_size = 0
+    file_count = 0
+    
+    if not path.exists():
+        return 0, 0
+    
+    for item in path.rglob("*"):
+        if item.is_file():
+            # 检查是否应该排除
+            should_exclude = False
+            for pattern in EXCLUDE_PATTERNS:
+                if pattern.startswith("*"):
+                    if item.name.endswith(pattern[1:]):
+                        should_exclude = True
+                        break
+                elif pattern in str(item.relative_to(path)):
+                    should_exclude = True
+                    break
+            
+            if not should_exclude:
+                total_size += item.stat().st_size
+                file_count += 1
+    
+    return total_size, file_count
+
+
+def format_size(size_bytes: int) -> str:
+    """格式化文件大小"""
+    for unit in ['B', 'KB', 'MB', 'GB']:
+        if size_bytes < 1024:
+            return f"{size_bytes:.2f} {unit}"
+        size_bytes /= 1024
+    return f"{size_bytes:.2f} TB"
+
+
+def analyze_project_structure():
+    """分析项目结构"""
+    print("=" * 70)
+    print("EG 项目打包分析")
+    print("=" * 70)
+    print()
+    
+    # 获取所有目录
+    all_dirs = [d for d in PROJECT_ROOT.iterdir() if d.is_dir()]
+    all_dir_names = {d.name for d in all_dirs}
+    
+    # 分类
+    packaged_dirs = []
+    dev_dirs = []
+    missing_dirs = []
+    unknown_dirs = []
+    
+    for d in all_dirs:
+        name = d.name
+        if name in DATA_DIRECTORIES:
+            packaged_dirs.append(d)
+        elif name in DEV_DIRECTORIES:
+            dev_dirs.append(d)
+        elif name.startswith(".") or name in ["new", "test_project"]:
+            # 特殊目录
+            if name not in DATA_DIRECTORIES:
+                missing_dirs.append(d)
+        else:
+            unknown_dirs.append(d)
+    
+    # 1. 已配置的打包目录
+    print("📦 已配置的打包目录:")
+    print("-" * 70)
+    total_size = 0
+    total_files = 0
+    
+    for d in sorted(packaged_dirs, key=lambda x: x.name):
+        size, count = get_directory_size(d)
+        total_size += size
+        total_files += count
+        status = "✓" if d.exists() else "✗ 不存在"
+        print(f"  {status:12} {d.name:25} {format_size(size):>12} ({count} 文件)")
+    
+    print("-" * 70)
+    print(f"  {'总计':12} {'':25} {format_size(total_size):>12} ({total_files} 文件)")
+    print()
+    
+    # 2. 开发目录 (正确排除)
+    if dev_dirs:
+        print("🔧 开发/构建目录 (正确排除):")
+        print("-" * 70)
+        for d in sorted(dev_dirs, key=lambda x: x.name):
+            size, count = get_directory_size(d)
+            print(f"  {'✓ 排除':12} {d.name:25} {format_size(size):>12}")
+        print()
+    
+    # 3. 未配置的目录 (需要检查)
+    if unknown_dirs:
+        print("⚠️  未配置的目录 (请检查是否需要打包):")
+        print("-" * 70)
+        for d in sorted(unknown_dirs, key=lambda x: x.name):
+            size, count = get_directory_size(d)
+            print(f"  {'? 待确认':12} {d.name:25} {format_size(size):>12} ({count} 文件)")
+            # 列出部分文件帮助判断
+            files = list(d.iterdir())[:5]
+            for f in files:
+                file_type = "📁" if f.is_dir() else "📄"
+                print(f"               {file_type} {f.name}")
+            if len(list(d.iterdir())) > 5:
+                print(f"               ... 等 {len(list(d.iterdir())) - 5} 个文件")
+        print()
+    
+    # 4. 检查配置中但目录不存在的
+    config_only = set(DATA_DIRECTORIES) - all_dir_names
+    if config_only:
+        print("⚠️  配置中有但目录不存在:")
+        print("-" * 70)
+        for name in config_only:
+            print(f"  ✗ 不存在     {name}")
+        print()
+    
+    # 5. 关键文件检查
+    print("🔑 关键文件检查:")
+    print("-" * 70)
+    
+    critical_files = {
+        "Start_Run.py": "入口脚本",
+        "main.py": "主程序",
+        "imgui.ini": "ImGui配置",
+        "icons/app.ico": "Windows图标",
+        "icons/app.png": "Linux图标",
+    }
+    
+    for file, desc in critical_files.items():
+        path = PROJECT_ROOT / file
+        exists = "✓" if path.exists() else "✗"
+        print(f"  {exists:3} {file:25} ({desc})")
+    print()
+    
+    # 6. 建议
+    print("💡 建议:")
+    print("-" * 70)
+    
+    suggestions = []
+    
+    if unknown_dirs:
+        suggestions.append(f"有 {len(unknown_dirs)} 个未配置目录，请检查是否需要打包")
+    
+    if config_only:
+        suggestions.append(f"配置中有 {len(config_only)} 个目录不存在，请更新配置")
+    
+    # 检查图标
+    if not (PROJECT_ROOT / "icons/app.ico").exists():
+        suggestions.append("缺少 Windows 图标: icons/app.ico")
+    if not (PROJECT_ROOT / "icons/app.png").exists():
+        suggestions.append("缺少 Linux 图标: icons/app.png")
+    
+    # 检查大小
+    if total_size > 500 * 1024 * 1024:
+        suggestions.append(f"打包内容较大 ({format_size(total_size)})，考虑优化")
+    
+    if not suggestions:
+        print("  ✓ 一切正常！")
+    else:
+        for s in suggestions:
+            print(f"  • {s}")
+    print()
+    
+    # 7. 预估输出
+    print("📊 预估输出:")
+    print("-" * 70)
+    
+    # Nuitka 编译后大约增加 20-50MB 开销
+    estimated_exe_size = total_size + 50 * 1024 * 1024
+    print(f"  数据文件:        {format_size(total_size)}")
+    print(f"  编译开销(预估):  ~50 MB")
+    print(f"  总计(预估):      {format_size(estimated_exe_size)}")
+    print(f"  压缩后(预估):    {format_size(estimated_exe_size * 0.6)}")  # 假设 60% 压缩率
+    print()
+    
+    # 返回是否有问题
+    has_issues = bool(unknown_dirs or config_only or not suggestions)
+    return has_issues
+
+
+def generate_updated_config():
+    """生成更新后的配置代码"""
+    print("=" * 70)
+    print("建议的配置更新")
+    print("=" * 70)
+    print()
+    
+    # 获取所有实际存在的目录
+    all_dirs = {d.name for d in PROJECT_ROOT.iterdir() if d.is_dir()}
+    
+    # 自动分类
+    auto_include = []
+    auto_exclude = []
+    
+    for name in sorted(all_dirs):
+        if name in DEV_DIRECTORIES:
+            auto_exclude.append(name)
+        elif name.startswith("."):
+            auto_exclude.append(name)
+        else:
+            auto_include.append(name)
+    
+    print("# 自动生成的配置建议:")
+    print()
+    print("DATA_DIRECTORIES = [")
+    for name in auto_include:
+        comment = ""
+        if name == "project":
+            comment = "  # 项目管理模块"
+        elif name == "new":
+            comment = "  # 默认项目模板"
+        elif name == "test_project":
+            comment = "  # 示例项目 (可选)"
+        print(f'    "{name}",{comment}')
+    print("]")
+    print()
+
+
+def main():
+    """主函数"""
+    import argparse
+    
+    parser = argparse.ArgumentParser(description="EG 打包分析工具")
+    parser.add_argument("--generate", action="store_true", help="生成配置建议")
+    
+    args = parser.parse_args()
+    
+    if args.generate:
+        generate_updated_config()
+    else:
+        has_issues = analyze_project_structure()
+        
+        print()
+        print("=" * 70)
+        if has_issues:
+            print("发现潜在问题，请检查上方输出")
+            sys.exit(1)
+        else:
+            print("✓ 分析完成，配置看起来正常")
+            sys.exit(0)
+
+
+if __name__ == "__main__":
+    main()

build_scripts/build_linux.sh

@@ -0,0 +1,525 @@
+#!/bin/bash
+# EG Linux 构建脚本 - 使用 Nuitka 编译 + AppImage 打包（保守白名单版）
+# 用法: ./build_linux.sh [版本号] [--clean] [--skip-compile] [--deb]
+
+set -e
+
+# 配置
+VERSION=${1:-"1.0.0"}
+CLEAN=false
+SKIP_COMPILE=false
+BUILD_DEB=false
+
+# 解析参数
+for arg in "$@"; do
+    case $arg in
+        --clean)
+            CLEAN=true
+            shift
+            ;;
+        --skip-compile)
+            SKIP_COMPILE=true
+            shift
+            ;;
+        --deb)
+            BUILD_DEB=true
+            shift
+            ;;
+    esac
+done
+
+# 路径配置
+PROJECT_ROOT="$(cd "$(dirname "$0")/.." && pwd)"
+DIST_DIR="$PROJECT_ROOT/dist"
+BUILD_DIR="$PROJECT_ROOT/build"
+NUITKA_DIR="$BUILD_DIR/nuitka"
+APPDIR="$BUILD_DIR/EG_$VERSION.AppDir"
+OUTPUT_NAME="EG_$VERSION"
+
+# 应用信息
+APP_NAME="EG"
+COMPANY_NAME="EG Team"
+APP_DESCRIPTION="EG 3D Editor and Game Engine"
+
+# 需要包含的数据目录
+DATA_DIRS=(
+    # 配置和资源
+    "config"
+    "icons"
+    "Resources"
+    "tex"
+    "templates"
+    
+    # 核心模块
+    "core"
+    "gui"
+    "project"           # 项目管理模块
+    "scene"
+    "scripts"
+    "ssbo_component"
+    "tools"
+    "TransformGizmo"
+    "ui"
+    
+    # 引擎和扩展
+    "RenderPipelineFile"
+    "vr_actions"
+    
+    # 项目模板
+    "new"                # 默认空项目模板
+    "demo"
+)
+
+# 需要包含的单个文件
+DATA_FILES=(
+    "imgui.ini"
+)
+
+REQUIRED_PACKAGES=(
+    "panda3d"
+    "direct"
+    "imgui_bundle"
+    "p3dimgui"
+    "openvr"
+    "RenderPipelineFile"
+    "rpcore"
+    "rplibs"
+    "rpplugins"
+    "core"
+    "scene"
+    "project"
+    "ui"
+    "gui"
+    "scripts"
+    "ssbo_component"
+    "tools"
+    "TransformGizmo"
+    "vr_actions"
+)
+
+OPTIONAL_PACKAGES=(
+    "plugins"
+    "gltf"
+    "playwright"
+    "PyQt5"
+)
+
+OPTIONAL_MODULES=(
+    "playwright.async_api"
+    "win32gui"
+    "win32con"
+    "PyQt5.QtCore"
+    "PyQt5.QtGui"
+    "PyQt5.QtWidgets"
+)
+
+# 颜色定义
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m' # No Color
+
+# 辅助函数
+log_info() {
+    echo -e "${BLUE}[INFO]${NC} $1"
+}
+
+log_success() {
+    echo -e "${GREEN}[OK]${NC} $1"
+}
+
+log_warning() {
+    echo -e "${YELLOW}[WARN]${NC} $1"
+}
+
+log_error() {
+    echo -e "${RED}[ERROR]${NC} $1"
+}
+
+log_step() {
+    local step=$1
+    local total=$2
+    local msg=$3
+    echo ""
+    echo -e "${YELLOW}[$step/$total]${NC} $msg"
+}
+
+python_importable() {
+    local module_name="$1"
+    python3 -c "import importlib.util, sys; sys.exit(0 if importlib.util.find_spec('$module_name') else 1)" >/dev/null 2>&1
+}
+
+# ==================== 主流程 ====================
+
+echo "========================================"
+echo "  EG Linux Builder v$VERSION"
+echo "========================================"
+echo "项目路径: $PROJECT_ROOT"
+
+# 步骤 1: 清理
+log_step 1 6 "清理旧构建..."
+
+if [ "$CLEAN" = true ]; then
+    rm -rf "$BUILD_DIR"
+    rm -rf "$DIST_DIR"
+    log_success "已清理所有构建文件"
+fi
+
+mkdir -p "$BUILD_DIR"
+mkdir -p "$DIST_DIR"
+
+# 步骤 2: 环境检查
+log_step 2 6 "检查构建环境..."
+
+# 检查必要的命令
+for cmd in python3 pip3 gcc g++; do
+    if ! command -v $cmd &> /dev/null; then
+        log_error "$cmd 未安装"
+        exit 1
+    fi
+done
+
+PYTHON_VERSION=$(python3 --version 2>&1)
+log_success "Python: $PYTHON_VERSION"
+
+# 检查并安装构建依赖
+log_info "检查构建依赖..."
+pip3 install -q nuitka pyinstaller 2>/dev/null || true
+log_success "构建依赖已就绪"
+
+# 安装项目依赖
+log_info "检查项目依赖..."
+pip3 install -q -r "$PROJECT_ROOT/requirements/requirements-minimal.txt" 2>/dev/null || log_warning "部分依赖可能安装失败"
+log_success "项目依赖已就绪"
+
+# 步骤 3: Nuitka 编译
+if [ "$SKIP_COMPILE" = false ]; then
+    log_step 3 6 "使用 Nuitka 编译主程序..."
+    
+    cd "$PROJECT_ROOT"
+    
+    REQUIRED_PACKAGE_ARGS=()
+    for package in "${REQUIRED_PACKAGES[@]}"; do
+        REQUIRED_PACKAGE_ARGS+=("--include-package=$package")
+    done
+
+    OPTIONAL_PACKAGE_ARGS=()
+    for package in "${OPTIONAL_PACKAGES[@]}"; do
+        if python_importable "$package"; then
+            OPTIONAL_PACKAGE_ARGS+=("--include-package=$package")
+            log_info "检测到可选包并纳入打包: $package"
+        fi
+    done
+
+    OPTIONAL_MODULE_ARGS=()
+    for module in "${OPTIONAL_MODULES[@]}"; do
+        if python_importable "$module"; then
+            OPTIONAL_MODULE_ARGS+=("--include-module=$module")
+            log_info "检测到可选模块并纳入打包: $module"
+        fi
+    done
+
+    # 构建 Nuitka 选项
+    NUITKA_OPTS=(
+        "--standalone"
+        "--follow-import-to=rpcore"
+        "--follow-import-to=rpplugins"
+        "--follow-import-to=core"
+        "--follow-import-to=scene"
+        "--follow-import-to=project"
+        "--follow-import-to=ui"
+        "--nofollow-import-to=tkinter,test,unittest,setuptools,pip"
+        "--linux-onefile-icon=icons/app.png"
+        "--lto=no"
+        "--jobs=$(nproc)"
+        "--output-dir=$NUITKA_DIR"
+        "--remove-output"
+    )
+
+    NUITKA_OPTS+=("${REQUIRED_PACKAGE_ARGS[@]}")
+    NUITKA_OPTS+=("${OPTIONAL_PACKAGE_ARGS[@]}")
+    NUITKA_OPTS+=("${OPTIONAL_MODULE_ARGS[@]}")
+    
+    log_info "开始编译 (可能需要几分钟)..."
+    python3 -m nuitka "${NUITKA_OPTS[@]}" Start_Run.py
+    
+    if [ $? -ne 0 ]; then
+        log_error "Nuitka 编译失败"
+        exit 1
+    fi
+    
+    log_success "编译成功"
+else
+    log_step 3 6 "跳过编译 (--skip-compile)"
+fi
+
+# 步骤 4: 准备 AppDir
+log_step 4 6 "准备 AppDir..."
+
+rm -rf "$APPDIR"
+mkdir -p "$APPDIR/usr/bin"
+mkdir -p "$APPDIR/usr/lib/EG"
+mkdir -p "$APPDIR/usr/share/applications"
+mkdir -p "$APPDIR/usr/share/icons/hicolor/256x256/apps"
+mkdir -p "$APPDIR/usr/share/icons/hicolor/128x128/apps"
+mkdir -p "$APPDIR/usr/share/icons/hicolor/64x64/apps"
+mkdir -p "$APPDIR/usr/share/metainfo"
+
+# 复制 Nuitka standalone 目录
+DIST_SOURCE="$NUITKA_DIR/Start_Run.dist"
+if [ ! -d "$DIST_SOURCE" ]; then
+    log_error "找不到编译后的 standalone 目录: $DIST_SOURCE"
+    exit 1
+fi
+
+cp -r "$DIST_SOURCE/"* "$APPDIR/usr/lib/EG/"
+
+if [ -f "$APPDIR/usr/lib/EG/Start_Run.bin" ]; then
+    cp "$APPDIR/usr/lib/EG/Start_Run.bin" "$APPDIR/usr/bin/EG"
+elif [ -f "$APPDIR/usr/lib/EG/Start_Run" ]; then
+    cp "$APPDIR/usr/lib/EG/Start_Run" "$APPDIR/usr/bin/EG"
+else
+    EXE_SOURCE=$(find "$APPDIR/usr/lib/EG" -maxdepth 1 -type f -executable | head -1)
+    if [ -z "$EXE_SOURCE" ]; then
+        log_error "找不到可执行文件"
+        exit 1
+    fi
+    cp "$EXE_SOURCE" "$APPDIR/usr/bin/EG"
+fi
+
+chmod +x "$APPDIR/usr/bin/EG"
+log_success "复制可执行文件和 standalone 运行时"
+
+# 复制数据目录
+COPIED=0
+for dir in "${DATA_DIRS[@]}"; do
+    if [ -d "$PROJECT_ROOT/$dir" ]; then
+        cp -r "$PROJECT_ROOT/$dir" "$APPDIR/usr/lib/EG/"
+        # 清理 Python 缓存
+        find "$APPDIR/usr/lib/EG/$dir" -type d -name "__pycache__" -exec rm -rf {} + 2>/dev/null || true
+        find "$APPDIR/usr/lib/EG/$dir" -name "*.pyc" -delete 2>/dev/null || true
+        log_success "复制目录: $dir"
+        ((COPIED++))
+    else
+        log_warning "目录不存在: $dir"
+    fi
+done
+log_success "共复制 $COPIED 个数据目录"
+
+# 复制单个文件
+for file in "${DATA_FILES[@]}"; do
+    if [ -f "$PROJECT_ROOT/$file" ]; then
+        cp "$PROJECT_ROOT/$file" "$APPDIR/usr/lib/EG/"
+        log_success "复制文件: $file"
+    fi
+done
+
+# 创建 .desktop 文件
+cat > "$APPDIR/usr/share/applications/EG.desktop" << EOF
+[Desktop Entry]
+Name=EG
+Name[zh_CN]=EG 编辑器
+Comment=$APP_DESCRIPTION
+Comment[zh_CN]=EG 3D 编辑器和游戏引擎
+Exec=EG
+Icon=EG
+Type=Application
+Categories=Graphics;3DGraphics;Development;Game;
+MimeType=application/x-eg-project;
+Terminal=false
+StartupNotify=true
+StartupWMClass=EG
+Keywords=3D;Editor;Game;Engine;Panda3D;
+EOF
+
+# 创建 AppStream metadata (用于软件中心)
+cat > "$APPDIR/usr/share/metainfo/EG.appdata.xml" << EOF
+<?xml version="1.0" encoding="UTF-8"?>
+<component type="desktop-application">
+  <id>com.yourcompany.EG</id>
+  <metadata_license>CC0-1.0</metadata_license>
+  <project_license>Proprietary</project_license>
+  <name>EG</name>
+  <summary>3D Editor and Game Engine</summary>
+  <description>
+    <p>EG is a powerful 3D editor and game engine based on Panda3D.</p>
+  </description>
+  <launchable type="desktop-id">EG.desktop</launchable>
+  <url type="homepage">https://your-website.com</url>
+  <provides>
+    <binary>EG</binary>
+  </provides>
+</component>
+EOF
+
+# 复制图标
+for size in 256 128 64; do
+    ICON_SRC="$PROJECT_ROOT/icons/app_${size}.png"
+    if [ -f "$ICON_SRC" ]; then
+        cp "$ICON_SRC" "$APPDIR/usr/share/icons/hicolor/${size}x${size}/apps/EG.png"
+    elif [ -f "$PROJECT_ROOT/icons/app.png" ]; then
+        # 如果只有一张图，复制到所有尺寸
+        cp "$PROJECT_ROOT/icons/app.png" "$APPDIR/usr/share/icons/hicolor/${size}x${size}/apps/EG.png"
+        break
+    fi
+done
+
+# 创建 AppRun
+cat > "$APPDIR/AppRun" << 'EOF'
+#!/bin/bash
+# AppRun for EG
+
+SELF=$(readlink -f "$0")
+HERE=${SELF%/*}
+
+# 设置环境变量
+export PATH="${HERE}/usr/bin:${PATH}"
+export EG_PROJECT_ROOT="${HERE}/usr/lib/EG"
+export LD_LIBRARY_PATH="${HERE}/usr/lib/EG:${HERE}/usr/lib:${LD_LIBRARY_PATH}"
+export PYTHONPATH="${HERE}/usr/lib/EG:${HERE}/usr/lib/EG/RenderPipelineFile:${PYTHONPATH}"
+
+# 设置 XDG 目录
+export XDG_DATA_DIRS="${HERE}/usr/share:${XDG_DATA_DIRS:-/usr/local/share:/usr/share}"
+
+# 切换到项目根目录，确保相对资源路径与 Windows 打包版保持一致
+cd "${EG_PROJECT_ROOT}" || exit 1
+
+# 执行主程序
+exec "${HERE}/usr/bin/EG" "$@"
+EOF
+chmod +x "$APPDIR/AppRun"
+
+# 复制 .desktop 和图标到根目录
+cp "$APPDIR/usr/share/applications/EG.desktop" "$APPDIR/"
+if [ -f "$APPDIR/usr/share/icons/hicolor/256x256/apps/EG.png" ]; then
+    cp "$APPDIR/usr/share/icons/hicolor/256x256/apps/EG.png" "$APPDIR/EG.png"
+elif [ -f "$APPDIR/usr/share/icons/hicolor/128x128/apps/EG.png" ]; then
+    cp "$APPDIR/usr/share/icons/hicolor/128x128/apps/EG.png" "$APPDIR/EG.png"
+elif [ -f "$APPDIR/usr/share/icons/hicolor/64x64/apps/EG.png" ]; then
+    cp "$APPDIR/usr/share/icons/hicolor/64x64/apps/EG.png" "$APPDIR/EG.png"
+fi
+
+log_success "AppDir 准备完成"
+
+# 步骤 5: 创建 AppImage
+log_step 5 6 "创建 AppImage..."
+
+cd "$DIST_DIR"
+
+# 下载 appimagetool
+APPIMAGETOOL="appimagetool-x86_64.AppImage"
+if [ ! -f "$APPIMAGETOOL" ]; then
+    log_info "下载 appimagetool..."
+    wget -q "https://github.com/AppImage/AppImageKit/releases/download/continuous/appimagetool-x86_64.AppImage" -O "$APPIMAGETOOL" || {
+        log_error "下载 appimagetool 失败"
+        exit 1
+    }
+    chmod +x "$APPIMAGETOOL"
+    log_success "下载完成"
+fi
+
+# 创建 AppImage
+APPIMAGE_NAME="EG-${VERSION}-x86_64.AppImage"
+log_info "构建 AppImage..."
+
+# 使用 --appimage-extract-and-run 避免 fuse 依赖
+ARCH=x86_64 ./"$APPIMAGETOOL" --appimage-extract-and-run "$APPDIR" "$APPIMAGE_NAME" 2>&1 | while read line; do
+    echo "  $line"
+done
+
+if [ -f "$APPIMAGE_NAME" ]; then
+    chmod +x "$APPIMAGE_NAME"
+    SIZE=$(du -h "$APPIMAGE_NAME" | cut -f1)
+    log_success "AppImage 创建成功: $APPIMAGE_NAME ($SIZE)"
+else
+    log_error "AppImage 创建失败"
+    exit 1
+fi
+
+# 可选: 创建 DEB 包
+if [ "$BUILD_DEB" = true ]; then
+    log_step 5 6 "创建 DEB 包..."
+    
+    DEB_DIR="$BUILD_DIR/eg_${VERSION}_amd64"
+    mkdir -p "$DEB_DIR/DEBIAN"
+    mkdir -p "$DEB_DIR/opt/EG"
+    mkdir -p "$DEB_DIR/usr/share/applications"
+    mkdir -p "$DEB_DIR/usr/share/icons/hicolor/256x256/apps"
+    mkdir -p "$DEB_DIR/usr/bin"
+    
+    # 控制文件
+    cat > "$DEB_DIR/DEBIAN/control" << EOF
+Package: eg
+Version: $VERSION
+Section: graphics
+Priority: optional
+Architecture: amd64
+Depends: libgl1, libopenal1, libglu1-mesa, python3 (>= 3.8)
+Maintainer: $COMPANY_NAME <support@your-website.com>
+Description: EG 3D Editor and Game Engine
+ EG is a powerful 3D editor and game engine
+ based on Panda3D with advanced rendering.
+EOF
+    
+    # 复制文件
+    cp -r "$APPDIR/usr/"* "$DEB_DIR/opt/EG/"
+    
+    # 创建启动器
+    cat > "$DEB_DIR/usr/bin/EG" << EOF
+#!/bin/bash
+cd /opt/EG
+./bin/EG "\$@"
+EOF
+    chmod +x "$DEB_DIR/usr/bin/EG"
+    
+    # 复制 desktop 文件和图标
+    cp "$APPDIR/usr/share/applications/EG.desktop" "$DEB_DIR/usr/share/applications/"
+    if [ -f "$APPDIR/usr/share/icons/hicolor/256x256/apps/EG.png" ]; then
+        cp "$APPDIR/usr/share/icons/hicolor/256x256/apps/EG.png" "$DEB_DIR/usr/share/icons/hicolor/256x256/apps/"
+    fi
+    
+    # 构建 deb
+    dpkg-deb --build "$DEB_DIR" "$DIST_DIR/eg_${VERSION}_amd64.deb"
+    
+    if [ -f "$DIST_DIR/eg_${VERSION}_amd64.deb" ]; then
+        SIZE=$(du -h "$DIST_DIR/eg_${VERSION}_amd64.deb" | cut -f1)
+        log_success "DEB 包创建成功: eg_${VERSION}_amd64.deb ($SIZE)"
+    fi
+fi
+
+# 步骤 6: 创建 tarball
+log_step 6 6 "创建 tarball..."
+
+cd "$BUILD_DIR"
+
+# 创建普通 tarball
+TARBALL="$DIST_DIR/EG_${VERSION}_Linux_Portable.tar.gz"
+tar -czf "$TARBALL" -C "$APPDIR/usr/lib/EG" .
+SIZE=$(du -h "$TARBALL" | cut -f1)
+log_success "Tarball 创建成功: EG_${VERSION}_Linux_Portable.tar.gz ($SIZE)"
+
+# ==================== 完成 ====================
+
+echo ""
+echo "========================================"
+echo -e "${GREEN}  构建完成!${NC}"
+echo "========================================"
+echo ""
+echo -e "${BLUE}输出文件:${NC}"
+
+for file in "$DIST_DIR"/EG-*; do
+    if [ -f "$file" ]; then
+        name=$(basename "$file")
+        size=$(du -h "$file" | cut -f1)
+        echo -e "  ${GREEN}✓${NC} $name (${size})"
+    fi
+done
+
+echo ""
+echo -e "${BLUE}输出目录:${NC} $DIST_DIR"
+echo ""
+
+# 清理
+rm -rf "$APPDIR"
+log_info "临时文件已清理"

build_scripts/build_windows.ps1

@@ -0,0 +1,201 @@
+# 元泰 EG Windows 构建脚本 - Nuitka
+param([string]$Version = "1.0.0")
+
+$ErrorActionPreference = "Stop"
+$ProjectRoot = "C:\Users\Tellme\apps\EG"
+$BuildDir = "$ProjectRoot\build"
+$DistDir = "$ProjectRoot\dist\EG_$Version"
+$DataDirs = @(
+    "config",
+    "icons",
+    "Resources",
+    "tex",
+    "templates",
+    "core",
+    "gui",
+    "project",
+    "scene",
+    "scripts",
+    "ssbo_component",
+    "tools",
+    "TransformGizmo",
+    "ui",
+    "RenderPipelineFile",
+    "vr_actions",
+    "new",
+    "demo"
+)
+$DataFiles = @(
+    "imgui.ini"
+)
+$RequiredPackages = @(
+    "panda3d",
+    "direct",
+    "imgui_bundle",
+    "p3dimgui",
+    "openvr",
+    "RenderPipelineFile",
+    "rpcore",
+    "rplibs",
+    "rpplugins",
+    "core",
+    "scene",
+    "project",
+    "ui",
+    "gui",
+    "scripts",
+    "ssbo_component",
+    "tools",
+    "TransformGizmo",
+    "vr_actions"
+)
+$OptionalPackages = @(
+    "plugins",
+    "gltf",
+    "playwright",
+    "PyQt5"
+)
+$OptionalModules = @(
+    "playwright.async_api",
+    "win32gui",
+    "win32con",
+    "PyQt5.QtCore",
+    "PyQt5.QtGui",
+    "PyQt5.QtWidgets"
+)
+
+Write-Host "========================================" -ForegroundColor Cyan
+Write-Host "  元泰 EG 构建 (Nuitka)" -ForegroundColor Cyan
+Write-Host "========================================" -ForegroundColor Cyan
+
+# 必须在 VS Dev Shell 中运行
+if (!(Get-Command cl -ErrorAction SilentlyContinue)) {
+    Write-Host "错误: 请在 VS Dev Shell 中运行此脚本" -ForegroundColor Red
+    exit 1
+}
+
+# 清理
+Write-Host "`n[1/4] 清理..." -ForegroundColor Yellow
+Remove-Item $BuildDir -Recurse -Force -ErrorAction SilentlyContinue
+Remove-Item $DistDir -Recurse -Force -ErrorAction SilentlyContinue
+New-Item -ItemType Directory -Force -Path $DistDir | Out-Null
+Write-Host "    完成" -ForegroundColor Green
+
+# 设置环境
+Write-Host "`n[2/4] 设置环境..." -ForegroundColor Yellow
+$env:PYTHONPATH = "$ProjectRoot;$ProjectRoot\RenderPipelineFile"
+$env:PYTHONDONTWRITEBYTECODE = "1"
+
+$imguiPath = (py -3.11 -c "import imgui_bundle, os; print(os.path.dirname(imgui_bundle.__file__))").Replace('\', '/')
+$glfwDll = "$imguiPath/glfw3.dll"
+Write-Host "    glfw3.dll: $glfwDll" -ForegroundColor Gray
+
+function Test-PythonImportable {
+    param([string]$ModuleName)
+
+    & py -3.11 -c "import importlib.util, sys; sys.exit(0 if importlib.util.find_spec('$ModuleName') else 1)" | Out-Null
+    return ($LASTEXITCODE -eq 0)
+}
+
+$requiredPackageArgs = @()
+foreach ($package in $RequiredPackages) {
+    $requiredPackageArgs += "--include-package=$package"
+}
+
+$detectedOptionalPackageArgs = @()
+foreach ($package in $OptionalPackages) {
+    if (Test-PythonImportable $package) {
+        $detectedOptionalPackageArgs += "--include-package=$package"
+        Write-Host "    检测到可选包并纳入打包: $package" -ForegroundColor Gray
+    }
+}
+
+$detectedOptionalModuleArgs = @()
+foreach ($module in $OptionalModules) {
+    if (Test-PythonImportable $module) {
+        $detectedOptionalModuleArgs += "--include-module=$module"
+        Write-Host "    检测到可选模块并纳入打包: $module" -ForegroundColor Gray
+    }
+}
+
+# Nuitka 编译
+Write-Host "`n[3/4] Nuitka 编译 (预计 20-40 分钟)..." -ForegroundColor Yellow
+Write-Host "    关键修复: 使用正斜杠作为目标路径" -ForegroundColor Gray
+Write-Host ""
+
+$nuitkaArgs = @(
+    "-m", "nuitka"
+    "--standalone"
+    "--follow-import-to=rpcore"
+    "--follow-import-to=rpplugins"
+    "--follow-import-to=core"
+    "--follow-import-to=scene"
+    "--follow-import-to=project"
+    "--follow-import-to=ui"
+    "--nofollow-import-to=tkinter,test,unittest,setuptools,pip"
+    # 关键: 正斜杠作为目标路径
+    "--include-data-files=$glfwDll=imgui_bundle/glfw3.dll"
+    "--windows-icon-from-ico=icons/app.ico"
+    "--windows-console-mode=disable"
+    "--windows-company-name=元泰"
+    "--windows-product-name=元泰 EG"
+    "--windows-file-version=$Version"
+    "--windows-product-version=$Version"
+    "--output-dir=$BuildDir"
+    "--jobs=8"
+    "--lto=no"
+    "Start_Run.py"
+)
+
+$nuitkaArgs += $requiredPackageArgs
+$nuitkaArgs += $detectedOptionalPackageArgs
+$nuitkaArgs += $detectedOptionalModuleArgs
+
+& py -3.11 @nuitkaArgs
+
+if ($LASTEXITCODE -ne 0) {
+    Write-Host "`n[错误] 编译失败!" -ForegroundColor Red
+    exit 1
+}
+
+Write-Host "`n    编译成功" -ForegroundColor Green
+
+# 复制到输出目录
+Write-Host "`n[4/4] 整理输出..." -ForegroundColor Yellow
+$sourceDir = "$BuildDir\Start_Run.dist"
+Copy-Item "$sourceDir\*" $DistDir -Recurse -Force
+
+# 复制项目运行时资源。Nuitka 会收集 Python 模块，但这些目录里的配置、
+# shader、模型、图标等仍然需要以真实文件形式存在于 dist 中。
+foreach ($dir in $DataDirs) {
+    $sourcePath = Join-Path $ProjectRoot $dir
+    if (Test-Path $sourcePath) {
+        Copy-Item $sourcePath (Join-Path $DistDir $dir) -Recurse -Force
+        Write-Host "    已复制目录: $dir" -ForegroundColor Gray
+    }
+}
+
+foreach ($file in $DataFiles) {
+    $sourcePath = Join-Path $ProjectRoot $file
+    if (Test-Path $sourcePath) {
+        Copy-Item $sourcePath $DistDir -Force
+        Write-Host "    已复制文件: $file" -ForegroundColor Gray
+    }
+}
+
+# 重命名 exe
+if (Test-Path "$DistDir\Start_Run.exe") {
+    Rename-Item "$DistDir\Start_Run.exe" "元泰.exe"
+}
+
+$exeSize = [math]::Round((Get-Item "$DistDir\元泰.exe").Length / 1MB, 2)
+Write-Host "    元泰.exe ($exeSize MB)" -ForegroundColor Green
+
+# 完成
+Write-Host "`n========================================" -ForegroundColor Green
+Write-Host "  构建完成!" -ForegroundColor Green
+Write-Host "========================================" -ForegroundColor Green
+Write-Host "输出: $DistDir\元泰.exe" -ForegroundColor White
+Write-Host ""
+Write-Host "测试运行:" -ForegroundColor Yellow
+Write-Host "  $DistDir\元泰.exe" -ForegroundColor Gray

build_scripts/check_env.ps1

@@ -0,0 +1,139 @@
+#!/usr/bin/env powershell
+# EG Windows 构建环境检查脚本
+
+Write-Host "========================================" -ForegroundColor Cyan
+Write-Host "  EG 构建环境检查" -ForegroundColor Cyan
+Write-Host "========================================" -ForegroundColor Cyan
+
+$Issues = @()
+$Warnings = @()
+
+# 检查 Python
+Write-Host "`n[1/6] 检查 Python..." -ForegroundColor Yellow
+try {
+    $PythonVersion = python --version 2>&1
+    if ($PythonVersion -match "Python (\d+)\.(\d+)") {
+        $Major = [int]$Matches[1]
+        $Minor = [int]$Matches[2]
+        if ($Major -ge 3 -and $Minor -ge 8) {
+            Write-Host "  ✓ $PythonVersion (符合要求)" -ForegroundColor Green
+        } else {
+            Write-Host "  ⚠ $PythonVersion (建议 3.8+)" -ForegroundColor Yellow
+            $Warnings += "Python 版本较旧"
+        }
+    }
+} catch {
+    Write-Host "  ✗ Python 未找到" -ForegroundColor Red
+    $Issues += "需要安装 Python 3.8+"
+}
+
+# 检查 pip
+Write-Host "`n[2/6] 检查 pip..." -ForegroundColor Yellow
+try {
+    $PipVersion = pip --version 2>&1
+    Write-Host "  ✓ $PipVersion" -ForegroundColor Green
+} catch {
+    Write-Host "  ✗ pip 未找到" -ForegroundColor Red
+    $Issues += "需要安装 pip"
+}
+
+# 检查 PowerShell 执行策略
+Write-Host "`n[3/6] 检查 PowerShell 执行策略..." -ForegroundColor Yellow
+$Policy = Get-ExecutionPolicy
+if ($Policy -eq "Restricted") {
+    Write-Host "  ✗ 当前策略: $Policy (需要修改)" -ForegroundColor Red
+    $Issues += "执行策略需要设置为 RemoteSigned: Set-ExecutionPolicy RemoteSigned -Scope CurrentUser"
+} else {
+    Write-Host "  ✓ 当前策略: $Policy" -ForegroundColor Green
+}
+
+# 检查必要模块
+Write-Host "`n[4/6] 检查 Python 模块..." -ForegroundColor Yellow
+$Modules = @("nuitka", "pyinstaller")
+foreach ($mod in $Modules) {
+    $result = pip show $mod 2>$null
+    if ($result) {
+        $version = ($result | Select-String "Version:").ToString().Split(":")[1].Trim()
+        Write-Host "  ✓ $mod $version" -ForegroundColor Green
+    } else {
+        Write-Host "  ⚠ $mod 未安装 (将自动安装)" -ForegroundColor Yellow
+    }
+}
+
+# 检查项目依赖
+Write-Host "`n[5/6] 检查项目依赖..." -ForegroundColor Yellow
+$ReqFile = "$PSScriptRoot\..\requirements\requirements.txt"
+if (Test-Path $ReqFile) {
+    Write-Host "  ✓ 找到 requirements.txt" -ForegroundColor Green
+    
+    # 检查几个关键包
+    $KeyPackages = @("Panda3D", "PyQt5", "openvr", "imgui-bundle")
+    foreach ($pkg in $KeyPackages) {
+        $result = pip show $pkg 2>$null
+        if ($result) {
+            Write-Host "  ✓ $pkg" -ForegroundColor Green
+        } else {
+            Write-Host "  ⚠ $pkg 未安装" -ForegroundColor Yellow
+        }
+    }
+} else {
+    Write-Host "  ✗ 未找到 requirements.txt" -ForegroundColor Red
+    $Warnings += "未找到依赖文件"
+}
+
+# 检查 Inno Setup
+Write-Host "`n[6/6] 检查 Inno Setup..." -ForegroundColor Yellow
+$ISCCPaths = @(
+    "C:\Program Files (x86)\Inno Setup 6\ISCC.exe",
+    "C:\Program Files\Inno Setup 6\ISCC.exe"
+)
+$Found = $false
+foreach ($path in $ISCCPaths) {
+    if (Test-Path $path) {
+        Write-Host "  ✓ 找到: $path" -ForegroundColor Green
+        $Found = $true
+        break
+    }
+}
+if (!$Found) {
+    Write-Host "  ⚠ Inno Setup 未找到 (可选，用于创建安装程序)" -ForegroundColor Yellow
+    Write-Host "    下载: https://jrsoftware.org/isdl.php" -ForegroundColor Gray
+    $Warnings += "Inno Setup 未安装 (可选)"
+}
+
+# 检查编译器
+Write-Host "`n[7/6] 检查 C++ 编译器..." -ForegroundColor Yellow
+$CL = Get-Command cl -ErrorAction SilentlyContinue
+$GCC = Get-Command gcc -ErrorAction SilentlyContinue
+
+if ($CL) {
+    Write-Host "  ✓ 找到 MSVC (cl)" -ForegroundColor Green
+} elseif ($GCC) {
+    Write-Host "  ✓ 找到 GCC" -ForegroundColor Green
+} else {
+    Write-Host "  ⚠ 未找到 C++ 编译器" -ForegroundColor Yellow
+    Write-Host "    Nuitka 将尝试自动安装 MinGW64 或使用已安装的 Visual Studio" -ForegroundColor Gray
+    $Warnings += "未找到 C++ 编译器 (Nuitka 会尝试自动解决)"
+}
+
+# 总结
+Write-Host "`n========================================" -ForegroundColor Cyan
+if ($Issues.Count -eq 0 -and $Warnings.Count -eq 0) {
+    Write-Host "  ✓ 环境检查通过!" -ForegroundColor Green
+    Write-Host "  可以开始构建" -ForegroundColor Green
+    exit 0
+} else {
+    if ($Issues.Count -gt 0) {
+        Write-Host "  发现问题 ($($Issues.Count)):" -ForegroundColor Red
+        foreach ($issue in $Issues) {
+            Write-Host "    - $issue" -ForegroundColor Red
+        }
+    }
+    if ($Warnings.Count -gt 0) {
+        Write-Host "  警告 ($($Warnings.Count)):" -ForegroundColor Yellow
+        foreach ($warn in $Warnings) {
+            Write-Host "    - $warn" -ForegroundColor Yellow
+        }
+    }
+    exit 1
+}

build_scripts/check_env.sh

@@ -0,0 +1,137 @@
+#!/bin/bash
+# EG Linux 构建环境检查脚本
+
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m'
+
+ISSUES=()
+WARNINGS=()
+
+echo "========================================"
+echo -e "${BLUE}  EG 构建环境检查${NC}"
+echo "========================================"
+
+# 检查 Python
+echo ""
+echo "[1/6] 检查 Python..."
+if command -v python3 &> /dev/null; then
+    VERSION=$(python3 --version 2>&1)
+    echo -e "  ${GREEN}✓${NC} $VERSION"
+else
+    echo -e "  ${RED}✗${NC} Python3 未找到"
+    ISSUES+=("需要安装 Python 3.8+")
+fi
+
+# 检查 pip
+echo ""
+echo "[2/6] 检查 pip..."
+if command -v pip3 &> /dev/null; then
+    VERSION=$(pip3 --version 2>&1 | awk '{print $2}')
+    echo -e "  ${GREEN}✓${NC} pip $VERSION"
+else
+    echo -e "  ${RED}✗${NC} pip3 未找到"
+    ISSUES+=("需要安装 pip3")
+fi
+
+# 检查 GCC
+echo ""
+echo "[3/6] 检查 GCC..."
+if command -v gcc &> /dev/null; then
+    VERSION=$(gcc --version 2>&1 | head -1)
+    echo -e "  ${GREEN}✓${NC} $VERSION"
+else
+    echo -e "  ${RED}✗${NC} GCC 未找到"
+    ISSUES+=("需要安装 GCC: sudo apt-get install build-essential")
+fi
+
+# 检查 G++
+echo ""
+echo "[4/6] 检查 G++..."
+if command -v g++ &> /dev/null; then
+    VERSION=$(g++ --version 2>&1 | head -1)
+    echo -e "  ${GREEN}✓${NC} $VERSION"
+else
+    echo -e "  ${RED}✗${NC} G++ 未找到"
+    ISSUES+=("需要安装 G++: sudo apt-get install build-essential")
+fi
+
+# 检查 Python 模块
+echo ""
+echo "[5/6] 检查 Python 模块..."
+MODULES=("nuitka" "pyinstaller")
+for mod in "${MODULES[@]}"; do
+    if pip3 show "$mod" &> /dev/null; then
+        VERSION=$(pip3 show "$mod" 2>/dev/null | grep Version | awk '{print $2}')
+        echo -e "  ${GREEN}✓${NC} $mod $VERSION"
+    else
+        echo -e "  ${YELLOW}⚠${NC} $mod 未安装 (将自动安装)"
+    fi
+done
+
+# 检查项目依赖
+echo ""
+echo "[6/6] 检查项目依赖..."
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+REQ_FILE="$SCRIPT_DIR/../requirements/requirements.txt"
+
+if [ -f "$REQ_FILE" ]; then
+    echo -e "  ${GREEN}✓${NC} 找到 requirements.txt"
+    
+    # 检查关键包
+    KEY_PKGS=("Panda3D" "PyQt5" "openvr" "imgui-bundle")
+    for pkg in "${KEY_PKGS[@]}"; do
+        # 转换为小写检查
+        pkg_lower=$(echo "$pkg" | tr '[:upper:]' '[:lower:]')
+        if pip3 show "$pkg_lower" &> /dev/null; then
+            echo -e "  ${GREEN}✓${NC} $pkg"
+        else
+            echo -e "  ${YELLOW}⚠${NC} $pkg 未安装"
+        fi
+    done
+else
+    echo -e "  ${YELLOW}⚠${NC} 未找到 requirements.txt"
+    WARNINGS+=("未找到依赖文件")
+fi
+
+# 检查额外工具
+echo ""
+echo "[7/6] 检查额外工具..."
+if command -v wget &> /dev/null; then
+    echo -e "  ${GREEN}✓${NC} wget"
+else
+    echo -e "  ${YELLOW}⚠${NC} wget 未安装 (用于下载 appimagetool)"
+    WARNINGS+=("建议安装 wget")
+fi
+
+# 检查 dpkg (用于构建 deb)
+if command -v dpkg-deb &> /dev/null; then
+    echo -e "  ${GREEN}✓${NC} dpkg-deb (可构建 DEB 包)"
+else
+    echo -e "  ${YELLOW}⚠${NC} dpkg-deb 未安装 (无法构建 DEB 包)"
+fi
+
+# 总结
+echo ""
+echo "========================================"
+if [ ${#ISSUES[@]} -eq 0 ] && [ ${#WARNINGS[@]} -eq 0 ]; then
+    echo -e "${GREEN}  ✓ 环境检查通过!${NC}"
+    echo -e "${GREEN}  可以开始构建${NC}"
+    exit 0
+else
+    if [ ${#ISSUES[@]} -gt 0 ]; then
+        echo -e "${RED}  发现问题 (${#ISSUES[@]}):${NC}"
+        for issue in "${ISSUES[@]}"; do
+            echo -e "    ${RED}- $issue${NC}"
+        done
+    fi
+    if [ ${#WARNINGS[@]} -gt 0 ]; then
+        echo -e "${YELLOW}  警告 (${#WARNINGS[@]}):${NC}"
+        for warn in "${WARNINGS[@]}"; do
+            echo -e "    ${YELLOW}- $warn${NC}"
+        done
+    fi
+    exit 1
+fi

build_scripts/check_python_version.py

@@ -0,0 +1,36 @@
+#!/usr/bin/env python3
+"""
+元泰 EG Python 版本检查脚本
+确保使用正确的 Python 版本
+"""
+
+import sys
+
+def check_python_version():
+    """检查 Python 版本是否为 3.11"""
+    version_info = sys.version_info
+    version_string = f"{version_info.major}.{version_info.minor}.{version_info.micro}"
+    
+    print("=" * 50)
+    print("元泰 EG Python 版本检查")
+    print("=" * 50)
+    print(f"当前 Python 版本: {version_string}")
+    print(f"可执行文件路径: {sys.executable}")
+    print()
+    
+    if version_info[:2] == (3, 11):
+        print("✅ Python 版本正确 (3.11.x)")
+        print()
+        print("可以继续安装依赖和构建:")
+        print("  pip install -r requirements/requirements-minimal.txt")
+        return 0
+    else:
+        print(f"❌ Python 版本错误!")
+        print(f"   需要: Python 3.11.x")
+        print(f"   当前: Python {version_string}")
+        print()
+        print("请参考 build_scripts/PYTHON_VERSION_REQUIREMENT.md 安装正确版本")
+        return 1
+
+if __name__ == "__main__":
+    sys.exit(check_python_version())

core/event_handler.py

@@ -450,7 +450,8 @@ class EventHandler:
             print(f"点击的是碰撞节点: {hitNode.getName()}")
             # 碰撞节点的父节点应该是模型
             parent = hitNode.getParent()
-            if parent in self.world.models:
+            model_list = self.world.models if hasattr(self.world, 'models') else []
+            if parent in model_list or parent.hasTag('is_model_root'):
                 selectedModel = parent
                 print(f"找到对应的模型: {selectedModel.getName()}")
             else:
@@ -460,13 +461,15 @@ class EventHandler:
             current = hitNode
             while current != self.world.render:
                 # 检查是否是模型
-                if current in self.world.models:
+                model_list = self.world.models if hasattr(self.world, 'models') else []
+                if current in model_list or current.hasTag('is_model_root'):
                     selectedModel = current
                     print(f"找到模型节点: {selectedModel.getName()}")
                     break
                     
                 # 检查是否是模型的子节点
-                for model in self.world.models:
+                model_list = self.world.models if hasattr(self.world, 'models') else []
+                for model in model_list:
                     if current.getParent() == model or current.isAncestorOf(model):
                         selectedModel = model
                         print(f"找到父模型: {selectedModel.getName()}")

core/imgui_webview.py

@@ -0,0 +1,225 @@
+"""
+imgui_webview.py
+后台 playwright 无头浏览器 + 截图 → Panda3D 纹理，供 ImGui 面板显示。
+"""
+from __future__ import annotations
+import threading
+import io
+import time
+
+
+class ImGuiWebView:
+    """
+    后台线程运行 playwright Chromium，定期截图并转换为 Panda3D 纹理。
+    ImGui 直接用 imgui.image() 显示纹理，鼠标/滚轮事件转发给浏览器。
+    """
+
+    def __init__(self, width: int = 1280, height: int = 720):
+        self.view_width  = width
+        self.view_height = height
+
+        # 截图数据（bytes）
+        self._screenshot: bytes | None = None
+        self._screenshot_lock = threading.Lock()
+        self.tex_dirty = False          # 有新截图待上传 GPU
+
+        # 状态
+        self.current_url  = ""
+        self.title        = ""
+        self.is_loading   = False
+        self.error: str | None = None
+
+        # 待处理指令（由 ImGui 线程写，浏览器线程读）
+        self._cmd_navigate: str | None = None
+        self._cmd_click:   tuple | None = None   # (x_ratio, y_ratio)
+        self._cmd_scroll:  float | None = None   # pixels
+        self._cmd_back    = False
+        self._cmd_forward = False
+        self._cmd_reload  = False
+        self._lock = threading.Lock()
+
+        self._running = False
+        self._thread: threading.Thread | None = None
+
+    # ------------------------------------------------------------------ #
+    #  公开控制 API（由 ImGui 线程调用，线程安全）
+    # ------------------------------------------------------------------ #
+
+    def start(self, url: str):
+        if self._running:
+            return
+        self._running = True
+        self._cmd_navigate = url
+        self._thread = threading.Thread(target=self._run, daemon=True,
+                                        name="imgui-webview")
+        self._thread.start()
+
+    def stop(self):
+        self._running = False
+
+    def navigate(self, url: str):
+        if not url.startswith(('http://', 'https://', 'file://')):
+            url = 'https://' + url
+        with self._lock:
+            self._cmd_navigate = url
+        self.is_loading = True
+
+    def click(self, x_ratio: float, y_ratio: float):
+        with self._lock:
+            self._cmd_click = (x_ratio, y_ratio)
+
+    def scroll(self, delta_px: float):
+        with self._lock:
+            self._cmd_scroll = delta_px
+
+    def go_back(self):
+        with self._lock:
+            self._cmd_back = True
+
+    def go_forward(self):
+        with self._lock:
+            self._cmd_forward = True
+
+    def reload(self):
+        with self._lock:
+            self._cmd_reload = True
+
+    def get_screenshot_bytes(self) -> bytes | None:
+        with self._screenshot_lock:
+            return self._screenshot
+
+    # ------------------------------------------------------------------ #
+    #  内部线程
+    # ------------------------------------------------------------------ #
+
+    def _run(self):
+        import asyncio
+        loop = asyncio.new_event_loop()
+        asyncio.set_event_loop(loop)
+        try:
+            loop.run_until_complete(self._async_run())
+        except Exception as exc:
+            self.error = f"WebView 线程异常: {exc}"
+            import traceback; traceback.print_exc()
+        finally:
+            loop.close()
+
+    async def _async_run(self):
+        try:
+            from playwright.async_api import async_playwright
+        except ImportError:
+            self.error = (
+                "playwright 未安装。\n"
+                "请运行: pip install playwright\n"
+                "然后运行: playwright install chromium"
+            )
+            self._running = False
+            return
+
+        try:
+            async with async_playwright() as pw:
+                browser = await pw.chromium.launch(headless=True)
+                ctx = await browser.new_context(
+                    viewport={"width": self.view_width,
+                               "height": self.view_height},
+                    user_agent=(
+                        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+                        "AppleWebKit/537.36 (KHTML, like Gecko) "
+                        "Chrome/121.0.0.0 Safari/537.36"
+                    ),
+                )
+                page = await ctx.new_page()
+
+                # 初次导航
+                start_url = self._cmd_navigate or "about:blank"
+                self._cmd_navigate = None
+                await self._goto(page, start_url)
+                await self._snap(page)
+
+                # 主循环
+                import asyncio
+                while self._running:
+                    await asyncio.sleep(0.05)
+
+                    with self._lock:
+                        nav_url = self._cmd_navigate; self._cmd_navigate = None
+                        clk     = self._cmd_click;    self._cmd_click   = None
+                        scr     = self._cmd_scroll;   self._cmd_scroll  = None
+                        do_back    = self._cmd_back;    self._cmd_back    = False
+                        do_fwd     = self._cmd_forward; self._cmd_forward = False
+                        do_reload  = self._cmd_reload;  self._cmd_reload  = False
+
+                    changed = False
+
+                    if nav_url:
+                        self.is_loading = True
+                        await self._goto(page, nav_url)
+                        changed = True
+                        self.is_loading = False
+
+                    if do_back:
+                        await page.go_back()
+                        await asyncio.sleep(0.5)
+                        changed = True
+
+                    if do_fwd:
+                        await page.go_forward()
+                        await asyncio.sleep(0.5)
+                        changed = True
+
+                    if do_reload:
+                        await page.reload(wait_until="domcontentloaded")
+                        changed = True
+
+                    if clk is not None:
+                        xr, yr = clk
+                        x = int(xr * self.view_width)
+                        y = int(yr * self.view_height)
+                        await page.mouse.click(x, y)
+                        await asyncio.sleep(0.4)
+                        changed = True
+
+                    if scr is not None:
+                        await page.evaluate(
+                            f"window.scrollBy(0, {int(scr)})"
+                        )
+                        await asyncio.sleep(0.15)
+                        changed = True
+
+                    if changed:
+                        self.current_url = page.url
+                        try:
+                            self.title = await page.title()
+                        except Exception:
+                            pass
+                        await self._snap(page)
+
+                await browser.close()
+
+        except Exception as exc:
+            self.error = str(exc)
+            import traceback; traceback.print_exc()
+        finally:
+            self._running = False
+
+    async def _goto(self, page, url: str):
+        import asyncio
+        try:
+            await page.goto(url, wait_until="domcontentloaded", timeout=20_000)
+            self.current_url = page.url
+            try:
+                self.title = await page.title()
+            except Exception:
+                pass
+        except Exception as exc:
+            print(f"[WebView] 导航失败 {url}: {exc}")
+
+    async def _snap(self, page):
+        """截图并更新 self._screenshot"""
+        try:
+            data = await page.screenshot(type="png", full_page=False)
+            with self._screenshot_lock:
+                self._screenshot = data
+            self.tex_dirty = True
+        except Exception as exc:
+            print(f"[WebView] 截图失败: {exc}")

core/patrol_system.py

@@ -1,495 +0,0 @@
-from direct.showbase.ShowBaseGlobal import globalClock
-from direct.task.TaskManagerGlobal import taskMgr
-from panda3d.core import Point3, Vec3
-import math
-
-
-class PatrolSystem:
-    """巡检系统类"""
-
-    def __init__(self, world):
-        """初始化巡检系统
-
-        Args:
-            world: 核心世界对象引用
-        """
-        self.world = world
-
-        # 巡检状态
-        self.is_patrolling = False
-        self.patrol_points = []  # 巡检点列表 [(pos, hpr, wait_time), ...]
-        self.current_patrol_index = 0
-        self.patrol_task = None
-
-        # 巡检参数
-        self.patrol_speed = 5.0  # 巡检移动速度（单位/秒）
-        self.patrol_turn_speed = 90.0  # 转向速度（度/秒）
-        self.patrol_wait_timer = 0.0
-        self.patrol_state = "moving"  # "moving", "turning_to_target", "waiting", "turning_back"
-
-        # 相机状态保存
-        self.original_cam_pos = None
-        self.original_cam_hpr = None
-
-        print("✓ 巡检系统初始化完成")
-
-    def add_patrol_point(self, position, heading=None, wait_time=3.0):
-        if heading is None:
-            if self.patrol_points:
-                last_pos = self.patrol_points[-1][0]
-                direction_x = position[0] - last_pos.x
-                direction_y = position[1] - last_pos.y
-                direction_z = position[2] - last_pos.z
-
-                import math
-                h=math.degrees(math.atan2(-direction_x,-direction_y))
-
-                distance_xy = math.sqrt(direction_x**2+direction_y**2)
-                p = math.degrees(math.atan2(direction_z,distance_xy))
-                p = max(-89,min(89,p))
-
-                r=0
-
-                heading = (h,p,r)
-
-            else:
-                # 使用当前相机朝向
-                current_hpr = self.world.cam.getHpr()
-                heading = (current_hpr.x, current_hpr.y, current_hpr.z)
-
-        pos = Point3(position[0], position[1], position[2])
-        hpr = Vec3(heading[0], heading[1], heading[2])
-
-        self.patrol_points.append((pos, hpr, wait_time))
-        print(f"✓ 添加巡检点 {len(self.patrol_points)}: 位置{position}, 朝向{heading}, 停留{wait_time}秒")
-
-    # 在 PatrolSystem 类中添加以下方法
-
-    def add_auto_heading_patrol_point(self, position, wait_time=3.0):
-        """添加自动计算朝向的巡检点（朝向路径前进方向）
-
-        Args:
-            position: 相机位置 (x, y, z)
-            wait_time: 在该点停留时间（秒）
-        """
-        heading = None  # 将自动计算朝向
-
-        # 复用原有的 add_patrol_point 方法
-        self.add_patrol_point(position, heading, wait_time)
-
-    def add_patrol_point_looking_at(self, position, look_at_position, wait_time=3.0):
-        """添加朝向指定位置的巡检点
-
-        Args:
-            position: 相机位置 (x, y, z)
-            look_at_position: 相机朝向的目标位置 (x, y, z)
-            wait_time: 在该点停留时间（秒）
-        """
-        import math
-
-        # 计算从当前位置到目标位置的方向向量
-        direction_x = look_at_position[0] - position[0]
-        direction_y = look_at_position[1] - position[1]
-        direction_z = look_at_position[2] - position[2]
-
-        # 计算HPR朝向
-        h = math.degrees(math.atan2(-direction_x, -direction_y))
-
-        distance_xy = math.sqrt(direction_x ** 2 + direction_y ** 2)
-        p = math.degrees(math.atan2(direction_z, distance_xy))
-        p = max(-89, min(89, p))  # 限制pitch角度在合理范围内
-
-        r = 0  # roll通常为0
-
-        heading = (h, p, r)
-        self.add_patrol_point(position, heading, wait_time)
-
-    def clear_patrol_points(self):
-        """清空所有巡检点"""
-        self.patrol_points = []
-        print("✓ 巡检点已清空")
-
-    def set_patrol_speed(self, move_speed, turn_speed=None):
-        """设置巡检速度
-
-        Args:
-            move_speed: 移动速度（单位/秒）
-            turn_speed: 转向速度（度/秒），如果为None则保持当前值
-        """
-        self.patrol_speed = move_speed
-        if turn_speed is not None:
-            self.patrol_turn_speed = turn_speed
-        print(f"✓ 巡检速度已设置: 移动{move_speed}, 转向{turn_speed or self.patrol_turn_speed}")
-
-    def start_patrol(self):
-        """开始巡检"""
-        if not self.patrol_points:
-            print("✗ 没有设置巡检点，无法开始巡检")
-            return False
-
-        if self.is_patrolling:
-            print("⚠ 巡检已在进行中")
-            return True
-
-        # 保存当前相机状态
-        self.original_cam_pos = Point3(self.world.cam.getPos())
-        self.original_cam_hpr = Vec3(self.world.cam.getHpr())
-
-        # 重置巡检状态
-        self.current_patrol_index = 0
-        self.patrol_state = "moving"
-        self.patrol_wait_timer = 0.0
-        self.is_patrolling = True
-
-        # 启动巡检任务
-        if self.patrol_task:
-            taskMgr.remove(self.patrol_task)
-        self.patrol_task = taskMgr.add(self._patrol_task, "patrol_task")
-
-        print(f"✓ 开始巡检，共{len(self.patrol_points)}个巡检点")
-        return True
-
-    def stop_patrol(self):
-        """停止巡检"""
-        if not self.is_patrolling:
-            print("⚠ 巡检未在进行中")
-            return False
-
-        # 停止巡检任务
-        if self.patrol_task:
-            taskMgr.remove(self.patrol_task)
-            self.patrol_task = None
-
-        self.is_patrolling = False
-        self.patrol_state = "moving"
-        self.patrol_wait_timer = 0.0
-
-        print("✓ 巡检已停止")
-        return True
-
-    def pause_patrol(self):
-        """暂停巡检"""
-        if not self.is_patrolling:
-            print("⚠ 巡检未在进行中")
-            return False
-
-        if self.patrol_task:
-            taskMgr.remove(self.patrol_task)
-            self.patrol_task = None
-
-        print("✓ 巡检已暂停")
-        return True
-
-    def resume_patrol(self):
-        """恢复巡检"""
-        if self.is_patrolling:
-            print("⚠ 巡检已在进行中")
-            return False
-
-        if not self.patrol_points:
-            print("✗ 没有设置巡检点")
-            return False
-
-        self.is_patrolling = True
-        self.patrol_task = taskMgr.add(self._patrol_task, "patrol_task")
-
-        print("✓ 巡检已恢复")
-        return True
-
-    def reset_to_original_position(self):
-        """重置相机到原始位置"""
-        if self.original_cam_pos and self.original_cam_hpr:
-            self.world.cam.setPos(self.original_cam_pos)
-            self.world.cam.setHpr(self.original_cam_hpr)
-            print("✓ 相机已重置到原始位置")
-            return True
-        else:
-            print("✗ 没有保存的原始位置")
-            return False
-
-    def _patrol_task(self, task):
-        """巡检主任务"""
-        try:
-            if not self.is_patrolling or not self.patrol_points:
-                return task.done
-
-            # 获取当前巡检点
-            current_point = self.patrol_points[self.current_patrol_index]
-            target_pos, target_hpr, wait_time = current_point
-
-            # 根据当前状态执行不同操作
-            if self.patrol_state == "moving":
-                self._handle_moving_state(target_pos)
-            elif self.patrol_state == "turning_to_target":
-                self._handle_turning_to_target_state(target_hpr)
-            elif self.patrol_state == "waiting":
-                self._handle_waiting_state(wait_time)
-            elif self.patrol_state == "turning_back":
-                self._handle_turning_back_state()
-
-            return task.cont
-
-        except Exception as e:
-            print(f"巡检任务出错: {e}")
-            import traceback
-            traceback.print_exc()
-            return task.done
-
-    def _handle_moving_state(self, target_pos):
-        """处理移动状态"""
-        current_pos = self.world.cam.getPos()
-        distance = (target_pos - current_pos).length()
-
-        if distance < 0.1:  # 到达目标点
-            print(f"✓ 到达巡检点 {self.current_patrol_index + 1}")
-            self.patrol_state = "turning_to_target"
-            return
-
-        # 计算移动方向和距离
-        direction = target_pos - current_pos
-        direction.normalize()
-
-        # 计算目标朝向（看向目标点）
-        target_hpr = self._look_at_to_hpr(direction)
-        current_hpr = self.world.cam.getHpr()
-
-        # 平滑转向到目标朝向
-        h_diff = self._normalize_angle(target_hpr.x - current_hpr.x)
-        p_diff = self._normalize_angle(target_hpr.y - current_hpr.y)
-        r_diff = self._normalize_angle(target_hpr.z - current_hpr.z)
-
-        # 计算本帧应转动的角度
-        dt = globalClock.getDt()
-        turn_amount = self.patrol_turn_speed * dt
-
-        # 逐步转向目标角度
-        new_hpr = Vec3(current_hpr)
-
-        if abs(h_diff) > turn_amount:
-            new_hpr.x += turn_amount if h_diff > 0 else -turn_amount
-        else:
-            new_hpr.x = target_hpr.x
-
-        if abs(p_diff) > turn_amount:
-            new_hpr.y += turn_amount if p_diff > 0 else -turn_amount
-        else:
-            new_hpr.y = target_hpr.y
-
-        if abs(r_diff) > turn_amount:
-            new_hpr.z += turn_amount if r_diff > 0 else -turn_amount
-        else:
-            new_hpr.z = target_hpr.z
-
-        self.world.cam.setHpr(new_hpr)
-
-        # 计算本帧应移动的距离
-        move_distance = self.patrol_speed * dt
-
-        # 如果移动距离大于剩余距离，则直接移动到目标点
-        if move_distance >= distance:
-            self.world.cam.setPos(target_pos)
-        else:
-            # 否则按方向移动
-            new_pos = current_pos + direction * move_distance
-            self.world.cam.setPos(new_pos)
-
-    def _handle_turning_to_target_state(self, target_hpr):
-        """处理转向目标状态"""
-        # 检查是否需要朝向下一个点
-        if target_hpr == "look_next":
-            # 计算朝向下一个点的方向
-            next_index = (self.current_patrol_index + 1) % len(self.patrol_points)
-            next_point_pos = self.patrol_points[next_index][0]
-
-            current_pos = self.world.cam.getPos()
-            direction = next_point_pos - current_pos
-            direction.normalize()
-
-            # 计算目标朝向
-            target_hpr = self._look_at_to_hpr(direction)
-
-        current_hpr = self.world.cam.getHpr()
-
-        # 计算角度差
-        h_diff = self._normalize_angle(target_hpr.x - current_hpr.x)
-        p_diff = self._normalize_angle(target_hpr.y - current_hpr.y)
-        r_diff = self._normalize_angle(target_hpr.z - current_hpr.z)
-
-        # 检查是否已完成转向
-        if abs(h_diff) < 1.0 and abs(p_diff) < 1.0 and abs(r_diff) < 1.0:
-            print(f"✓ 完成转向，开始停留")
-            self.patrol_state = "waiting"
-            self.patrol_wait_timer = 0.0
-            return
-
-        # 计算本帧应转动的角度
-        dt = globalClock.getDt()
-        turn_amount = self.patrol_turn_speed * dt
-
-        # 逐步转向目标角度
-        new_hpr = Vec3(current_hpr)
-
-        if abs(h_diff) > turn_amount:
-            new_hpr.x += turn_amount if h_diff > 0 else -turn_amount
-        else:
-            new_hpr.x = target_hpr.x
-
-        if abs(p_diff) > turn_amount:
-            new_hpr.y += turn_amount if p_diff > 0 else -turn_amount
-        else:
-            new_hpr.y = target_hpr.y
-
-        if abs(r_diff) > turn_amount:
-            new_hpr.z += turn_amount if r_diff > 0 else -turn_amount
-        else:
-            new_hpr.z = target_hpr.z
-
-        self.world.cam.setHpr(new_hpr)
-
-    def _handle_waiting_state(self, wait_time):
-        """处理等待状态"""
-        self.patrol_wait_timer += globalClock.getDt()
-
-        if self.patrol_wait_timer >= wait_time:
-            print(f"✓ 停留结束，准备转回原朝向")
-            self.patrol_state = "turning_back"
-
-    # 修改 core/patrol_system.py 中的 _handle_turning_back_state 方法
-
-    def _handle_turning_back_state(self):
-        """处理转回原朝向状态"""
-        # 直接完成转向状态，进入移动状态
-        print(f"✓ 停留结束，开始移动到下一个点")
-        # 移动到下一个巡检点
-        next_index = (self.current_patrol_index + 1) % len(self.patrol_points)
-        self.current_patrol_index = next_index
-        self.patrol_state = "moving"
-        return
-
-    def _normalize_angle(self, angle):
-        """规范化角度到-180到180度之间"""
-        while angle > 180:
-            angle -= 360
-        while angle < -180:
-            angle += 360
-        return angle
-
-    def _look_at_to_hpr(self, direction):
-        """将方向向量转换为HPR角度"""
-        # 简化的转换，实际应用中可能需要更精确的计算
-        h = math.degrees(math.atan2(-direction.x, -direction.y))
-        p = math.degrees(math.asin(direction.z))
-        return Vec3(h, p, 0)
-
-    def get_patrol_status(self):
-        """获取巡检状态信息"""
-        return {
-            "is_patrolling": self.is_patrolling,
-            "current_point": self.current_patrol_index,
-            "total_points": len(self.patrol_points),
-            "state": self.patrol_state,
-            "wait_timer": self.patrol_wait_timer
-        }
-
-    def list_patrol_points(self):
-        """列出所有巡检点"""
-        if not self.patrol_points:
-            print("没有设置巡检点")
-            return
-
-        print(f"巡检点列表 (共{len(self.patrol_points)}个):")
-        for i, (pos, hpr, wait_time) in enumerate(self.patrol_points):
-            current_marker = " >>>" if i == self.current_patrol_index and self.is_patrolling else ""
-            print(f"  {i + 1}. 位置:({pos.x:.1f}, {pos.y:.1f}, {pos.z:.1f}) "
-                  f"朝向:({hpr.x:.1f}, {hpr.y:.1f}, {hpr.z:.1f}) "
-                  f"停留:{wait_time}秒{current_marker}")
-
-    def remove_patrol_point(self, index):
-        """移除指定索引的巡检点"""
-        if 0 <= index < len(self.patrol_points):
-            removed_point = self.patrol_points.pop(index)
-            print(
-                f"✓ 移除巡检点 {index + 1}: 位置({removed_point[0].x:.1f}, {removed_point[0].y:.1f}, {removed_point[0].z:.1f})")
-
-            # 调整当前索引
-            if self.current_patrol_index >= len(self.patrol_points) and self.patrol_points:
-                self.current_patrol_index = len(self.patrol_points) - 1
-            elif self.current_patrol_index >= len(self.patrol_points):
-                self.current_patrol_index = 0
-        else:
-            print(f"✗ 无效的巡检点索引: {index}")
-
-    def insert_patrol_point(self, index, position, heading=None, wait_time=3.0):
-        """在指定位置插入巡检点"""
-        if index < 0 or index > len(self.patrol_points):
-            print(f"✗ 无效的插入位置: {index}")
-            return
-
-        if heading is None:
-            # 使用当前相机朝向
-            current_hpr = self.world.cam.getHpr()
-            heading = (current_hpr.x, current_hpr.y, current_hpr.z)
-
-        pos = Point3(position[0], position[1], position[2])
-        hpr = Vec3(heading[0], heading[1], heading[2])
-
-        self.patrol_points.insert(index, (pos, hpr, wait_time))
-        print(f"✓ 在位置 {index + 1} 插入巡检点: 位置{position}, 朝向{heading}, 停留{wait_time}秒")
-
-    def update_patrol_point(self, index, position=None, heading=None, wait_time=None):
-        """更新指定巡检点的信息"""
-        if 0 <= index < len(self.patrol_points):
-            pos, hpr, wt = self.patrol_points[index]
-
-            if position is not None:
-                pos = Point3(position[0], position[1], position[2])
-            if heading is not None:
-                hpr = Vec3(heading[0], heading[1], heading[2])
-            if wait_time is not None:
-                wt = wait_time
-
-            self.patrol_points[index] = (pos, hpr, wt)
-            print(f"✓ 更新巡检点 {index + 1}")
-        else:
-            print(f"✗ 无效的巡检点索引: {index}")
-
-    def goto_patrol_point(self, index):
-        """直接跳转到指定巡检点"""
-        if not self.patrol_points:
-            print("✗ 没有设置巡检点")
-            return False
-
-        if 0 <= index < len(self.patrol_points):
-            pos, hpr, _ = self.patrol_points[index]
-            self.world.cam.setPos(pos)
-            self.world.cam.setHpr(hpr)
-            self.current_patrol_index = index
-            print(f"✓ 跳转到巡检点 {index + 1}")
-            return True
-        else:
-            print(f"✗ 无效的巡检点索引: {index}")
-            return False
-
-    def cleanup(self):
-        """清理巡检系统资源"""
-        self.stop_patrol()
-        self.clear_patrol_points()
-        self.original_cam_pos = None
-        self.original_cam_hpr = None
-        print("✓ 巡检系统资源已清理")
-
-
-# 使用示例和便捷函数
-def create_default_patrol_route(patrol_system):
-    """创建默认的巡检路线（示例）"""
-    # 清空现有巡检点
-    patrol_system.clear_patrol_points()
-
-    # 添加一些示例巡检点
-    patrol_system.add_patrol_point((0, -20, 5), (0, -15, 0), 2.0)  # 点1：前方低位置
-    patrol_system.add_patrol_point((0, 0, 10), (0, -30, 0), 3.0)  # 点2：中央高位置
-    patrol_system.add_patrol_point((15, 10, 5), (-45, -10, 0), 2.5)  # 点3：右侧位置
-    patrol_system.add_patrol_point((-15, 10, 5), (45, -10, 0), 2.5)  # 点4：左侧位置
-
-    print("✓ 默认巡检路线已创建")
-

core/resource_manager.py

@@ -389,6 +389,12 @@ class ResourceManager:
                     shutil.copytree(src, dst)
                 else:
                     shutil.copy2(src, dst)
+                    if src.suffix.lower() == '.fbx':
+                        fbm_src = src.with_name(src.stem + '.fbm')
+                        if fbm_src.exists() and fbm_src.is_dir():
+                            fbm_dst = destination_root / fbm_src.name
+                            if not fbm_dst.exists():
+                                shutil.copytree(fbm_src, fbm_dst)
                 imported.append(dst)
             except Exception as e:
                 errors.append(f"导入失败 {src}: {e}")

core/selection.py

@@ -14,6 +14,7 @@ from panda3d.core import (Vec3, Point3, Point2, LineSegs, ColorAttrib, RenderSta
                           TransparencyAttrib, Vec4, CollisionCapsule)
 from direct.task.TaskManagerGlobal import taskMgr
 import math
+from core.selection_outline import SelectionOutlineManager
 
 class SelectionSystem:
     """选择和变换系统类"""
@@ -30,6 +31,17 @@ class SelectionSystem:
         self.selectedNode = None
         self.selectionBox = None  # 选择框
         self.selectionBoxTarget = None  # 选择框跟踪的目标节点
+        self.show_selection_box = False
+        self.enable_unity_outline = True
+        self.outline_manager = getattr(self.world, "_selection_outline_manager", None)
+        if not self.outline_manager:
+            self.outline_manager = SelectionOutlineManager(
+                self.world,
+                enabled=self.enable_unity_outline,
+            )
+            setattr(self.world, "_selection_outline_manager", self.outline_manager)
+        else:
+            self.outline_manager.set_enabled(self.enable_unity_outline)
 
         # 坐标轴工具（Gizmo）相关
         self.gizmo = None  # 坐标轴
@@ -207,6 +219,21 @@ class SelectionSystem:
             import traceback
             traceback.print_exc()
 
+    def _updateSelectionOutline(self, nodePath):
+        """Update Unity-like selection outline visuals."""
+        try:
+            if not getattr(self, "outline_manager", None):
+                return
+            if not self.enable_unity_outline:
+                self.outline_manager.clear()
+                return
+            if nodePath and not nodePath.isEmpty():
+                self.outline_manager.set_targets([nodePath])
+            else:
+                self.outline_manager.clear()
+        except Exception as e:
+            print(f"update selection outline failed: {e}")
+
     def updateSelectionBoxGeometry(self):
         """更新选择框的几何形状和位置"""
         try:
@@ -1963,10 +1990,30 @@ class SelectionSystem:
                         command = RotateNodeCommand(self.gizmoTarget, self.gizmoTargetStartHpr, current_hpr)
                         self.world.command_manager.execute_command(command)
                         print(f"创建旋转命令: {self.gizmoTargetStartHpr} -> {current_hpr}")
+                        print(f"创建旋转命令: {self.gizmoTargetStartHpr} -> {current_hpr}")
                         
             except Exception as e:
                 print(f"创建撤销命令时出错: {e}")
         
+        # 同步碰撞体
+        try:
+            target = self.gizmoTarget
+            if target:
+                # 寻找它的所属模型根节点
+                root_model = target
+                while root_model and root_model != self.world.render:
+                    model_list = self.world.models if hasattr(self.world, 'models') else []
+                    if root_model in model_list or root_model.hasTag('is_model_root'):
+                        break
+                    root_model = root_model.getParent()
+                
+                # 如果这个节点属于某个模型，或者是模型自己，更新该模型的碰撞边界
+                if root_model and hasattr(self.world, 'models') and root_model in self.world.models:
+                    if hasattr(self.world, 'scene_manager') and hasattr(self.world.scene_manager, 'refreshCollisionBounds'):
+                        self.world.scene_manager.refreshCollisionBounds(root_model)
+        except Exception as e:
+            print(f"同步模型碰撞体失败: {e}")
+
         # 恢复所有轴的颜色
         for axis_name in ["x", "y", "z"]:
             self.setGizmoAxisColor(axis_name, self.gizmo_colors[axis_name])
@@ -2045,8 +2092,12 @@ class SelectionSystem:
 
                 # 创建选择框
                 #print("创建选择框...")
-                self.createSelectionBox(nodePath)
+                if self.show_selection_box:
-                if self.selectionBox:
+                    self.createSelectionBox(nodePath)
+                else:
+                    self.clearSelectionBox()
+
+                if (not self.show_selection_box) or self.selectionBox:
                     box_name = "Unknown"
                     if self.selectionBox and not self.selectionBox.isEmpty():
                         box_name = self.selectionBox.getName()
@@ -2056,6 +2107,7 @@ class SelectionSystem:
 
                 # 创建坐标轴
                 #print("创建坐标轴...")
+                self._updateSelectionOutline(nodePath)
                 self.createGizmo(nodePath)
                 if self.gizmo:
                     gizmo_name = "Unknown"
@@ -2068,6 +2120,7 @@ class SelectionSystem:
             else:
                 print("清除选择...")
                 self.clearSelectionBox()
+                self._updateSelectionOutline(None)
                 self.clearGizmo()
                 print("✓ 取消选择")
 
@@ -2145,6 +2198,40 @@ class SelectionSystem:
         """获取当前选中的节点"""
         return self.selectedNode
 
+    def deleteSelectedNode(self):
+        """兼容旧接口：删除当前选中节点。"""
+        node = self.selectedNode
+        if not node or node.isEmpty():
+            return False
+
+        try:
+            if node.getName() == "render":
+                return False
+        except Exception:
+            return False
+
+        self._deleting_node = True
+        try:
+            # 优先走应用层统一删除链路（命令系统/SSBO清理/消息等）。
+            if hasattr(self.world, "_delete_node") and callable(self.world._delete_node):
+                deleted = bool(self.world._delete_node(node))
+            else:
+                deleted = False
+                scene_manager = getattr(self.world, "scene_manager", None)
+                if scene_manager and hasattr(scene_manager, "models") and node in scene_manager.models:
+                    scene_manager.models.remove(node)
+                try:
+                    node.removeNode()
+                    deleted = True
+                except Exception:
+                    deleted = False
+
+            if deleted:
+                self.clearSelection()
+            return deleted
+        finally:
+            self._deleting_node = False
+
     def hasSelection(self):
         """检查是否有选中的节点"""
         return self.selectedNode is not None
@@ -2156,6 +2243,9 @@ class SelectionSystem:
         if (self.selectionBoxTarget and self.selectionBoxTarget.isEmpty()):
             self.clearSelectionBox()
 
+        if self.selectedNode and self.selectedNode.isEmpty():
+            self._updateSelectionOutline(None)
+
     def setupGizmoCollision(self):
         if not self.gizmo or not self.gizmoXAxis or not self.gizmoYAxis or not self.gizmoZAxis:
             return
@@ -2875,6 +2965,7 @@ class SelectionSystem:
 
         # 清理其他资源
         self.clearSelectionBox()
+        self._updateSelectionOutline(None)
         self.clearGizmo()
     def clearSelection(self):
         """清除当前选择"""
@@ -2882,6 +2973,7 @@ class SelectionSystem:
             self.selectedNode = None
             self.selectedObject = None
             self.clearSelectionBox()
+            self._updateSelectionOutline(None)
             self.clearGizmo()
 
             # 清除树形控件中的选择

core/selection_outline.py

@@ -0,0 +1,303 @@
+from direct.task.TaskManagerGlobal import taskMgr
+from panda3d.core import (
+    BitMask32,
+    Camera,
+    FrameBufferProperties,
+    GraphicsOutput,
+    GraphicsPipe,
+    NodePath,
+    Shader,
+    Texture,
+    Vec4,
+    WindowProperties,
+)
+
+
+class SelectionOutlineManager:
+    """Selection mask manager feeding RenderPipeline SelectionOutlineStage."""
+
+    OUTLINE_PREFIX = "selectionOutline"
+
+    def __init__(
+        self,
+        app,
+        enabled=True,
+        outline_color=Vec4(1.0, 0.55, 0.0, 1.0),
+        outline_width_px=2.0,
+        fill_alpha=0.0,
+        max_targets=128,
+    ):
+        self.app = app
+        self.enabled = bool(enabled)
+        self.outline_color = Vec4(outline_color)
+        self.outline_width_px = max(0.0, float(outline_width_px))
+        self.fill_alpha = max(0.0, min(1.0, float(fill_alpha)))
+        self.max_targets = max(1, int(max_targets))
+
+        self._task_name = "selection_outline_mask_sync"
+        self._tracked = []  # [(source_np, clone_np), ...]
+        self._stage_missing_warned = False
+
+        self._mask_root = NodePath(f"{self.OUTLINE_PREFIX}_mask_root")
+        self._mask_buffer = None
+        self._mask_texture = None
+        self._mask_cam = None
+        self._mask_cam_np = None
+        self._mask_shader = self._build_mask_shader()
+        self._buffer_size = (0, 0)
+
+    @staticmethod
+    def _is_empty(np):
+        if not np:
+            return True
+        if hasattr(np, "isEmpty"):
+            return np.isEmpty()
+        if hasattr(np, "is_empty"):
+            return np.is_empty()
+        return False
+
+    @classmethod
+    def is_outline_node(cls, node_path):
+        if not node_path or cls._is_empty(node_path):
+            return False
+        name = node_path.getName() if hasattr(node_path, "getName") else ""
+        if name.startswith(cls.OUTLINE_PREFIX):
+            return True
+        try:
+            if node_path.hasPythonTag("selection_outline"):
+                return True
+        except Exception:
+            pass
+        return False
+
+    def set_enabled(self, enabled):
+        self.enabled = bool(enabled)
+        if not self.enabled:
+            self.clear()
+        self._apply_stage_inputs()
+
+    def set_targets(self, targets):
+        if not self.enabled:
+            self.clear()
+            self._apply_stage_inputs()
+            return
+
+        self._ensure_mask_resources()
+        self.clear()
+
+        if not targets:
+            self._apply_stage_inputs()
+            return
+
+        seen = set()
+        valid = []
+        for target in targets:
+            if self._is_empty(target) or self.is_outline_node(target):
+                continue
+            key = str(target)
+            if key in seen:
+                continue
+            seen.add(key)
+            valid.append(target)
+            if len(valid) >= self.max_targets:
+                break
+
+        for source in valid:
+            self._clone_target(source)
+
+        if self._tracked:
+            self._start_sync_task()
+            self._sync_once()
+            print(f"[SelectionOutline] targets={len(self._tracked)} active")
+        else:
+            print("[SelectionOutline] no valid targets for outline")
+
+        self._apply_stage_inputs()
+
+    def clear(self):
+        self._stop_sync_task()
+        for _, clone_np in self._tracked:
+            if not self._is_empty(clone_np):
+                clone_np.removeNode()
+        self._tracked = []
+        self._apply_stage_inputs()
+
+    def cleanup(self):
+        self.clear()
+        self._destroy_mask_resources()
+
+    def _build_mask_shader(self):
+        return Shader.make(
+            Shader.SL_GLSL,
+            """
+#version 430
+in vec4 p3d_Vertex;
+uniform mat4 p3d_ModelViewProjectionMatrix;
+void main() {
+    gl_Position = p3d_ModelViewProjectionMatrix * p3d_Vertex;
+}
+""",
+            """
+#version 430
+out vec4 result;
+void main() {
+    result = vec4(1.0, 1.0, 1.0, 1.0);
+}
+""",
+        )
+
+    def _clone_target(self, source):
+        try:
+            clone_np = source.copyTo(self._mask_root)
+            if self._is_empty(clone_np):
+                return
+            if not self._node_has_geom(clone_np):
+                clone_np.removeNode()
+                return
+            clone_np.setName(f"{self.OUTLINE_PREFIX}_{source.getName()}")
+            clone_np.setPythonTag("selection_outline", True)
+            clone_np.setCollideMask(BitMask32.allOff())
+            clone_np.setMat(self.app.render, source.getMat(self.app.render))
+            self._tracked.append((source, clone_np))
+        except Exception as exc:
+            print(f"[SelectionOutline] clone failed: {exc}")
+
+    def _node_has_geom(self, np):
+        if self._is_empty(np):
+            return False
+        try:
+            node = np.node()
+            if node and hasattr(node, "isGeomNode") and node.isGeomNode():
+                return True
+        except Exception:
+            pass
+        try:
+            return not np.find("**/+GeomNode").isEmpty()
+        except Exception:
+            return False
+
+    def _start_sync_task(self):
+        taskMgr.remove(self._task_name)
+        taskMgr.add(self._sync_task, self._task_name)
+
+    def _stop_sync_task(self):
+        taskMgr.remove(self._task_name)
+
+    def _sync_task(self, task):
+        self._sync_once()
+        return task.cont
+
+    def _sync_once(self):
+        if not self.enabled:
+            self._apply_stage_inputs()
+            return
+
+        self._ensure_mask_resources()
+        alive = []
+        for source, clone_np in self._tracked:
+            if self._is_empty(source) or self._is_empty(clone_np):
+                if not self._is_empty(clone_np):
+                    clone_np.removeNode()
+                continue
+            clone_np.setMat(self.app.render, source.getMat(self.app.render))
+            alive.append((source, clone_np))
+        self._tracked = alive
+        self._apply_stage_inputs()
+
+    def _get_stage(self):
+        rp = getattr(self.app, "render_pipeline", None)
+        if not rp or not getattr(rp, "stage_mgr", None):
+            return None
+        return rp.stage_mgr.get_stage("SelectionOutlineStage")
+
+    def _apply_stage_inputs(self):
+        stage = self._get_stage()
+        if not stage:
+            if not self._stage_missing_warned:
+                print("[SelectionOutline] SelectionOutlineStage not found; plugin may be disabled.")
+                self._stage_missing_warned = True
+            return
+
+        self._stage_missing_warned = False
+        stage.set_outline_style(
+            color=self.outline_color,
+            width_px=self.outline_width_px,
+            fill_alpha=self.fill_alpha,
+        )
+        stage.set_mask_texture(self._mask_texture)
+        stage.set_enabled_outline(self.enabled and bool(self._tracked))
+
+    def _get_window_size(self):
+        if not getattr(self.app, "win", None):
+            return 1, 1
+        return max(1, self.app.win.getXSize()), max(1, self.app.win.getYSize())
+
+    def _ensure_mask_resources(self):
+        size = self._get_window_size()
+        if size != self._buffer_size:
+            self._destroy_mask_resources()
+            self._buffer_size = size
+        if self._mask_buffer:
+            return
+
+        if not getattr(self.app, "graphicsEngine", None) or not getattr(self.app, "win", None):
+            return
+
+        w, h = self._buffer_size
+        win_props = WindowProperties()
+        win_props.setSize(w, h)
+        fb_props = FrameBufferProperties()
+        fb_props.setRgbaBits(8, 8, 8, 8)
+        fb_props.setDepthBits(24)
+
+        self._mask_buffer = self.app.graphicsEngine.make_output(
+            self.app.pipe,
+            "selection_outline_mask",
+            -80,
+            fb_props,
+            win_props,
+            GraphicsPipe.BFRefuseWindow,
+            self.app.win.getGsg(),
+            self.app.win,
+        )
+        if not self._mask_buffer:
+            print("[SelectionOutline] failed to create mask buffer")
+            return
+
+        self._mask_texture = Texture("selection_outline_mask_tex")
+        self._mask_texture.setMinfilter(Texture.FTNearest)
+        self._mask_texture.setMagfilter(Texture.FTNearest)
+        self._mask_buffer.addRenderTexture(self._mask_texture, GraphicsOutput.RTMBindOrCopy)
+        self._mask_buffer.setClearColor(Vec4(0, 0, 0, 0))
+        self._mask_buffer.setClearColorActive(True)
+        self._mask_buffer.setActive(True)
+
+        self._mask_cam = Camera("selection_outline_mask_camera")
+        self._mask_cam.setScene(self._mask_root)
+        self._mask_cam.setLens(self.app.camLens)
+        self._mask_cam_np = self.app.cam.attachNewNode(self._mask_cam)
+
+        dr = self._mask_buffer.makeDisplayRegion()
+        dr.setCamera(self._mask_cam_np)
+
+        state_np = NodePath("selection_outline_mask_state")
+        state_np.setShader(self._mask_shader, 10000)
+        state_np.setLightOff(1)
+        state_np.setMaterialOff(1)
+        state_np.setTextureOff(1)
+        state_np.setColorOff(1)
+        self._mask_cam.setInitialState(state_np.getState())
+
+    def _destroy_mask_resources(self):
+        if self._mask_cam_np and not self._is_empty(self._mask_cam_np):
+            self._mask_cam_np.removeNode()
+        self._mask_cam_np = None
+        self._mask_cam = None
+
+        if self._mask_buffer and getattr(self.app, "graphicsEngine", None):
+            try:
+                self.app.graphicsEngine.removeWindow(self._mask_buffer)
+            except Exception:
+                pass
+        self._mask_buffer = None
+        self._mask_texture = None

core/world.py

@@ -14,8 +14,9 @@ from direct.showbase.ShowBase import ShowBase
 from direct.showbase.ShowBaseGlobal import globalClock
 from scene.scene_manager import SceneManager
 
-# 设置 RenderPipelineFile 路径
+# 设置 RenderPipelineFile 路径。打包后可通过环境变量显式指定项目根目录，
-project_root = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+# 避免 AppImage / 独立发行目录的布局差异导致资源定位错误。
+project_root = os.environ.get("EG_PROJECT_ROOT") or os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 render_pipeline_path = os.path.join(project_root, "RenderPipelineFile")
 sys.path.insert(0, render_pipeline_path)
 
@@ -65,8 +66,16 @@ class CoreWorld(ShowBase):
         if is_fullscreen:
             loadPrcFileData("", "fullscreen #t")
 
-        # 创建渲染管线
+        # 创建渲染管线。打包后 Nuitka 会把 rpcore/rplibs 平铺到发行目录，
+        # 自动探测到的 base_path 可能变成 dist 根目录，导致找不到
+        # RenderPipelineFile/config/pipeline.yaml，这里显式指回真实管线目录。
         self.render_pipeline = RenderPipeline()
+        rp_base_dir = os.path.join(project_root, "RenderPipelineFile")
+        rp_config_dir = os.path.join(rp_base_dir, "config")
+        if os.path.isdir(rp_base_dir):
+            self.render_pipeline.mount_mgr.base_path = rp_base_dir
+        if os.path.isdir(rp_config_dir):
+            self.render_pipeline.mount_mgr.config_dir = rp_config_dir
         self.render_pipeline.pre_showbase_init()
         
         # 强制开启多线程支持 (Video播放必需)

imgui.ini

@@ -1,222 +0,0 @@
-[Window][Debug##Default]
-Pos=28,465
-Size=400,400
-Collapsed=0
-
-[Window][Explore: render]
-Pos=60,60
-Size=410,761
-Collapsed=0
-
-[Window][Node Placer "camera"]
-Pos=486,83
-Size=472,216
-Collapsed=0
-
-[Window][Time Slider "pandaPace"]
-Pos=-69,89
-Size=745,76
-Collapsed=0
-
-[Window][Dear ImGui Demo]
-Pos=241,149
-Size=832,45
-Collapsed=0
-
-[Window][工具栏]
-Pos=241,20
-Size=908,74
-Collapsed=0
-DockId=0x0000000D,0
-
-[Window][场景树]
-Pos=0,20
-Size=239,468
-Collapsed=0
-DockId=0x00000007,0
-
-[Window][属性面板]
-Pos=1151,20
-Size=229,730
-Collapsed=0
-DockId=0x00000003,0
-
-[Window][控制台]
-Pos=0,490
-Size=239,260
-Collapsed=0
-DockId=0x00000008,0
-
-[Window][脚本管理]
-Pos=1540,20
-Size=380,390
-Collapsed=0
-DockId=0x00000003,1
-
-[Window][中文显示测试]
-Pos=60,60
-Size=135,263
-Collapsed=0
-
-[Window][WindowOverViewport_11111111]
-Pos=0,20
-Size=1380,730
-Collapsed=0
-
-[Window][测试窗口1]
-Pos=60,60
-Size=121,82
-Collapsed=0
-
-[Window][测试窗口2]
-Pos=60,60
-Size=121,82
-Collapsed=0
-
-[Window][测试窗口3]
-Pos=60,60
-Size=93,65
-Collapsed=0
-
-[Window][新建项目]
-Pos=760,354
-Size=400,300
-Collapsed=0
-
-[Window][选择路径]
-Pos=660,254
-Size=600,500
-Collapsed=0
-
-[Window][打开项目]
-Pos=710,304
-Size=500,400
-Collapsed=0
-
-[Window][导入模型]
-Pos=660,254
-Size=600,500
-Collapsed=0
-
-[Window][资源管理器]
-Pos=241,464
-Size=908,286
-Collapsed=0
-DockId=0x00000006,0
-
-[Window][创建3D文本]
-Pos=60,60
-Size=88,202
-Collapsed=0
-
-[Window][创建GUI按钮]
-Pos=60,60
-Size=93,226
-Collapsed=0
-
-[Window][创建3D图片]
-Pos=60,60
-Size=88,226
-Collapsed=0
-
-[Window][创建点光源]
-Pos=60,60
-Size=109,274
-Collapsed=0
-
-[Window][创建GUI标签]
-Pos=60,60
-Size=93,226
-Collapsed=0
-
-[Window][创建聚光灯]
-Pos=60,60
-Size=89,250
-Collapsed=0
-
-[Window][颜色选择器]
-Pos=810,304
-Size=300,400
-Collapsed=0
-
-[Window][选择diffuse纹理文件##texture_dialog]
-Pos=660,304
-Size=600,400
-Collapsed=0
-
-[Window][创建GUI图片]
-Pos=60,60
-Size=101,226
-Collapsed=0
-
-[Window][LUI编辑器]
-Pos=1628,412
-Size=292,597
-Collapsed=0
-DockId=0x00000004,0
-
-[Window][LUI测试控制面板]
-Pos=6,10
-Size=300,200
-Collapsed=1
-
-[Window][选择高度图文件]
-Pos=60,60
-Size=596,498
-Collapsed=0
-
-[Window][创建平面地形]
-Pos=61,60
-Size=238,176
-Collapsed=0
-
-[Window][选择normal纹理文件##texture_dialog]
-Pos=660,304
-Size=600,400
-Collapsed=0
-
-[Window][选择metallic纹理文件##texture_dialog]
-Pos=660,304
-Size=600,400
-Collapsed=0
-
-[Window][选择roughness纹理文件##texture_dialog]
-Pos=660,304
-Size=600,400
-Collapsed=0
-
-[Window][VR状态]
-Pos=162,152
-Size=138,296
-Collapsed=0
-
-[Window][VR设置]
-Pos=474,130
-Size=120,384
-Collapsed=0
-
-[Window][选择emission纹理文件##texture_dialog]
-Pos=660,304
-Size=600,400
-Collapsed=0
-
-[Window][Web面板]
-Pos=373,98
-Size=942,580
-Collapsed=0
-
-[Docking][Data]
-DockSpace         ID=0x08BD597D Window=0x1BBC0F80 Pos=0,20 Size=1380,730 Split=X
-  DockNode        ID=0x00000001 Parent=0x08BD597D SizeRef=1689,989 Split=X
-    DockNode      ID=0x00000009 Parent=0x00000001 SizeRef=239,989 Split=Y Selected=0xE0015051
-      DockNode    ID=0x00000007 Parent=0x00000009 SizeRef=271,634 Selected=0xE0015051
-      DockNode    ID=0x00000008 Parent=0x00000009 SizeRef=271,353 Selected=0x5428E753
-    DockNode      ID=0x0000000A Parent=0x00000001 SizeRef=1448,989 Split=Y
-      DockNode    ID=0x0000000D Parent=0x0000000A SizeRef=1318,74 HiddenTabBar=1 Selected=0x43A39006
-      DockNode    ID=0x0000000E Parent=0x0000000A SizeRef=1318,913 Split=Y
-        DockNode  ID=0x00000005 Parent=0x0000000E SizeRef=1341,625 CentralNode=1
-        DockNode  ID=0x00000006 Parent=0x0000000E SizeRef=1341,286 Selected=0x3A2E05C3
-  DockNode        ID=0x00000002 Parent=0x08BD597D SizeRef=229,989 Split=Y Selected=0x3188AB8D
-    DockNode      ID=0x00000003 Parent=0x00000002 SizeRef=351,390 Selected=0x5DB6FF37
-    DockNode      ID=0x00000004 Parent=0x00000002 SizeRef=351,597 Selected=0x1EB923B7
-

project/project_manager.py

@@ -4,6 +4,7 @@ import json
 import datetime
 import subprocess
 import shutil
+from project.webgl_packager import WebGLPackager
 
 class ProjectManager:
 
@@ -11,6 +12,7 @@ class ProjectManager:
         self.world = world
         self.current_project_path = None
         self.project_config = None
+        self.last_webgl_export_report = None
         
         print("✓ 项目管理系统初始化完成")
     
@@ -300,6 +302,52 @@ class ProjectManager:
             return False
     
     # ==================== 项目打包功能 ====================
+
+    def buildWebGLPackage(self, output_dir):
+        """将当前项目打包为 WebGL 静态站点目录。
+
+        Args:
+            output_dir (str): 用户选择的输出根目录
+
+        Returns:
+            bool: success/partial 返回 True，failed 返回 False
+        """
+        try:
+            if not self.current_project_path:
+                print("错误: 请先创建或打开一个项目！")
+                return False
+
+            if not output_dir:
+                print("错误: 请指定 WebGL 打包输出目录！")
+                return False
+
+            project_path = os.path.normpath(self.current_project_path)
+            output_dir = os.path.normpath(output_dir)
+
+            packager = WebGLPackager(self.world)
+            report = packager.package(project_path, output_dir)
+            self.last_webgl_export_report = report
+
+            status = report.get("status", "failed")
+            report_path = os.path.join(
+                report.get("output_dir", ""),
+                "reports",
+                "export_report.json",
+            )
+
+            if status in ("success", "partial"):
+                print(f"WebGL打包完成: {status}")
+                print(f"输出目录: {report.get('output_dir', '')}")
+                print(f"报告路径: {report_path}")
+                return True
+
+            print("WebGL打包失败")
+            print(f"报告路径: {report_path}")
+            return False
+
+        except Exception as e:
+            print(f"WebGL打包过程出错: {str(e)}")
+            return False
     
     def buildPackage(self, build_dir):
         """打包项目为可执行文件 - 按照Panda3D官方标准方法
@@ -478,7 +526,6 @@ class ProjectManager:
     def _copyScriptSystemToBuild(self,build_dir):
             core_files = [
                 "script_system.py",
-                "InfoPanelManager.py",
                 "CustomMouseController.py"
             ]
 
@@ -840,4 +887,4 @@ setup(
             
         except Exception as e:
             print(f"执行打包命令失败: {str(e)}")
-            return False
+            return False

project/webgl_packager.py

@@ -0,0 +1,995 @@
+"""WebGL project packager for EG editor (Three.js static scene export)."""
+
+from __future__ import annotations
+
+import datetime
+import json
+import os
+import re
+import shutil
+import stat
+import subprocess
+import tempfile
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Tuple
+
+
+class WebGLPackager:
+    """Export current EG scene into a static WebGL package directory."""
+
+    BASIS_MATRIX_ROW_MAJOR = [
+        1.0,
+        0.0,
+        0.0,
+        0.0,
+        0.0,
+        0.0,
+        1.0,
+        0.0,
+        0.0,
+        -1.0,
+        0.0,
+        0.0,
+        0.0,
+        0.0,
+        0.0,
+        1.0,
+    ]
+
+    ENERGY_TO_INTENSITY_SCALE = 0.001
+
+    def __init__(self, world):
+        self.world = world
+        self.scene_manager = getattr(world, "scene_manager", None)
+
+        self._project_path = ""
+        self._output_root = ""
+        self._assets_model_dir = ""
+        self._assets_texture_dir = ""
+
+        self._copied_source_to_uri: Dict[str, str] = {}
+        self._name_counter: Dict[str, int] = {}
+        self._node_id_by_pointer: Dict[int, str] = {}
+
+        self.report: Dict[str, Any] = {
+            "status": "failed",
+            "warnings": [],
+            "missing_assets": [],
+            "unsupported_assets": [],
+            "converted_assets": [],
+            "copied_assets": [],
+            "output_dir": "",
+        }
+
+    def package(self, project_path: str, output_dir: str) -> Dict[str, Any]:
+        """Export project as WebGL static site and return export report."""
+        self._project_path = os.path.normpath(project_path)
+
+        project_name = os.path.basename(self._project_path.rstrip(os.sep)) or "project"
+        self._output_root = os.path.normpath(os.path.join(output_dir, f"{project_name}_webgl"))
+        self._assets_model_dir = os.path.join(self._output_root, "assets", "models")
+        self._assets_texture_dir = os.path.join(self._output_root, "assets", "textures")
+        self.report["output_dir"] = self._output_root
+
+        try:
+            if not os.path.isdir(self._project_path):
+                self._fail(f"项目路径不存在: {self._project_path}")
+                return self.report
+
+            self._prepare_output_dir()
+            self._copy_templates()
+
+            scene_manifest = self._build_scene_manifest(project_name)
+            self._write_json(
+                os.path.join(self._output_root, "scene", "scene_webgl.json"),
+                scene_manifest,
+            )
+
+            self._write_preview_scripts()
+
+            status = "success"
+            if self.report["missing_assets"] or self.report["unsupported_assets"]:
+                status = "partial"
+            self.report["status"] = status
+
+        except Exception as exc:
+            self._fail(f"WebGL打包失败: {exc}")
+
+        finally:
+            self._write_json(
+                os.path.join(self._output_root, "reports", "export_report.json"),
+                self.report,
+            )
+
+        return self.report
+
+    def _prepare_output_dir(self) -> None:
+        if os.path.isdir(self._output_root):
+            shutil.rmtree(self._output_root)
+
+        os.makedirs(os.path.join(self._output_root, "js"), exist_ok=True)
+        os.makedirs(os.path.join(self._output_root, "vendor"), exist_ok=True)
+        os.makedirs(self._assets_model_dir, exist_ok=True)
+        os.makedirs(self._assets_texture_dir, exist_ok=True)
+        os.makedirs(os.path.join(self._output_root, "scene"), exist_ok=True)
+        os.makedirs(os.path.join(self._output_root, "reports"), exist_ok=True)
+
+    def _copy_templates(self) -> None:
+        repo_root = Path(__file__).resolve().parent.parent
+        template_root = repo_root / "templates" / "webgl"
+        if not template_root.exists():
+            raise FileNotFoundError(f"模板目录不存在: {template_root}")
+
+        file_mapping = {
+            "index.html": "index.html",
+            "style.css": "style.css",
+            "viewer.js": os.path.join("js", "viewer.js"),
+        }
+        for src_name, dst_rel in file_mapping.items():
+            src = template_root / src_name
+            dst = Path(self._output_root) / dst_rel
+            if not src.exists():
+                raise FileNotFoundError(f"模板文件不存在: {src}")
+            dst.parent.mkdir(parents=True, exist_ok=True)
+            shutil.copy2(str(src), str(dst))
+
+        vendor_src = template_root / "vendor"
+        vendor_dst = Path(self._output_root) / "vendor"
+        if vendor_src.exists():
+            for entry in vendor_src.iterdir():
+                if entry.is_file():
+                    shutil.copy2(str(entry), str(vendor_dst / entry.name))
+
+        self._try_resolve_vendor_files(vendor_dst)
+
+        # Placeholder marker warning
+        placeholder_file = vendor_dst / "three.module.min.js"
+        if placeholder_file.exists():
+            content = placeholder_file.read_text(encoding="utf-8", errors="ignore")
+            if "EG_VENDOR_PLACEHOLDER" in content:
+                self.report["warnings"].append(
+                    "当前 vendor 为占位文件，请替换为官方 three.module.min.js / OrbitControls.js / GLTFLoader.js 后再预览。"
+                )
+
+    def _try_resolve_vendor_files(self, vendor_dst: Path) -> None:
+        """Try to replace template placeholders with system-installed Three.js modules."""
+        lookup_roots = [
+            Path("/usr/share/javascript/three"),
+            Path("/usr/share/nodejs/three"),
+            Path("/usr/lib/node_modules/three"),
+            Path.home() / ".local/lib/node_modules/three",
+        ]
+
+        targets = {
+            "three.module.min.js": [
+                "build/three.module.min.js",
+                "build/three.module.js",
+            ],
+            "OrbitControls.js": [
+                "examples/jsm/controls/OrbitControls.js",
+            ],
+            "GLTFLoader.js": [
+                "examples/jsm/loaders/GLTFLoader.js",
+            ],
+        }
+
+        for dst_name, rel_candidates in targets.items():
+            dst_path = vendor_dst / dst_name
+            if not dst_path.exists():
+                continue
+
+            try:
+                content = dst_path.read_text(encoding="utf-8", errors="ignore")
+            except Exception:
+                content = ""
+
+            # Only replace placeholders.
+            if "EG_VENDOR_PLACEHOLDER" not in content:
+                continue
+
+            found_source = None
+            for root in lookup_roots:
+                if not root.exists():
+                    continue
+                for rel in rel_candidates:
+                    candidate = root / rel
+                    if candidate.exists() and candidate.is_file():
+                        found_source = candidate
+                        break
+                if found_source:
+                    break
+
+            if found_source:
+                shutil.copy2(str(found_source), str(dst_path))
+
+    def _build_scene_manifest(self, project_name: str) -> Dict[str, Any]:
+        render = getattr(self.world, "render", None)
+        if not render:
+            raise RuntimeError("world.render 不可用")
+
+        export_nodes: List[Any] = []
+
+        model_nodes = self._collect_model_nodes()
+        spot_nodes = self._collect_valid_nodes(getattr(self.scene_manager, "Spotlight", []))
+        point_nodes = self._collect_valid_nodes(getattr(self.scene_manager, "Pointlight", []))
+        ground_node = self._get_default_ground_node()
+
+        export_nodes.extend(model_nodes)
+        export_nodes.extend(spot_nodes)
+        export_nodes.extend(point_nodes)
+        if ground_node is not None:
+            export_nodes.append(ground_node)
+
+        # Unique by pointer
+        uniq: Dict[int, Any] = {}
+        for node in export_nodes:
+            uniq[id(node)] = node
+        export_nodes = list(uniq.values())
+
+        for index, node in enumerate(export_nodes, start=1):
+            self._node_id_by_pointer[id(node)] = f"node_{index:04d}"
+
+        nodes_json: List[Dict[str, Any]] = []
+
+        for node in model_nodes:
+            entry = self._build_model_node_entry(node)
+            if entry:
+                nodes_json.append(entry)
+
+        for node in point_nodes:
+            entry = self._build_light_node_entry(node, kind="point_light")
+            if entry:
+                nodes_json.append(entry)
+
+        for node in spot_nodes:
+            entry = self._build_light_node_entry(node, kind="spot_light")
+            if entry:
+                nodes_json.append(entry)
+
+        if ground_node is not None:
+            entry = self._build_ground_node_entry(ground_node)
+            if entry:
+                nodes_json.append(entry)
+
+        manifest = {
+            "meta": {
+                "format_version": "1.0",
+                "project_name": project_name,
+                "exported_at": datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
+            },
+            "coordinate": {
+                "source": "panda3d_zup",
+                "target": "threejs_yup",
+                "matrix_convention": "panda_row_vector_row_major",
+                "basis_matrix": self.BASIS_MATRIX_ROW_MAJOR,
+            },
+            "camera": self._build_camera_entry(),
+            "environment": self._build_environment_entry(),
+            "nodes": nodes_json,
+        }
+        return manifest
+
+    def _collect_model_nodes(self) -> List[Any]:
+        if not self.scene_manager:
+            return []
+
+        all_models = self._collect_valid_nodes(getattr(self.scene_manager, "models", []))
+        light_ptrs = {id(n) for n in self._collect_valid_nodes(getattr(self.scene_manager, "Spotlight", []))}
+        light_ptrs |= {id(n) for n in self._collect_valid_nodes(getattr(self.scene_manager, "Pointlight", []))}
+
+        models: List[Any] = []
+        for node in all_models:
+            if id(node) in light_ptrs:
+                continue
+            if node.hasTag("light_type"):
+                continue
+            if node.getName() in {"render", "camera", "cam"}:
+                continue
+            models.append(node)
+        return models
+
+    @staticmethod
+    def _collect_valid_nodes(nodes: List[Any]) -> List[Any]:
+        valid = []
+        for node in nodes or []:
+            if not node:
+                continue
+            try:
+                if node.isEmpty():
+                    continue
+            except Exception:
+                continue
+            valid.append(node)
+        return valid
+
+    def _get_default_ground_node(self):
+        ground = getattr(self.world, "ground", None)
+        if not ground:
+            return None
+        try:
+            if ground.isEmpty():
+                return None
+            return ground
+        except Exception:
+            return None
+
+    def _build_camera_entry(self) -> Dict[str, Any]:
+        render = getattr(self.world, "render", None)
+        cam = getattr(self.world, "cam", None) or getattr(self.world, "camera", None)
+
+        default = {
+            "matrix_local_row_major": [
+                1.0,
+                0.0,
+                0.0,
+                0.0,
+                0.0,
+                1.0,
+                0.0,
+                0.0,
+                0.0,
+                0.0,
+                1.0,
+                0.0,
+                0.0,
+                -50.0,
+                20.0,
+                1.0,
+            ],
+            "fov_deg": 80.0,
+            "near": 0.1,
+            "far": 10000.0,
+        }
+
+        if not cam or not render:
+            return default
+
+        try:
+            cam_mat = cam.getMat(render)
+            fov = 80.0
+            near = 0.1
+            far = 10000.0
+            lens = cam.node().getLens() if cam.node() else None
+            if lens:
+                try:
+                    fov = float(lens.getFov()[0])
+                except Exception:
+                    pass
+                try:
+                    near = float(lens.getNear())
+                except Exception:
+                    pass
+                try:
+                    far = float(lens.getFar())
+                except Exception:
+                    pass
+            return {
+                "matrix_local_row_major": self._mat4_to_row_major_list(cam_mat),
+                "fov_deg": fov,
+                "near": near,
+                "far": far,
+            }
+        except Exception:
+            return default
+
+    def _build_environment_entry(self) -> Dict[str, Any]:
+        ambient = getattr(self.world, "ambient_light", None)
+        directional = getattr(self.world, "directional_light", None)
+
+        ambient_color = [0.2, 0.2, 0.2]
+        directional_color = [0.8, 0.8, 0.8]
+        directional_dir = [0.0, 0.0, -1.0]
+
+        if ambient and not ambient.isEmpty():
+            try:
+                c = ambient.node().getColor()
+                ambient_color = [float(c[0]), float(c[1]), float(c[2])]
+            except Exception:
+                pass
+
+        if directional and not directional.isEmpty():
+            try:
+                c = directional.node().getColor()
+                directional_color = [float(c[0]), float(c[1]), float(c[2])]
+            except Exception:
+                pass
+            try:
+                q = directional.getQuat(getattr(self.world, "render", directional.getParent()))
+                fwd = q.getForward()
+                directional_dir = [float(fwd[0]), float(fwd[1]), float(fwd[2])]
+            except Exception:
+                pass
+
+        return {
+            "include_default_ground": True,
+            "ambient_light": {
+                "color": ambient_color,
+                "intensity": 1.0,
+            },
+            "directional_light": {
+                "color": directional_color,
+                "intensity": 1.0,
+                "direction": directional_dir,
+            },
+        }
+
+    def _build_model_node_entry(self, node) -> Optional[Dict[str, Any]]:
+        node_id = self._node_id_by_pointer.get(id(node))
+        if not node_id:
+            return None
+
+        parent_id, mat = self._get_parent_and_matrix(node)
+        node_name = node.getName() or node_id
+
+        model_source, source_tag = self._resolve_model_source(node)
+        if not model_source:
+            self.report["missing_assets"].append(
+                {
+                    "node": node_name,
+                    "reason": "model_path_not_found",
+                    "tags_checked": ["model_path", "saved_model_path", "original_path", "file"],
+                }
+            )
+            return None
+
+        model_uri = self._prepare_model_asset(model_source, node_name)
+        if not model_uri:
+            self.report["unsupported_assets"].append(
+                {
+                    "node": node_name,
+                    "source": model_source,
+                    "reason": "model_conversion_failed",
+                }
+            )
+            return None
+
+        material_override = self._extract_material_override(node)
+
+        textures = self._collect_and_copy_texture_overrides(node, model_source)
+
+        entry: Dict[str, Any] = {
+            "id": node_id,
+            "name": node_name,
+            "kind": "model",
+            "parent_id": parent_id,
+            "matrix_local_row_major": self._mat4_to_row_major_list(mat),
+            "model": {"uri": model_uri},
+            "material_override": material_override,
+            "source_model_tag": source_tag,
+        }
+        if textures:
+            entry["texture_overrides"] = textures
+        return entry
+
+    def _build_light_node_entry(self, node, kind: str) -> Optional[Dict[str, Any]]:
+        node_id = self._node_id_by_pointer.get(id(node))
+        if not node_id:
+            return None
+
+        parent_id, mat = self._get_parent_and_matrix(node)
+        node_name = node.getName() or node_id
+
+        light_obj = node.getPythonTag("rp_light_object") if node.hasPythonTag("rp_light_object") else None
+
+        color = [1.0, 1.0, 1.0]
+        energy = self._safe_float(node.getTag("light_energy"), 5000.0) if node.hasTag("light_energy") else 5000.0
+        radius = self._safe_float(node.getTag("light_radius"), 30.0) if node.hasTag("light_radius") else 30.0
+        spot_fov = self._safe_float(node.getTag("light_fov"), 45.0) if node.hasTag("light_fov") else 45.0
+        inner_ratio = 0.4
+
+        if light_obj is not None:
+            try:
+                c = getattr(light_obj, "color", None)
+                if c is not None:
+                    color = [float(c[0]), float(c[1]), float(c[2])]
+            except Exception:
+                pass
+            try:
+                energy = float(getattr(light_obj, "energy", energy))
+            except Exception:
+                pass
+            try:
+                radius = float(getattr(light_obj, "radius", radius))
+            except Exception:
+                pass
+            try:
+                if hasattr(light_obj, "fov"):
+                    spot_fov = float(getattr(light_obj, "fov", spot_fov))
+            except Exception:
+                pass
+            try:
+                if hasattr(light_obj, "inner_radius"):
+                    inner_ratio = float(getattr(light_obj, "inner_radius", inner_ratio))
+            except Exception:
+                pass
+
+        intensity = max(0.0, energy * self.ENERGY_TO_INTENSITY_SCALE)
+
+        return {
+            "id": node_id,
+            "name": node_name,
+            "kind": kind,
+            "parent_id": parent_id,
+            "matrix_local_row_major": self._mat4_to_row_major_list(mat),
+            "light": {
+                "color": color,
+                "intensity": intensity,
+                "range": max(0.0, radius),
+                "spot_angle_deg": max(1.0, spot_fov),
+                "inner_cone_ratio": max(0.0, min(1.0, inner_ratio)),
+            },
+        }
+
+    def _build_ground_node_entry(self, node) -> Optional[Dict[str, Any]]:
+        node_id = self._node_id_by_pointer.get(id(node))
+        if not node_id:
+            return None
+
+        parent_id, mat = self._get_parent_and_matrix(node)
+        node_name = node.getName() or "ground"
+
+        color = [0.8, 0.8, 0.8, 1.0]
+        roughness = 1.0
+        metallic = 0.0
+
+        material = None
+        try:
+            material = node.getMaterial()
+        except Exception:
+            material = None
+
+        if material:
+            try:
+                if material.hasBaseColor():
+                    c = material.getBaseColor()
+                    color = [float(c[0]), float(c[1]), float(c[2]), float(c[3])]
+            except Exception:
+                pass
+            try:
+                roughness = float(material.getRoughness())
+            except Exception:
+                pass
+            try:
+                metallic = float(material.getMetallic())
+            except Exception:
+                pass
+
+        return {
+            "id": node_id,
+            "name": node_name,
+            "kind": "ground",
+            "parent_id": parent_id,
+            "matrix_local_row_major": self._mat4_to_row_major_list(mat),
+            "ground": {
+                "width": 100.0,
+                "height": 100.0,
+            },
+            "material_override": {
+                "base_color": [color[0], color[1], color[2], 1.0],
+                "roughness": roughness,
+                "metallic": metallic,
+                "opacity": 1.0,
+            },
+        }
+
+    def _resolve_model_source(self, node) -> Tuple[Optional[str], str]:
+        tags = ["model_path", "saved_model_path", "original_path", "file"]
+        for tag in tags:
+            if not node.hasTag(tag):
+                continue
+            value = (node.getTag(tag) or "").strip()
+            if not value:
+                continue
+            resolved = self._resolve_asset_path(value)
+            if resolved:
+                return resolved, tag
+        return None, ""
+
+    def _resolve_asset_path(self, candidate: str) -> Optional[str]:
+        candidate = (candidate or "").strip()
+        if not candidate:
+            return None
+
+        if candidate.startswith(("http://", "https://")):
+            return None
+
+        # Absolute path: use directly.
+        if os.path.isabs(candidate):
+            abs_candidate = os.path.normpath(candidate)
+            if os.path.exists(abs_candidate):
+                return abs_candidate
+            return None
+
+        # Relative path resolution policy.
+        search_roots = [
+            self._project_path,
+            os.path.join(self._project_path, "models"),
+            os.path.join(self._project_path, "Resources"),
+            os.path.join(self._project_path, "scenes", "resources"),
+            os.getcwd(),
+        ]
+
+        for root in search_roots:
+            full = os.path.normpath(os.path.join(root, candidate))
+            if os.path.exists(full):
+                return full
+
+        return None
+
+    def _prepare_model_asset(self, source_path: str, node_name: str) -> Optional[str]:
+        source_path = os.path.normpath(source_path)
+        ext = os.path.splitext(source_path)[1].lower()
+
+        if ext in {".gltf", ".glb"}:
+            return self._copy_asset_to_models(source_path)
+
+        with tempfile.TemporaryDirectory(prefix="eg_webgl_conv_") as temp_dir:
+            conversion_source = source_path
+
+            if ext == ".bam":
+                temp_egg = os.path.join(temp_dir, self._sanitize_filename(Path(source_path).stem) + ".egg")
+                if not self._run_tool_command(["bam2egg", source_path, temp_egg], timeout=120):
+                    return None
+
+                temp_obj = os.path.join(temp_dir, self._sanitize_filename(Path(source_path).stem) + ".obj")
+                if not self._run_tool_command(["egg2obj", temp_egg, temp_obj], timeout=120):
+                    return None
+                conversion_source = temp_obj
+
+            elif ext == ".egg":
+                temp_obj = os.path.join(temp_dir, self._sanitize_filename(Path(source_path).stem) + ".obj")
+                if not self._run_tool_command(["egg2obj", source_path, temp_obj], timeout=120):
+                    return None
+                conversion_source = temp_obj
+
+            target_filename = self._unique_filename(node_name or Path(source_path).stem, ".glb")
+            target_abs = os.path.join(self._assets_model_dir, target_filename)
+
+            converter = self._convert_to_glb(conversion_source, target_abs)
+            if not converter:
+                return None
+
+            self.report["converted_assets"].append(
+                {
+                    "source": source_path,
+                    "converted_from": conversion_source,
+                    "target": os.path.relpath(target_abs, self._output_root).replace("\\", "/"),
+                    "converter": converter,
+                }
+            )
+            return os.path.relpath(target_abs, self._output_root).replace("\\", "/")
+
+    def _convert_to_glb(self, source_path: str, target_path: str) -> str:
+        scene_manager = self.scene_manager
+        if scene_manager is None:
+            return ""
+
+        conversion_order = [
+            "_convertWithBlender",
+            "_convertWithFBX2glTF",
+            "_convertWithAssimp",
+        ]
+
+        for method_name in conversion_order:
+            method = getattr(scene_manager, method_name, None)
+            if not callable(method):
+                continue
+            try:
+                ok = method(source_path, target_path, None)
+            except TypeError:
+                ok = method(source_path, target_path)
+            except Exception:
+                ok = False
+
+            if ok and os.path.exists(target_path):
+                return method_name
+
+        return ""
+
+    def _run_tool_command(self, args: List[str], timeout: int) -> bool:
+        executable = shutil.which(args[0])
+        if not executable:
+            self.report["warnings"].append(f"缺少转换工具: {args[0]}")
+            return False
+
+        try:
+            result = subprocess.run(
+                args,
+                check=False,
+                capture_output=True,
+                text=True,
+                timeout=timeout,
+            )
+        except Exception as exc:
+            self.report["warnings"].append(f"执行命令失败: {' '.join(args)} ({exc})")
+            return False
+
+        if result.returncode != 0:
+            stderr = (result.stderr or "").strip()
+            stdout = (result.stdout or "").strip()
+            detail = stderr or stdout or f"exit={result.returncode}"
+            self.report["warnings"].append(f"命令失败: {' '.join(args)} -> {detail}")
+            return False
+        return True
+
+    def _copy_asset_to_models(self, source_path: str) -> str:
+        source_path = os.path.normpath(source_path)
+        if source_path in self._copied_source_to_uri:
+            return self._copied_source_to_uri[source_path]
+
+        ext = os.path.splitext(source_path)[1].lower() or ".bin"
+        safe_name = self._unique_filename(Path(source_path).stem, ext)
+        target_abs = os.path.join(self._assets_model_dir, safe_name)
+        shutil.copy2(source_path, target_abs)
+
+        rel_uri = os.path.relpath(target_abs, self._output_root).replace("\\", "/")
+        self._copied_source_to_uri[source_path] = rel_uri
+        self.report["copied_assets"].append(
+            {
+                "source": source_path,
+                "target": rel_uri,
+                "type": "model",
+            }
+        )
+        return rel_uri
+
+    def _collect_and_copy_texture_overrides(self, node, model_source: str) -> List[Dict[str, Any]]:
+        textures: List[Dict[str, Any]] = []
+        texture_pairs = self._extract_texture_stage_and_paths(node)
+        if not texture_pairs:
+            return textures
+
+        model_dir = os.path.dirname(model_source)
+        for stage_name, tex_path in texture_pairs:
+            abs_path = self._resolve_texture_path(tex_path, model_dir)
+            if not abs_path:
+                continue
+            rel_uri = self._copy_asset_to_textures(abs_path)
+            if rel_uri:
+                textures.append({"stage": stage_name, "uri": rel_uri})
+
+        return textures
+
+    def _extract_texture_stage_and_paths(self, node) -> List[Tuple[str, str]]:
+        pairs: List[Tuple[str, str]] = []
+        seen: set = set()
+
+        nodes_to_scan = [node]
+        try:
+            geom_paths = node.findAllMatches("**/+GeomNode")
+            for i in range(geom_paths.getNumPaths()):
+                nodes_to_scan.append(geom_paths.getPath(i))
+        except Exception:
+            pass
+
+        for np in nodes_to_scan:
+            try:
+                stages = np.findAllTextureStages()
+            except Exception:
+                continue
+
+            try:
+                stage_count = stages.getNumTextureStages()
+            except Exception:
+                stage_count = 0
+
+            for idx in range(stage_count):
+                try:
+                    stage = stages.getTextureStage(idx)
+                    texture = np.getTexture(stage)
+                except Exception:
+                    continue
+                if not texture:
+                    continue
+
+                tex_path = ""
+                try:
+                    if texture.hasFullpath():
+                        tex_path = texture.getFullpath().toOsSpecific()
+                except Exception:
+                    tex_path = ""
+
+                if not tex_path:
+                    continue
+
+                stage_name = stage.getName() if stage else f"stage_{idx}"
+                key = (stage_name, tex_path)
+                if key in seen:
+                    continue
+                seen.add(key)
+                pairs.append(key)
+
+        return pairs
+
+    def _resolve_texture_path(self, path_hint: str, model_dir: str) -> Optional[str]:
+        path_hint = (path_hint or "").strip()
+        if not path_hint:
+            return None
+
+        if os.path.isabs(path_hint) and os.path.exists(path_hint):
+            return os.path.normpath(path_hint)
+
+        search_roots = [
+            model_dir,
+            self._project_path,
+            os.path.join(self._project_path, "scenes", "resources"),
+            os.getcwd(),
+        ]
+        for root in search_roots:
+            full = os.path.normpath(os.path.join(root, path_hint))
+            if os.path.exists(full):
+                return full
+
+        return None
+
+    def _copy_asset_to_textures(self, source_path: str) -> Optional[str]:
+        source_path = os.path.normpath(source_path)
+        cache_key = f"texture::{source_path}"
+        if cache_key in self._copied_source_to_uri:
+            return self._copied_source_to_uri[cache_key]
+
+        ext = os.path.splitext(source_path)[1].lower() or ".png"
+        safe_name = self._unique_filename(Path(source_path).stem, ext)
+        target_abs = os.path.join(self._assets_texture_dir, safe_name)
+        try:
+            shutil.copy2(source_path, target_abs)
+        except Exception:
+            return None
+
+        rel_uri = os.path.relpath(target_abs, self._output_root).replace("\\", "/")
+        self._copied_source_to_uri[cache_key] = rel_uri
+        self.report["copied_assets"].append(
+            {
+                "source": source_path,
+                "target": rel_uri,
+                "type": "texture",
+            }
+        )
+        return rel_uri
+
+    def _extract_material_override(self, node) -> Dict[str, Any]:
+        base_color = [1.0, 1.0, 1.0, 1.0]
+        roughness = 0.5
+        metallic = 0.0
+
+        material = None
+        try:
+            material = node.getMaterial()
+        except Exception:
+            material = None
+
+        if material:
+            try:
+                if material.hasBaseColor():
+                    c = material.getBaseColor()
+                    base_color = [float(c[0]), float(c[1]), float(c[2]), float(c[3])]
+            except Exception:
+                pass
+            try:
+                roughness = float(material.getRoughness())
+            except Exception:
+                pass
+            try:
+                metallic = float(material.getMetallic())
+            except Exception:
+                pass
+        else:
+            try:
+                c = node.getColor()
+                base_color = [float(c[0]), float(c[1]), float(c[2]), float(c[3])]
+            except Exception:
+                pass
+
+        opacity = max(0.0, min(1.0, float(base_color[3])))
+        if opacity >= 0.999:
+            opacity = 1.0
+        elif opacity <= 0.001:
+            opacity = 0.0
+
+        return {
+            "base_color": base_color,
+            "roughness": roughness,
+            "metallic": metallic,
+            "opacity": opacity,
+        }
+
+    def _get_parent_and_matrix(self, node) -> Tuple[Optional[str], Any]:
+        render = getattr(self.world, "render", None)
+        parent_id = None
+
+        try:
+            parent = node.getParent()
+        except Exception:
+            parent = None
+
+        if parent and not parent.isEmpty() and id(parent) in self._node_id_by_pointer:
+            parent_id = self._node_id_by_pointer[id(parent)]
+            try:
+                return parent_id, node.getMat(parent)
+            except Exception:
+                pass
+
+        if render:
+            try:
+                return None, node.getMat(render)
+            except Exception:
+                pass
+
+        try:
+            return None, node.getMat()
+        except Exception:
+            return None, node.getTransform().getMat()
+
+    @staticmethod
+    def _mat4_to_row_major_list(mat4_obj) -> List[float]:
+        try:
+            values = []
+            for r in range(4):
+                for c in range(4):
+                    values.append(float(mat4_obj.getCell(r, c)))
+            return values
+        except Exception:
+            return [
+                1.0,
+                0.0,
+                0.0,
+                0.0,
+                0.0,
+                1.0,
+                0.0,
+                0.0,
+                0.0,
+                0.0,
+                1.0,
+                0.0,
+                0.0,
+                0.0,
+                0.0,
+                1.0,
+            ]
+
+    @staticmethod
+    def _safe_float(value: Any, default: float) -> float:
+        try:
+            return float(value)
+        except Exception:
+            return float(default)
+
+    def _unique_filename(self, stem: str, suffix: str) -> str:
+        safe_stem = self._sanitize_filename(stem) or "asset"
+        key = f"{safe_stem}{suffix.lower()}"
+        index = self._name_counter.get(key, 0)
+        self._name_counter[key] = index + 1
+        if index == 0:
+            return f"{safe_stem}{suffix}"
+        return f"{safe_stem}_{index:03d}{suffix}"
+
+    @staticmethod
+    def _sanitize_filename(name: str) -> str:
+        normalized = re.sub(r"[^A-Za-z0-9._-]+", "_", str(name or "")).strip("._")
+        return normalized or "asset"
+
+    def _write_preview_scripts(self) -> None:
+        sh_path = os.path.join(self._output_root, "run_preview.sh")
+        bat_path = os.path.join(self._output_root, "run_preview.bat")
+
+        sh_content = "#!/usr/bin/env bash\npython3 -m http.server 8000\n"
+        bat_content = "@echo off\npython -m http.server 8000\n"
+
+        with open(sh_path, "w", encoding="utf-8") as f:
+            f.write(sh_content)
+        with open(bat_path, "w", encoding="utf-8") as f:
+            f.write(bat_content)
+
+        current_mode = os.stat(sh_path).st_mode
+        os.chmod(sh_path, current_mode | stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH)
+
+    @staticmethod
+    def _write_json(path: str, payload: Dict[str, Any]) -> None:
+        os.makedirs(os.path.dirname(path), exist_ok=True)
+        with open(path, "w", encoding="utf-8") as f:
+            json.dump(payload, f, ensure_ascii=False, indent=2)
+
+    def _fail(self, message: str) -> None:
+        self.report["status"] = "failed"
+        self.report["warnings"].append(message)

requirements/requirements-minimal.txt

@@ -0,0 +1,30 @@
+# 元泰 EG 最小依赖配置 (Python 3.11)
+# 注意: 必须使用 Python 3.11，其他版本不保证兼容性
+
+# Python 版本要求
+# python_version == "3.11"
+
+# 核心 3D 引擎
+Panda3D==1.10.16
+panda3d-frame==22.10.post1
+panda3d-imgui==1.1.0
+
+# ImGui 绑定
+imgui-bundle==1.92.4
+
+# VR 支持
+openvr>=2.5.101
+
+# 图像处理
+Pillow==11.3.0
+
+# YAML 配置解析
+PyYAML==5.4.1
+
+# 其他工具
+psutil==5.9.0
+
+# 注意: 已移除以下未使用的依赖
+# - PyQt5 (未使用，节省 ~50MB)
+# - PySide6 (未使用，节省 ~100MB)
+# - PyQt5-Qt5, PyQt5_sip, PySide6_Addons, PySide6_Essentials

scene/scene_manager_convert_tiles_mixin.py

@@ -0,0 +1,516 @@
+"""Scene manager conversion and tileset operations."""
+
+import os
+import shutil
+import time
+import json
+import aiohttp
+import asyncio
+import inspect
+from pathlib import Path
+
+from panda3d.core import (
+    ModelPool, ModelRoot, Filename, NodePath, GeomNode, Material, Vec4, Vec3,
+    MaterialAttrib, ColorAttrib, Point3, CollisionNode, CollisionSphere, CollisionBox,
+    BitMask32, TransparencyAttrib, LColor, TransformState, RenderModeAttrib
+)
+from panda3d.egg import EggData, EggVertexPool
+from direct.actor.Actor import Actor
+from RenderPipelineFile.rpplugins.smaa.jitters import halton_seq
+from scene import util
+
+class SceneManagerConvertTilesMixin:
+    def _get_tree_widget(self):
+        """安全获取树形控件"""
+        try:
+            if (hasattr(self.world, 'interface_manager') and
+                    hasattr(self.world.interface_manager, 'treeWidget')):
+                return self.world.interface_manager.treeWidget
+        except AttributeError:
+            pass
+        return None
+
+    def _shouldConvertToGLB(self, filepath):
+        """判断是否应该转换为GLB格式"""
+        ext = os.path.splitext(filepath)[1].lower()
+        # 需要转换的格式：FBX, OBJ, DAE等（但不转换已经是GLB/GLTF的）
+        convert_formats = ['.fbx', '.obj', '.dae', '.3ds', '.blend']
+        return ext in convert_formats
+
+    def _convertToGLBWithProgress(self, filepath):
+        """带进度显示的GLB转换"""
+        try:
+            from PyQt5.QtWidgets import QProgressDialog, QApplication
+            from PyQt5.QtCore import Qt
+            
+            # 创建进度对话框
+            progress = QProgressDialog("正在转换模型格式以获得更好的动画支持...", "取消", 0, 100)
+            progress.setWindowTitle("模型格式转换")
+            progress.setWindowModality(Qt.WindowModal)
+            progress.show()
+            QApplication.processEvents()
+            
+            try:
+                result = self._convertToGLB(filepath, progress)
+                progress.hide()
+                return result
+            except Exception as e:
+                progress.hide()
+                print(f"转换过程出错: {e}")
+                return None
+                
+        except ImportError:
+            # 如果没有 PyQt5，直接转换
+            return self._convertToGLB(filepath)
+
+    def _convertToGLB(self, filepath, progress=None):
+        """将模型文件转换为GLB格式"""
+        try:
+            print(f"[GLB转换] 开始转换: {filepath}")
+            
+            if progress:
+                progress.setValue(10)
+                progress.setLabelText("准备转换...")
+                from PyQt5.QtWidgets import QApplication
+                QApplication.processEvents()
+            
+            # 准备输出路径
+            base_name = os.path.splitext(os.path.basename(filepath))[0]
+            output_dir = os.path.dirname(filepath)
+            glb_path = os.path.join(output_dir, f"{base_name}_auto_converted.glb")
+            
+            # 如果已经存在转换后的文件，直接使用
+            if os.path.exists(glb_path):
+                # 检查文件时间，如果原文件更新则重新转换
+                original_time = os.path.getmtime(filepath)
+                converted_time = os.path.getmtime(glb_path)
+                if converted_time > original_time:
+                    print(f"[GLB转换] 使用现有转换文件: {glb_path}")
+                    if progress:
+                        progress.setValue(100)
+                    return glb_path
+            
+            if progress:
+                progress.setValue(20)
+                progress.setLabelText("尝试 Blender 转换...")
+                QApplication.processEvents()
+            
+            # 方法1: 使用 Blender 进行转换
+            if self._convertWithBlender(filepath, glb_path, progress):
+                return glb_path
+            
+            if progress:
+                progress.setValue(60)
+                progress.setLabelText("尝试 FBX2glTF 转换...")
+                QApplication.processEvents()
+            
+            # 方法2: 使用 FBX2glTF (如果可用)
+            if self._convertWithFBX2glTF(filepath, glb_path, progress):
+                return glb_path
+            
+            if progress:
+                progress.setValue(80)
+                progress.setLabelText("尝试 Assimp 转换...")
+                QApplication.processEvents()
+            
+            # 方法3: 使用 Assimp
+            if self._convertWithAssimp(filepath, glb_path, progress):
+                return glb_path
+            
+            #print(f"[GLB转换] 所有转换方法都失败，既然没有可以转换格式的工具和环境那么就用原始文件，不一定非要转换")
+            return None
+            
+        except Exception as e:
+            print(f"[GLB转换] 转换过程出错: {e}")
+            return None
+
+    def _convertWithBlender(self, input_path, output_path, progress=None):
+        """使用 Blender 进行转换"""
+        try:
+            import subprocess
+            import tempfile
+            
+            print(f"[Blender转换] {input_path} → {output_path}")
+            
+            # 创建 Blender 脚本
+            script_content = f'''
+import bpy
+import sys
+import os
+
+# 清理默认场景
+bpy.ops.object.select_all(action='SELECT')
+bpy.ops.object.delete(use_global=False)
+
+print("开始导入文件...")
+
+# 根据文件类型选择导入方法
+input_file = "{input_path}"
+output_file = "{output_path}"
+
+try:
+    ext = os.path.splitext(input_file)[1].lower()
+    
+    if ext == '.fbx':
+        bpy.ops.import_scene.fbx(filepath=input_file)
+    elif ext == '.obj':
+        bpy.ops.import_scene.obj(filepath=input_file)
+    elif ext == '.dae':
+        bpy.ops.wm.collada_import(filepath=input_file)
+    elif ext == '.blend':
+        bpy.ops.wm.open_mainfile(filepath=input_file)
+    else:
+        print(f"不支持的格式: {{ext}}")
+        sys.exit(1)
+    
+    print("导入成功，开始导出GLB...")
+    
+    # 导出为 GLB，保留动画
+    bpy.ops.export_scene.gltf(
+        filepath=output_file,
+        export_format='GLB',
+        export_animations=True,
+        export_force_sampling=True,
+        export_frame_range=True,
+        export_current_frame=False,
+        export_skins=True,
+        export_morph=True,
+        export_lights=True,
+        export_cameras=False
+    )
+    
+    print("GLB导出成功！")
+    
+except Exception as e:
+    print(f"转换失败: {{e}}")
+    sys.exit(1)
+'''
+            
+            # 写入临时脚本文件
+            with tempfile.NamedTemporaryFile(mode='w', suffix='.py', delete=False) as temp_file:
+                temp_file.write(script_content)
+                script_path = temp_file.name
+            
+            try:
+                # 执行 Blender 转换
+                result = subprocess.run([
+                    'blender', '--background', '--python', script_path
+                ], capture_output=True, text=True, timeout=180)
+                
+                # 清理临时文件
+                os.unlink(script_path)
+                
+                if result.returncode == 0 and os.path.exists(output_path):
+                    print(f"[Blender转换] 转换成功")
+                    return True
+                else:
+                    print(f"[Blender转换] 转换失败: {result.stderr}")
+                    return False
+                    
+            except subprocess.TimeoutExpired:
+                print(f"[Blender转换] 转换超时")
+                return False
+            except FileNotFoundError:
+                print(f"[Blender转换] Blender 未安装")
+                return False
+                
+        except Exception as e:
+            print(f"[Blender转换] 转换过程出错: {e}")
+            return False
+
+    def _convertWithFBX2glTF(self, input_path, output_path, progress=None):
+        """使用 FBX2glTF 进行转换（仅支持FBX）"""
+        try:
+            import subprocess
+            
+            if not input_path.lower().endswith('.fbx'):
+                return False
+                
+            print(f"[FBX2glTF转换] {input_path} → {output_path}")
+            
+            # 使用 FBX2glTF 转换
+            result = subprocess.run([
+                'FBX2glTF', input_path, '--output', output_path, '--binary'
+            ], capture_output=True, text=True, timeout=120)
+            
+            if result.returncode == 0 and os.path.exists(output_path):
+                print(f"[FBX2glTF转换] 转换成功")
+                return True
+            else:
+                print(f"[FBX2glTF转换] 转换失败: {result.stderr}")
+                return False
+                
+        except subprocess.TimeoutExpired:
+            print(f"[FBX2glTF转换] 转换超时")
+            return False
+        except FileNotFoundError:
+            print(f"[FBX2glTF转换] FBX2glTF 未安装")
+            return False
+        except Exception as e:
+            print(f"[FBX2glTF转换] 转换过程出错: {e}")
+            return False
+
+    def _convertWithAssimp(self, input_path, output_path, progress=None):
+        """使用 PyAssimp 进行转换"""
+        try:
+            import pyassimp
+            
+            print(f"[PyAssimp转换] {input_path} → {output_path}")
+            
+            # 加载模型
+            scene = pyassimp.load(input_path)
+            if not scene:
+                print(f"[PyAssimp转换] 加载模型失败")
+                return False
+            
+            if progress:
+                progress.setValue(30)
+            
+            # 导出为GLB格式
+            pyassimp.export(scene, output_path, "glb2")
+            
+            if progress:
+                progress.setValue(80)
+            
+            # 释放资源
+            pyassimp.release(scene)
+            
+            if os.path.exists(output_path):
+                print(f"[PyAssimp转换] 转换成功")
+                return True
+            else:
+                print(f"[PyAssimp转换] 转换失败: 输出文件未生成")
+                return False
+                
+        except ImportError:
+            print(f"[PyAssimp转换] PyAssimp 未安装")
+            return False
+        except Exception as e:
+            print(f"[PyAssimp转换] 转换过程出错: {e}")
+            return False
+
+    def _check_async_task(self, panda3d_task):
+        # 检查 asyncio 任务是否完成
+        if hasattr(self, '_current_asyncio_task'):
+            if self._current_asyncio_task.done():
+                try:
+                    self._current_asyncio_task.result()
+                except Exception as e:
+                    print(f"异步任务出错：{e}")
+                # 返回 Panda3D 任务管理器的完成状态
+                return panda3d_task.done  # 注意是 done 而不是 DONE
+        # 返回 Panda3D 任务管理器的继续状态
+        return panda3d_task.cont  # 注意是 cont 而不是 CONTINUE
+
+    def _parse_tileset(self,tileset_data,tileset_info):
+        try:
+            root = tileset_data.get('root',{})
+            self._parse_tile(root,tileset_info['node'],tileset_info)
+            print("✓ Tileset 解析完成")
+        except Exception as e:
+            print(f"✗ Tileset 解析出错: {e}")
+
+    def _parse_tile(self, tile_data, parent_node, tileset_info):
+        try:
+            # 获取tileID
+            tile_id = f"tile_{len(tileset_info['tiles'])}"
+            print(f"创建tile节点: {tile_id}")
+            # 创建tile节点
+            tile_node = parent_node.attachNewNode(tile_id)
+
+            tileset_info['tiles'][tile_id] = {
+                'node': tile_node,
+                'data': tile_data,
+                'loaded': False
+            }
+
+            # 如果有内容，创建占位几何体
+            if 'content' in tile_data:
+                print(f"为tile {tile_id} 创建几何体")
+                self._create_tile_geometry(tile_node)
+            # 递归解析子tiles
+            children = tile_data.get('children', [])
+            print(f"Tile {tile_id} 有 {len(children)} 个子节点")
+            for child_data in children:
+                self._parse_tile(child_data, tile_node, tileset_info)
+        except Exception as e:
+            print(f"✗ Tile 解析出错: {e}")
+            import traceback
+            traceback.print_exc()
+
+    def _create_tile_geometry(self,parent_node):
+        """为 tile 创建占位几何体"""
+        try:
+            # 创建一个简单的立方体作为占位符
+            from panda3d.core import GeomVertexFormat, GeomVertexData, GeomVertexWriter
+            from panda3d.core import Geom, GeomTriangles, GeomNode
+
+            format = GeomVertexFormat.getV3n3c4()
+            vdata = GeomVertexData('tile_cube', format, Geom.UHStatic)
+
+            vertex = GeomVertexWriter(vdata, 'vertex')
+            normal = GeomVertexWriter(vdata, 'normal')
+            color = GeomVertexWriter(vdata, 'color')
+
+            # 定义立方体顶点
+            vertices = [
+                (-0.5, -0.5, -0.5), (0.5, -0.5, -0.5), (0.5, 0.5, -0.5), (-0.5, 0.5, -0.5),
+                (-0.5, -0.5, 0.5), (0.5, -0.5, 0.5), (0.5, 0.5, 0.5), (-0.5, 0.5, 0.5)
+            ]
+
+            for vert in vertices:
+                vertex.addData3f(*vert)
+                normal.addData3f(0, 0, 1)
+                color.addData4f(0.2, 0.6, 0.8, 1.0)
+
+            # 创建几何体
+            geom = Geom(vdata)
+
+            # 创建面
+            prim = GeomTriangles(Geom.UHStatic)
+            # 底面
+            prim.addVertices(0, 1, 2)
+            prim.addVertices(0, 2, 3)
+            # 顶面
+            prim.addVertices(4, 7, 6)
+            prim.addVertices(4, 6, 5)
+            # 前面
+            prim.addVertices(0, 4, 5)
+            prim.addVertices(0, 5, 1)
+            # 后面
+            prim.addVertices(2, 6, 7)
+            prim.addVertices(2, 7, 3)
+            # 左面
+            prim.addVertices(0, 3, 7)
+            prim.addVertices(0, 7, 4)
+            # 右面
+            prim.addVertices(1, 5, 6)
+            prim.addVertices(1, 6, 2)
+
+            prim.closePrimitive()
+            geom.addPrimitive(prim)
+
+            # 创建几何节点
+            geom_node = GeomNode('tile_geometry')
+            geom_node.addGeom(geom)
+
+            # 添加到场景
+            cube_node = parent_node.attachNewNode(geom_node)
+            cube_node.setScale(1000)  # 放大以便观察
+
+            # 添加材质
+            material = Material()
+            material.setBaseColor((0.2, 0.6, 0.8, 1.0))
+            material.setSpecular((0.1, 0.1, 0.1, 1.0))
+            material.setShininess(10.0)
+            cube_node.setMaterial(material)
+
+        except Exception as e:
+            print(f"✗ 创建 tile 几何体出错: {e}")
+
+    def _create_placeholder_geometry(self, parent_node):
+        """创建一个简单的占位符几何体，让用户能看到节点"""
+        try:
+            from panda3d.core import GeomVertexFormat, GeomVertexData, GeomVertexWriter
+            from panda3d.core import Geom, GeomTriangles, GeomNode
+
+            # 创建简单的立方体作为占位符
+            format = GeomVertexFormat.getV3n3c4()
+            vdata = GeomVertexData('placeholder_cube', format, Geom.UHStatic)
+
+            vertex = GeomVertexWriter(vdata, 'vertex')
+            normal = GeomVertexWriter(vdata, 'normal')
+            color = GeomVertexWriter(vdata, 'color')
+
+            # 定义立方体顶点
+            size = 1.0
+            vertices = [
+                # 前面 (Z+)
+                (-size, -size, size), (size, -size, size), (size, size, size), (-size, size, size),
+                # 后面 (Z-)
+                (-size, -size, -size), (-size, size, -size), (size, size, -size), (size, -size, -size),
+                # 左面 (X-)
+                (-size, -size, -size), (-size, -size, size), (-size, size, size), (-size, size, -size),
+                # 右面 (X+)
+                (size, -size, -size), (size, size, -size), (size, size, size), (size, -size, size),
+                # 上面 (Y+)
+                (-size, size, -size), (-size, size, size), (size, size, size), (size, size, -size),
+                # 下面 (Y-)
+                (-size, -size, -size), (size, -size, -size), (size, -size, size), (-size, -size, size)
+            ]
+
+            normals = [
+                # 前面法线
+                (0, 0, 1), (0, 0, 1), (0, 0, 1), (0, 0, 1),
+                # 后面法线
+                (0, 0, -1), (0, 0, -1), (0, 0, -1), (0, 0, -1),
+                # 左面法线
+                (-1, 0, 0), (-1, 0, 0), (-1, 0, 0), (-1, 0, 0),
+                # 右面法线
+                (1, 0, 0), (1, 0, 0), (1, 0, 0), (1, 0, 0),
+                # 上面法线
+                (0, 1, 0), (0, 1, 0), (0, 1, 0), (0, 1, 0),
+                # 下面法线
+                (0, -1, 0), (0, -1, 0), (0, -1, 0), (0, -1, 0)
+            ]
+
+            # 青色
+            face_colors = [
+                (0.0, 1.0, 1.0, 1.0), (0.0, 1.0, 1.0, 1.0), (0.0, 1.0, 1.0, 1.0), (0.0, 1.0, 1.0, 1.0),  # 前面 - 青色
+                (0.0, 0.8, 0.8, 1.0), (0.0, 0.8, 0.8, 1.0), (0.0, 0.8, 0.8, 1.0), (0.0, 0.8, 0.8, 1.0),  # 后面 - 稍暗青色
+                (0.0, 0.9, 0.9, 1.0), (0.0, 0.9, 0.9, 1.0), (0.0, 0.9, 0.9, 1.0), (0.0, 0.9, 0.9, 1.0),  # 左面 - 中等青色
+                (0.0, 0.7, 0.7, 1.0), (0.0, 0.7, 0.7, 1.0), (0.0, 0.7, 0.7, 1.0), (0.0, 0.7, 0.7, 1.0),  # 右面 - 稍暗青色
+                (0.0, 1.0, 1.0, 1.0), (0.0, 1.0, 1.0, 1.0), (0.0, 1.0, 1.0, 1.0), (0.0, 1.0, 1.0, 1.0),  # 上面 - 青色
+                (0.0, 0.6, 0.6, 1.0), (0.0, 0.6, 0.6, 1.0), (0.0, 0.6, 0.6, 1.0), (0.0, 0.6, 0.6, 1.0)  # 下面 - 更暗青色
+            ]
+
+            for i, vert in enumerate(vertices):
+                vertex.addData3f(*vert)
+                normal.addData3f(*normals[i])
+                color.addData4f(*face_colors[i])
+
+            # 创建几何体
+            geom = Geom(vdata)
+
+            # 创建面（每个面两个三角形）
+            prim = GeomTriangles(Geom.UHStatic)
+
+            # 每个面4个顶点，生成2个三角形
+            for face in range(6):  # 6个面
+                base_index = face * 4
+                # 第一个三角形
+                prim.addVertices(base_index, base_index + 1, base_index + 2)
+                # 第二个三角形
+                prim.addVertices(base_index, base_index + 2, base_index + 3)
+
+            prim.closePrimitive()
+            geom.addPrimitive(prim)
+
+            # 创建几何节点
+            geom_node = GeomNode('tileset_placeholder')
+            geom_node.addGeom(geom)
+
+            # 添加到场景
+            cube_node = parent_node.attachNewNode(geom_node)
+            cube_node.setScale(5)  # 设置合适的大小
+
+            # 设置双面渲染
+            cube_node.setTwoSided(True)
+
+            # 添加材质
+            material = Material()
+            material.setBaseColor((0.0, 1.0, 1.0, 1.0))  # 青色
+            material.setSpecular((0.5, 0.5, 0.5, 1.0))
+            material.setShininess(32.0)
+            cube_node.setMaterial(material)
+
+            # 添加标识标签
+            cube_node.setTag("element_type", "cesium_placeholder")
+
+            print("✓ 占位符几何体创建完成")
+            return cube_node
+        except Exception as e:
+            print(f"✗ 创建占位符几何体出错: {e}")
+            import traceback
+            traceback.print_exc()
+            return None

scene/scene_manager_impl.py

@@ -0,0 +1,19 @@
+"""Scene manager implementation composed from focused mixins."""
+
+from scene.scene_manager_convert_tiles_mixin import SceneManagerConvertTilesMixin
+from scene.scene_manager_io_mixin import SceneManagerIOMixin
+from scene.scene_manager_light_mixin import SceneManagerLightMixin
+from scene.scene_manager_model_mixin import SceneManagerModelMixin
+from scene.scene_manager_serialization_mixin import SceneManagerSerializationMixin
+
+
+class SceneManagerImpl(
+    SceneManagerSerializationMixin,
+    SceneManagerConvertTilesMixin,
+    SceneManagerLightMixin,
+    SceneManagerIOMixin,
+    SceneManagerModelMixin,
+):
+    """Composed scene manager implementation."""
+
+    pass

scene/scene_manager_light_mixin.py

@@ -0,0 +1,455 @@
+"""Scene manager light operations."""
+
+import os
+import shutil
+import time
+import json
+import aiohttp
+import asyncio
+import inspect
+from pathlib import Path
+
+from panda3d.core import (
+    ModelPool, ModelRoot, Filename, NodePath, GeomNode, Material, Vec4, Vec3,
+    MaterialAttrib, ColorAttrib, Point3, CollisionNode, CollisionSphere, CollisionBox,
+    BitMask32, TransparencyAttrib, LColor, TransformState, RenderModeAttrib
+)
+from panda3d.egg import EggData, EggVertexPool
+from direct.actor.Actor import Actor
+from RenderPipelineFile.rpplugins.smaa.jitters import halton_seq
+from scene import util
+
+class SceneManagerLightMixin:
+    def _recreateSpotLight(self, light_node):
+        """重新创建聚光灯"""
+        try:
+            from RenderPipelineFile.rpcore import SpotLight
+            from panda3d.core import Vec3
+            from core.render_pipeline_utils import get_render_pipeline
+
+            # 创建聚光灯对象
+            light = SpotLight()
+            light.direction = Vec3(0, 0, -1)
+            light.fov = 70
+            light.set_color_from_temperature(5 * 1000.0)
+
+            # 恢复保存的属性
+            if light_node.hasTag("light_energy"):
+                light.energy = float(light_node.getTag("light_energy"))
+            else:
+                light.energy = 5000
+
+            light.radius = 1000
+            light.casts_shadows = True
+            light.shadow_map_resolution = 256
+
+            light_pos = light_node.getPos()
+            light.setPos(light_pos)
+
+            # 添加到渲染管线
+            render_pipeline = get_render_pipeline()
+            render_pipeline.add_light(light)
+
+            # 保存光源对象引用
+            light_node.setPythonTag("rp_light_object", light)
+
+            # 添加到管理列表（去重）
+            if light_node not in self.Spotlight:
+                self.Spotlight.append(light_node)
+
+            # 确保灯光节点有正确的标签，以便在场景树更新时被识别
+            if not light_node.hasTag("is_scene_element"):
+                light_node.setTag("is_scene_element", "1")
+            light_node.setTag("is_scene_element", "1")
+            light_node.setTag("element_type", "spotlight")
+            light_node.setTag("tree_item_type", "LIGHT_NODE")
+
+            if light_node.hasTag("stored_energy"):
+                stored_energy = float(light_node.getTag("stored_energy"))
+                if stored_energy > 0:
+                    light_node.setTag("stored_energy", str(stored_energy))
+
+            user_visible = True
+            if light_node.hasTag("user_visible"):
+                user_visible = light_node.getTag("user_visible").lower() == "true"
+
+            light_node.setPythonTag("user_visible",user_visible)
+            if not user_visible:
+                self.toggleLightVisibility(light_node,False)
+        except Exception as e:
+            print(f"重新创建聚光灯失败: {str(e)}")
+            import traceback
+            traceback.print_exc()
+
+    def _recreatePointLight(self, light_node):
+        """重新创建点光源"""
+        try:
+            from RenderPipelineFile.rpcore import PointLight
+            from core.render_pipeline_utils import get_render_pipeline
+
+            # 创建点光源对象
+            light = PointLight()
+
+            # 恢复保存的属性
+            if light_node.hasTag("light_energy"):
+                light.energy = float(light_node.getTag("light_energy"))
+            else:
+                light.energy = 5000
+
+            light.radius = 1000
+            light.inner_radius = 0.4
+            light.set_color_from_temperature(5 * 1000.0)
+            light.casts_shadows = True
+            light.shadow_map_resolution = 256
+
+            # 设置位置
+            light.setPos(light_node.getPos())
+
+            # 添加到渲染管线
+            render_pipeline = get_render_pipeline()
+            render_pipeline.add_light(light)
+
+            # 保存光源对象引用
+            light_node.setPythonTag("rp_light_object", light)
+
+            # 添加到管理列表（去重）
+            if light_node not in self.Pointlight:
+                self.Pointlight.append(light_node)
+
+            # 确保灯光节点有正确的标签，以便在场景树更新时被识别
+            if not light_node.hasTag("is_scene_element"):
+                light_node.setTag("is_scene_element", "1")
+
+            light_node.setTag("is_scene_element", "1")
+            light_node.setTag("element_type", "pointlight")
+            light_node.setTag("tree_item_type", "LIGHT_NODE")
+
+            if light_node.hasTag("stored_energy"):
+                stored_energy = float(light_node.getTag("stored_energy"))
+                if stored_energy > 0:
+                    light_node.setTag("stored_energy", str(stored_energy))
+
+            user_visible = True
+            if light_node.hasTag("user_visible"):
+                user_visible = light_node.getTag("user_visible").lower()=="true"
+
+            light_node.setPythonTag("user_visible",user_visible)
+
+            if not user_visible:
+                self.toggleLightVisibility(light_node,False)
+        except Exception as e:
+            print(f"重新创建点光源失败: {str(e)}")
+            import traceback
+            traceback.print_exc()
+
+    def isLightObject(self, nodePath):
+        """检查是否为灯光对象"""
+        try:
+            if not nodePath:
+                return False
+
+
+            # 方法1: 检查PythonTag
+            if nodePath.hasPythonTag("rp_light_object"):
+                rp_light = nodePath.getPythonTag("rp_light_object")
+                if rp_light is not None:
+                    return True
+
+            # 方法2: 检查element_type标签
+            if nodePath.hasTag("element_type"):
+                element_type = nodePath.getTag("element_type")
+                if element_type in ["spotlight", "pointlight"]:
+                    return True
+
+            # 方法3: 检查tree_item_type标签
+            if nodePath.hasTag("tree_item_type"):
+                tree_item_type = nodePath.getTag("tree_item_type")
+                if tree_item_type == "LIGHT_NODE":
+                    return True
+
+            # 方法4: 通过名称模式匹配（作为后备方案）
+            node_name = nodePath.getName().lower()
+            if "spotlight" in node_name or "pointlight" in node_name:
+                return True
+
+            return False
+        except Exception as e:
+            print(f"检查灯光对象时出错: {e}")
+            import traceback
+            traceback.print_exc()
+            return False
+
+    def toggleLightVisibility(self, light_node, visible):
+        """切换灯光可见性"""
+        try:
+            print(f"切换灯光可见性: {light_node.getName()}, 可见={visible}")
+
+            # 保存用户可见性状态到该特定节点
+            light_node.setPythonTag("user_visible", visible)
+
+            # 获取该特定灯光对象
+            rp_light_object = light_node.getPythonTag("rp_light_object")
+            if not rp_light_object:
+                print(f"错误: {light_node.getName()} 未找到RP灯光对象引用")
+                return
+
+            # 获取RenderPipeline实例
+            from core.render_pipeline_utils import get_render_pipeline
+            render_pipeline = get_render_pipeline()
+
+            if not render_pipeline:
+                print("错误: 无法获取RenderPipeline实例")
+                return
+
+            try:
+                if visible:
+                    if light_node.hasTag("stored_energy"):
+                        stored_energy = float(light_node.getTag("stored_energy"))
+                        rp_light_object.energy=stored_energy
+                        print(f"已恢复灯光强度: {light_node.getName()}, 能量={stored_energy}")
+                    # 启用特定灯光
+                    # render_pipeline.add_light(rp_light_object)
+                    # print(f"已添加灯光到渲染管线: {light_node.getName()}")
+                else:
+                    # 禁用特定灯光
+                    current_energy = rp_light_object.energy
+                    if current_energy != 0.0:
+                        light_node.setTag("stored_energy", str(current_energy))
+                    elif light_node.hasTag("stored_energy"):
+                        stored_energy = float(light_node.getTag("stored_energy"))
+                        current_energy = stored_energy
+                    else:
+                        current_energy = 0.0
+                    rp_light_object.energy = 0.0
+                    print(f"已禁用灯光: {light_node.getName()}, 保存的能量={current_energy}")
+                    # render_pipeline.remove_light(rp_light_object)
+                    # print(f"已从渲染管线移除灯光: {light_node.getName()}")
+            except Exception as e:
+                print(f"操作RenderPipeline灯光时出错: {e}")
+
+            # 控制节点显示状态（可选，主要是视觉上的）
+            if visible:
+                light_node.show()
+            else:
+                light_node.hide()
+
+            print(f"灯光可见性设置完成: {visible}")
+        except Exception as e:
+            print(f"切换灯光可见性失败: {str(e)}")
+            import traceback
+            traceback.print_exc()
+
+    def _recreateLightFromData(self, node_data, parent_node, name):
+        """根据数据重建光源"""
+        try:
+            light_type = node_data.get('tags', {}).get('light_type', 'spot_light')
+
+            # 创建光源
+            if light_type == 'spot_light':
+                light_node = self.createSpotLight(pos=node_data.get('pos', (0, 0, 0)))
+            else:  # point_light
+                light_node = self.createPointLight(pos=node_data.get('pos', (0, 0, 0)))
+
+            if light_node:
+                # 设置名称
+                light_node.setName(name)
+
+                # 恢复其他属性
+                light_data = node_data.get('light_data', {})
+                rp_light = light_node.getPythonTag("rp_light_object")
+                if rp_light and light_data:
+                    if 'energy' in light_data:
+                        rp_light.energy = light_data['energy']
+                    if 'radius' in light_data:
+                        rp_light.radius = light_data['radius']
+                    if 'fov' in light_data and hasattr(rp_light, 'fov'):
+                        rp_light.fov = light_data['fov']
+                    if 'inner_radius' in light_data and hasattr(rp_light, 'inner_radius'):
+                        rp_light.inner_radius = light_data['inner_radius']
+                    if 'casts_shadows' in light_data and hasattr(rp_light, 'casts_shadows'):
+                        rp_light.casts_shadows = light_data['casts_shadows']
+                    if 'shadow_map_resolution' in light_data and hasattr(rp_light, 'shadow_map_resolution'):
+                        rp_light.shadow_map_resolution = light_data['shadow_map_resolution']
+
+                # 恢复其他标签
+                for tag_key, tag_value in node_data.get('tags', {}).items():
+                    if tag_key not in ['name', 'light_type']:
+                        light_node.setTag(tag_key, str(tag_value))
+
+            return light_node
+
+        except Exception as e:
+            print(f"重建光源失败: {e}")
+            return None
+
+    def createSpotLight(self, pos=(0, 0, 5)):
+        """创建聚光灯
+        
+        Args:
+            pos (tuple): 光源位置 (x, y, z)
+            
+        Returns:
+            NodePath: 创建的聚光灯节点
+        """
+        try:
+            # 检查是否使用RenderPipeline
+            if hasattr(self.world, 'render_pipeline') and self.world.render_pipeline:
+                from RenderPipelineFile.rpcore import SpotLight
+                from core.render_pipeline_utils import get_render_pipeline
+                
+                # 创建RenderPipeline聚光灯
+                from panda3d.core import Vec3
+                spotlight = SpotLight()
+                spotlight.direction = Vec3(0, 0, -1)  # 向下照射
+                spotlight.fov = 70  # 聚光角度
+                spotlight.set_color_from_temperature(6500)  # 日光色温
+                spotlight.energy = 5000  # 光照强度
+                spotlight.radius = 1000  # 光照范围
+                spotlight.casts_shadows = True  # 启用阴影
+                spotlight.shadow_map_resolution = 256  # 阴影分辨率
+                spotlight.setPos(pos)
+                
+                # 添加到RenderPipeline
+                render_pipeline = get_render_pipeline()
+                if render_pipeline:
+                    render_pipeline.add_light(spotlight)
+                    print(f"✓ RenderPipeline聚光灯创建成功，位置: {pos}")
+                    
+                    # 创建包装节点用于场景树显示
+                    light_name = f"Spotlight_{len(self.Spotlight)}"
+                    spotlight_node = self.world.render.attachNewNode(light_name)
+                    spotlight_node.setPos(*pos)
+                    spotlight_node.setTag("light_type", "spot_light")
+                    spotlight_node.setTag("is_scene_element", "1")
+                    spotlight_node.setTag("tree_item_type", "LIGHT_NODE")
+                    spotlight_node.setTag("light_energy", str(getattr(spotlight, "energy", 5000)))
+                    spotlight_node.setTag("created_by_user", "1")
+                    spotlight_node.setTag("element_type", "spotlight")
+                    spotlight_node.setPythonTag("rp_light_object", spotlight)
+                    self.Spotlight.append(spotlight_node)
+                    return spotlight_node
+                else:
+                    print("✗ 无法获取RenderPipeline实例")
+                    return None
+            else:
+                # 使用标准Panda3D光源
+                from panda3d.core import Spotlight, PerspectiveLens
+                
+                # 创建聚光灯
+                spotlight = Spotlight('spotlight')
+                spotlight.setColor((1, 1, 1, 1))  # 白色光
+                spotlight.setLens(PerspectiveLens())
+                
+                # 创建光源节点
+                spotlight_node = self.world.render.attachNewNode(spotlight)
+                spotlight_node.setPos(pos)
+                spotlight_node.setTag("light_type", "spot_light")
+                spotlight_node.setTag("is_scene_element", "1")
+                spotlight_node.setTag("tree_item_type", "LIGHT_NODE")
+                spotlight_node.setTag("created_by_user", "1")
+                spotlight_node.setTag("element_type", "spotlight")
+                
+                # 设置聚光灯方向（向下照射）
+                spotlight_node.lookAt(pos[0], pos[1], pos[2] - 5)  # 向下看5个单位
+                
+                # 设置聚光灯范围
+                spotlight.setExponent(1.0)  # 聚光指数
+                spotlight.setAttenuation((1, 0.1, 0.01))  # 衰减
+                
+                # 添加到光源列表
+                self.Spotlight.append(spotlight_node)
+                
+                # 启用光源
+                self.world.render.setLight(spotlight_node)
+                
+                # 启用阴影
+                if hasattr(spotlight, 'setShadowCaster'):
+                    spotlight.setShadowCaster(True)
+                
+                print(f"✓ 标准聚光灯创建成功，位置: {pos}")
+                return spotlight_node
+            
+        except Exception as e:
+            print(f"✗ 创建聚光灯失败: {e}")
+            return None
+
+    def createPointLight(self, pos=(0, 0, 5)):
+        """创建点光源
+        
+        Args:
+            pos (tuple): 光源位置 (x, y, z)
+            
+        Returns:
+            NodePath: 创建的点光源节点
+        """
+        try:
+            # 检查是否使用RenderPipeline
+            if hasattr(self.world, 'render_pipeline') and self.world.render_pipeline:
+                from RenderPipelineFile.rpcore import PointLight
+                from core.render_pipeline_utils import get_render_pipeline
+                
+                # 创建RenderPipeline点光源
+                pointlight = PointLight()
+                pointlight.set_color_from_temperature(6500)  # 日光色温
+                pointlight.energy = 3000  # 光照强度
+                pointlight.radius = 1000  # 光照范围
+                pointlight.casts_shadows = True  # 启用阴影
+                pointlight.shadow_map_resolution = 256  # 阴影分辨率
+                pointlight.setPos(pos)
+                
+                # 添加到RenderPipeline
+                render_pipeline = get_render_pipeline()
+                if render_pipeline:
+                    render_pipeline.add_light(pointlight)
+                    print(f"✓ RenderPipeline点光源创建成功，位置: {pos}")
+                    
+                    # 创建包装节点用于场景树显示
+                    light_name = f"Pointlight_{len(self.Pointlight)}"
+                    pointlight_node = self.world.render.attachNewNode(light_name)
+                    pointlight_node.setPos(*pos)
+                    pointlight_node.setTag("light_type", "point_light")
+                    pointlight_node.setTag("is_scene_element", "1")
+                    pointlight_node.setTag("tree_item_type", "LIGHT_NODE")
+                    pointlight_node.setTag("light_energy", str(getattr(pointlight, "energy", 3000)))
+                    pointlight_node.setTag("created_by_user", "1")
+                    pointlight_node.setTag("element_type", "pointlight")
+                    pointlight_node.setPythonTag("rp_light_object", pointlight)
+                    self.Pointlight.append(pointlight_node)
+                    return pointlight_node
+                else:
+                    print("✗ 无法获取RenderPipeline实例")
+                    return None
+            else:
+                # 使用标准Panda3D光源
+                from panda3d.core import PointLight
+                
+                # 创建点光源
+                pointlight = PointLight('pointlight')
+                pointlight.setColor((1, 1, 1, 1))  # 白色光
+                pointlight.setAttenuation((1, 0.1, 0.01))  # 衰减设置
+                
+                # 创建光源节点
+                pointlight_node = self.world.render.attachNewNode(pointlight)
+                pointlight_node.setPos(pos)
+                pointlight_node.setTag("light_type", "point_light")
+                pointlight_node.setTag("is_scene_element", "1")
+                pointlight_node.setTag("tree_item_type", "LIGHT_NODE")
+                pointlight_node.setTag("created_by_user", "1")
+                pointlight_node.setTag("element_type", "pointlight")
+                
+                # 添加到光源列表
+                self.Pointlight.append(pointlight_node)
+                
+                # 启用光源
+                self.world.render.setLight(pointlight_node)
+                
+                # 启用阴影
+                if hasattr(pointlight, 'setShadowCaster'):
+                    pointlight.setShadowCaster(True)
+                
+                print(f"✓ 标准点光源创建成功，位置: {pos}")
+                return pointlight_node
+            
+        except Exception as e:
+            print(f"✗ 创建点光源失败: {e}")
+            return None

scene/scene_manager_serialization_mixin.py

@@ -0,0 +1,863 @@
+"""Scene manager node serialization operations."""
+
+import os
+import shutil
+import time
+import json
+import aiohttp
+import asyncio
+import inspect
+from pathlib import Path
+
+from panda3d.core import (
+    ModelPool, ModelRoot, Filename, NodePath, GeomNode, Material, Vec4, Vec3,
+    MaterialAttrib, ColorAttrib, Point3, CollisionNode, CollisionSphere, CollisionBox,
+    BitMask32, TransparencyAttrib, LColor, TransformState, RenderModeAttrib
+)
+from panda3d.egg import EggData, EggVertexPool
+from direct.actor.Actor import Actor
+from RenderPipelineFile.rpplugins.smaa.jitters import halton_seq
+from scene import util
+
+class SceneManagerSerializationMixin:
+    def serializeNode(self, node):
+        """序列化节点为字典数据"""
+        try:
+            node_data = {
+                'name': node.getName(),
+                'type': type(node.node()).__name__,
+                'pos': (node.getX(), node.getY(), node.getZ()),
+                'hpr': (node.getH(), node.getP(), node.getR()),
+                'scale': (node.getSx(), node.getSy(), node.getSz()),
+                'tags': {},
+                'children': []
+            }
+
+            # 保存所有标签
+            for tag_key in node.getTagKeys():
+                node_data['tags'][tag_key] = node.getTag(tag_key)
+
+            # 特殊处理不同类型的节点
+            if hasattr(node.node(), 'getClassType'):
+                node_class = node.node().getClassType().getName()
+                node_data['node_class'] = node_class
+
+            # 递归序列化子节点
+            for child in node.getChildren():
+                # 跳过辅助节点
+                if not child.getName().startswith(('gizmo', 'selectionBox', 'grid')):
+                    child_data = self.serializeNode(child)
+                    if child_data:
+                        node_data['children'].append(child_data)
+
+            return node_data
+
+        except Exception as e:
+            print(f"序列化节点 {node.getName()} 失败: {e}")
+            import traceback
+            traceback.print_exc()
+            return None
+
+    def deserializeNode(self, node_data, parent_node):
+        """从字典数据反序列化节点"""
+        try:
+            # 创建新节点
+            node_name = node_data.get('name', 'node')
+            new_node = parent_node.attachNewNode(node_name)
+
+            # 设置变换
+            pos = node_data.get('pos', (0, 0, 0))
+            hpr = node_data.get('hpr', (0, 0, 0))
+            scale = node_data.get('scale', (1, 1, 1))
+
+            new_node.setPos(*pos)
+            new_node.setHpr(*hpr)
+            new_node.setScale(*scale)
+
+            # 恢复标签
+            for tag_key, tag_value in node_data.get('tags', {}).items():
+                new_node.setTag(tag_key, tag_value)
+
+            # 根据节点类型进行特殊处理
+            node_type = node_data.get('type', '')
+            node_class = node_data.get('node_class', '')
+
+            # 特殊处理光源节点
+            if 'light_type' in node_data.get('tags', {}):
+                light_type = node_data['tags']['light_type']
+                if light_type == 'spot_light':
+                    self._recreateSpotLight(new_node)
+                elif light_type == 'point_light':
+                    self._recreatePointLight(new_node)
+
+            # 递归创建子节点
+            for child_data in node_data.get('children', []):
+                self.deserializeNode(child_data, new_node)
+
+            return new_node
+
+        except Exception as e:
+            print(f"反序列化节点 {node_data.get('name', 'unknown')} 失败: {e}")
+            import traceback
+            traceback.print_exc()
+            return None
+
+    def serializeNodeForCopy(self, node):
+        """序列化节点用于复制操作，完整保存视觉属性"""
+        try:
+            if not node or node.isEmpty():
+                return None
+
+            node_data = {
+                'name': node.getName(),
+                'type': type(node.node()).__name__,
+                'pos': (node.getX(), node.getY(), node.getZ()),
+                'hpr': (node.getH(), node.getP(), node.getR()),
+                'scale': (node.getSx(), node.getSy(), node.getSz()),
+                'tags': {},
+                'children': []
+            }
+
+            # 保存所有标签
+            try:
+                if hasattr(node, 'getTagKeys'):
+                    for tag_key in node.getTagKeys():
+                        node_data['tags'][tag_key] = node.getTag(tag_key)
+            except Exception as e:
+                print(f"获取标签时出错: {e}")
+
+            # 保存视觉属性
+            try:
+                # 保存颜色属性
+                if hasattr(node, 'getColor'):
+                    color = node.getColor()
+                    node_data['color'] = (color.getX(), color.getY(), color.getZ(), color.getW())
+
+                # 保存材质属性
+                if hasattr(node, 'getMaterial'):
+                    material = node.getMaterial()
+                    if material:
+                        material_data = {}
+                        material_data['base_color'] = (
+                            material.getBaseColor().getX(),
+                            material.getBaseColor().getY(),
+                            material.getBaseColor().getZ(),
+                            material.getBaseColor().getW()
+                        )
+                        material_data['ambient'] = (
+                            material.getAmbient().getX(),
+                            material.getAmbient().getY(),
+                            material.getAmbient().getZ(),
+                            material.getAmbient().getW()
+                        )
+                        material_data['diffuse'] = (
+                            material.getDiffuse().getX(),
+                            material.getDiffuse().getY(),
+                            material.getDiffuse().getZ(),
+                            material.getDiffuse().getW()
+                        )
+                        material_data['specular'] = (
+                            material.getSpecular().getX(),
+                            material.getSpecular().getY(),
+                            material.getSpecular().getZ(),
+                            material.getSpecular().getW()
+                        )
+                        material_data['shininess'] = material.getShininess()
+                        node_data['material'] = material_data
+
+            except Exception as e:
+                print(f"保存视觉属性时出错: {e}")
+
+            # 根据节点类型保存特定信息
+            if node.hasTag("tree_item_type"):
+                node_type = node.getTag("tree_item_type")
+                node_data['node_type'] = node_type
+
+                # 保存特定类型节点的额外信息
+                if node_type in ["LIGHT_NODE", "SPOT_LIGHT_NODE", "POINT_LIGHT_NODE"]:
+                    # 保存光源特定信息
+                    rp_light = node.getPythonTag("rp_light_object")
+                    if rp_light:
+                        node_data['light_data'] = {
+                            'energy': getattr(rp_light, 'energy', 5000),
+                            'radius': getattr(rp_light, 'radius', 1000),
+                            'fov': getattr(rp_light, 'fov', 70) if hasattr(rp_light, 'fov') else None,
+                            'inner_radius': getattr(rp_light, 'inner_radius', 0.4) if hasattr(rp_light,
+                                                                                              'inner_radius') else None,
+                            'casts_shadows': getattr(rp_light, 'casts_shadows', True) if hasattr(rp_light,
+                                                                                                 'casts_shadows') else True,
+                            'shadow_map_resolution': getattr(rp_light, 'shadow_map_resolution', 256) if hasattr(
+                                rp_light, 'shadow_map_resolution') else 256
+                        }
+                elif node_type in ["GUI_BUTTON", "GUI_LABEL", "GUI_ENTRY", "GUI_IMAGE",
+                                   "GUI_3D_TEXT", "GUI_3D_IMAGE", "GUI_VIRTUAL_SCREEN"]:
+                    # 保存GUI元素特定信息
+                    node_data['gui_data'] = self._serializeGUIData(node)
+                elif node_type == "IMPORTED_MODEL_NODE":
+                    # 保存模型特定信息
+                    node_data['model_data'] = self._serializeModelData(node)
+
+            return node_data
+
+        except Exception as e:
+            print(f"序列化节点失败: {e}")
+            import traceback
+            traceback.print_exc()
+            return None
+
+    def _serializeGUIData(self, node):
+        """序列化GUI元素数据"""
+        try:
+            gui_data = {}
+
+            # 保存GUI相关的通用属性
+            if node.hasTag("gui_type"):
+                gui_data['gui_type'] = node.getTag("gui_type")
+
+            # 保存文本内容（如果有的话）
+            if node.hasTag("text"):
+                gui_data['text'] = node.getTag("text")
+
+            # 保存其他GUI相关标签
+            gui_tags = ['font', 'font_size', 'text_color', 'bg_color', 'size']
+            for tag in gui_tags:
+                if node.hasTag(tag):
+                    gui_data[tag] = node.getTag(tag)
+
+            return gui_data
+        except Exception as e:
+            print(f"序列化GUI数据失败: {e}")
+            return {}
+
+    def _serializeModelTextures(self, node):
+        """序列化模型纹理信息"""
+        try:
+            texture_data = {}
+            
+            # 获取节点的所有纹理阶段
+            from panda3d.core import TextureStage
+            texture_stages = node.findAllTextureStages()
+            
+            if texture_stages.getNumTextureStages() > 0:
+                texture_data['textures'] = {}
+                
+                # 遍历所有纹理阶段
+                for i in range(texture_stages.getNumTextureStages()):
+                    stage = texture_stages.getTextureStage(i)
+                    stage_name = stage.getName()
+                    
+                    # 获取该阶段的纹理
+                    texture = node.getTexture(stage)
+                    if texture:
+                        # 保存纹理信息
+                        texture_info = {
+                            'stage_name': stage_name,
+                            'stage_mode': stage.getMode(),
+                            'stage_sort': stage.getSort(),  # 保存纹理阶段排序
+                            'texture_path': texture.getFullpath().toOsSpecific() if texture.hasFullpath() else '',
+                            'texture_name': texture.getName(),
+                            'wrap_u': texture.getWrapU(),
+                            'wrap_v': texture.getWrapV(),
+                            'minfilter': texture.getMinfilter(),
+                            'magfilter': texture.getMagfilter(),
+                            'anisotropic_degree': texture.getAnisotropicDegree()
+                        }
+                        
+                        # 保存颜色比例和偏移（使用安全的方法）
+                        try:
+                            texture_info['color_scale'] = tuple(node.getTextureScale(stage))
+                        except:
+                            texture_info['color_scale'] = (1.0, 1.0, 1.0, 1.0)
+                            
+                        try:
+                            texture_info['color_offset'] = tuple(node.getTextureOffset(stage))
+                        except:
+                            texture_info['color_offset'] = (0.0, 0.0, 0.0, 0.0)
+                        
+                        texture_data['textures'][stage_name] = texture_info
+                        
+            return texture_data
+        except Exception as e:
+            print(f"序列化模型纹理时出错: {e}")
+            return {}
+
+    def _serializeModelData(self, node):
+        """序列化模型数据，包括材质和纹理信息"""
+        try:
+            model_data = {}
+
+            # 保存模型相关的标签
+            model_tags = ['model_path', 'file', 'element_type']
+            for tag in model_tags:
+                if node.hasTag(tag):
+                    model_data[tag] = node.getTag(tag)
+
+            # 保存材质信息
+            try:
+                # 获取模型的材质信息
+                if hasattr(node, 'getState'):
+                    state = node.getState()
+                    if state:
+                        # 保存基础颜色信息（使用正确的方法）
+                        from panda3d.core import ColorAttrib
+                        color_attrib = state.getAttrib(ColorAttrib)
+                        if color_attrib and not color_attrib.isOff():
+                            color = color_attrib.getColor()
+                            model_data['base_color'] = (
+                                color.getX(),
+                                color.getY(),
+                                color.getZ(),
+                                color.getW()
+                            )
+
+                        # 保存其他材质属性
+                        from panda3d.core import MaterialAttrib
+                        material_attrib = state.getAttrib(MaterialAttrib.getClassType())
+                        if material_attrib:
+                            material = material_attrib.getMaterial()
+                            if material:
+                                # 保存基础颜色
+                                base_color = material.getBaseColor()
+                                model_data['material_base_color'] = (
+                                    base_color.getX(), base_color.getY(), base_color.getZ(), base_color.getW()
+                                )
+
+                                # 保存环境光颜色
+                                ambient_color = material.getAmbient()
+                                model_data['material_ambient_color'] = (
+                                    ambient_color.getX(), ambient_color.getY(), ambient_color.getZ(),
+                                    ambient_color.getW()
+                                )
+
+                                # 保存漫反射颜色
+                                diffuse_color = material.getDiffuse()
+                                model_data['material_diffuse_color'] = (
+                                    diffuse_color.getX(), diffuse_color.getY(), diffuse_color.getZ(),
+                                    diffuse_color.getW()
+                                )
+
+                                # 保存高光颜色
+                                specular_color = material.getSpecular()
+                                model_data['material_specular_color'] = (
+                                    specular_color.getX(), specular_color.getY(), specular_color.getZ(),
+                                    specular_color.getW()
+                                )
+
+                                # 保存粗糙度和金属度等参数
+                                model_data['material_roughness'] = material.getRoughness()
+                                model_data['material_metallic'] = material.getMetallic()
+                                
+                                # 保存自发光颜色
+                                emission_color = material.getEmission()
+                                model_data['material_emission_color'] = (
+                                    emission_color.getX(), emission_color.getY(), emission_color.getZ(),
+                                    emission_color.getW()
+                                )
+                                
+                                # 保存光泽度
+                                model_data['material_shininess'] = material.getShininess()
+                                
+                        # 保存透明度信息
+                        from panda3d.core import TransparencyAttrib
+                        transparency_attrib = state.getAttrib(TransparencyAttrib.getClassType())
+                        if transparency_attrib:
+                            model_data['transparency_mode'] = transparency_attrib.get_mode()
+
+            except Exception as e:
+                print(f"保存材质信息时出错: {e}")
+                
+            # 保存纹理信息
+            try:
+                texture_data = self._serializeModelTextures(node)
+                if texture_data:
+                    model_data['texture_data'] = texture_data
+            except Exception as e:
+                print(f"保存纹理信息时出错: {e}")
+
+            return model_data
+        except Exception as e:
+            print(f"序列化模型数据失败: {e}")
+            return {}
+
+    def recreateNodeFromData(self, node_data, parent_node):
+        """根据数据重建节点，并确保在场景树中显示"""
+        try:
+            if not node_data or not parent_node or parent_node.isEmpty():
+                return None
+
+            print(f"正在重建节点 {node_data}")
+            node_type = node_data.get('node_type', '')
+            original_name = node_data.get('name', 'node')
+
+            # 生成唯一名称
+            unique_name = self._generateUniqueName(original_name, parent_node)
+
+            # 根据节点类型调用相应的重建方法
+            new_node = None
+            if node_type in ["LIGHT_NODE", "SPOT_LIGHT_NODE", "POINT_LIGHT_NODE"]:
+                new_node = self._recreateLightFromData(node_data, parent_node, unique_name)
+            elif node_type == "CESIUM_TILESET_NODE":
+                new_node = self._recreateTilesetFromData(node_data, parent_node, unique_name)
+            elif node_type in ["GUI_BUTTON", "GUI_LABEL", "GUI_ENTRY", "GUI_IMAGE",
+                               "GUI_3DTEXT", "GUI_3DIMAGE", "GUI_VIDEO_SCREEN","GUI_2D_VIDEO_SCREEN"]:
+                new_node = self._recreateGUIFromData(node_data, parent_node, unique_name)
+            elif node_type == "IMPORTED_MODEL_NODE":
+                new_node = self._recreateModelFromData(node_data, parent_node, unique_name)
+            else:
+                # 创建普通节点
+                new_node = self._createBasicNodeFromData(node_data, parent_node, unique_name)
+
+            # 如果成功创建节点，确保它在场景树中显示
+            if new_node:
+                # 尝试更新场景树以显示新节点
+                try:
+                    if hasattr(self.world, 'interface_manager') and self.world.interface_manager:
+                        # 查找父节点在场景树中的对应项
+                        parent_item = self._findTreeItemForNode(parent_node)
+                        if parent_item:
+                            # 添加新节点到场景树
+                            tree_widget = self.world.interface_manager.treeWidget
+                            if tree_widget:
+                                tree_widget.add_node_to_tree_widget(new_node, parent_item, node_type or "NODE")
+                except Exception as e:
+                    print(f"添加节点到场景树时出错: {e}")
+
+            return new_node
+
+        except Exception as e:
+            print(f"重建节点失败: {e}")
+            import traceback
+            traceback.print_exc()
+            return None
+
+    def _findTreeItemForNode(self, node):
+        """根据节点查找对应的场景树项"""
+        try:
+            if hasattr(self.world, 'interface_manager') and self.world.interface_manager:
+                tree_widget = self.world.interface_manager.treeWidget
+                if tree_widget:
+                    # 遍历场景树查找匹配的节点项
+                    for i in range(tree_widget.topLevelItemCount()):
+                        item = tree_widget.topLevelItem(i)
+                        result = self._findTreeItemForNodeRecursive(item, node)
+                        if result:
+                            return result
+            return None
+        except Exception as e:
+            print(f"查找场景树项时出错: {e}")
+            return None
+
+    def _findTreeItemForNodeRecursive(self, item, target_node):
+        """递归查找场景树项"""
+        try:
+            # 检查当前项是否匹配
+            item_node = getattr(item, 'node_path', None)
+            if not item_node:
+                item_node = getattr(item, 'node', None)
+
+            if item_node and item_node == target_node:
+                return item
+
+            # 递归检查子项
+            for i in range(item.childCount()):
+                child_item = item.child(i)
+                result = self._findTreeItemForNodeRecursive(child_item, target_node)
+                if result:
+                    return result
+
+            return None
+        except Exception as e:
+            print(f"递归查找场景树项时出错: {e}")
+            return None
+
+    def _recreateTilesetFromData(self, node_data, parent_node, name):
+        """根据数据重建Tileset"""
+        try:
+            tileset_url = node_data.get('tileset_url', '')
+            if not tileset_url:
+                return None
+
+            # 使用现有方法加载tileset
+            position = node_data.get('pos', (0, 0, 0))
+            tileset_node = self.load_cesium_tileset(tileset_url, position)
+
+            if tileset_node:
+                # 设置名称
+                tileset_node.setName(name)
+
+                # 恢复其他标签
+                for tag_key, tag_value in node_data.get('tags', {}).items():
+                    if tag_key not in ['name']:
+                        tileset_node.setTag(tag_key, str(tag_value))
+
+            return tileset_node
+
+        except Exception as e:
+            print(f"重建Tileset失败: {e}")
+            return None
+
+    def _recreateGUIFromData(self, node_data, parent_node, name):
+        """根据数据重建GUI元素"""
+        try:
+            gui_data = node_data.get('gui_data', {})
+            #gui_type = gui_data.get('gui_type', '')
+            gui_type = node_data.get("tags").get("gui_type", "")
+
+            print(f"正在重建GUI元素: {gui_type}")
+            print(f"正在重建GUI元素: {node_data}")
+
+            # 根据GUI类型调用相应的创建方法
+            new_gui_element = None
+
+            if gui_type == "button" and hasattr(self.world, 'createGUIButton'):
+                pos = node_data.get('pos', (0, 0, 0))
+                text = node_data.get('tags').get('gui_text', '')
+                size = node_data.get('scale', 1)
+                print(pos,text,size)
+                new_gui_element = self.world.createGUIButton(pos,text,size)
+            elif gui_type == "label" and hasattr(self.world, 'createGUILabel'):
+                pos = node_data.get('pos', (0, 0, 0))
+                text = node_data.get('tags').get('gui_text', '')
+                size = node_data.get('scale', 1)
+                new_gui_element = self.world.createGUILabel(pos,text,size)
+            elif gui_type == "entry" and hasattr(self.world, 'createGUIEntry'):
+                pos = node_data.get('pos', (0, 0, 0))
+                text = node_data.get('tags').get('gui_text', '')
+                size = node_data.get('scale', 1)
+                new_gui_element = self.world.createGUIEntry(pos,text,size)
+            elif gui_type == "2d_image" and hasattr(self.world, 'createGUI2DImage'):
+                pos = node_data.get('pos', (0, 0, 0))
+                image_path = node_data.get('tags').get('image_path', '')
+                size = node_data.get('size', 1)
+                new_gui_element = self.world.createGUI2DImage(pos, image_path, size)
+            elif gui_type == "3d_text" and hasattr(self.world, 'createGUI3DText'):
+                print("正在创建3D文本!!!")
+                pos = node_data.get('pos', (0, 0, 0))
+                text = node_data.get('tags', {}).get('gui_text', '')
+                scale = node_data.get('scale', 1)
+                if isinstance(scale, (list, tuple)):
+                    scale = scale[0] if len(scale) > 0 else 1
+                print(f"正在创建3D文本: 位置={pos}, 文本={text}, 大小={scale}")
+                new_gui_element = self.world.createGUI3DText(pos, text, scale)
+            elif gui_type == "3d_image" and hasattr(self.world, 'createGUI3DImage'):
+                pos = node_data.get('pos', (0, 0, 0))
+                image_path = node_data.get('tags').get('gui_image_path', '')
+                scale = node_data.get('scale', (1, 1))
+                if isinstance(scale, (int, float)):
+                    scale = (scale, scale)
+                elif isinstance(scale, (list, tuple)) and len(scale) >= 2:
+                    scale = (scale[0], scale[1])
+                else:
+                    scale = (1, 1)
+                print(f"正在创建3D图片: 位置={pos}, 路径={image_path}, 大小={scale}")
+                new_gui_element = self.world.gui_manager.createGUI3DImage(pos, image_path, scale)
+            elif gui_type == "video_screen" and hasattr(self.world.gui_manager, 'createVideoScreen'):
+                pos = node_data.get('pos', (0, 0, 0))
+                video_path = node_data.get('tags').get('video_path', '')
+                scale = node_data.get('scale', (1, 1,1))
+                new_gui_element = self.world.gui_manager.createVideoScreen(pos,scale,video_path)
+            elif gui_type == "2d_video_screen" and hasattr(self.world.gui_manager, 'createGUI2DVideoScreen'):
+                pos = node_data.get('pos', (0, 0, 0))
+                video_path = node_data.get('tags').get('video_path', '')
+                scale = node_data.get('scale', (1, 1, 1))
+                new_gui_element = self.world.gui_manager.createGUI2DVideoScreen(pos,scale,video_path)
+
+            if new_gui_element:
+                # 设置名称和变换
+                if hasattr(new_gui_element, 'setName'):
+                    new_gui_element.setName(name)
+
+                # 设置位置、旋转、缩放
+                pos = node_data.get('pos', (0, 0, 0))
+                hpr = node_data.get('hpr', (0, 0, 0))
+                scale = node_data.get('scale', (1, 1, 1))
+
+                if hasattr(new_gui_element, 'setPos'):
+                    new_gui_element.setPos(*pos)
+                if hasattr(new_gui_element, 'setHpr'):
+                    new_gui_element.setHpr(*hpr)
+                if hasattr(new_gui_element, 'setScale'):
+                    new_gui_element.setScale(*scale)
+
+                # 恢复文本内容
+                if 'text' in gui_data and hasattr(new_gui_element, 'setText'):
+                    new_gui_element.setText(gui_data['text'])
+
+                # 恢复其他标签
+                for tag_key, tag_value in node_data.get('tags', {}).items():
+                    if hasattr(new_gui_element, 'setTag') and tag_key not in ['name']:
+                        new_gui_element.setTag(tag_key, str(tag_value))
+
+                print(f"GUI元素重建成功: {name}")
+
+            return new_gui_element
+
+        except Exception as e:
+            print(f"重建GUI元素失败: {e}")
+            import traceback
+            traceback.print_exc()
+            return None
+
+    def _recreateModelFromData(self, node_data, parent_node, name):
+        """根据数据重建模型，保持材质"""
+        try:
+            model_data = node_data.get('model_data', {})
+            model_path = model_data.get('model_path', model_data.get('file', ''))
+
+            if not model_path or not os.path.exists(model_path):
+                # 如果原始模型文件不存在，创建一个基本节点
+                return self._createBasicNodeFromData(node_data, parent_node, name)
+
+            # 导入模型，保持原有参数
+            model = self.importModel(
+                model_path,
+                apply_unit_conversion=False,  # 已经处理过的模型不需要再次转换
+                normalize_scales=False,  # 保持原有缩放
+                auto_convert_to_glb=False  # 已经处理过的模型不需要再次转换
+            )
+
+            if model:
+                # 设置名称
+                model.setName(name)
+
+                # 设置变换
+                pos = node_data.get('pos', (0, 0, 0))
+                hpr = node_data.get('hpr', (0, 0, 0))
+                scale = node_data.get('scale', (1, 1, 1))
+
+                model.setPos(*pos)
+                model.setHpr(*hpr)
+                model.setScale(*scale)
+
+                # 恢复材质和纹理信息
+                try:
+                    self._restoreModelMaterial(model, model_data)
+                except Exception as e:
+                    print(f"恢复模型材质时出错: {e}")
+
+                # 恢复标签
+                for tag_key, tag_value in node_data.get('tags', {}).items():
+                    if tag_key not in ['name']:
+                        model.setTag(tag_key, str(tag_value))
+
+                # 添加到模型列表
+                if model not in self.models:
+                    self.models.append(model)
+
+            return model
+
+        except Exception as e:
+            print(f"重建模型失败: {e}")
+            # 出错时创建基本节点
+            return self._createBasicNodeFromData(node_data, parent_node, name)
+
+    def _restoreModelTextures(self, model, texture_data):
+        """恢复模型纹理"""
+        try:
+            if not texture_data or 'textures' not in texture_data:
+                return
+                
+            from panda3d.core import TextureStage, SamplerState
+            
+            textures_info = texture_data['textures']
+            
+            # 为每个纹理阶段恢复纹理
+            for stage_name, texture_info in textures_info.items():
+                # 创建纹理阶段
+                stage = TextureStage(stage_name)
+                stage.setMode(texture_info.get('stage_mode', TextureStage.M_modulate))
+                # 恢复纹理阶段排序
+                stage.setSort(texture_info.get('stage_sort', 0))  # 默认为0（p3d_Texture0）
+                
+                # 加载纹理
+                texture_path = texture_info['texture_path']
+                if texture_path and os.path.exists(texture_path):
+                    texture = self.world.loader.loadTexture(texture_path)
+                    if texture:
+                        # 设置纹理属性
+                        texture.setWrapU(texture_info.get('wrap_u', SamplerState.WM_repeat))
+                        texture.setWrapV(texture_info.get('wrap_v', SamplerState.WM_repeat))
+                        texture.setMinfilter(texture_info.get('minfilter', SamplerState.FT_linear))
+                        texture.setMagfilter(texture_info.get('magfilter', SamplerState.FT_linear))
+                        texture.setAnisotropicDegree(texture_info.get('anisotropic_degree', 1))
+                        
+                        # 应用纹理到模型
+                        model.setTexture(stage, texture, 1)  # 1 表示强制应用
+                        
+                        # 恢复颜色比例和偏移（使用安全的方法）
+                        if 'color_scale' in texture_info:
+                            try:
+                                model.setTextureScale(stage, *texture_info['color_scale'])
+                            except Exception as e:
+                                print(f"恢复纹理比例失败: {e}")
+                            
+                        if 'color_offset' in texture_info:
+                            try:
+                                model.setTextureOffset(stage, *texture_info['color_offset'])
+                            except Exception as e:
+                                print(f"恢复纹理偏移失败: {e}")
+                            
+                        print(f"恢复纹理: {stage_name} <- {texture_path}")
+                else:
+                    print(f"纹理文件不存在或路径为空: {texture_path}")
+                    
+        except Exception as e:
+            print(f"恢复模型纹理时出错: {e}")
+
+    def _restoreModelMaterial(self, model, model_data):
+        """恢复模型材质和纹理"""
+        try:
+            # 恢复基础颜色
+            if 'base_color' in model_data:
+                from panda3d.core import ColorAttrib
+                base_color = model_data['base_color']
+                color = (base_color[0], base_color[1], base_color[2], base_color[3])
+                model.setColor(color)
+
+            # 恢复复杂材质属性
+            if any(key.startswith('material_') for key in model_data.keys()):
+                from panda3d.core import Material
+
+                # 创建新材质或获取现有材质
+                material = Material()
+
+                # 恢复基础颜色
+                if 'material_base_color' in model_data:
+                    base_color = model_data['material_base_color']
+                    material.setBaseColor((base_color[0], base_color[1], base_color[2], base_color[3]))
+
+                # 恢复环境光颜色
+                if 'material_ambient_color' in model_data:
+                    ambient_color = model_data['material_ambient_color']
+                    material.setAmbient((ambient_color[0], ambient_color[1], ambient_color[2], ambient_color[3]))
+
+                # 恢复漫反射颜色
+                if 'material_diffuse_color' in model_data:
+                    diffuse_color = model_data['material_diffuse_color']
+                    material.setDiffuse((diffuse_color[0], diffuse_color[1], diffuse_color[2], diffuse_color[3]))
+
+                # 恢复高光颜色
+                if 'material_specular_color' in model_data:
+                    specular_color = model_data['material_specular_color']
+                    material.setSpecular((specular_color[0], specular_color[1], specular_color[2], specular_color[3]))
+                    
+                # 恢复自发光颜色
+                if 'material_emission_color' in model_data:
+                    emission_color = model_data['material_emission_color']
+                    material.setEmission((emission_color[0], emission_color[1], emission_color[2], emission_color[3]))
+
+                # 恢复粗糙度和金属度
+                if 'material_roughness' in model_data:
+                    material.setRoughness(model_data['material_roughness'])
+
+                if 'material_metallic' in model_data:
+                    material.setMetallic(model_data['material_metallic'])
+                    
+                # 恢复光泽度
+                if 'material_shininess' in model_data:
+                    material.setShininess(model_data['material_shininess'])
+
+                # 应用材质到模型
+                model.setMaterial(material)
+                
+            # 恢复透明度设置
+            if 'transparency_mode' in model_data:
+                from panda3d.core import TransparencyAttrib
+                transparency_mode = model_data['transparency_mode']
+                model.setTransparency(transparency_mode)
+                
+            # 恢复纹理信息
+            if 'texture_data' in model_data:
+                self._restoreModelTextures(model, model_data['texture_data'])
+
+        except Exception as e:
+            print(f"恢复材质失败: {e}")
+
+    def _createBasicNodeFromData(self, node_data, parent_node, name):
+        """创建基本节点，保持视觉属性"""
+        try:
+            new_node = parent_node.attachNewNode(name)
+
+            # 设置变换
+            pos = node_data.get('pos', (0, 0, 0))
+            hpr = node_data.get('hpr', (0, 0, 0))
+            scale = node_data.get('scale', (1, 1, 1))
+
+            new_node.setPos(*pos)
+            new_node.setHpr(*hpr)
+            new_node.setScale(*scale)
+
+            # 恢复视觉属性
+            try:
+                # 恢复颜色
+                if 'color' in node_data:
+                    color_data = node_data['color']
+                    new_node.setColor(color_data[0], color_data[1], color_data[2], color_data[3])
+
+                # 恢复材质
+                if 'material' in node_data:
+                    from panda3d.core import Material
+                    material_data = node_data['material']
+                    material = Material()
+
+                    if 'base_color' in material_data:
+                        bc = material_data['base_color']
+                        material.setBaseColor((bc[0], bc[1], bc[2], bc[3]))
+
+                    if 'ambient' in material_data:
+                        ac = material_data['ambient']
+                        material.setAmbient((ac[0], ac[1], ac[2], ac[3]))
+
+                    if 'diffuse' in material_data:
+                        dc = material_data['diffuse']
+                        material.setDiffuse((dc[0], dc[1], dc[2], dc[3]))
+
+                    if 'specular' in material_data:
+                        sc = material_data['specular']
+                        material.setSpecular((sc[0], sc[1], sc[2], sc[3]))
+
+                    if 'shininess' in material_data:
+                        material.setShininess(material_data['shininess'])
+
+                    new_node.setMaterial(material)
+
+            except Exception as e:
+                print(f"恢复视觉属性时出错: {e}")
+
+            # 恢复标签
+            for tag_key, tag_value in node_data.get('tags', {}).items():
+                if tag_key not in ['name']:
+                    new_node.setTag(tag_key, str(tag_value))
+
+            return new_node
+
+        except Exception as e:
+            print(f"创建基本节点失败: {e}")
+            return None
+
+    def _generateUniqueName(self, base_name, parent_node):
+        """生成唯一节点名称"""
+        try:
+            # 移除可能的数字后缀
+            import re
+            import time
+            name_base = re.sub(r'_\d+$', '', base_name)
+
+            # 查找现有同名节点
+            counter = 1
+            unique_name = base_name
+            while True:
+                # 检查父节点下是否已存在同名子节点
+                existing_node = parent_node.find(unique_name)
+                if existing_node.isEmpty():
+                    break
+                unique_name = f"{name_base}_{counter}"
+                counter += 1
+                if counter > 1000:  # 防止无限循环
+                    break
+            
+            return unique_name
+
+        except Exception as e:
+            print(f"生成唯一名称时出错: {e}")
+            return f"{base_name}_{int(time.time())}"

scene/tree_roles.py

@@ -0,0 +1,5 @@
+"""Shared tree data-role constants (Qt-independent)."""
+
+# Qt.ItemDataRole.UserRole
+TREE_USER_ROLE = 256
+

scene/util.py

@@ -56,16 +56,29 @@ class CrossPlatformPathHandler:
         exists = os.path.exists(filepath)
         return exists
 
-    def _panda3d_normalize(self, filepath):
+    def _panda3d_normalize(self, filepath):
-        """使用Panda3D标准化路径"""
+        """使用Panda3D标准化路径"""
-        try:
+        for ctor_name in ("fromOsSpecificW", "from_os_specific_w", "fromOsSpecific", "from_os_specific"):
-            panda_filename = Filename.from_os_specific(filepath)
+            ctor = getattr(Filename, ctor_name, None)
-            normalized_path = panda_filename.get_fullpath()
+            if not ctor:
-            print(f"✓ Panda3D标准化: {normalized_path}")
+                continue
-            return normalized_path
+            try:
-        except Exception as e:
+                panda_filename = ctor(filepath)
-            print(f"⚠️ Panda3D标准化失败: {e}")
+                normalized_path = panda_filename.get_fullpath()
-            return filepath
+                if normalized_path:
+                    print(f"✓ Panda3D标准化: {normalized_path}")
+                    return normalized_path
+            except Exception:
+                continue
+        try:
+            panda_filename = Filename(filepath)
+            normalized_path = panda_filename.get_fullpath()
+            if normalized_path:
+                print(f"✓ Panda3D标准化: {normalized_path}")
+                return normalized_path
+        except Exception as e:
+            print(f"⚠️ Panda3D标准化失败: {e}")
+        return filepath
 
     def _attempt_path_fixes(self, filepath):
         """尝试各种路径修复方法"""
@@ -293,4 +306,4 @@ def normalize_model_path(filepath):
 
 def suggest_file_placement(filename):
     """便捷函数：建议文件放置位置"""
-    return path_handler.suggest_file_placement(filename)
+    return path_handler.suggest_file_placement(filename)

ssbo_component/ssbo_controller.py

@@ -1,51 +1,193 @@
 
+import math
 from panda3d.core import (
     GeomVertexFormat, GeomVertexWriter, GeomVertexReader, GeomVertexRewriter,
     InternalName, Vec3, Vec4, LMatrix4f, ShaderBuffer, GeomEnums,
     BoundingSphere, NodePath, GeomNode, Texture, SamplerState,
     Point3, BoundingBox, Quat
 )
-import struct
 import time
 
 class ObjectController:
     """
-    物体控制器 (No Custom Shader Mode)
+    混合架构控制器 (Chunked Static + Dynamic Editing)
-    ====================================
+    ================================================
-    Uses RP's default rendering (no rp.set_effect) for maximum FPS.
+    - 默认: 每个 chunk 使用 flatten 后的静态表示
-    Vertex colors baked for picking. Movement modifies vertex data directly.
+    - 编辑: 被选中对象所属 chunk 切换为动态表示，直接改 NodePath 变换
-    Stores original vertex positions per object for rotation/translation.
+    - 提交: 离开 chunk 时仅重建该 chunk 的静态表示
     """
-    def __init__(self):
+
+    def __init__(self, chunk_size=64, chunk_world_size=40.0):
+        self.chunk_size = max(8, int(chunk_size))
+        self.chunk_world_size = max(8.0, float(chunk_world_size))
+        self._reset_state()
+
+    def _reset_state(self):
         self.name_to_ids = {}
         self.id_to_name = {}
         self.key_to_node = {}
         self.node_list = []
         self.display_names = {}
-        self.global_transforms = []   # Original transforms (for center/position)
+        self.global_transforms = []
-        
-        self.id_to_chunk = {}         # global_id -> (chunk_key, local_idx)
-        self.chunks = {}              # chunk_key -> dict with 'node' key
-        
-        # Vertex index: local_id -> list of (geom_node_np, geom_idx, [row_indices])
-        self.vertex_index = {}
-        
-        # Original vertex positions: local_id -> list of (Vec3,) matching row order
-        self.original_positions = {}
-        
-        # Current position offsets: local_id -> Vec3 delta
         self.position_offsets = {}
+        self.vertex_index = {}
+        self.original_positions = {}
         self.local_to_global_id = {}
         self.local_transform_state = {}
         self.local_transform_base_positions = {}
+        self.pick_vertex_index = {}
         self.virtual_tree = None
         self.virtual_tree_meta = None
-        
+
         self.model = None
         self.pick_model = None
-        self.chunk_node = None  # Single chunk node
+        self.id_to_chunk = {}      # global_id -> chunk_id
-        self._source_model_name = ""
+        self.id_to_object_np = {}  # global_id -> dynamic object nodepath
-        self._source_model_stem = ""
+        self.id_to_pick_np = {}    # global_id -> pick-scene nodepath
+
+        # chunk_id -> {
+        #   "dynamic_np": NodePath,
+        #   "static_np": NodePath or None,
+        #   "members": [global_id],
+        #   "dirty": bool,
+        #   "dynamic_enabled": bool
+        # }
+        self.chunks = {}
+        self.active_chunks = set()
+        self._next_chunk_id = 0
+        # spatial cell key -> [chunk_id, ...]
+        self._cell_to_chunks = {}
+
+        # UI hierarchy metadata (matches source model parent/child structure)
+        self.tree_root_key = None
+        self.tree_nodes = {}
+        self._path_to_tree_key = {}
+
+    def _register_tree_node(self, key, display_name, parent_key):
+        self.tree_nodes[key] = {
+            "name": display_name,
+            "parent": parent_key,
+            "children": [],
+            "local_ids": [],
+        }
+        self.display_names[key] = display_name
+        self.name_to_ids[key] = []
+        if parent_key is not None and parent_key in self.tree_nodes:
+            self.tree_nodes[parent_key]["children"].append(key)
+
+    def _build_scene_tree(self, root_np):
+        """Capture source model hierarchy for UI (independent from render batching)."""
+        self.tree_root_key = "0"
+
+        def walk(np, parent_key, key):
+            display_name = np.get_name() or "Unnamed"
+            self._register_tree_node(key, display_name, parent_key)
+            self._path_to_tree_key[str(np)] = key
+
+            children = list(np.get_children())
+            for i, child in enumerate(children):
+                walk(child, key, f"{key}/{i}")
+
+        walk(root_np, None, self.tree_root_key)
+
+    def _get_model_world_mat(self):
+        """Return current model net transform matrix (to top/root)."""
+        if not self.model:
+            return LMatrix4f.ident_mat()
+        try:
+            if self.model.isEmpty():
+                return LMatrix4f.ident_mat()
+        except Exception:
+            try:
+                if self.model.is_empty():
+                    return LMatrix4f.ident_mat()
+            except Exception:
+                pass
+
+        try:
+            return LMatrix4f(self.model.getNetTransform().getMat())
+        except Exception:
+            try:
+                # snake_case fallback in newer Panda3D bindings
+                return LMatrix4f(self.model.get_net_transform().get_mat())
+            except Exception:
+                pass
+
+        try:
+            top = self.model.getTop()
+            if top and not top.isEmpty():
+                return LMatrix4f(self.model.getMat(top))
+        except Exception:
+            pass
+
+        try:
+            return LMatrix4f(self.model.getMat())
+        except Exception:
+            return LMatrix4f.ident_mat()
+
+    def get_model_world_mat(self):
+        """Public accessor for current model net transform matrix."""
+        return self._get_model_world_mat()
+
+    def _local_point_to_world(self, local_pos):
+        """Convert a local-space point to world-space based on model net transform."""
+        mat = self._get_model_world_mat()
+        p = Point3(float(local_pos.x), float(local_pos.y), float(local_pos.z))
+        wp = mat.xformPoint(p)
+        return Vec3(wp.x, wp.y, wp.z)
+
+    def _world_vec_to_local(self, world_vec):
+        """Convert a world-space vector to model-local space."""
+        mat = self._get_model_world_mat()
+        inv = LMatrix4f(mat)
+        try:
+            inv.invertInPlace()
+        except Exception:
+            try:
+                inv.invert_in_place()
+            except Exception:
+                return Vec3(world_vec)
+        v = Vec3(world_vec)
+        lv = inv.xformVec(v)
+        return Vec3(lv.x, lv.y, lv.z)
+
+    def world_vector_to_model_local(self, world_vec):
+        """Public converter from world delta vector to model-local delta vector."""
+        return self._world_vec_to_local(world_vec)
+
+    def get_model_world_quat(self):
+        """Return current model world quaternion."""
+        if not self.model:
+            return Quat.identQuat()
+        try:
+            if self.model.isEmpty():
+                return Quat.identQuat()
+        except Exception:
+            pass
+        try:
+            top = self.model.getTop()
+            if top and not top.isEmpty():
+                return Quat(self.model.getQuat(top))
+        except Exception:
+            pass
+        try:
+            return Quat(self.model.getQuat())
+        except Exception:
+            return Quat.identQuat()
+
+    def world_quat_delta_to_model_local(self, delta_quat_world):
+        """
+        Convert world-space delta quaternion to model-local delta quaternion.
+        local = inv(model_world_rot) * world_delta * model_world_rot
+        """
+        if delta_quat_world is None:
+            return Quat.identQuat()
+        model_q = self.get_model_world_quat()
+        inv_model_q = Quat(model_q)
+        inv_model_q.invertInPlace()
+        local_q = inv_model_q * Quat(delta_quat_world) * model_q
+        local_q.normalize()
+        return local_q
 
     def _build_original_hierarchy_key(self, np, model_root):
         """Capture hierarchy path before flatten/reparent."""
@@ -60,6 +202,71 @@ class ObjectController:
         if not parts:
             return np.get_name() or "Unnamed"
         return "/".join(parts)
+    def _aggregate_tree_ids(self, key):
+        node = self.tree_nodes[key]
+        agg_ids = list(node["local_ids"])
+        for child_key in node["children"]:
+            agg_ids.extend(self._aggregate_tree_ids(child_key))
+        self.name_to_ids[key] = agg_ids
+        return agg_ids
+
+    def _build_tree_preorder(self, key, out):
+        out.append(key)
+        for child_key in self.tree_nodes[key]["children"]:
+            self._build_tree_preorder(child_key, out)
+
+    def should_hide_tree_node(self, key):
+        """
+        Hide a redundant wrapper node directly below the file root, e.g. ROOT.
+        This keeps `model.glb` as the visible root in the UI.
+        """
+        node = self.tree_nodes.get(key)
+        if not node:
+            return False
+        if node["parent"] != self.tree_root_key:
+            return False
+
+        name = (node["name"] or "").strip().lower()
+        if name != "root":
+            return False
+
+        # Keep node visible if it actually carries direct geoms.
+        if node["local_ids"]:
+            return False
+
+        return len(node["children"]) > 0
+
+    def _encode_id_color(self, vdata, object_id):
+        if not vdata.has_column("color"):
+            new_fmt = vdata.get_format().get_union_format(GeomVertexFormat.get_v3c4())
+            vdata.set_format(new_fmt)
+
+        low = object_id & 0xFF
+        high = (object_id >> 8) & 0xFF
+        r = low / 255.0
+        g = high / 255.0
+
+        writer = GeomVertexWriter(vdata, InternalName.make("color"))
+        for row in range(vdata.get_num_rows()):
+            writer.set_row(row)
+            writer.set_data4f(r, g, 0.0, 1.0)
+
+    def _ensure_chunk(self, root_np, chunk_id):
+        if chunk_id in self.chunks:
+            return self.chunks[chunk_id]
+
+        dynamic_np = root_np.attach_new_node(f"chunk_{chunk_id:04d}_dynamic")
+        dynamic_np.stash()
+
+        chunk_data = {
+            "dynamic_np": dynamic_np,
+            "static_np": None,
+            "members": [],
+            "dirty": False,
+            "dynamic_enabled": False,
+        }
+        self.chunks[chunk_id] = chunk_data
+        return chunk_data
 
     def _is_wrapper_segment(self, segment):
         s = (segment or "").strip().lower()
@@ -99,20 +306,21 @@ class ObjectController:
         self.local_to_global_id = {}
         self.local_transform_state = {}
         self.local_transform_base_positions = {}
+        self.pick_vertex_index = {}
         self.virtual_tree = None
         self.virtual_tree_meta = None
         self.pick_model = None
         model_name = (model.get_name() or "").strip()
         self._source_model_name = model_name.lower()
         self._source_model_stem = model_name.rsplit(".", 1)[0].lower() if "." in model_name else model_name.lower()
-        
+
         global_id_counter = 0
         chunk_key = model.get_name() or "default"
-        
+
         # No chunk wrapper — flatten directly on model (same as load_jyc_flatten.py)
         self.chunk_node = model
         self.chunks[chunk_key] = {'node': model, 'base_id': 0}
-        
+
         # Cache original hierarchy path BEFORE flatten/reparent.
         original_keys = {}
         for np in geom_nodes:
@@ -121,12 +329,12 @@ class ObjectController:
         # Flatten hierarchy
         for np in geom_nodes:
             np.wrt_reparent_to(model)
-        
+
         local_idx = 0
-        
+
         for np in geom_nodes:
             gnode = np.node()
-            
+
             if gnode.get_num_parents() > 1:
                 parent = np.get_parent()
                 if not parent.is_empty():
@@ -134,57 +342,57 @@ class ObjectController:
                     np.detach_node()
                     np = new_np
                     gnode = np.node()
-            
+
             unique_key = original_keys.get(id(np), str(np))
             display_name = np.get_name() or f"Object_{global_id_counter}"
-            
+
             if unique_key not in self.name_to_ids:
                 self.name_to_ids[unique_key] = []
                 self.key_to_node[unique_key] = np
                 self.node_list.append(unique_key)
                 self.display_names[unique_key] = display_name
-            
+
             # Save original transform
             mat_double = np.get_mat()
             original_transform = LMatrix4f(mat_double)
-            
+
             for i in range(gnode.get_num_geoms()):
                 geom = gnode.modify_geom(i)
                 vdata = geom.modify_vertex_data()
-                
+
                 if not vdata.has_column("color"):
                     new_format = vdata.get_format().get_union_format(GeomVertexFormat.get_v3c4())
                     vdata.set_format(new_format)
-                
+
                 # Encode Local ID in R/G
                 low = local_idx % 256
                 high = local_idx // 256
                 r = low / 255.0
                 g = high / 255.0
-                
+
                 writer = GeomVertexWriter(vdata, InternalName.make("color"))
                 for row in range(vdata.get_num_rows()):
                     writer.set_row(row)
                     writer.set_data4f(r, g, 0.0, 1.0)
-                
+
                 self.global_transforms.append(original_transform)
                 self.id_to_chunk[global_id_counter] = (chunk_key, local_idx)
                 self.name_to_ids[unique_key].append(global_id_counter)
                 self.id_to_name[global_id_counter] = unique_key
                 self.local_to_global_id[local_idx] = global_id_counter
                 self.position_offsets[local_idx] = Vec3(0, 0, 0)
-                
+
                 global_id_counter += 1
                 local_idx += 1
-            
+
             # DO NOT reset transform — keep world-space positions
-        
+
         # Flatten directly on model — NO set_final, allows per-geom frustum culling
         model.flatten_strong()
-        
+
         t1 = time.time()
         print(f"[控制器] Flatten took {(t1-t0)*1000:.0f}ms")
-        
+
         # Build vertex index AFTER flatten
         self._build_vertex_index(model)
         self._init_local_transform_state()
@@ -192,6 +400,7 @@ class ObjectController:
 
         # Keep ID colors only in picking clone to avoid affecting visible shading.
         self.pick_model = model.copy_to(NodePath("ssbo_pick_root"))
+        self._build_pick_vertex_index(self.pick_model)
         self._set_uniform_vertex_color(model, 1.0, 1.0, 1.0, 1.0)
         
         t2 = time.time()
@@ -219,6 +428,114 @@ class ObjectController:
                     writer.set_row(row)
                     writer.set_data4f(r, g, b, a)
 
+    def _build_tree_preorder(self, key, out):
+        out.append(key)
+        for child_key in self.tree_nodes[key]["children"]:
+            self._build_tree_preorder(child_key, out)
+
+    def should_hide_tree_node(self, key):
+        """
+        Hide a redundant wrapper node directly below the file root, e.g. ROOT.
+        This keeps `model.glb` as the visible root in the UI.
+        """
+        node = self.tree_nodes.get(key)
+        if not node:
+            return False
+        if node["parent"] != self.tree_root_key:
+            return False
+
+        name = (node["name"] or "").strip().lower()
+        if name != "root":
+            return False
+
+        # Keep node visible if it actually carries direct geoms.
+        if node["local_ids"]:
+            return False
+
+        return len(node["children"]) > 0
+
+    def _encode_id_color(self, vdata, object_id):
+        if not vdata.has_column("color"):
+            new_fmt = vdata.get_format().get_union_format(GeomVertexFormat.get_v3c4())
+            vdata.set_format(new_fmt)
+
+        low = object_id & 0xFF
+        high = (object_id >> 8) & 0xFF
+        r = low / 255.0
+        g = high / 255.0
+
+        writer = GeomVertexWriter(vdata, InternalName.make("color"))
+        for row in range(vdata.get_num_rows()):
+            writer.set_row(row)
+            writer.set_data4f(r, g, 0.0, 1.0)
+
+    def _ensure_chunk(self, root_np, chunk_id):
+        if chunk_id in self.chunks:
+            return self.chunks[chunk_id]
+
+        dynamic_np = root_np.attach_new_node(f"chunk_{chunk_id:04d}_dynamic")
+        dynamic_np.stash()
+
+        chunk_data = {
+            "dynamic_np": dynamic_np,
+            "static_np": None,
+            "members": [],
+            "dirty": False,
+            "dynamic_enabled": False,
+        }
+        self.chunks[chunk_id] = chunk_data
+        return chunk_data
+
+    def _get_cell_key_from_pos(self, pos):
+        inv = 1.0 / self.chunk_world_size
+        return (
+            int(math.floor(pos.x * inv)),
+            int(math.floor(pos.y * inv)),
+            int(math.floor(pos.z * inv)),
+        )
+
+    def _allocate_spatial_chunk(self, root_np, world_pos):
+        """
+        Allocate object into a spatially-local chunk for better frustum culling.
+        Objects in the same world cell are grouped together, and overflow creates
+        another chunk for that same cell.
+        """
+        cell_key = self._get_cell_key_from_pos(world_pos)
+        chunk_ids = self._cell_to_chunks.setdefault(cell_key, [])
+
+        for chunk_id in chunk_ids:
+            chunk = self.chunks.get(chunk_id)
+            if chunk and len(chunk["members"]) < self.chunk_size:
+                return chunk_id, chunk
+
+        chunk_id = self._next_chunk_id
+        self._next_chunk_id += 1
+        chunk_ids.append(chunk_id)
+        return chunk_id, self._ensure_chunk(root_np, chunk_id)
+
+    def _rebuild_static_chunk(self, chunk_id):
+        chunk = self.chunks.get(chunk_id)
+        if not chunk:
+            return
+
+        old_static = chunk.get("static_np")
+        if old_static and not old_static.is_empty():
+            old_static.remove_node()
+
+        static_np = chunk["dynamic_np"].copy_to(self.model)
+        static_np.set_name(f"chunk_{chunk_id:04d}_static")
+        static_np.unstash()
+        static_np.flatten_strong()
+
+        chunk["static_np"] = static_np
+        chunk["dirty"] = False
+
+        # Keep visibility coherent with current mode after rebuild.
+        if chunk["dynamic_enabled"]:
+            static_np.stash()
+        else:
+            static_np.unstash()
+
     def build_virtual_hierarchy(self):
         """Build a readonly virtual tree from node_list path keys."""
         root = {
@@ -403,6 +720,228 @@ class ObjectController:
                     self.vertex_index[uid].append((gn_np, gi, rows))
                     self.original_positions[uid].append(pos.copy())
 
+    def _set_chunk_dynamic(self, chunk_id, enabled):
+        chunk = self.chunks.get(chunk_id)
+        if not chunk:
+            return
+
+        if enabled:
+            if chunk["dynamic_enabled"]:
+                return
+            chunk["dynamic_np"].unstash()
+            if chunk["static_np"] and not chunk["static_np"].is_empty():
+                chunk["static_np"].stash()
+            chunk["dynamic_enabled"] = True
+            self.active_chunks.add(chunk_id)
+            return
+
+        if not chunk["dynamic_enabled"]:
+            return
+        if chunk["static_np"] and not chunk["static_np"].is_empty():
+            chunk["static_np"].unstash()
+        chunk["dynamic_np"].stash()
+        chunk["dynamic_enabled"] = False
+        self.active_chunks.discard(chunk_id)
+
+    def _resolve_chunk_and_local_idx(self, global_id):
+        """
+        Compatibility helper for merged branches:
+        - legacy: id_to_chunk[gid] -> (chunk_id, local_idx)
+        - current: id_to_chunk[gid] -> chunk_id (local_idx defaults to gid)
+        """
+        mapping = self.id_to_chunk.get(global_id)
+        if mapping is None:
+            return None, None
+        if isinstance(mapping, (tuple, list)):
+            if not mapping:
+                return None, None
+            chunk_id = mapping[0]
+            local_idx = mapping[1] if len(mapping) > 1 else global_id
+            return chunk_id, local_idx
+        return mapping, global_id
+
+    def set_active_ids(self, active_ids):
+        """切换编辑激活集合，仅保留 active_ids 对应 chunk 为动态模式。"""
+        target_chunks = set()
+        for obj_id in active_ids:
+            chunk_id, _ = self._resolve_chunk_and_local_idx(obj_id)
+            if chunk_id is not None:
+                target_chunks.add(chunk_id)
+
+        # Demote no-longer-active chunks. Dirty chunks are re-baked before demotion.
+        for chunk_id in list(self.active_chunks):
+            if chunk_id in target_chunks:
+                continue
+            if self.chunks[chunk_id]["dirty"]:
+                self._rebuild_static_chunk(chunk_id)
+            self._set_chunk_dynamic(chunk_id, False)
+
+        # Promote target chunks.
+        for chunk_id in target_chunks:
+            self._set_chunk_dynamic(chunk_id, True)
+
+    def bake_ids_and_collect(self, model):
+        """
+        构建混合架构:
+        1) 把每个 geom 拆成可独立编辑的动态对象
+        2) 按 chunk 生成 flatten 后的静态副本
+        """
+        t0 = time.time()
+        self._reset_state()
+
+        geom_nodes = list(model.find_all_matches("**/+GeomNode"))
+        print(f"[控制器] 找到 {len(geom_nodes)} 个 GeomNode")
+
+        # Build hierarchy metadata first so UI can mirror source model tree.
+        self._build_scene_tree(model)
+
+        root_name = model.get_name() or "scene"
+        scene_root = NodePath(root_name)
+        pick_root = NodePath(root_name + "_pick")
+        self.model = scene_root
+        self.pick_model = pick_root
+
+        global_id = 0
+        for np in geom_nodes:
+            gnode = np.node()
+            owner_key = self._path_to_tree_key.get(str(np), self.tree_root_key)
+
+            world_mat = LMatrix4f(np.get_mat(model))
+
+            for gi in range(gnode.get_num_geoms()):
+                # Render geometry stays untouched (keep original material/color behavior).
+                render_geom = gnode.get_geom(gi).make_copy()
+                render_gnode = GeomNode(f"obj_{global_id}")
+                render_gnode.add_geom(render_geom, gnode.get_geom_state(gi))
+
+                # Picking geometry gets encoded ID in vertex color.
+                pick_geom = gnode.get_geom(gi).make_copy()
+                pick_vdata = pick_geom.modify_vertex_data()
+                self._encode_id_color(pick_vdata, global_id)
+                pick_gnode = GeomNode(f"pick_{global_id}")
+                pick_gnode.add_geom(pick_geom, gnode.get_geom_state(gi))
+
+                world_pos = world_mat.get_row3(3)
+                chunk_id, chunk = self._allocate_spatial_chunk(scene_root, world_pos)
+
+                obj_np = chunk["dynamic_np"].attach_new_node(render_gnode)
+                obj_np.set_mat(world_mat)
+                pick_np = pick_root.attach_new_node(pick_gnode)
+                pick_np.set_mat(world_mat)
+
+                chunk["members"].append(global_id)
+                self.id_to_chunk[global_id] = chunk_id
+                self.id_to_object_np[global_id] = obj_np
+                self.id_to_pick_np[global_id] = pick_np
+                self.tree_nodes[owner_key]["local_ids"].append(global_id)
+                self.id_to_name[global_id] = owner_key
+                self.global_transforms.append(LMatrix4f(world_mat))
+                self.position_offsets[global_id] = Vec3(0, 0, 0)
+                global_id += 1
+
+        t1 = time.time()
+        print(f"[控制器] Dynamic object build took {(t1 - t0) * 1000:.0f}ms")
+
+        for chunk_id in sorted(self.chunks):
+            self._rebuild_static_chunk(chunk_id)
+            self._set_chunk_dynamic(chunk_id, False)
+
+        t2 = time.time()
+        print(f"[控制器] Static chunk flatten took {(t2 - t1) * 1000:.0f}ms")
+        print(f"[控制器] Built {len(self.chunks)} chunks, {global_id} objects")
+        print(f"[控制器] Spatial chunking: cell={self.chunk_world_size:.1f}, max_members={self.chunk_size}")
+
+        # Fill per-node aggregate IDs and build deterministic preorder list for UI.
+        self._aggregate_tree_ids(self.tree_root_key)
+        self.node_list = []
+        self._build_tree_preorder(self.tree_root_key, self.node_list)
+
+        model.remove_node()
+        return global_id
+
+    def _build_pick_vertex_index(self, pick_root):
+        """
+        Build local_id -> [(geom_node_np, geom_idx, row_indices_array)] for pick model.
+        This keeps GPU-picking geometry writable in sync with visible geometry edits.
+        """
+        import numpy as np
+
+        self.pick_vertex_index = {}
+        if not pick_root:
+            return
+
+        for gn_np in pick_root.find_all_matches("**/+GeomNode"):
+            gnode = gn_np.node()
+            for gi in range(gnode.get_num_geoms()):
+                geom = gnode.get_geom(gi)
+                vdata = geom.get_vertex_data()
+                num_rows = vdata.get_num_rows()
+                if num_rows == 0:
+                    continue
+
+                fmt = vdata.get_format()
+                color_col = fmt.get_column(InternalName.make("color"))
+                if color_col is None:
+                    continue
+
+                color_array_idx = fmt.get_array_with(InternalName.make("color"))
+                color_start = color_col.get_start()
+                color_array_format = fmt.get_array(color_array_idx)
+                color_stride = color_array_format.get_stride()
+
+                color_handle = vdata.get_array(color_array_idx).get_handle()
+                color_raw = bytes(color_handle.get_data())
+                color_buf = np.frombuffer(color_raw, dtype=np.uint8).reshape(num_rows, color_stride)
+
+                num_components = color_col.get_num_components()
+                component_bytes = color_col.get_component_bytes()
+
+                if component_bytes == 4:
+                    color_data = np.ndarray(
+                        (num_rows, num_components),
+                        dtype=np.float32,
+                        buffer=color_buf[:, color_start:color_start + num_components * 4].tobytes()
+                    )
+                    r_vals = (color_data[:, 0] * 255.0 + 0.5).astype(np.int32)
+                    g_vals = (color_data[:, 1] * 255.0 + 0.5).astype(np.int32)
+                elif component_bytes == 1:
+                    color_bytes = color_buf[:, color_start:color_start + num_components].copy()
+                    r_vals = color_bytes[:, 0].astype(np.int32)
+                    g_vals = color_bytes[:, 1].astype(np.int32)
+                else:
+                    continue
+
+                local_ids = r_vals + (g_vals << 8)
+                sort_idx = np.argsort(local_ids)
+                sorted_ids = local_ids[sort_idx]
+                boundaries = np.where(np.diff(sorted_ids) != 0)[0] + 1
+                id_groups = np.split(sort_idx, boundaries)
+                group_ids = sorted_ids[np.concatenate([[0], boundaries])]
+
+                for k in range(len(group_ids)):
+                    uid = int(group_ids[k])
+                    rows = id_groups[k]
+                    if uid not in self.pick_vertex_index:
+                        self.pick_vertex_index[uid] = []
+                    self.pick_vertex_index[uid].append((gn_np, gi, rows))
+
+    def _apply_vertices_to_pick(self, local_idx, entry_idx, new_pos):
+        """Mirror one transformed vertex group to pick-model geometry."""
+        pick_entries = self.pick_vertex_index.get(local_idx)
+        if not pick_entries or entry_idx >= len(pick_entries):
+            return
+
+        pick_gn_np, pick_gi, pick_rows = pick_entries[entry_idx]
+        gnode = pick_gn_np.node()
+        geom = gnode.modify_geom(pick_gi)
+        vdata = geom.modify_vertex_data()
+        writer = GeomVertexWriter(vdata, "vertex")
+
+        max_rows = min(len(pick_rows), len(new_pos))
+        for j in range(max_rows):
+            writer.set_row(int(pick_rows[j]))
+            writer.set_data3f(float(new_pos[j, 0]), float(new_pos[j, 1]), float(new_pos[j, 2]))
+
     def _init_local_transform_state(self):
         """Initialize transform state for each local_idx after vertex index is ready."""
         self.local_transform_state = {}
@@ -424,10 +963,9 @@ class ObjectController:
             return local_indices
         seen = set()
         for global_id in global_ids:
-            mapping = self.id_to_chunk.get(global_id)
+            _, local_idx = self._resolve_chunk_and_local_idx(global_id)
-            if not mapping:
+            if local_idx is None:
                 continue
-            _, local_idx = mapping
             if local_idx in seen:
                 continue
             if local_idx not in self.vertex_index:
@@ -437,7 +975,7 @@ class ObjectController:
         return local_indices
 
     def get_local_pivot(self, local_idx):
-        """Get pivot for one local object (world-space center)."""
+        """Get pivot for one local object (model-local center)."""
         global_id = self.local_to_global_id.get(local_idx)
         if global_id is None:
             return Vec3(0, 0, 0)
@@ -457,7 +995,8 @@ class ObjectController:
             valid += 1
         if valid == 0:
             return Vec3(0, 0, 0)
-        return acc / float(valid)
+        center_local = acc / float(valid)
+        return self._local_point_to_world(center_local)
 
     def begin_transform_session(self, local_indices):
         """Create immutable baseline snapshot for one gizmo drag session."""
@@ -543,6 +1082,7 @@ class ObjectController:
                 for j in range(len(rows)):
                     writer.set_row(int(rows[j]))
                     writer.set_data3f(float(new_pos[j, 0]), float(new_pos[j, 1]), float(new_pos[j, 2]))
+                self._apply_vertices_to_pick(local_idx, i, new_pos)
 
     def _quat_to_np_mat3(self, quat):
         """Convert Panda3D Quat to 3x3 numpy rotation matrix."""
@@ -571,7 +1111,7 @@ class ObjectController:
         ], dtype=np.float32)
 
     def create_ssbo(self):
-        """No SSBO needed — using RP default rendering."""
+        """No SSBO needed in hybrid mode."""
         return None
 
     def move_object(self, global_id, delta):
@@ -584,10 +1124,28 @@ class ObjectController:
         
         if global_id not in self.id_to_chunk:
             return
-            
+
-        _, local_idx = self.id_to_chunk[global_id]
+        chunk_id, local_idx = self._resolve_chunk_and_local_idx(global_id)
-        
+        if local_idx is None:
-        if local_idx not in self.vertex_index:
+            return
+
+        # Hybrid chunk mode (current) may move NodePaths directly without
+        # vertex_index/original_positions populated.
+        if local_idx not in self.vertex_index or local_idx not in self.original_positions:
+            obj_np = self.id_to_object_np.get(global_id)
+            if not obj_np or obj_np.is_empty():
+                return
+            next_pos = obj_np.get_pos() + delta
+            if hasattr(obj_np, "set_fluid_pos"):
+                obj_np.set_fluid_pos(next_pos)
+            else:
+                obj_np.set_pos(next_pos)
+            pick_np = self.id_to_pick_np.get(global_id)
+            if pick_np and not pick_np.is_empty():
+                pick_np.set_mat(self.model, obj_np.get_mat(self.model))
+            if chunk_id is not None and chunk_id in self.chunks:
+                self.chunks[chunk_id]["dirty"] = True
+            self.position_offsets[local_idx] = self.position_offsets.get(local_idx, Vec3(0)) + delta
             return
         
         # Accumulate offset
@@ -611,28 +1169,35 @@ class ObjectController:
             for j in range(len(rows)):
                 writer.set_row(int(rows[j]))
                 writer.set_data3f(float(new_pos[j, 0]), float(new_pos[j, 1]), float(new_pos[j, 2]))
+            self._apply_vertices_to_pick(local_idx, i, new_pos)
 
     def get_world_pos(self, global_id):
-        """Get current world position of an object."""
+        if not self.model:
-        if global_id not in self.id_to_chunk:
+            return Vec3(0, 0, 0)
+
+        obj_np = self.id_to_object_np.get(global_id)
+        if obj_np and not obj_np.is_empty():
+            p = obj_np.get_pos(self.model)
+            return self._local_point_to_world(Vec3(p))
+
+        _, local_idx = self._resolve_chunk_and_local_idx(global_id)
+        if local_idx is None:
             return Vec3(0, 0, 0)
-        _, local_idx = self.id_to_chunk[global_id]
         
         original_mat = self.global_transforms[global_id]
         original_pos = original_mat.get_row3(3)
         offset = self.position_offsets.get(local_idx, Vec3(0))
-        
+
-        return Vec3(original_pos) + offset
+        local_pos = Vec3(original_pos) + offset
+        return self._local_point_to_world(local_pos)
 
     def get_object_center(self, global_id):
-        """Get the original center position of an object (for rotation pivot)."""
         if global_id >= len(self.global_transforms):
             return Vec3(0, 0, 0)
         mat = self.global_transforms[global_id]
         return Vec3(mat.get_row3(3))
 
     def get_transform(self, global_id):
-        """Get original transform."""
         if global_id >= len(self.global_transforms):
             return LMatrix4f.ident_mat()
         return self.global_transforms[global_id]

templates/main_template.py

@@ -2,8 +2,7 @@
 # -*- coding: utf-8 -*-
 
 """
-测试动画模型 - Panda3D应用程序
+打包
-使用Panda3D引擎编辑器创建
 """
 
 from __future__ import print_function
@@ -14,7 +13,6 @@ from direct.actor.Actor import Actor
 from direct.showbase.ShowBaseGlobal import globalClock
 from panda3d.core import TextNode, CardMaker, TextureStage, NodePath, Texture, TransparencyAttrib, CollisionTraverser, \
     Point3
-from core.InfoPanelManager import InfoPanelManager
 # 获取渲染管线路径
 # 在文件开头添加sys导入（如果还没有的话）
 import sys
@@ -98,8 +96,6 @@ class MainApp(ShowBase):
         # 加载所有脚本e
         self.script_manager.load_all_scripts_from_directory()
 
-        self.info_panel_manager = InfoPanelManager(self)
-
         try:
             # 再导入controller模块
             from rpcore.util.movement_controller import MovementController
@@ -539,8 +535,6 @@ class MainApp(ShowBase):
                             video_path=video_path,
                             size=absolute_scale
                         )
-                    elif gui_type == "info_panel":
-                        new_element = self.info_panel_manager.onCreateSampleInfoPanel()
 
                     if "scripts" in gui_info and new_element:
                         self.processScripts(new_element,gui_info["scripts"])

templates/webgl/index.html

@@ -0,0 +1,16 @@
+<!doctype html>
+<html lang="en">
+<head>
+  <meta charset="utf-8" />
+  <meta name="viewport" content="width=device-width, initial-scale=1" />
+  <title>EG WebGL Scene</title>
+  <link rel="stylesheet" href="./style.css" />
+</head>
+<body>
+  <div id="app">
+    <canvas id="scene-canvas"></canvas>
+    <div id="status" class="status">Loading scene...</div>
+  </div>
+  <script type="module" src="./js/viewer.js"></script>
+</body>
+</html>

templates/webgl/style.css

@@ -0,0 +1,61 @@
+:root {
+  --bg: #0f1115;
+  --panel: rgba(16, 20, 28, 0.88);
+  --text: #d6dde9;
+  --ok: #77d0b9;
+  --warn: #e2b272;
+  --err: #f27878;
+}
+
+* {
+  box-sizing: border-box;
+}
+
+html,
+body,
+#app {
+  width: 100%;
+  height: 100%;
+  margin: 0;
+  padding: 0;
+  overflow: hidden;
+  background: radial-gradient(circle at 20% 20%, #1b2230 0%, var(--bg) 50%, #090b10 100%);
+  color: var(--text);
+  font-family: "Segoe UI", "SF Pro Text", "PingFang SC", sans-serif;
+}
+
+#scene-canvas {
+  width: 100%;
+  height: 100%;
+  display: block;
+}
+
+.status {
+  position: fixed;
+  left: 16px;
+  bottom: 16px;
+  max-width: min(640px, calc(100vw - 32px));
+  padding: 10px 12px;
+  background: var(--panel);
+  border: 1px solid rgba(255, 255, 255, 0.12);
+  border-radius: 8px;
+  backdrop-filter: blur(4px);
+  line-height: 1.45;
+  white-space: pre-wrap;
+  font-size: 13px;
+}
+
+.status.ok {
+  border-color: rgba(119, 208, 185, 0.5);
+  color: var(--ok);
+}
+
+.status.warn {
+  border-color: rgba(226, 178, 114, 0.55);
+  color: var(--warn);
+}
+
+.status.error {
+  border-color: rgba(242, 120, 120, 0.65);
+  color: var(--err);
+}

templates/webgl/viewer.js

@@ -0,0 +1,386 @@
+const statusEl = document.getElementById("status");
+const canvas = document.getElementById("scene-canvas");
+
+function setStatus(message, level = "warn") {
+  if (!statusEl) return;
+  statusEl.textContent = message;
+  statusEl.className = `status ${level}`;
+}
+
+function rowMajorToMatrix4(THREE, m) {
+  const mat = new THREE.Matrix4();
+  mat.set(
+    m[0], m[1], m[2], m[3],
+    m[4], m[5], m[6], m[7],
+    m[8], m[9], m[10], m[11],
+    m[12], m[13], m[14], m[15],
+  );
+  return mat;
+}
+
+function pandaRowMajorToMatrix4(THREE, m) {
+  const mat = new THREE.Matrix4();
+  // Panda's matrix data uses row-vector convention (translation on last row).
+  // Three.js expects column-vector convention (translation on last column).
+  mat.set(
+    m[0], m[4], m[8], m[12],
+    m[1], m[5], m[9], m[13],
+    m[2], m[6], m[10], m[14],
+    m[3], m[7], m[11], m[15],
+  );
+  return mat;
+}
+
+function convertNodeMatrix(THREE, sourceMatRowMajor, basis, basisInv, matrixConvention = "panda_row_vector_row_major") {
+  const src = matrixConvention === "panda_row_vector_row_major"
+    ? pandaRowMajorToMatrix4(THREE, sourceMatRowMajor)
+    : rowMajorToMatrix4(THREE, sourceMatRowMajor);
+  return basis.clone().multiply(src).multiply(basisInv);
+}
+
+function toColorArray(color, fallback = [1, 1, 1]) {
+  if (!Array.isArray(color) || color.length < 3) return fallback;
+  return [Number(color[0]) || 0, Number(color[1]) || 0, Number(color[2]) || 0];
+}
+
+function applyMaterialOverride(THREE, root, override) {
+  if (!override) return;
+
+  root.traverse((obj) => {
+    if (!obj.isMesh || !obj.material) return;
+
+    const list = Array.isArray(obj.material) ? obj.material : [obj.material];
+    for (const mat of list) {
+      if (mat.color && Array.isArray(override.base_color)) {
+        const [r, g, b] = override.base_color;
+        mat.color.setRGB(r ?? 1, g ?? 1, b ?? 1);
+      }
+      if (Object.prototype.hasOwnProperty.call(override, "roughness") && "roughness" in mat) {
+        mat.roughness = Number(override.roughness);
+      }
+      if (Object.prototype.hasOwnProperty.call(override, "metallic") && "metalness" in mat) {
+        mat.metalness = Number(override.metallic);
+      }
+      if (Object.prototype.hasOwnProperty.call(override, "opacity")) {
+        const opacity = THREE.MathUtils.clamp(Number(override.opacity), 0, 1);
+        const isTransparent = opacity < 0.999;
+        mat.opacity = opacity;
+        mat.transparent = isTransparent;
+        // Prevent "see-through solid mesh" when source GLTF had transparent pipeline state.
+        mat.depthWrite = !isTransparent;
+        mat.depthTest = true;
+        mat.blending = isTransparent ? THREE.NormalBlending : THREE.NoBlending;
+        if (!isTransparent && "alphaTest" in mat) {
+          mat.alphaTest = 0;
+        }
+        if (!isTransparent && "transmission" in mat) {
+          mat.transmission = 0;
+        }
+      }
+      mat.needsUpdate = true;
+    }
+  });
+}
+
+function textureSlotByStage(stageName) {
+  const key = String(stageName || "").toLowerCase();
+  if (key.includes("normal")) return "normalMap";
+  if (key.includes("rough")) return "roughnessMap";
+  if (key.includes("metal")) return "metalnessMap";
+  if (key.includes("emission") || key.includes("emissive")) return "emissiveMap";
+  if (key.includes("ao")) return "aoMap";
+  if (key.includes("alpha") || key.includes("opacity")) return "alphaMap";
+  return "map";
+}
+
+function applyTextureOverrides(THREE, root, textureOverrides, textureLoader) {
+  if (!Array.isArray(textureOverrides) || textureOverrides.length === 0) return;
+
+  const texBySlot = new Map();
+  for (const item of textureOverrides) {
+    if (!item || !item.uri) continue;
+    const slot = textureSlotByStage(item.stage);
+    if (texBySlot.has(slot)) continue;
+
+    try {
+      const tex = textureLoader.load(item.uri);
+      tex.flipY = false;
+      texBySlot.set(slot, tex);
+    } catch (err) {
+      console.warn("Texture load failed:", item.uri, err);
+    }
+  }
+
+  if (texBySlot.size === 0) return;
+
+  root.traverse((obj) => {
+    if (!obj.isMesh || !obj.material) return;
+
+    const list = Array.isArray(obj.material) ? obj.material : [obj.material];
+    for (const mat of list) {
+      for (const [slot, tex] of texBySlot.entries()) {
+        if (slot in mat) {
+          mat[slot] = tex;
+        }
+      }
+      mat.needsUpdate = true;
+    }
+  });
+}
+
+function directionToThree(THREE, direction, basis) {
+  const d = new THREE.Vector3(
+    Number(direction?.[0] ?? 0),
+    Number(direction?.[1] ?? 0),
+    Number(direction?.[2] ?? -1),
+  );
+  d.applyMatrix4(basis);
+  if (d.lengthSq() < 1e-6) d.set(0, 0, -1);
+  return d.normalize();
+}
+
+async function bootstrap() {
+  setStatus("Loading WebGL dependencies...");
+
+  let THREE;
+  let OrbitControls;
+  let GLTFLoader;
+
+  try {
+    THREE = await import("../vendor/three.module.min.js");
+    ({ OrbitControls } = await import("../vendor/OrbitControls.js"));
+    ({ GLTFLoader } = await import("../vendor/GLTFLoader.js"));
+  } catch (err) {
+    setStatus(
+      [
+        "Failed to load local Three.js vendor files.",
+        "Please replace vendor placeholders with official files:",
+        "- vendor/three.module.min.js",
+        "- vendor/OrbitControls.js",
+        "- vendor/GLTFLoader.js",
+        "",
+        String(err),
+      ].join("\n"),
+      "error",
+    );
+    throw err;
+  }
+
+  setStatus("Loading scene manifest...");
+
+  const response = await fetch("../scene/scene_webgl.json", { cache: "no-cache" });
+  if (!response.ok) {
+    throw new Error(`Failed to load scene manifest: HTTP ${response.status}`);
+  }
+
+  const data = await response.json();
+
+  const renderer = new THREE.WebGLRenderer({
+    canvas,
+    antialias: true,
+    alpha: false,
+  });
+  renderer.setPixelRatio(window.devicePixelRatio || 1);
+  renderer.setSize(window.innerWidth, window.innerHeight);
+  renderer.outputColorSpace = THREE.SRGBColorSpace;
+
+  const scene = new THREE.Scene();
+  scene.background = new THREE.Color(0x11151c);
+
+  const cameraData = data.camera || {};
+  const camera = new THREE.PerspectiveCamera(
+    Number(cameraData.fov_deg ?? 80),
+    window.innerWidth / Math.max(1, window.innerHeight),
+    Number(cameraData.near ?? 0.1),
+    Number(cameraData.far ?? 10000),
+  );
+
+  const controls = new OrbitControls(camera, renderer.domElement);
+  controls.enableDamping = true;
+  controls.target.set(0, 0, 0);
+
+  const basis = rowMajorToMatrix4(
+    THREE,
+    Array.isArray(data.coordinate?.basis_matrix)
+      ? data.coordinate.basis_matrix
+      : [1, 0, 0, 0, 0, 0, 1, 0, 0, -1, 0, 0, 0, 0, 0, 1],
+  );
+  const basisInv = basis.clone().invert();
+  const matrixConvention = String(data.coordinate?.matrix_convention || "panda_row_vector_row_major");
+
+  if (Array.isArray(cameraData.matrix_local_row_major) && cameraData.matrix_local_row_major.length === 16) {
+    const camMat = convertNodeMatrix(
+      THREE,
+      cameraData.matrix_local_row_major,
+      basis,
+      basisInv,
+      matrixConvention,
+    );
+    camera.matrix.copy(camMat);
+    camera.matrix.decompose(camera.position, camera.quaternion, camera.scale);
+    camera.matrixAutoUpdate = true;
+    camera.updateMatrix();
+  } else {
+    camera.position.set(0, -50, 20);
+    camera.lookAt(0, 0, 0);
+  }
+
+  const env = data.environment || {};
+
+  if (env.ambient_light) {
+    const c = toColorArray(env.ambient_light.color, [0.2, 0.2, 0.2]);
+    const amb = new THREE.AmbientLight(new THREE.Color(c[0], c[1], c[2]), Number(env.ambient_light.intensity ?? 1));
+    scene.add(amb);
+  }
+
+  if (env.directional_light) {
+    const c = toColorArray(env.directional_light.color, [0.8, 0.8, 0.8]);
+    const dirLight = new THREE.DirectionalLight(
+      new THREE.Color(c[0], c[1], c[2]),
+      Number(env.directional_light.intensity ?? 1),
+    );
+
+    const dir = directionToThree(THREE, env.directional_light.direction, basis);
+    dirLight.position.copy(dir.clone().multiplyScalar(-40));
+    dirLight.target.position.set(0, 0, 0);
+
+    scene.add(dirLight);
+    scene.add(dirLight.target);
+  }
+
+  const nodeMap = new Map();
+  const pendingModelLoads = [];
+
+  const gltfLoader = new GLTFLoader();
+  const textureLoader = new THREE.TextureLoader();
+
+  for (const node of Array.isArray(data.nodes) ? data.nodes : []) {
+    let obj;
+
+    if (node.kind === "point_light") {
+      const light = node.light || {};
+      const c = toColorArray(light.color, [1, 1, 1]);
+      obj = new THREE.PointLight(
+        new THREE.Color(c[0], c[1], c[2]),
+        Number(light.intensity ?? 1),
+        Number(light.range ?? 0),
+      );
+    } else if (node.kind === "spot_light") {
+      const light = node.light || {};
+      const c = toColorArray(light.color, [1, 1, 1]);
+      const angle = THREE.MathUtils.degToRad(Number(light.spot_angle_deg ?? 45));
+      const spot = new THREE.SpotLight(
+        new THREE.Color(c[0], c[1], c[2]),
+        Number(light.intensity ?? 1),
+        Number(light.range ?? 0),
+        angle,
+        1 - Number(light.inner_cone_ratio ?? 0.4),
+      );
+      spot.target.position.set(0, 0, -1);
+      obj = spot;
+    } else if (node.kind === "ground") {
+      const g = node.ground || {};
+      const width = Number(g.width ?? 100);
+      const height = Number(g.height ?? 100);
+      const m = node.material_override || {};
+      const bc = Array.isArray(m.base_color) ? m.base_color : [0.8, 0.8, 0.8, 1];
+      const mat = new THREE.MeshStandardMaterial({
+        color: new THREE.Color(Number(bc[0] ?? 0.8), Number(bc[1] ?? 0.8), Number(bc[2] ?? 0.8)),
+        roughness: Number(m.roughness ?? 1),
+        metalness: Number(m.metallic ?? 0),
+        transparent: Number(m.opacity ?? 1) < 1,
+        opacity: Number(m.opacity ?? 1),
+        side: THREE.DoubleSide,
+      });
+      obj = new THREE.Mesh(new THREE.PlaneGeometry(width, height), mat);
+      obj.receiveShadow = true;
+    } else {
+      obj = new THREE.Group();
+      const modelUri = node.model?.uri;
+      if (modelUri) {
+        const p = new Promise((resolve) => {
+          gltfLoader.load(
+            modelUri,
+            (gltf) => {
+              const root = gltf.scene || (Array.isArray(gltf.scenes) ? gltf.scenes[0] : null);
+              if (root) {
+                applyMaterialOverride(THREE, root, node.material_override || null);
+                applyTextureOverrides(THREE, root, node.texture_overrides || [], textureLoader);
+                obj.add(root);
+              }
+              resolve();
+            },
+            undefined,
+            (err) => {
+              console.warn(`Failed to load model ${modelUri}:`, err);
+              resolve();
+            },
+          );
+        });
+        pendingModelLoads.push(p);
+      }
+    }
+
+    obj.name = node.name || node.id || "node";
+
+    if (Array.isArray(node.matrix_local_row_major) && node.matrix_local_row_major.length === 16) {
+      const converted = convertNodeMatrix(THREE, node.matrix_local_row_major, basis, basisInv, matrixConvention);
+      obj.matrixAutoUpdate = false;
+      obj.matrix.copy(converted);
+      obj.matrix.decompose(obj.position, obj.quaternion, obj.scale);
+    }
+
+    nodeMap.set(node.id, obj);
+  }
+
+  for (const node of Array.isArray(data.nodes) ? data.nodes : []) {
+    const obj = nodeMap.get(node.id);
+    if (!obj) continue;
+
+    const parent = node.parent_id ? nodeMap.get(node.parent_id) : null;
+    if (parent) {
+      parent.add(obj);
+    } else {
+      scene.add(obj);
+    }
+
+    if (obj.isSpotLight && obj.target) {
+      obj.add(obj.target);
+    }
+  }
+
+  await Promise.all(pendingModelLoads);
+
+  const resize = () => {
+    const w = window.innerWidth;
+    const h = Math.max(1, window.innerHeight);
+    camera.aspect = w / h;
+    camera.updateProjectionMatrix();
+    renderer.setSize(w, h);
+  };
+
+  window.addEventListener("resize", resize);
+  resize();
+
+  setStatus(
+    `Scene ready. Nodes: ${(data.nodes || []).length}.\nUse mouse to orbit, wheel to zoom.`,
+    "ok",
+  );
+
+  const clock = new THREE.Clock();
+  const tick = () => {
+    requestAnimationFrame(tick);
+    const dt = clock.getDelta();
+    if (dt >= 0) {
+      controls.update();
+      renderer.render(scene, camera);
+    }
+  };
+
+  tick();
+}
+
+bootstrap().catch((err) => {
+  console.error(err);
+  setStatus(`Viewer bootstrap failed:\n${String(err)}`, "error");
+});

ui/LUI/__init__.py

@@ -0,0 +1 @@
+"""LUI split modules."""

ui/LUI/lui_shared.py

@@ -0,0 +1,69 @@
+"""Shared LUI bootstrap and imports."""
+
+import os
+import sys
+from pathlib import Path
+
+import panda3d.core as p3d
+from panda3d.core import NodePath, CardMaker
+from imgui_bundle import imgui, imgui_ctx
+
+UI_DIR = Path(__file__).resolve().parents[1]
+if str(UI_DIR) not in sys.path:
+    sys.path.insert(0, str(UI_DIR))
+
+BUILTIN_DIR = UI_DIR / "Builtin"
+if str(BUILTIN_DIR) not in sys.path:
+    sys.path.insert(0, str(BUILTIN_DIR))
+
+import panda3d
+panda_dir = os.path.dirname(panda3d.__file__)
+if hasattr(os, "add_dll_directory"):
+    try:
+        os.add_dll_directory(panda_dir)
+        os.add_dll_directory(str(UI_DIR))
+    except Exception as e:
+        print(f"Warning: Failed to add DLL directory: {e}")
+
+try:
+    import lui
+    panda3d.lui = lui
+    sys.modules["panda3d.lui"] = lui
+
+    from Builtin.LUIRegion import LUIRegion
+    from Builtin.LUIInputHandler import LUIInputHandler
+    from Builtin.LUIButton import LUIButton
+    from Builtin.LUILabel import LUILabel
+    from Builtin.LUIInputField import LUIInputField
+    from Builtin.LUISlider import LUISlider
+    from Builtin.LUIFrame import LUIFrame
+    from Builtin.LUISkin import LUIDefaultSkin
+    from Builtin.LUISprite import LUISprite
+    from Builtin.LUIObject import LUIObject
+    from Builtin.LUICheckbox import LUICheckbox
+    from Builtin.LUIProgressbar import LUIProgressbar
+    from Builtin.LUISelectbox import LUISelectbox
+    from Builtin.LUIScrollableRegion import LUIScrollableRegion
+    from Builtin.LUITabbedFrame import LUITabbedFrame
+    from Builtin.LUIVerticalLayout import LUIVerticalLayout
+    from Builtin.LUIHorizontalLayout import LUIHorizontalLayout
+except ImportError as e:
+    print(f"Error: Failed to import LUI: {e}")
+    lui = None
+    LUIRegion = None
+    LUIInputHandler = None
+    LUIButton = None
+    LUILabel = None
+    LUIInputField = None
+    LUISlider = None
+    LUIFrame = None
+    LUIDefaultSkin = None
+    LUISprite = None
+    LUIObject = None
+    LUICheckbox = None
+    LUIProgressbar = None
+    LUISelectbox = None
+    LUIScrollableRegion = None
+    LUITabbedFrame = None
+    LUIVerticalLayout = None
+    LUIHorizontalLayout = None

ui/panels/animation_tools.py

@@ -3,7 +3,7 @@ from pathlib import Path
 
 from direct.actor.Actor import Actor
 from direct.task.TaskManagerGlobal import taskMgr
-from panda3d.core import NodePath, PartSubset
+from panda3d.core import NodePath, PartSubset, Filename
 
 
 class _BoundAnimationProxy:
@@ -704,7 +704,22 @@ class AnimationTools:
 
         def _try_create_actor_from_source(source, source_desc):
             try:
-                actor = Actor(source)
+                resolved_source = source
+                if isinstance(source, (str, os.PathLike)):
+                    src_text = os.fspath(source)
+                    resolved_source = src_text
+                    for ctor_name in ("fromOsSpecificW", "from_os_specific_w", "fromOsSpecific", "from_os_specific"):
+                        ctor = getattr(Filename, ctor_name, None)
+                        if not ctor:
+                            continue
+                        try:
+                            candidate = ctor(src_text).get_fullpath()
+                            if candidate:
+                                resolved_source = candidate
+                                break
+                        except Exception:
+                            continue
+                actor = Actor(resolved_source)
                 # 无论是否已检测到动画名，都显式绑定一次，避免“有名字但无可播放控制”
                 try:
                     actor.bindAllAnims(allowAsyncBind=False)
@@ -1139,7 +1154,10 @@ class AnimationTools:
 
             # 路径 Actor 失败后，再尝试把文件作为普通模型加载并 autoBind
             try:
-                loaded_model = self.loader.loadModel(p)
+                model_source = p
+                if isinstance(p, (str, os.PathLike)):
+                    model_source = Filename.from_os_specific(os.fspath(p))
+                loaded_model = self.loader.loadModel(model_source)
                 if loaded_model and not loaded_model.isEmpty():
                     proxy = _try_create_autobind_proxy(loaded_model, f"文件路径({p})", owns_node=True)
                     if proxy:

ui/panels/app_actions.py

@@ -256,12 +256,105 @@ class AppActions:
         self.add_info_message("另存为项目（功能待实现）")
         # TODO: 实现另存为对话框
         # self.show_save_as_dialog = True
+
+    def _on_build_webgl_package(self):
+        """处理“打包为 WebGL”菜单项。"""
+        if not hasattr(self, "project_manager") or not self.project_manager:
+            self.add_error_message("项目管理器未初始化")
+            return
+
+        if not self.project_manager.current_project_path:
+            self.add_warning_message("请先创建或打开项目，再进行WebGL打包")
+            return
+
+        # 导出前先保存当前项目
+        if not self._save_project_impl():
+            self.add_error_message("打包前自动保存失败，已取消WebGL打包")
+            return
+
+        current_project_path = self.project_manager.current_project_path
+        initial_dir = os.path.dirname(current_project_path) if current_project_path else os.getcwd()
+        selected_dir = self._select_directory_system_dialog("选择 WebGL 打包输出目录", initial_dir)
+        if not selected_dir:
+            dialog_error = getattr(self, "_last_directory_dialog_error", "")
+            if dialog_error:
+                # Fallback: when system dialog is unavailable (e.g. missing tkinter),
+                # reuse the built-in path browser so user can still choose directory.
+                self.add_warning_message(f"系统目录选择器不可用: {dialog_error}")
+                self.path_browser_mode = "webgl_build"
+                self.path_browser_current_path = initial_dir if os.path.isdir(initial_dir) else os.getcwd()
+                self.path_browser_selected_path = self.path_browser_current_path
+                self.show_path_browser = True
+                self._pending_webgl_package = True
+                self._refresh_path_browser()
+                self.add_info_message("已切换到内置路径浏览器，请选择输出目录并点击确定")
+            else:
+                self.add_info_message("已取消WebGL打包")
+            return
+
+        self._execute_webgl_package(selected_dir)
+
+    def _execute_webgl_package(self, selected_dir):
+        """执行WebGL打包并输出消息。"""
+        ok = self.project_manager.buildWebGLPackage(selected_dir)
+        report = getattr(self.project_manager, "last_webgl_export_report", None) or {}
+        status = report.get("status", "failed")
+        out_dir = report.get("output_dir", "")
+        report_path = os.path.join(out_dir, "reports", "export_report.json") if out_dir else ""
+
+        if ok:
+            missing_count = len(report.get("missing_assets", []))
+            unsupported_count = len(report.get("unsupported_assets", []))
+            if status == "partial":
+                self.add_warning_message(
+                    f"WebGL打包部分成功: 缺失资源 {missing_count}，不支持资源 {unsupported_count}"
+                )
+            else:
+                self.add_success_message("WebGL打包成功")
+
+            if out_dir:
+                self.add_info_message(f"输出目录: {out_dir}")
+            if report_path:
+                self.add_info_message(f"报告: {report_path}")
+        else:
+            self.add_error_message("WebGL打包失败")
+            if report_path:
+                self.add_warning_message(f"请检查报告: {report_path}")
     
 
     def _on_exit(self):
         """处理退出菜单项"""
         self.add_info_message("退出应用程序")
         self.userExit()
+
+    def _select_directory_system_dialog(self, title, initial_dir=""):
+        """打开系统目录选择器并返回目录路径。"""
+        self._last_directory_dialog_error = ""
+        try:
+            import tkinter as tk
+            from tkinter import filedialog
+
+            if not initial_dir or (not os.path.isdir(initial_dir)):
+                initial_dir = os.getcwd()
+
+            root = tk.Tk()
+            root.withdraw()
+            try:
+                root.attributes("-topmost", True)
+            except Exception:
+                pass
+
+            selected = filedialog.askdirectory(
+                title=title or "选择目录",
+                initialdir=initial_dir,
+                mustexist=True,
+            )
+            root.destroy()
+            return os.path.normpath(selected) if selected else ""
+        except Exception as e:
+            print(f"目录选择器打开失败: {e}")
+            self._last_directory_dialog_error = str(e)
+            return ""
     
     # ==================== 键盘事件处理函数 ====================
     
@@ -735,6 +828,9 @@ class AppActions:
             
             node_name = node.getName() or "未命名节点"
             parent = node.getParent()
+            ssbo_editor = getattr(self, "ssbo_editor", None)
+            ssbo_model = getattr(ssbo_editor, "model", None) if ssbo_editor else None
+            deleting_ssbo_root = bool(ssbo_model and (node == ssbo_model))
             
             # 创建删除命令
             if hasattr(self, 'command_manager') and self.command_manager:
@@ -747,6 +843,12 @@ class AppActions:
                 print(f"[删除] 命令管理器不可用，直接删除节点: {node_name}")
                 self._perform_node_cleanup(node)
                 node.removeNode()
+
+            if deleting_ssbo_root and ssbo_editor:
+                try:
+                    ssbo_editor.on_model_deleted(node)
+                except Exception as e:
+                    print(f"[SSBO] 删除模型后清理失败: {e}")
             
             print(f"[删除] 成功删除节点: {node_name}")
             return True
@@ -921,6 +1023,8 @@ class AppActions:
             # 检查场景文件
             scene_file = os.path.join(project_path, "scenes", "scene.bam")
             if os.path.exists(scene_file):
+                if getattr(self, "use_ssbo_mouse_picking", False) and callable(getattr(self, "_import_model_for_runtime", None)):
+                    self.use_ssbo_scene_import = True
                 # 加载场景
                 try:
                     if self.scene_manager.loadScene(scene_file):
@@ -1027,17 +1131,12 @@ class AppActions:
 
     def _import_model_for_runtime(self, file_path, prefer_scene_manager=False):
         """Import model through the active runtime path.
-        SSBO mode: load via SSBOEditor only (avoid duplicate SceneManager model).
+        SSBO mode: load via SSBOEditor always (regardless of prefer_scene_manager).
         Legacy mode: load via SceneManager.
         """
-        if prefer_scene_manager:
-            if hasattr(self, 'scene_manager') and self.scene_manager:
-                return self.scene_manager.importModel(file_path)
-            return None
-
         if self.use_ssbo_mouse_picking and getattr(self, 'ssbo_editor', None):
             try:
-                # Clear selection/gizmo first to avoid dangling references to soon-to-be removed nodes.
+                # Clear selection/gizmo first, then import as an additional scene model.
                 if hasattr(self, 'selection') and self.selection:
                     try:
                         self.selection.clearSelection()
@@ -1047,32 +1146,24 @@ class AppActions:
                         except Exception:
                             pass
 
-                # Remove legacy scene-manager models to avoid duplicate rendering
-                if hasattr(self, 'scene_manager') and self.scene_manager and hasattr(self.scene_manager, 'models'):
-                    for m in list(self.scene_manager.models):
-                        try:
-                            if m and not m.isEmpty():
-                                m.removeNode()
-                        except Exception:
-                            pass
-                    self.scene_manager.models = []
-
-                # Replace previous SSBO model
-                old_model = getattr(self.ssbo_editor, 'model', None)
-                if old_model is not None:
-                    try:
-                        if not old_model.isEmpty():
-                            old_model.removeNode()
-                    except Exception:
-                        pass
-
                 self.ssbo_editor.load_model(file_path)
                 model_np = getattr(self.ssbo_editor, 'model', None)
+                normalized_import_path = str(file_path).replace("\\", "/").lower()
+                is_project_scene_bam = (
+                    normalized_import_path.endswith(".bam")
+                    and "/scenes/" in normalized_import_path
+                )
                 # Keep legacy ray-pick fallback usable by adding a collision body.
                 if model_np:
+                    try:
+                        from scene import util as scene_util
+                        normalized_model_path = scene_util.normalize_model_path(file_path)
+                    except Exception:
+                        normalized_model_path = file_path
                     # Apply vital tags manually since SSBO overrides SceneManager loader
-                    model_np.setTag("model_path", file_path)
+                    model_np.setTag("model_path", normalized_model_path)
                     model_np.setTag("original_path", file_path)
+                    model_np.setTag("saved_model_path", normalized_model_path)
                     model_np.setTag("is_model_root", "1")
                     model_np.setTag("is_scene_element", "1")
                     model_np.setTag("file", os.path.basename(file_path))
@@ -1080,8 +1171,14 @@ class AppActions:
                     
                     if hasattr(self, 'scene_manager') and self.scene_manager:
                         try:
-                            self.scene_manager.setupCollision(model_np)
+                            if not is_project_scene_bam:
-                            self.scene_manager._processModelAnimations(model_np)
+                                self.scene_manager.setupCollision(model_np)
+                                self.scene_manager._processModelAnimations(model_np)
+                            else:
+                                model_np.setTag("has_animations", "false")
+                                model_np.setTag("has_animations_checked", "true")
+                            if hasattr(self.scene_manager, "models") and model_np not in self.scene_manager.models:
+                                self.scene_manager.models.append(model_np)
                         except Exception as e:
                             print(f"[SSBO] setup components failed: {e}")
                 return model_np
@@ -1131,33 +1228,35 @@ class AppActions:
                 self.add_error_message("模型导入失败")
                 return None
 
-            if hasattr(self.scene_manager, 'processMaterials'):
+            # SSBO 模式下保留原始材质/纹理，避免模型发黑。
-                self.scene_manager.processMaterials(model_node)
+            if not getattr(self, "use_ssbo_mouse_picking", False):
-                if show_info_message:
-                    self.add_info_message("已应用默认材质")
-
-            try:
-                model_node.clearMaterial()
-                model_node.clearTexture()
-
                 if hasattr(self.scene_manager, 'processMaterials'):
                     self.scene_manager.processMaterials(model_node)
+                    if show_info_message:
+                        self.add_info_message("已应用默认材质")
 
                 try:
-                    color = model_node.getColor()
+                    model_node.clearMaterial()
-                    if color and len(color) >= 4 and color == (1, 1, 1, 1):
+                    model_node.clearTexture()
+
+                    if hasattr(self.scene_manager, 'processMaterials'):
+                        self.scene_manager.processMaterials(model_node)
+
+                    try:
+                        color = model_node.getColor()
+                        if color and len(color) >= 4 and color == (1, 1, 1, 1):
+                            model_node.setColor(0.8, 0.8, 0.8, 1.0)
+                        elif not color:
+                            model_node.setColor(0.8, 0.8, 0.8, 1.0)
+                    except Exception:
                         model_node.setColor(0.8, 0.8, 0.8, 1.0)
-                    elif not color:
+                except Exception as e:
-                        model_node.setColor(0.8, 0.8, 0.8, 1.0)
+                    self.add_warning_message(f"材质处理警告: {e}")
-                except Exception:
-                    model_node.setColor(0.8, 0.8, 0.8, 1.0)
-            except Exception as e:
-                self.add_warning_message(f"材质处理警告: {e}")
 
             if set_origin:
                 model_node.setPos(0, 0, 0)
 
-            if hasattr(self.scene_manager, 'models'):
+            if hasattr(self.scene_manager, 'models') and model_node not in self.scene_manager.models:
                 self.scene_manager.models.append(model_node)
 
             if select_model and hasattr(self, 'selection') and self.selection:
@@ -1172,7 +1271,7 @@ class AppActions:
 
     def _on_import_model(self):
         """处理导入模型菜单项"""
-        self.add_info_message("打开导入模型对话框")
+        self.add_info_message("打开系统文件选择器")
         self.show_import_dialog = True
     
 

ui/panels/create_actions.py

@@ -78,57 +78,6 @@ class CreateActions:
         except Exception as e:
             self.add_error_message(f"创建平面失败: {str(e)}")
     
-
-    def _on_create_3d_text(self):
-        """创建3D文本"""
-        self.show_3d_text_dialog = True
-    
-
-    def _on_create_3d_image(self):
-        """创建3D图片"""
-        self.show_3d_image_dialog = True
-    
-
-    def _on_create_gui_button(self):
-        """创建GUI按钮"""
-        self.show_gui_button_dialog = True
-    
-
-    def _on_create_gui_label(self):
-        """创建GUI标签"""
-        self.show_gui_label_dialog = True
-    
-
-    def _on_create_gui_entry(self):
-        """创建GUI输入框"""
-        self.show_gui_entry_dialog = True
-    
-
-    def _on_create_gui_image(self):
-        """创建GUI图片"""
-        self.show_gui_image_dialog = True
-    
-
-    def _on_create_video_screen(self):
-        """创建视频屏幕"""
-        self.show_video_screen_dialog = True
-    
-
-    def _on_create_2d_video_screen(self):
-        """创建2D视频屏幕"""
-        self.show_2d_video_screen_dialog = True
-    
-
-    def _on_create_spherical_video(self):
-        """创建球形视频"""
-        self.show_spherical_video_dialog = True
-    
-
-    def _on_create_virtual_screen(self):
-        """创建虚拟屏幕"""
-        self.show_virtual_screen_dialog = True
-    
-
     def _on_create_spot_light(self):
         """创建聚光灯"""
         self.show_spot_light_dialog = True

ui/panels/dialog_panels.py

@@ -265,88 +265,80 @@ class DialogPanels:
     
 
     def _draw_import_dialog(self):
-        """绘制导入模型对话框"""
+        """使用系统文件选择器导入模型。"""
         if not self.show_import_dialog:
             return
-        
+
-        # 设置对话框标志
+        # 立即关闭标记，防止每帧重复弹窗
-        flags = (imgui.WindowFlags_.no_resize | 
+        self.show_import_dialog = False
-                imgui.WindowFlags_.no_collapse |
+
-                imgui.WindowFlags_.modal)
+        selected_path = self._select_model_file_system_dialog()
-        
+        if not selected_path:
-        # 获取屏幕尺寸，居中显示对话框
+            self.add_info_message("已取消导入模型")
-        display_size = imgui.get_io().display_size
+            return
-        dialog_width = 600
+
-        dialog_height = 500
+        self.import_file_path = selected_path
-        imgui.set_next_window_size((dialog_width, dialog_height))
+        self._import_model()
-        imgui.set_next_window_pos(
+
-            ((display_size.x - dialog_width) / 2, (display_size.y - dialog_height) / 2)
+    def _select_model_file_system_dialog(self):
-        )
+        """弹出系统文件选择器并返回模型路径。"""
-        
+        try:
-        with imgui_ctx.begin("导入模型", True, flags) as window:
+            import tkinter as tk
-            if not window.opened:
+            from tkinter import filedialog
-                self.show_import_dialog = False
+
-                return
+            initial_dir = os.path.dirname(self.import_file_path) if self.import_file_path else ""
-            
+            if not initial_dir or (not os.path.isdir(initial_dir)):
-            imgui.text("选择要导入的模型文件")
+                candidates = [
-            imgui.separator()
+                    os.path.join(os.getcwd(), "Resources", "models"),
-            
+                    os.path.join(os.getcwd(), "Resources"),
-            # 文件路径输入
+                    os.getcwd(),
-            imgui.text("文件路径:")
+                ]
-            changed, file_path = imgui.input_text("##import_file_path", self.import_file_path, 512)
+                initial_dir = next((p for p in candidates if os.path.isdir(p)), os.getcwd())
-            if changed:
+
-                self.import_file_path = file_path
+            normalized_exts = []
-            
+            for ext in getattr(self, "supported_formats", []):
-            imgui.same_line()
+                ext = str(ext).strip().lower()
-            if imgui.button("浏览..."):
+                if not ext:
-                self.path_browser_mode = "import_model"
+                    continue
-                self.path_browser_current_path = os.path.dirname(self.import_file_path) if self.import_file_path else os.getcwd()
+                if not ext.startswith("."):
-                self.show_path_browser = True
+                    ext = f".{ext}"
-                self._refresh_path_browser()
+                if ext not in normalized_exts:
-            
+                    normalized_exts.append(ext)
-            imgui.separator()
+
-            
+            model_patterns = " ".join(f"*{ext}" for ext in normalized_exts) or "*.*"
-            # 支持的格式说明
+            filetypes = [
-            imgui.text("支持的文件格式:")
+                ("模型文件", model_patterns),
-            formats_text = ", ".join(self.supported_formats)
+                ("All Files", "*.*"),
-            imgui.text_colored((0.7, 0.7, 0.7, 1.0), formats_text)
+            ]
-            
+
-            imgui.separator()
+            root = tk.Tk()
-            
+            root.withdraw()
-            # 文件预览信息
+            try:
-            if self.import_file_path and os.path.exists(self.import_file_path):
+                root.attributes("-topmost", True)
-                file_size = os.path.getsize(self.import_file_path)
+            except Exception:
-                imgui.text(f"文件大小: {file_size / 1024:.2f} KB")
+                pass
-                
+
-                file_ext = os.path.splitext(self.import_file_path)[1].lower()
+            selected_path = filedialog.askopenfilename(
-                if file_ext in self.supported_formats:
+                title="选择要导入的模型文件",
-                    imgui.text_colored((0.176, 1.0, 0.769, 1.0), "✓ 文件格式支持")
+                initialdir=initial_dir,
-                else:
+                filetypes=filetypes,
-                    imgui.text_colored((1.0, 0.3, 0.3, 1.0), "✗ 不支持的文件格式")
+            )
-            else:
+            root.destroy()
-                imgui.text_colored((0.7, 0.7, 0.7, 1.0), "请选择有效的文件路径")
+
-            
+            if not selected_path:
-            imgui.separator()
+                return ""
-            
+
-            # 按钮区域
+            selected_path = os.path.normpath(selected_path)
-            can_import = (self.import_file_path and 
+            file_ext = os.path.splitext(selected_path)[1].lower()
-                         os.path.exists(self.import_file_path) and
+            if normalized_exts and file_ext not in normalized_exts:
-                         os.path.splitext(self.import_file_path)[1].lower() in self.supported_formats)
+                self.add_error_message(f"不支持的文件格式: {file_ext}")
-            
+                return ""
-            # 根据状态设置按钮颜色
+
-            if can_import:
+            self.add_info_message(f"已选择文件: {selected_path}")
-                if imgui.button("导入"):
+            return selected_path
-                    self._import_model()
+        except Exception as e:
-                    self.show_import_dialog = False
+            self.add_error_message(f"打开系统文件选择器失败: {e}")
-            else:
+            return ""
-                # 禁用状态的按钮（灰色显示）
-                imgui.push_style_color(imgui.Col_.button, (0.3, 0.3, 0.3, 1.0))
-                imgui.button("导入")
-                imgui.pop_style_color()
-            
-            imgui.same_line()
-            if imgui.button("取消"):
-                self.show_import_dialog = False
     
 
     def _refresh_path_browser(self):
@@ -410,435 +402,16 @@ class DialogPanels:
                 # 导入模型模式：使用选择的文件路径
                 self.import_file_path = self.path_browser_selected_path
                 self.add_info_message(f"已选择文件: {self.import_file_path}")
+            elif self.path_browser_mode == "webgl_build":
+                output_dir = self.path_browser_current_path
+                self.add_info_message(f"已选择WebGL输出目录: {output_dir}")
+                if getattr(self, "_pending_webgl_package", False):
+                    self._pending_webgl_package = False
+                    if hasattr(self, "_execute_webgl_package"):
+                        self._execute_webgl_package(output_dir)
         except Exception as e:
             self.add_error_message(f"应用路径失败: {e}")
-    
-    # ==================== 创建功能对话框实现 ====================
-    
 
-    def _draw_gui_button_dialog(self):
-        """绘制GUI按钮创建对话框"""
-        if not self.show_gui_button_dialog:
-            return
-            
-        flags = (imgui.WindowFlags_.no_resize | 
-                imgui.WindowFlags_.no_collapse |
-                imgui.WindowFlags_.modal)
-        
-        with imgui_ctx.begin("创建GUI按钮", self.show_gui_button_dialog, flags) as window:
-            if not window:
-                self.show_gui_button_dialog = False
-                return
-            
-            # 初始化参数
-            if 'button_text' not in self.dialog_params:
-                self.dialog_params['button_text'] = "按钮"
-            if 'button_pos' not in self.dialog_params:
-                self.dialog_params['button_pos'] = [0.0, 0.0, 0.0]
-            if 'button_size' not in self.dialog_params:
-                self.dialog_params['button_size'] = [0.1, 0.1, 0.1]
-            
-            imgui.text("GUI按钮参数设置")
-            imgui.separator()
-            
-            # 文本输入
-            changed, self.dialog_params['button_text'] = imgui.input_text("按钮文本", self.dialog_params['button_text'], 256)
-            
-            # 位置输入
-            changed, x = imgui.input_float("X坐标", self.dialog_params['button_pos'][0])
-            if changed:
-                self.dialog_params['button_pos'][0] = x
-            changed, y = imgui.input_float("Y坐标", self.dialog_params['button_pos'][1])
-            if changed:
-                self.dialog_params['button_pos'][1] = y
-            changed, z = imgui.input_float("Z坐标", self.dialog_params['button_pos'][2])
-            if changed:
-                self.dialog_params['button_pos'][2] = z
-            
-            # 大小输入
-            changed, width = imgui.input_float("宽度", self.dialog_params['button_size'][0])
-            if changed:
-                self.dialog_params['button_size'][0] = width
-            changed, height = imgui.input_float("高度", self.dialog_params['button_size'][1])
-            if changed:
-                self.dialog_params['button_size'][1] = height
-            
-            imgui.separator()
-            
-            # 按钮
-            if imgui.button("创建"):
-                try:
-                    pos = tuple(self.dialog_params['button_pos'])
-                    text = self.dialog_params['button_text']
-                    size = tuple(self.dialog_params['button_size'][:2])
-                    
-                    result = self.createGUIButton(pos, text, size)
-                    self.add_success_message(f"GUI按钮创建成功: {text}")
-                    self.show_gui_button_dialog = False
-                except Exception as e:
-                    self.add_error_message(f"创建GUI按钮失败: {str(e)}")
-            
-            imgui.same_line()
-            if imgui.button("取消"):
-                self.show_gui_button_dialog = False
-    
-
-    def _draw_gui_label_dialog(self):
-        """绘制GUI标签创建对话框"""
-        if not self.show_gui_label_dialog:
-            return
-            
-        flags = (imgui.WindowFlags_.no_resize | 
-                imgui.WindowFlags_.no_collapse |
-                imgui.WindowFlags_.modal)
-        
-        with imgui_ctx.begin("创建GUI标签", self.show_gui_label_dialog, flags) as window:
-            if not window:
-                self.show_gui_label_dialog = False
-                return
-            
-            # 初始化参数
-            if 'label_text' not in self.dialog_params:
-                self.dialog_params['label_text'] = "标签"
-            if 'label_pos' not in self.dialog_params:
-                self.dialog_params['label_pos'] = [0.0, 0.0, 0.0]
-            if 'label_size' not in self.dialog_params:
-                self.dialog_params['label_size'] = [0.1, 0.1, 0.1]
-            
-            imgui.text("GUI标签参数设置")
-            imgui.separator()
-            
-            # 文本输入
-            changed, self.dialog_params['label_text'] = imgui.input_text("标签文本", self.dialog_params['label_text'], 256)
-            
-            # 位置输入
-            changed, x = imgui.input_float("X坐标", self.dialog_params['label_pos'][0])
-            if changed:
-                self.dialog_params['label_pos'][0] = x
-            changed, y = imgui.input_float("Y坐标", self.dialog_params['label_pos'][1])
-            if changed:
-                self.dialog_params['label_pos'][1] = y
-            changed, z = imgui.input_float("Z坐标", self.dialog_params['label_pos'][2])
-            if changed:
-                self.dialog_params['label_pos'][2] = z
-            
-            # 大小输入
-            changed, width = imgui.input_float("宽度", self.dialog_params['label_size'][0])
-            if changed:
-                self.dialog_params['label_size'][0] = width
-            changed, height = imgui.input_float("高度", self.dialog_params['label_size'][1])
-            if changed:
-                self.dialog_params['label_size'][1] = height
-            
-            imgui.separator()
-            
-            # 按钮
-            if imgui.button("创建"):
-                try:
-                    pos = tuple(self.dialog_params['label_pos'])
-                    text = self.dialog_params['label_text']
-                    size = tuple(self.dialog_params['label_size'][:2])
-                    
-                    result = self.createGUILabel(pos, text, size)
-                    self.add_success_message(f"GUI标签创建成功: {text}")
-                    self.show_gui_label_dialog = False
-                except Exception as e:
-                    self.add_error_message(f"创建GUI标签失败: {str(e)}")
-            
-            imgui.same_line()
-            if imgui.button("取消"):
-                self.show_gui_label_dialog = False
-    
-
-    def _draw_gui_entry_dialog(self):
-        """绘制GUI输入框创建对话框"""
-        if not self.show_gui_entry_dialog:
-            return
-            
-        flags = (imgui.WindowFlags_.no_resize | 
-                imgui.WindowFlags_.no_collapse |
-                imgui.WindowFlags_.modal)
-        
-        with imgui_ctx.begin("创建GUI输入框", self.show_gui_entry_dialog, flags) as window:
-            if not window:
-                self.show_gui_entry_dialog = False
-                return
-            
-            # 初始化参数
-            if 'entry_pos' not in self.dialog_params:
-                self.dialog_params['entry_pos'] = [0.0, 0.0, 0.0]
-            if 'entry_size' not in self.dialog_params:
-                self.dialog_params['entry_size'] = [0.2, 0.05, 0.1]
-            
-            imgui.text("GUI输入框参数设置")
-            imgui.separator()
-            
-            # 位置输入
-            changed, x = imgui.input_float("X坐标", self.dialog_params['entry_pos'][0])
-            if changed:
-                self.dialog_params['entry_pos'][0] = x
-            changed, y = imgui.input_float("Y坐标", self.dialog_params['entry_pos'][1])
-            if changed:
-                self.dialog_params['entry_pos'][1] = y
-            changed, z = imgui.input_float("Z坐标", self.dialog_params['entry_pos'][2])
-            if changed:
-                self.dialog_params['entry_pos'][2] = z
-            
-            # 大小输入
-            changed, width = imgui.input_float("宽度", self.dialog_params['entry_size'][0])
-            if changed:
-                self.dialog_params['entry_size'][0] = width
-            changed, height = imgui.input_float("高度", self.dialog_params['entry_size'][1])
-            if changed:
-                self.dialog_params['entry_size'][1] = height
-            
-            imgui.separator()
-            
-            # 按钮
-            if imgui.button("创建"):
-                try:
-                    pos = tuple(self.dialog_params['entry_pos'])
-                    size = tuple(self.dialog_params['entry_size'][:2])
-                    
-                    result = self.createGUIEntry(pos, size)
-                    self.add_success_message("GUI输入框创建成功")
-                    self.show_gui_entry_dialog = False
-                except Exception as e:
-                    self.add_error_message(f"创建GUI输入框失败: {str(e)}")
-            
-            imgui.same_line()
-            if imgui.button("取消"):
-                self.show_gui_entry_dialog = False
-    
-
-    def _draw_gui_image_dialog(self):
-        """绘制GUI图片创建对话框"""
-        if not self.show_gui_image_dialog:
-            return
-            
-        flags = (imgui.WindowFlags_.no_resize | 
-                imgui.WindowFlags_.no_collapse |
-                imgui.WindowFlags_.modal)
-        
-        with imgui_ctx.begin("创建GUI图片", self.show_gui_image_dialog, flags) as window:
-            if not window:
-                self.show_gui_image_dialog = False
-                return
-            
-            # 初始化参数
-            if 'image_pos' not in self.dialog_params:
-                self.dialog_params['image_pos'] = [0.0, 0.0, 0.0]
-            if 'image_size' not in self.dialog_params:
-                self.dialog_params['image_size'] = [0.2, 0.2, 0.1]
-            if 'image_path' not in self.dialog_params:
-                self.dialog_params['image_path'] = ""
-            
-            imgui.text("GUI图片参数设置")
-            imgui.separator()
-            
-            # 位置输入
-            changed, x = imgui.input_float("X坐标", self.dialog_params['image_pos'][0])
-            if changed:
-                self.dialog_params['image_pos'][0] = x
-            changed, y = imgui.input_float("Y坐标", self.dialog_params['image_pos'][1])
-            if changed:
-                self.dialog_params['image_pos'][1] = y
-            changed, z = imgui.input_float("Z坐标", self.dialog_params['image_pos'][2])
-            if changed:
-                self.dialog_params['image_pos'][2] = z
-            
-            # 大小输入
-            changed, width = imgui.input_float("宽度", self.dialog_params['image_size'][0])
-            if changed:
-                self.dialog_params['image_size'][0] = width
-            changed, height = imgui.input_float("高度", self.dialog_params['image_size'][1])
-            if changed:
-                self.dialog_params['image_size'][1] = height
-            
-            # 图片路径
-            changed, self.dialog_params['image_path'] = imgui.input_text("图片路径", self.dialog_params['image_path'], 512)
-            
-            imgui.separator()
-            
-            # 按钮
-            if imgui.button("创建"):
-                try:
-                    pos = tuple(self.dialog_params['image_pos'])
-                    size = tuple(self.dialog_params['image_size'][:2])
-                    image_path = self.dialog_params['image_path']
-                    
-                    result = self.createGUIImage(pos, image_path, size)
-                    self.add_success_message("GUI图片创建成功")
-                    self.show_gui_image_dialog = False
-                except Exception as e:
-                    self.add_error_message(f"创建GUI图片失败: {str(e)}")
-            
-            imgui.same_line()
-            if imgui.button("取消"):
-                self.show_gui_image_dialog = False
-    
-
-    def _draw_3d_text_dialog(self):
-        """绘制3D文本创建对话框"""
-        if not self.show_3d_text_dialog:
-            return
-            
-        flags = (imgui.WindowFlags_.no_resize | 
-                imgui.WindowFlags_.no_collapse |
-                imgui.WindowFlags_.modal)
-        
-        with imgui_ctx.begin("创建3D文本", self.show_3d_text_dialog, flags) as window:
-            if not window:
-                self.show_3d_text_dialog = False
-                return
-            
-            # 初始化参数
-            if 'text3d_text' not in self.dialog_params:
-                self.dialog_params['text3d_text'] = "3D文本"
-            if 'text3d_pos' not in self.dialog_params:
-                self.dialog_params['text3d_pos'] = [0.0, 0.0, 0.0]
-            if 'text3d_size' not in self.dialog_params:
-                self.dialog_params['text3d_size'] = 1.0
-            
-            imgui.text("3D文本参数设置")
-            imgui.separator()
-            
-            # 文本输入
-            changed, self.dialog_params['text3d_text'] = imgui.input_text("文本内容", self.dialog_params['text3d_text'], 256)
-            
-            # 位置输入
-            changed, x = imgui.input_float("X坐标", self.dialog_params['text3d_pos'][0])
-            if changed:
-                self.dialog_params['text3d_pos'][0] = x
-            changed, y = imgui.input_float("Y坐标", self.dialog_params['text3d_pos'][1])
-            if changed:
-                self.dialog_params['text3d_pos'][1] = y
-            changed, z = imgui.input_float("Z坐标", self.dialog_params['text3d_pos'][2])
-            if changed:
-                self.dialog_params['text3d_pos'][2] = z
-            
-            # 大小输入
-            changed, self.dialog_params['text3d_size'] = imgui.input_float("文本大小", self.dialog_params['text3d_size'])
-            
-            imgui.separator()
-            
-            # 按钮
-            if imgui.button("创建"):
-                try:
-                    pos = tuple(self.dialog_params['text3d_pos'])
-                    text = self.dialog_params['text3d_text']
-                    size = self.dialog_params['text3d_size']
-                    
-                    result = self.create3DText(pos, text, size)
-                    self.add_success_message(f"3D文本创建成功: {text}")
-                    self.show_3d_text_dialog = False
-                except Exception as e:
-                    self.add_error_message(f"创建3D文本失败: {str(e)}")
-            
-            imgui.same_line()
-            if imgui.button("取消"):
-                self.show_3d_text_dialog = False
-    
-
-    def _draw_3d_image_dialog(self):
-        """绘制3D图片创建对话框"""
-        if not self.show_3d_image_dialog:
-            return
-            
-        flags = (imgui.WindowFlags_.no_resize | 
-                imgui.WindowFlags_.no_collapse |
-                imgui.WindowFlags_.modal)
-        
-        with imgui_ctx.begin("创建3D图片", self.show_3d_image_dialog, flags) as window:
-            if not window:
-                self.show_3d_image_dialog = False
-                return
-            
-            # 初始化参数
-            if 'image3d_pos' not in self.dialog_params:
-                self.dialog_params['image3d_pos'] = [0.0, 0.0, 0.0]
-            if 'image3d_size' not in self.dialog_params:
-                self.dialog_params['image3d_size'] = [1.0, 1.0, 1.0]
-            if 'image3d_path' not in self.dialog_params:
-                self.dialog_params['image3d_path'] = ""
-            
-            imgui.text("3D图片参数设置")
-            imgui.separator()
-            
-            # 位置输入
-            changed, x = imgui.input_float("X坐标", self.dialog_params['image3d_pos'][0])
-            if changed:
-                self.dialog_params['image3d_pos'][0] = x
-            changed, y = imgui.input_float("Y坐标", self.dialog_params['image3d_pos'][1])
-            if changed:
-                self.dialog_params['image3d_pos'][1] = y
-            changed, z = imgui.input_float("Z坐标", self.dialog_params['image3d_pos'][2])
-            if changed:
-                self.dialog_params['image3d_pos'][2] = z
-            
-            # 大小输入
-            changed, width = imgui.input_float("宽度", self.dialog_params['image3d_size'][0])
-            if changed:
-                self.dialog_params['image3d_size'][0] = width
-            changed, height = imgui.input_float("高度", self.dialog_params['image3d_size'][1])
-            if changed:
-                self.dialog_params['image3d_size'][1] = height
-            
-            # 图片路径
-            changed, self.dialog_params['image3d_path'] = imgui.input_text("图片路径", self.dialog_params['image3d_path'], 512)
-            
-            imgui.separator()
-            
-            # 按钮
-            if imgui.button("创建"):
-                try:
-                    pos = tuple(self.dialog_params['image3d_pos'])
-                    size = tuple(self.dialog_params['image3d_size'][:2])
-                    image_path = self.dialog_params['image3d_path']
-                    
-                    result = self.create3DImage(pos, image_path, size)
-                    self.add_success_message("3D图片创建成功")
-                    self.show_3d_image_dialog = False
-                except Exception as e:
-                    self.add_error_message(f"创建3D图片失败: {str(e)}")
-            
-            imgui.same_line()
-            if imgui.button("取消"):
-                self.show_3d_image_dialog = False
-    
-    # 添加其他创建对话框的占位符方法
-
-    def _draw_video_screen_dialog(self):
-        """绘制视频屏幕创建对话框"""
-        if not self.show_video_screen_dialog:
-            return
-        self.show_video_screen_dialog = False
-        self.add_info_message("视频屏幕创建功能开发中...")
-    
-
-    def _draw_2d_video_screen_dialog(self):
-        """绘制2D视频屏幕创建对话框"""
-        if not self.show_2d_video_screen_dialog:
-            return
-        self.show_2d_video_screen_dialog = False
-        self.add_info_message("2D视频屏幕创建功能开发中...")
-    
-
-    def _draw_spherical_video_dialog(self):
-        """绘制球形视频创建对话框"""
-        if not self.show_spherical_video_dialog:
-            return
-        self.show_spherical_video_dialog = False
-        self.add_info_message("球形视频创建功能开发中...")
-    
-
-    def _draw_virtual_screen_dialog(self):
-        """绘制虚拟屏幕创建对话框"""
-        if not self.show_virtual_screen_dialog:
-            return
-        self.show_virtual_screen_dialog = False
-        self.add_info_message("虚拟屏幕创建功能开发中...")
-    
 
     def _draw_spot_light_dialog(self):
         """绘制聚光灯创建对话框"""
@@ -1457,5 +1030,3 @@ class DialogPanels:
         except Exception as e:
             print(f"刷新高度图浏览器时出错: {e}")
             self.heightmap_browser_items = []
-    
-    # ==================== 导入功能实现 ====================

ui/panels/editor_panels.py

@@ -40,7 +40,7 @@ class EditorPanels:
 
     def _ensure_web_panel_state(self):
         if not hasattr(self.app, "web_panel_url_input") or not self.app.web_panel_url_input:
-            self.app.web_panel_url_input = "https://www.example.com"
+            self.app.web_panel_url_input = "https://www.baidu.com"
         if not hasattr(self.app, "_imgui_webview"):
             self.app._imgui_webview = None
         if not hasattr(self.app, "_imgui_webview_tex_id"):
@@ -171,6 +171,8 @@ class EditorPanels:
                                 self.app._on_save_project()
                             if imgui.menu_item("另存为", "", False, True)[1]:
                                 self.app._on_save_as_project()
+                            if imgui.menu_item("打包为 WebGL", "", False, True)[1]:
+                                self.app._on_build_webgl_package()
                             imgui.separator()
                             if imgui.menu_item("退出", "Alt+F4", False, True)[1]:
                                 self.app._on_exit()
@@ -217,33 +219,33 @@ class EditorPanels:
                                         self.app._on_create_plane()
                     
                             # 3D GUI子菜单
-                            with imgui_ctx.begin_menu("3D GUI") as three_d_gui_menu:
+                            # with imgui_ctx.begin_menu("3D GUI") as three_d_gui_menu:
-                                if three_d_gui_menu:
+                            #     if three_d_gui_menu:
-                                    if imgui.menu_item("3D文本", "", False, True)[1]:
+                            #         if imgui.menu_item("3D文本", "", False, True)[1]:
-                                        self.app._on_create_3d_text()
+                            #             self.app._on_create_3d_text()
-                                    if imgui.menu_item("3D图片", "", False, True)[1]:
+                            #         if imgui.menu_item("3D图片", "", False, True)[1]:
-                                        self.app._on_create_3d_image()
+                            #             self.app._on_create_3d_image()
                     
-                            # GUI子菜单
+                            # # GUI子菜单
-                            with imgui_ctx.begin_menu("GUI") as gui_menu:
+                            # with imgui_ctx.begin_menu("GUI") as gui_menu:
-                                if gui_menu:
+                            #     if gui_menu:
-                                    if imgui.menu_item("创建按钮", "", False, True)[1]:
+                            #         if imgui.menu_item("创建按钮", "", False, True)[1]:
-                                        self.app._on_create_gui_button()
+                            #             self.app._on_create_gui_button()
-                                    if imgui.menu_item("创建标签", "", False, True)[1]:
+                            #         if imgui.menu_item("创建标签", "", False, True)[1]:
-                                        self.app._on_create_gui_label()
+                            #             self.app._on_create_gui_label()
-                                    if imgui.menu_item("创建输入框", "", False, True)[1]:
+                            #         if imgui.menu_item("创建输入框", "", False, True)[1]:
-                                        self.app._on_create_gui_entry()
+                            #             self.app._on_create_gui_entry()
-                                    if imgui.menu_item("创建图片", "", False, True)[1]:
+                            #         if imgui.menu_item("创建图片", "", False, True)[1]:
-                                        self.app._on_create_gui_image()
+                            #             self.app._on_create_gui_image()
-                                    imgui.separator()
+                            #         imgui.separator()
-                                    if imgui.menu_item("创建视频屏幕", "", False, True)[1]:
+                            #         if imgui.menu_item("创建视频屏幕", "", False, True)[1]:
-                                        self.app._on_create_video_screen()
+                            #             self.app._on_create_video_screen()
-                                    if imgui.menu_item("创建2D视频屏幕", "", False, True)[1]:
+                            #         if imgui.menu_item("创建2D视频屏幕", "", False, True)[1]:
-                                        self.app._on_create_2d_video_screen()
+                            #             self.app._on_create_2d_video_screen()
-                                    if imgui.menu_item("创建球形视频", "", False, True)[1]:
+                            #         if imgui.menu_item("创建球形视频", "", False, True)[1]:
-                                        self.app._on_create_spherical_video()
+                            #             self.app._on_create_spherical_video()
-                                    if imgui.menu_item("创建虚拟屏幕", "", False, True)[1]:
+                            #         if imgui.menu_item("创建虚拟屏幕", "", False, True)[1]:
-                                        self.app._on_create_virtual_screen()
+                            #             self.app._on_create_virtual_screen()
                     
                             # 光源子菜单
                             with imgui_ctx.begin_menu("光源") as light_menu:
@@ -278,10 +280,10 @@ class EditorPanels:
                             # 信息面板子菜单
                             with imgui_ctx.begin_menu("信息面板") as info_panel_menu:
                                 if info_panel_menu:
-                                    if imgui.menu_item("创建2D示例面板", "", False, True)[1]:
+                                    # if imgui.menu_item("创建2D示例面板", "", False, True)[1]:
-                                        self.app._on_create_2d_sample_panel()
+                                    #     self.app._on_create_2d_sample_panel()
-                                    if imgui.menu_item("创建3D实例面板", "", False, True)[1]:
+                                    # if imgui.menu_item("创建3D实例面板", "", False, True)[1]:
-                                        self.app._on_create_3d_sample_panel()
+                                    #     self.app._on_create_3d_sample_panel()
                                     if imgui.menu_item("Web面板", "", False, True)[1]:
                                         self.app._on_create_web_panel()
             
@@ -318,10 +320,10 @@ class EditorPanels:
                             imgui.separator()
                     
                             # 编辑工具
-                            if imgui.menu_item("光照编辑", "", False, True)[1]:
+                            # if imgui.menu_item("光照编辑", "", False, True)[1]:
-                                self.app.tool_manager.setCurrentTool("光照编辑")
+                            #     self.app.tool_manager.setCurrentTool("光照编辑")
-                            if imgui.menu_item("图形编辑", "", False, True)[1]:
+                            # if imgui.menu_item("图形编辑", "", False, True)[1]:
-                                self.app.tool_manager.setCurrentTool("图形编辑")
+                            #     self.app.tool_manager.setCurrentTool("图形编辑")
                     
                             imgui.separator()
                     
@@ -610,8 +612,33 @@ class EditorPanels:
         
             # 处理节点选择
             if imgui.is_item_clicked():
+                # In SSBO mode, clicking model root should finally bind gizmo to
+                # SSBO group proxy (not legacy model root). So run legacy
+                # selection first, then force SSBO root selection at the end.
+                force_ssbo_root_key = None
+                ssbo_editor_ref = None
+                try:
+                    ssbo_editor = getattr(self.app, "ssbo_editor", None)
+                    controller = getattr(ssbo_editor, "controller", None) if ssbo_editor else None
+                    ssbo_model = getattr(ssbo_editor, "model", None) if ssbo_editor else None
+                    root_key = getattr(controller, "tree_root_key", None) if controller else None
+                    if (
+                        ssbo_editor and controller and ssbo_model and node == ssbo_model
+                        and root_key and root_key in controller.tree_nodes
+                    ):
+                        force_ssbo_root_key = root_key
+                        ssbo_editor_ref = ssbo_editor
+                except Exception:
+                    pass
+
                 if hasattr(self.app, 'selection') and self.app.selection:
                     self.app.selection.updateSelection(node)
+
+                if force_ssbo_root_key and ssbo_editor_ref:
+                    try:
+                        ssbo_editor_ref.select_node(force_ssbo_root_key)
+                    except Exception:
+                        pass
                 # Clear LUI selection when a scene node is selected
                 if hasattr(self.app, 'lui_manager'):
                     self.app.lui_manager.selected_index = -1
@@ -655,7 +682,7 @@ class EditorPanels:
                 imgui.pop_style_color()
 
     def _draw_ssbo_virtual_children(self, node):
-        """Draw SSBO controller nodes as virtual children for scene tree."""
+        """Draw SSBO controller tree_nodes as virtual children for scene tree."""
         ssbo_editor = getattr(self.app, "ssbo_editor", None)
         if not ssbo_editor:
             return False
@@ -664,66 +691,59 @@ class EditorPanels:
         if not model or model != node or not controller:
             return False
 
-        tree_root = controller.get_virtual_hierarchy() if hasattr(controller, "get_virtual_hierarchy") else None
+        root_key = getattr(controller, "tree_root_key", None)
-        if not tree_root or not tree_root.get("children"):
+        if not root_key or root_key not in controller.tree_nodes:
             imgui.text_disabled("(无可用子节点)")
             return True
 
-        for name in sorted(tree_root["children"].keys()):
+        root_node = controller.tree_nodes[root_key]
-            child = tree_root["children"][name]
+        if not root_node["children"]:
-            self._draw_ssbo_virtual_tree_node(ssbo_editor, child, "ssbo_root")
+            imgui.text_disabled("(无可用子节点)")
+            return True
+
+        for child_key in root_node["children"]:
+            self._draw_ssbo_virtual_tree_node(ssbo_editor, controller, child_key)
         return True
 
-    def _draw_ssbo_virtual_tree_node(self, ssbo_editor, tree_node, unique_id_prefix, depth=0):
+    def _draw_ssbo_virtual_tree_node(self, ssbo_editor, controller, key):
-        """Recursively draw virtual SSBO hierarchy in scene tree."""
+        """Recursively draw SSBO tree_nodes hierarchy in scene tree."""
-        if not tree_node:
+        node_data = controller.tree_nodes.get(key)
+        if not node_data:
             return
-        path = tree_node.get("path", "")
-        display = tree_node.get("display_name") or tree_node.get("name") or path
-        leaf_key = tree_node.get("leaf_key")
-        group_key = tree_node.get("group_key")
-        children = tree_node.get("children", {}) or {}
 
-        label = f"{display}##{unique_id_prefix}_{path}"
+        # Skip redundant wrapper nodes (e.g. ROOT), show their children instead.
+        if controller.should_hide_tree_node(key):
+            for child_key in node_data["children"]:
+                self._draw_ssbo_virtual_tree_node(ssbo_editor, controller, child_key)
+            return
 
-        # Leaf: selectable to trigger SSBO selection.
+        display = controller.display_names.get(key, key)
-        if not children and leaf_key:
+        obj_count = len(controller.name_to_ids.get(key, []))
-            is_selected = (getattr(ssbo_editor, "selected_name", None) == leaf_key)
+        children = node_data["children"]
+        is_selected = (getattr(ssbo_editor, "selected_name", None) == key)
+
+        if not children:
+            # Leaf node: selectable
+            label = f"{display} ({obj_count})##{key}"
             if imgui.selectable(label, is_selected)[0]:
-                ssbo_editor.select_node(leaf_key)
+                ssbo_editor.select_node(key)
                 if hasattr(self.app, "lui_manager"):
                     self.app.lui_manager.selected_index = -1
-            return
+        else:
-
+            # Branch node: tree node
-        # Non-leaf: tree node only for hierarchy display.
+            flags = imgui.TreeNodeFlags_.open_on_arrow
-        opened = imgui.tree_node(label)
+            if is_selected:
-        # Clicking non-leaf row selects its aggregate group so parent transform affects children.
+                flags |= imgui.TreeNodeFlags_.selected
-        if group_key and imgui.is_item_clicked(0):
+            label = f"{display} ({obj_count})##{key}"
-            ssbo_editor.select_node(group_key)
+            opened = imgui.tree_node_ex(label, flags)
-            if hasattr(self.app, "lui_manager"):
+            if imgui.is_item_clicked(0):
-                self.app.lui_manager.selected_index = -1
+                ssbo_editor.select_node(key)
-        if opened:
+                if hasattr(self.app, "lui_manager"):
-            # If this node is also a selectable leaf, render selectable entry first.
+                    self.app.lui_manager.selected_index = -1
-            if group_key:
+            if opened:
-                is_group_selected = (getattr(ssbo_editor, "selected_name", None) == group_key)
+                for child_key in children:
-                if imgui.selectable(f"[整体] {display}##group_{unique_id_prefix}_{path}", is_group_selected)[0]:
+                    self._draw_ssbo_virtual_tree_node(ssbo_editor, controller, child_key)
-                    ssbo_editor.select_node(group_key)
+                imgui.tree_pop()
-                    if hasattr(self.app, "lui_manager"):
-                        self.app.lui_manager.selected_index = -1
-            if leaf_key:
-                is_selected = (getattr(ssbo_editor, "selected_name", None) == leaf_key)
-                if imgui.selectable(f"[节点] {display}##leaf_{unique_id_prefix}_{path}", is_selected)[0]:
-                    ssbo_editor.select_node(leaf_key)
-                    if hasattr(self.app, "lui_manager"):
-                        self.app.lui_manager.selected_index = -1
-            for child_name in sorted(children.keys()):
-                self._draw_ssbo_virtual_tree_node(
-                    ssbo_editor,
-                    children[child_name],
-                    unique_id_prefix,
-                    depth + 1,
-                )
-            imgui.tree_pop()
     def _show_node_context_menu(self, node, name, node_type):
         """显示节点右键菜单"""
         self.app._context_menu_node = True
@@ -2211,7 +2231,10 @@ class EditorPanels:
         # 删除对象
         imgui.same_line()
         if imgui.button("删除"):
-            if hasattr(self, 'selection') and self.selection:
+            if hasattr(self.app, "_on_delete") and callable(self.app._on_delete):
+                self.app._on_delete()
+            elif hasattr(self, 'selection') and self.selection and hasattr(self.selection, "deleteSelectedNode"):
+                # 兼容旧选择系统接口
                 self.selection.deleteSelectedNode()
     
 
@@ -2439,4 +2462,3 @@ class EditorPanels:
         except Exception as e:
             print(f"绘制着色模型面板失败: {e}")
 
-

ui/panels/object_factory.py

@@ -339,51 +339,6 @@ class ObjectFactory:
         except Exception as e:
             print(f"✗ 创建平面失败: {e}")
             return None
-    
-
-    def create2DSamplePanel(self):
-        """创建2D示例面板"""
-        try:
-            from core.InfoPanelManager import createSampleInfoPanel
-            result = createSampleInfoPanel(self.render)
-            return result
-        except Exception as e:
-            print(f"创建2D示例面板失败: {e}")
-            return None
-    
-
-    def create3DSamplePanel(self):
-        """创建3D实例面板"""
-        try:
-            if hasattr(self, 'info_panel_manager') and self.info_panel_manager:
-                # 创建3D信息面板
-                panel_id = f"3d_sample_{int(time.time())}"
-                result = self.info_panel_manager.create3DInfoPanel(
-                    panel_id=panel_id,
-                    position=(0, 0, 2),
-                    size=(1.0, 0.6),
-                    bg_color=(0.15, 0.25, 0.35, 0.95),
-                    border_color=(0.3, 0.5, 0.7, 1.0),
-                    title_color=(0.7, 0.9, 1.0, 1.0),
-                    content_color=(0.95, 0.95, 0.95, 1.0)
-                )
-                
-                # 添加示例内容
-                if result:
-                    sample_data = {
-                        "标题": "3D信息面板",
-                        "状态": "运行中",
-                        "创建时间": time.strftime("%Y-%m-%d %H:%M:%S"),
-                        "位置": f"X:0, Y:0, Z:2"
-                    }
-                    self.info_panel_manager.updatePanelContent(panel_id, content=sample_data)
-                
-                return result
-            return None
-        except Exception as e:
-            print(f"创建3D示例面板失败: {e}")
-            return None
-    
 
     def createWebPanel(self, url="https://www.example.com"):
         """创建Web面板"""

ui/panels/panel_delegates.py

@@ -0,0 +1,753 @@
+"""Delegated panel/runtime methods for MyWorld."""
+
+
+class PanelDelegates:
+    def _draw_menu_bar(self):
+        self.editor_panels.draw_menu_bar()
+
+    def _draw_toolbar(self):
+        self.editor_panels.draw_toolbar()
+
+    def _draw_scene_tree(self, *args, **kwargs):
+        return self.editor_panels._draw_scene_tree(*args, **kwargs)
+
+    def _build_scene_tree(self, *args, **kwargs):
+        return self.editor_panels._build_scene_tree(*args, **kwargs)
+
+    def _draw_scene_node(self, *args, **kwargs):
+        return self.editor_panels._draw_scene_node(*args, **kwargs)
+
+    def _show_node_context_menu(self, *args, **kwargs):
+        return self.editor_panels._show_node_context_menu(*args, **kwargs)
+
+    def _draw_resource_manager(self, *args, **kwargs):
+        return self.editor_panels._draw_resource_manager(*args, **kwargs)
+
+    def _draw_property_panel(self, *args, **kwargs):
+        return self.editor_panels._draw_property_panel(*args, **kwargs)
+
+    def _draw_web_panel(self, *args, **kwargs):
+        return self.editor_panels._draw_web_panel(*args, **kwargs)
+
+    def _draw_node_properties(self, *args, **kwargs):
+        return self.editor_panels._draw_node_properties(*args, **kwargs)
+
+    def _getActor(self, *args, **kwargs):
+        return self.animation_tools._getActor(*args, **kwargs)
+
+    def _getModelFormat(self, *args, **kwargs):
+        return self.animation_tools._getModelFormat(*args, **kwargs)
+
+    def _processAnimationNames(self, *args, **kwargs):
+        return self.animation_tools._processAnimationNames(*args, **kwargs)
+
+    def _isLikelyBoneGroup(self, *args, **kwargs):
+        return self.animation_tools._isLikelyBoneGroup(*args, **kwargs)
+
+    def _analyzeAnimationQuality(self, *args, **kwargs):
+        return self.animation_tools._analyzeAnimationQuality(*args, **kwargs)
+
+    def _playAnimation(self, *args, **kwargs):
+        return self.animation_tools._playAnimation(*args, **kwargs)
+
+    def _pauseAnimation(self, *args, **kwargs):
+        return self.animation_tools._pauseAnimation(*args, **kwargs)
+
+    def _stopAnimation(self, *args, **kwargs):
+        return self.animation_tools._stopAnimation(*args, **kwargs)
+
+    def _loopAnimation(self, *args, **kwargs):
+        return self.animation_tools._loopAnimation(*args, **kwargs)
+
+    def _setAnimationSpeed(self, *args, **kwargs):
+        return self.animation_tools._setAnimationSpeed(*args, **kwargs)
+
+    def _clear_animation_cache(self, *args, **kwargs):
+        return self.animation_tools._clear_animation_cache(*args, **kwargs)
+    def _get_node_type_from_node(self, node):
+        """从节点判断其类型"""
+        # 检查是否为GUI元素
+        if hasattr(node, 'getPythonTag') and node.getPythonTag('gui_element'):
+            return "GUI元素"
+        
+        # 检查是否为光源
+        node_name = node.getName() or ""
+        if "light" in node_name.lower() or "Light" in node_name:
+            return "光源"
+        
+        # 检查是否为相机
+        if "camera" in node_name.lower() or "Camera" in node_name:
+            return "相机"
+        
+        # 检查是否为模型
+        if hasattr(self, 'scene_manager') and self.scene_manager:
+            if hasattr(self.scene_manager, 'models') and node in self.scene_manager.models:
+                return "模型"
+        
+        # 默认为几何体
+        return "几何体"
+    
+    def _draw_status_badges(self, *args, **kwargs):
+        return self.editor_panels._draw_status_badges(*args, **kwargs)
+
+    def _draw_transform_properties(self, *args, **kwargs):
+        return self.editor_panels._draw_transform_properties(*args, **kwargs)
+
+    def _draw_gui_properties(self, *args, **kwargs):
+        return self.editor_panels._draw_gui_properties(*args, **kwargs)
+
+    def _draw_light_properties(self, *args, **kwargs):
+        return self.editor_panels._draw_light_properties(*args, **kwargs)
+
+    def _draw_model_properties(self, *args, **kwargs):
+        return self.editor_panels._draw_model_properties(*args, **kwargs)
+
+    def _draw_animation_properties(self, *args, **kwargs):
+        return self.editor_panels._draw_animation_properties(*args, **kwargs)
+
+    def _draw_collision_properties(self, *args, **kwargs):
+        return self.editor_panels._draw_collision_properties(*args, **kwargs)
+
+    def _draw_shape_specific_parameters(self, *args, **kwargs):
+        return self.editor_panels._draw_shape_specific_parameters(*args, **kwargs)
+
+    def _draw_sphere_parameters(self, *args, **kwargs):
+        return self.editor_panels._draw_sphere_parameters(*args, **kwargs)
+
+    def _draw_box_parameters(self, *args, **kwargs):
+        return self.editor_panels._draw_box_parameters(*args, **kwargs)
+
+    def _draw_capsule_parameters(self, *args, **kwargs):
+        return self.editor_panels._draw_capsule_parameters(*args, **kwargs)
+
+    def _draw_plane_parameters(self, *args, **kwargs):
+        return self.editor_panels._draw_plane_parameters(*args, **kwargs)
+
+    def _draw_property_actions(self, *args, **kwargs):
+        return self.editor_panels._draw_property_actions(*args, **kwargs)
+
+    def _draw_appearance_properties(self, *args, **kwargs):
+        return self.editor_panels._draw_appearance_properties(*args, **kwargs)
+
+    def _draw_material_properties(self, *args, **kwargs):
+        return self.editor_panels._draw_material_properties(*args, **kwargs)
+
+    def _draw_shading_model_panel(self, *args, **kwargs):
+        return self.editor_panels._draw_shading_model_panel(*args, **kwargs)
+
+
+    def _apply_gui_font(self, *args, **kwargs):
+        return self.property_helpers._apply_gui_font(*args, **kwargs)
+
+    def _apply_gui_font_size(self, *args, **kwargs):
+        return self.property_helpers._apply_gui_font_size(*args, **kwargs)
+
+    def _apply_gui_font_style(self, *args, **kwargs):
+        return self.property_helpers._apply_gui_font_style(*args, **kwargs)
+
+    def _has_collision(self, *args, **kwargs):
+        return self.property_helpers._has_collision(*args, **kwargs)
+
+    def _get_current_collision_shape(self, *args, **kwargs):
+        return self.property_helpers._get_current_collision_shape(*args, **kwargs)
+
+    def _get_current_collision_shape_type(self, *args, **kwargs):
+        return self.property_helpers._get_current_collision_shape_type(*args, **kwargs)
+
+    def _get_collision_position_offset(self, *args, **kwargs):
+        return self.property_helpers._get_collision_position_offset(*args, **kwargs)
+
+    def _is_collision_visible(self, *args, **kwargs):
+        return self.property_helpers._is_collision_visible(*args, **kwargs)
+
+    def _add_collision_to_node(self, *args, **kwargs):
+        return self.property_helpers._add_collision_to_node(*args, **kwargs)
+
+    def _remove_collision_from_node(self, *args, **kwargs):
+        return self.property_helpers._remove_collision_from_node(*args, **kwargs)
+
+    def _toggle_collision_visibility(self, *args, **kwargs):
+        return self.property_helpers._toggle_collision_visibility(*args, **kwargs)
+
+    def _update_collision_position(self, *args, **kwargs):
+        return self.property_helpers._update_collision_position(*args, **kwargs)
+
+    def _get_shape_type_from_name(self, *args, **kwargs):
+        return self.property_helpers._get_shape_type_from_name(*args, **kwargs)
+
+    def _get_sphere_radius(self, *args, **kwargs):
+        return self.property_helpers._get_sphere_radius(*args, **kwargs)
+
+    def _update_sphere_radius(self, *args, **kwargs):
+        return self.property_helpers._update_sphere_radius(*args, **kwargs)
+
+    def _get_box_size(self, *args, **kwargs):
+        return self.property_helpers._get_box_size(*args, **kwargs)
+
+    def _update_box_size(self, *args, **kwargs):
+        return self.property_helpers._update_box_size(*args, **kwargs)
+
+    def _get_capsule_radius(self, *args, **kwargs):
+        return self.property_helpers._get_capsule_radius(*args, **kwargs)
+
+    def _update_capsule_radius(self, *args, **kwargs):
+        return self.property_helpers._update_capsule_radius(*args, **kwargs)
+
+    def _get_capsule_height(self, *args, **kwargs):
+        return self.property_helpers._get_capsule_height(*args, **kwargs)
+
+    def _update_capsule_height(self, *args, **kwargs):
+        return self.property_helpers._update_capsule_height(*args, **kwargs)
+
+    def _get_plane_normal(self, *args, **kwargs):
+        return self.property_helpers._get_plane_normal(*args, **kwargs)
+
+    def _update_plane_normal(self, *args, **kwargs):
+        return self.property_helpers._update_plane_normal(*args, **kwargs)
+
+    def _manual_collision_detection(self, *args, **kwargs):
+        return self.property_helpers._manual_collision_detection(*args, **kwargs)
+
+    def _update_node_name(self, *args, **kwargs):
+        return self.property_helpers._update_node_name(*args, **kwargs)
+
+    def _get_material_base_color(self, *args, **kwargs):
+        return self.property_helpers._get_material_base_color(*args, **kwargs)
+
+    def _update_material_base_color(self, *args, **kwargs):
+        return self.property_helpers._update_material_base_color(*args, **kwargs)
+
+    def _update_material_roughness(self, *args, **kwargs):
+        return self.property_helpers._update_material_roughness(*args, **kwargs)
+
+    def _update_material_metallic(self, *args, **kwargs):
+        return self.property_helpers._update_material_metallic(*args, **kwargs)
+
+    def _update_material_ior(self, *args, **kwargs):
+        return self.property_helpers._update_material_ior(*args, **kwargs)
+
+    def _apply_material_preset(self, *args, **kwargs):
+        return self.property_helpers._apply_material_preset(*args, **kwargs)
+
+    def _apply_material_to_node(self, *args, **kwargs):
+        return self.property_helpers._apply_material_to_node(*args, **kwargs)
+
+    def _reset_material(self, *args, **kwargs):
+        return self.property_helpers._reset_material(*args, **kwargs)
+
+    def _select_texture_for_material(self, *args, **kwargs):
+        return self.property_helpers._select_texture_for_material(*args, **kwargs)
+
+    def _apply_texture_to_material(self, *args, **kwargs):
+        return self.property_helpers._apply_texture_to_material(*args, **kwargs)
+
+    def _clear_all_textures(self, *args, **kwargs):
+        return self.property_helpers._clear_all_textures(*args, **kwargs)
+
+    def _display_current_textures(self, *args, **kwargs):
+        return self.property_helpers._display_current_textures(*args, **kwargs)
+
+    def _update_shading_model(self, *args, **kwargs):
+        return self.property_helpers._update_shading_model(*args, **kwargs)
+
+    def _update_transparency(self, *args, **kwargs):
+        return self.property_helpers._update_transparency(*args, **kwargs)
+
+    def _draw_texture_file_dialog(self, *args, **kwargs):
+        return self.property_helpers._draw_texture_file_dialog(*args, **kwargs)
+
+    def start_transform_monitoring(self, *args, **kwargs):
+        return self.property_helpers.start_transform_monitoring(*args, **kwargs)
+
+    def stop_transform_monitoring(self, *args, **kwargs):
+        return self.property_helpers.stop_transform_monitoring(*args, **kwargs)
+
+    def _update_last_transform_values(self, *args, **kwargs):
+        return self.property_helpers._update_last_transform_values(*args, **kwargs)
+
+    def _check_transform_changes(self, *args, **kwargs):
+        return self.property_helpers._check_transform_changes(*args, **kwargs)
+
+    def update_transform_monitoring(self, *args, **kwargs):
+        return self.property_helpers.update_transform_monitoring(*args, **kwargs)
+
+    def show_color_picker(self, *args, **kwargs):
+        return self.property_helpers.show_color_picker(*args, **kwargs)
+
+    def _draw_color_picker(self, *args, **kwargs):
+        return self.property_helpers._draw_color_picker(*args, **kwargs)
+
+    def _apply_color_selection(self, *args, **kwargs):
+        return self.property_helpers._apply_color_selection(*args, **kwargs)
+
+    def _draw_color_button(self, *args, **kwargs):
+        return self.property_helpers._draw_color_button(*args, **kwargs)
+
+    def _refresh_available_fonts(self, *args, **kwargs):
+        return self.property_helpers._refresh_available_fonts(*args, **kwargs)
+
+    def show_font_selector(self, *args, **kwargs):
+        return self.property_helpers.show_font_selector(*args, **kwargs)
+
+    def _draw_font_selector(self, *args, **kwargs):
+        return self.property_helpers._draw_font_selector(*args, **kwargs)
+
+    def _apply_font_selection(self, *args, **kwargs):
+        return self.property_helpers._apply_font_selection(*args, **kwargs)
+
+    def _draw_font_selector_button(self, *args, **kwargs):
+        return self.property_helpers._draw_font_selector_button(*args, **kwargs)
+    def _draw_console(self, *args, **kwargs):
+        return self.script_panels._draw_console(*args, **kwargs)
+
+    def _draw_script_panel(self, *args, **kwargs):
+        return self.script_panels._draw_script_panel(*args, **kwargs)
+
+    def _draw_script_status_group(self, *args, **kwargs):
+        return self.script_panels._draw_script_status_group(*args, **kwargs)
+
+    def _draw_create_script_group(self, *args, **kwargs):
+        return self.script_panels._draw_create_script_group(*args, **kwargs)
+
+    def _draw_available_scripts_group(self, *args, **kwargs):
+        return self.script_panels._draw_available_scripts_group(*args, **kwargs)
+
+    def _draw_script_mounting_group(self, *args, **kwargs):
+        return self.script_panels._draw_script_mounting_group(*args, **kwargs)
+
+
+    def _toggle_hot_reload(self, *args, **kwargs):
+        return self.app_actions._toggle_hot_reload(*args, **kwargs)
+
+    def _create_new_script(self, *args, **kwargs):
+        return self.app_actions._create_new_script(*args, **kwargs)
+
+    def _refresh_scripts_list(self, *args, **kwargs):
+        return self.app_actions._refresh_scripts_list(*args, **kwargs)
+
+    def _reload_all_scripts(self, *args, **kwargs):
+        return self.app_actions._reload_all_scripts(*args, **kwargs)
+
+    def _on_script_selected(self, *args, **kwargs):
+        return self.app_actions._on_script_selected(*args, **kwargs)
+
+    def _edit_script(self, *args, **kwargs):
+        return self.app_actions._edit_script(*args, **kwargs)
+
+    def _mount_script_to_selected(self, *args, **kwargs):
+        return self.app_actions._mount_script_to_selected(*args, **kwargs)
+
+    def _unmount_script_from_selected(self, *args, **kwargs):
+        return self.app_actions._unmount_script_from_selected(*args, **kwargs)
+
+    def _on_new_project(self, *args, **kwargs):
+        return self.app_actions._on_new_project(*args, **kwargs)
+
+    def _on_open_project(self, *args, **kwargs):
+        return self.app_actions._on_open_project(*args, **kwargs)
+
+    def _on_save_project(self, *args, **kwargs):
+        return self.app_actions._on_save_project(*args, **kwargs)
+
+    def _on_save_as_project(self, *args, **kwargs):
+        return self.app_actions._on_save_as_project(*args, **kwargs)
+
+    def _on_build_webgl_package(self, *args, **kwargs):
+        return self.app_actions._on_build_webgl_package(*args, **kwargs)
+
+    def _execute_webgl_package(self, *args, **kwargs):
+        return self.app_actions._execute_webgl_package(*args, **kwargs)
+
+    def _on_exit(self, *args, **kwargs):
+        return self.app_actions._on_exit(*args, **kwargs)
+
+    def _on_ctrl_pressed(self, *args, **kwargs):
+        return self.app_actions._on_ctrl_pressed(*args, **kwargs)
+
+    def _on_ctrl_released(self, *args, **kwargs):
+        return self.app_actions._on_ctrl_released(*args, **kwargs)
+
+    def _on_alt_pressed(self, *args, **kwargs):
+        return self.app_actions._on_alt_pressed(*args, **kwargs)
+
+    def _on_alt_released(self, *args, **kwargs):
+        return self.app_actions._on_alt_released(*args, **kwargs)
+
+    def _on_n_pressed(self, *args, **kwargs):
+        return self.app_actions._on_n_pressed(*args, **kwargs)
+
+    def _on_o_pressed(self, *args, **kwargs):
+        return self.app_actions._on_o_pressed(*args, **kwargs)
+
+    def _on_f4_pressed(self, *args, **kwargs):
+        return self.app_actions._on_f4_pressed(*args, **kwargs)
+
+    def _on_delete_pressed(self, *args, **kwargs):
+        return self.app_actions._on_delete_pressed(*args, **kwargs)
+
+    def _on_escape_pressed(self, *args, **kwargs):
+        return self.app_actions._on_escape_pressed(*args, **kwargs)
+
+    def _on_wheel_up(self, *args, **kwargs):
+        return self.app_actions._on_wheel_up(*args, **kwargs)
+
+    def _on_wheel_down(self, *args, **kwargs):
+        return self.app_actions._on_wheel_down(*args, **kwargs)
+
+    def _is_mouse_over_imgui(self, *args, **kwargs):
+        return self.app_actions._is_mouse_over_imgui(*args, **kwargs)
+
+    def processImGuiMouseClick(self, *args, **kwargs):
+        return self.app_actions.processImGuiMouseClick(*args, **kwargs)
+
+    def add_message(self, *args, **kwargs):
+        return self.app_actions.add_message(*args, **kwargs)
+
+    def add_success_message(self, *args, **kwargs):
+        return self.app_actions.add_success_message(*args, **kwargs)
+
+    def add_error_message(self, *args, **kwargs):
+        return self.app_actions.add_error_message(*args, **kwargs)
+
+    def add_warning_message(self, *args, **kwargs):
+        return self.app_actions.add_warning_message(*args, **kwargs)
+
+    def add_info_message(self, *args, **kwargs):
+        return self.app_actions.add_info_message(*args, **kwargs)
+
+    def _on_undo(self, *args, **kwargs):
+        return self.app_actions._on_undo(*args, **kwargs)
+
+    def _on_redo(self, *args, **kwargs):
+        return self.app_actions._on_redo(*args, **kwargs)
+
+    def _on_copy(self, *args, **kwargs):
+        return self.app_actions._on_copy(*args, **kwargs)
+
+    def _on_cut(self, *args, **kwargs):
+        return self.app_actions._on_cut(*args, **kwargs)
+
+    def _on_paste(self, *args, **kwargs):
+        return self.app_actions._on_paste(*args, **kwargs)
+
+    def _on_delete(self, *args, **kwargs):
+        return self.app_actions._on_delete(*args, **kwargs)
+
+    def _delete_node(self, *args, **kwargs):
+        return self.app_actions._delete_node(*args, **kwargs)
+
+    def _perform_node_cleanup(self, *args, **kwargs):
+        return self.app_actions._perform_node_cleanup(*args, **kwargs)
+
+    def _create_new_project(self, *args, **kwargs):
+        return self.app_actions._create_new_project(*args, **kwargs)
+
+    def _open_project_path(self, *args, **kwargs):
+        return self.app_actions._open_project_path(*args, **kwargs)
+
+    def _save_project_impl(self, *args, **kwargs):
+        return self.app_actions._save_project_impl(*args, **kwargs)
+
+    def _open_project_impl(self, *args, **kwargs):
+        return self.app_actions._open_project_impl(*args, **kwargs)
+
+    def _create_new_project_impl(self, *args, **kwargs):
+        return self.app_actions._create_new_project_impl(*args, **kwargs)
+
+    def _update_window_title(self, *args, **kwargs):
+        return self.app_actions._update_window_title(*args, **kwargs)
+
+    def _import_model_for_runtime(self, *args, **kwargs):
+        return self.app_actions._import_model_for_runtime(*args, **kwargs)
+
+    def _on_import_model(self, *args, **kwargs):
+        return self.app_actions._on_import_model(*args, **kwargs)
+
+    def _import_model(self, *args, **kwargs):
+        return self.app_actions._import_model(*args, **kwargs)
+
+    def _import_model_with_menu_logic(self, *args, **kwargs):
+        return self.app_actions._import_model_with_menu_logic(*args, **kwargs)
+
+    def _draw_new_project_dialog(self, *args, **kwargs):
+        return self.dialog_panels._draw_new_project_dialog(*args, **kwargs)
+
+    def _draw_open_project_dialog(self, *args, **kwargs):
+        return self.dialog_panels._draw_open_project_dialog(*args, **kwargs)
+
+    def _draw_path_browser(self, *args, **kwargs):
+        return self.dialog_panels._draw_path_browser(*args, **kwargs)
+
+    def _draw_import_dialog(self, *args, **kwargs):
+        return self.dialog_panels._draw_import_dialog(*args, **kwargs)
+
+    def _refresh_path_browser(self, *args, **kwargs):
+        return self.dialog_panels._refresh_path_browser(*args, **kwargs)
+
+    def _apply_selected_path(self, *args, **kwargs):
+        return self.dialog_panels._apply_selected_path(*args, **kwargs)
+
+    def _draw_spot_light_dialog(self, *args, **kwargs):
+        return self.dialog_panels._draw_spot_light_dialog(*args, **kwargs)
+
+    def _draw_point_light_dialog(self, *args, **kwargs):
+        return self.dialog_panels._draw_point_light_dialog(*args, **kwargs)
+
+    def _draw_terrain_dialog(self, *args, **kwargs):
+        return self.dialog_panels._draw_terrain_dialog(*args, **kwargs)
+
+    def _draw_script_dialog(self, *args, **kwargs):
+        return self.dialog_panels._draw_script_dialog(*args, **kwargs)
+
+    def _draw_script_browser(self, *args, **kwargs):
+        return self.dialog_panels._draw_script_browser(*args, **kwargs)
+
+    def _refresh_script_browser(self, *args, **kwargs):
+        return self.dialog_panels._refresh_script_browser(*args, **kwargs)
+
+    def _draw_heightmap_browser(self, *args, **kwargs):
+        return self.dialog_panels._draw_heightmap_browser(*args, **kwargs)
+
+    def _refresh_heightmap_browser(self, *args, **kwargs):
+        return self.dialog_panels._refresh_heightmap_browser(*args, **kwargs)
+
+    def setup_drag_drop_support(self):
+        """初始化拖拽支持。"""
+        return self.model_drag_drop.setup_drag_drop_support()
+
+    def _is_point_in_resource_manager(self, drop_pos):
+        """判断投放坐标是否命中资源管理器窗口。"""
+        return self.model_drag_drop.is_point_in_resource_manager(drop_pos)
+
+    def _resolve_resource_drop_target_dir(self, drop_pos):
+        """根据投放坐标解析资源管理器中的目标目录。"""
+        return self.model_drag_drop.resolve_resource_drop_target_dir(drop_pos)
+
+    def _process_external_drop_events(self):
+        """处理外部拖入（系统文件拖拽）事件。"""
+        return self.model_drag_drop.process_external_drop_events()
+
+    def _handle_external_drop(self, file_paths, drop_pos=None):
+        """把外部拖入文件分发到资源管理器或场景。"""
+        return self.model_drag_drop.handle_external_drop(file_paths, drop_pos)
+
+    def add_dragged_file(self, file_path):
+        """添加拖拽的文件"""
+        return self.model_drag_drop.add_dragged_file(file_path)
+    
+    def clear_dragged_files(self):
+        """清空拖拽文件列表"""
+        return self.model_drag_drop.clear_dragged_files()
+    
+    def process_dragged_files(self):
+        """处理拖拽的文件"""
+        return self.model_drag_drop.process_dragged_files()
+    
+    def _import_model_from_path(self, file_path):
+        """从路径导入模型的内部方法"""
+        return self.model_drag_drop.import_model_from_path(file_path)
+    
+    def _draw_drag_drop_interface(self, *args, **kwargs):
+        return self.interaction_panels._draw_drag_drop_interface(*args, **kwargs)
+
+    def _handle_drag_drop_completion(self, *args, **kwargs):
+        return self.interaction_panels._handle_drag_drop_completion(*args, **kwargs)
+
+    def _draw_drag_overlay(self, *args, **kwargs):
+        return self.interaction_panels._draw_drag_overlay(*args, **kwargs)
+
+    def _draw_drag_status(self, *args, **kwargs):
+        return self.interaction_panels._draw_drag_status(*args, **kwargs)
+
+    def _draw_context_menus(self, *args, **kwargs):
+        return self.interaction_panels._draw_context_menus(*args, **kwargs)
+
+    def _delete_node_simple(self, *args, **kwargs):
+        return self.interaction_panels._delete_node_simple(*args, **kwargs)
+
+    def _copy_node(self, *args, **kwargs):
+        return self.interaction_panels._copy_node(*args, **kwargs)
+
+
+    def _on_create_empty_object(self, *args, **kwargs):
+        return self.create_actions._on_create_empty_object(*args, **kwargs)
+
+    def _on_create_cube(self, *args, **kwargs):
+        return self.create_actions._on_create_cube(*args, **kwargs)
+
+    def _on_create_sphere(self, *args, **kwargs):
+        return self.create_actions._on_create_sphere(*args, **kwargs)
+
+    def _on_create_cylinder(self, *args, **kwargs):
+        return self.create_actions._on_create_cylinder(*args, **kwargs)
+
+    def _on_create_plane(self, *args, **kwargs):
+        return self.create_actions._on_create_plane(*args, **kwargs)
+
+    def _on_create_3d_text(self, *args, **kwargs):
+        return self.create_actions._on_create_3d_text(*args, **kwargs)
+
+    def _on_create_3d_image(self, *args, **kwargs):
+        return self.create_actions._on_create_3d_image(*args, **kwargs)
+
+    def _on_create_gui_button(self, *args, **kwargs):
+        return self.create_actions._on_create_gui_button(*args, **kwargs)
+
+    def _on_create_gui_label(self, *args, **kwargs):
+        return self.create_actions._on_create_gui_label(*args, **kwargs)
+
+    def _on_create_gui_entry(self, *args, **kwargs):
+        return self.create_actions._on_create_gui_entry(*args, **kwargs)
+
+    def _on_create_gui_image(self, *args, **kwargs):
+        return self.create_actions._on_create_gui_image(*args, **kwargs)
+
+    def _on_create_video_screen(self, *args, **kwargs):
+        return self.create_actions._on_create_video_screen(*args, **kwargs)
+
+    def _on_create_2d_video_screen(self, *args, **kwargs):
+        return self.create_actions._on_create_2d_video_screen(*args, **kwargs)
+
+    def _on_create_spherical_video(self, *args, **kwargs):
+        return self.create_actions._on_create_spherical_video(*args, **kwargs)
+
+    def _on_create_virtual_screen(self, *args, **kwargs):
+        return self.create_actions._on_create_virtual_screen(*args, **kwargs)
+
+    def _on_create_spot_light(self, *args, **kwargs):
+        return self.create_actions._on_create_spot_light(*args, **kwargs)
+
+    def _on_create_point_light(self, *args, **kwargs):
+        return self.create_actions._on_create_point_light(*args, **kwargs)
+
+    def _on_create_flat_terrain(self, *args, **kwargs):
+        return self.create_actions._on_create_flat_terrain(*args, **kwargs)
+
+    def _on_create_heightmap_terrain(self, *args, **kwargs):
+        return self.create_actions._on_create_heightmap_terrain(*args, **kwargs)
+
+    def _on_create_script(self, *args, **kwargs):
+        return self.create_actions._on_create_script(*args, **kwargs)
+
+    def _on_load_script(self, *args, **kwargs):
+        return self.create_actions._on_load_script(*args, **kwargs)
+
+    def _on_reload_all_scripts(self, *args, **kwargs):
+        return self.create_actions._on_reload_all_scripts(*args, **kwargs)
+
+    def _on_open_scripts_manager(self, *args, **kwargs):
+        return self.create_actions._on_open_scripts_manager(*args, **kwargs)
+
+    def _on_create_2d_sample_panel(self, *args, **kwargs):
+        return self.create_actions._on_create_2d_sample_panel(*args, **kwargs)
+
+    def _on_create_3d_sample_panel(self, *args, **kwargs):
+        return self.create_actions._on_create_3d_sample_panel(*args, **kwargs)
+
+    def _on_create_web_panel(self, *args, **kwargs):
+        return self.create_actions._on_create_web_panel(*args, **kwargs)
+
+
+    def createEmptyObject(self, *args, **kwargs):
+        return self.object_factory.createEmptyObject(*args, **kwargs)
+
+    def create3DText(self, *args, **kwargs):
+        return self.object_factory.create3DText(*args, **kwargs)
+
+    def create3DImage(self, *args, **kwargs):
+        return self.object_factory.create3DImage(*args, **kwargs)
+
+    def createCube(self, *args, **kwargs):
+        return self.object_factory.createCube(*args, **kwargs)
+
+    def createSphere(self, *args, **kwargs):
+        return self.object_factory.createSphere(*args, **kwargs)
+
+    def createCylinder(self, *args, **kwargs):
+        return self.object_factory.createCylinder(*args, **kwargs)
+
+    def createPlane(self, *args, **kwargs):
+        return self.object_factory.createPlane(*args, **kwargs)
+
+    def create2DSamplePanel(self, *args, **kwargs):
+        return self.object_factory.create2DSamplePanel(*args, **kwargs)
+
+    def create3DSamplePanel(self, *args, **kwargs):
+        return self.object_factory.create3DSamplePanel(*args, **kwargs)
+
+    def createWebPanel(self, *args, **kwargs):
+        return self.object_factory.createWebPanel(*args, **kwargs)
+    def createGUIButton(self, pos=(0, 0, 0), text="按钮", size=0.1):
+        return self.runtime_actions.createGUIButton(pos, text, size)
+
+    def _create_simple_gui_button(self, pos=(0, 0, 0), text="按钮", size=0.1):
+        return self.runtime_actions._create_simple_gui_button(pos, text, size)
+
+    def _on_gui_button_click(self):
+        return self.runtime_actions._on_gui_button_click()
+
+    def _toggle_vr_mode(self):
+        return self.runtime_actions._toggle_vr_mode()
+
+    def _exit_vr_mode(self):
+        return self.runtime_actions._exit_vr_mode()
+
+    def _show_vr_status(self):
+        return self.runtime_actions._show_vr_status()
+
+    def _show_vr_settings(self):
+        return self.runtime_actions._show_vr_settings()
+
+    def _show_vr_performance_report(self):
+        return self.runtime_actions._show_vr_performance_report()
+
+    def _get_chinese_font(self):
+        return self.runtime_actions._get_chinese_font()
+
+    def createGUILabel(self, pos=(0, 0, 0), text="标签", size=0.08):
+        return self.runtime_actions.createGUILabel(pos, text, size)
+
+    def _create_simple_gui_label(self, pos=(0, 0, 0), text="标签", size=0.08):
+        return self.runtime_actions._create_simple_gui_label(pos, text, size)
+
+    def createGUIEntry(self, pos=(0, 0, 0), placeholder="输入文本...", size=0.08):
+        return self.runtime_actions.createGUIEntry(pos, placeholder, size)
+
+    def _create_simple_gui_entry(self, pos=(0, 0, 0), placeholder="输入文本...", size=0.08):
+        return self.runtime_actions._create_simple_gui_entry(pos, placeholder, size)
+
+    def createGUIImage(self, pos=(0, 0, 0), image_path=None, size=2):
+        return self.runtime_actions.createGUIImage(pos, image_path, size)
+
+    def _create_simple_gui_image(self, pos=(0, 0, 0), image_path=None, size=2):
+        return self.runtime_actions._create_simple_gui_image(pos, image_path, size)
+
+    def createVideoScreen(self, pos=(0, 0, 0), size=1, video_path=None):
+        return self.runtime_actions.createVideoScreen(pos, size, video_path)
+
+    def create2DVideoScreen(self, pos=(0, 0, 0), size=0.2, video_path=None):
+        return self.runtime_actions.create2DVideoScreen(pos, size, video_path)
+
+    def createSphericalVideo(self, pos=(0, 0, 0), radius=5.0, video_path=None):
+        return self.runtime_actions.createSphericalVideo(pos, radius, video_path)
+
+    def createVirtualScreen(self, pos=(0, 0, 0), size=(2, 1), text="虚拟屏幕"):
+        return self.runtime_actions.createVirtualScreen(pos, size, text)
+
+    def createSpotLight(self, pos=(0, 0, 5)):
+        return self.runtime_actions.createSpotLight(pos)
+
+    def createPointLight(self, pos=(0, 0, 5)):
+        return self.runtime_actions.createPointLight(pos)
+
+    def createFlatTerrain(self, size=(10, 10), resolution=129):
+        return self.runtime_actions.createFlatTerrain(size, resolution)
+
+    def createTerrainFromHeightMap(self, heightmap_path, scale=(1.0, 1.0, 10.0)):
+        return self.runtime_actions.createTerrainFromHeightMap(heightmap_path, scale)
+
+    def createScript(self, script_name, template="basic"):
+        return self.runtime_actions.createScript(script_name, template)
+
+    def loadScript(self, script_path):
+        return self.runtime_actions.loadScript(script_path)

ui/panels/runtime_actions.py

@@ -0,0 +1,396 @@
+import os
+from pathlib import Path
+
+
+class RuntimeActions:
+    """Runtime-facing creation and utility actions extracted from main world."""
+
+    def __init__(self, app):
+        self.app = app
+
+    def __getattr__(self, name):
+        return getattr(self.app, name)
+
+    def __setattr__(self, name, value):
+        if name == "app" or name in self.__dict__ or hasattr(type(self), name):
+            object.__setattr__(self, name, value)
+        else:
+            setattr(self.app, name, value)
+
+    def createGUIButton(self, pos=(0, 0, 0), text="按钮", size=0.1):
+        """创建2D GUI按钮"""
+        try:
+            if hasattr(self, "gui_manager") and self.gui_manager:
+                return self._create_simple_gui_button(pos, text, size)
+            return None
+        except Exception as e:
+            print(f"创建GUI按钮失败: {e}")
+            return None
+
+    def _create_simple_gui_button(self, pos=(0, 0, 0), text="按钮", size=0.1):
+        """创建简单的GUI按钮，不依赖QT树形控件"""
+        try:
+            from direct.gui.DirectGui import DirectButton
+
+            gui_pos = (pos[0] * 0.1, 0, pos[2] * 0.1)
+            font = self._get_chinese_font()
+
+            if isinstance(size, (list, tuple)) and len(size) >= 2:
+                scale_value = size[0]
+            else:
+                scale_value = size
+
+            button = DirectButton(
+                text=text,
+                pos=gui_pos,
+                scale=scale_value,
+                text_font=font,
+                command=self._on_gui_button_click
+            )
+
+            button_wrapper = type("GUIElement", (), {})()
+            button_wrapper.node = button
+            button_wrapper.name = text
+            button_wrapper.gui_type = "GUI_BUTTON"
+            button_wrapper.position = pos
+            button_wrapper.size = size
+
+            self.gui_manager.gui_elements.append(button_wrapper)
+
+            print(f"✓ GUI按钮创建成功: {text}")
+            return button_wrapper
+        except Exception as e:
+            print(f"✗ 创建简单GUI按钮失败: {e}")
+            return None
+
+    def _on_gui_button_click(self):
+        """GUI按钮点击事件处理"""
+        print("GUI按钮被点击了")
+
+    def _toggle_vr_mode(self):
+        """切换VR模式"""
+        if self.vr_manager:
+            if self.vr_manager.is_enabled():
+                self._exit_vr_mode()
+            else:
+                self.vr_manager.enable()
+                self.add_info_message("已进入VR模式")
+        else:
+            self.add_error_message("VR管理器未初始化")
+
+    def _exit_vr_mode(self):
+        """退出VR模式"""
+        if self.vr_manager:
+            self.vr_manager.disable()
+            self.add_info_message("已退出VR模式")
+
+    def _show_vr_status(self):
+        """显示VR状态"""
+        if self.vr_manager:
+            status = "已启用" if self.vr_manager.is_enabled() else "未启用"
+            self.add_info_message(f"VR状态: {status}")
+
+            if self.vr_manager.is_enabled():
+                devices = self.vr_manager.get_connected_devices()
+                if devices:
+                    self.add_info_message(f"连接的设备: {', '.join(devices)}")
+                else:
+                    self.add_info_message("未检测到VR设备")
+        else:
+            self.add_error_message("VR管理器未初始化")
+
+    def _show_vr_settings(self):
+        """显示VR设置"""
+        if self.vr_manager:
+            self.add_info_message("VR设置对话框待实现")
+        else:
+            self.add_error_message("VR管理器未初始化")
+
+    def _show_vr_performance_report(self):
+        """显示VR性能报告"""
+        if self.vr_manager and self.vr_manager.is_enabled():
+            report = self.vr_manager.get_performance_report()
+            self.add_info_message(f"VR性能报告: {report}")
+        else:
+            self.add_info_message("VR未启用或管理器未初始化")
+
+    def _get_chinese_font(self):
+        """获取中文字体"""
+        try:
+            from panda3d.core import TextNode
+
+            font_paths = [
+                "/usr/share/fonts/truetype/wqy/wqy-microhei.ttc",
+                "/usr/share/fonts/opentype/noto/NotoSansCJK-Regular.ttc",
+                "/System/Library/Fonts/PingFang.ttc",
+                "C:/Windows/Fonts/simhei.ttf",
+                "C:/Windows/Fonts/msyh.ttc",
+            ]
+
+            for font_path in font_paths:
+                if Path(font_path).exists():
+                    try:
+                        font = self.loader.loadFont(font_path)
+                        print(f"✓ 为GUI加载中文字体成功: {font_path}")
+                        return font
+                    except Exception:
+                        print(f"⚠️ 字体加载失败，尝试下一个: {font_path}")
+                        continue
+
+            print("⚠️ 无法加载中文字体，使用默认字体")
+            return TextNode.getDefaultFont()
+
+        except Exception as e:
+            print(f"⚠️ 获取中文字体失败: {e}")
+            from panda3d.core import TextNode
+            return TextNode.getDefaultFont()
+
+    def createGUILabel(self, pos=(0, 0, 0), text="标签", size=0.08):
+        """创建2D GUI标签"""
+        try:
+            if hasattr(self, "gui_manager") and self.gui_manager:
+                return self._create_simple_gui_label(pos, text, size)
+            return None
+        except Exception as e:
+            print(f"创建GUI标签失败: {e}")
+            return None
+
+    def _create_simple_gui_label(self, pos=(0, 0, 0), text="标签", size=0.08):
+        """创建简单的GUI标签，不依赖QT树形控件"""
+        try:
+            from direct.gui.DirectGui import DirectLabel
+
+            gui_pos = (pos[0] * 0.1, 0, pos[2] * 0.1)
+            font = self._get_chinese_font()
+
+            if isinstance(size, (list, tuple)) and len(size) >= 2:
+                scale_value = size[0]
+            else:
+                scale_value = size
+
+            label = DirectLabel(
+                text=text,
+                pos=gui_pos,
+                scale=scale_value,
+                text_font=font,
+                text_fg=(1, 1, 1, 1)
+            )
+
+            label_wrapper = type("GUIElement", (), {})()
+            label_wrapper.node = label
+            label_wrapper.name = text
+            label_wrapper.gui_type = "GUI_LABEL"
+            label_wrapper.position = pos
+            label_wrapper.size = size
+
+            self.gui_manager.gui_elements.append(label_wrapper)
+
+            print(f"✓ GUI标签创建成功: {text}")
+            return label_wrapper
+        except Exception as e:
+            print(f"✗ 创建简单GUI标签失败: {e}")
+            return None
+
+    def createGUIEntry(self, pos=(0, 0, 0), placeholder="输入文本...", size=0.08):
+        """创建2D GUI文本输入框"""
+        try:
+            if hasattr(self, "gui_manager") and self.gui_manager:
+                return self._create_simple_gui_entry(pos, placeholder, size)
+            return None
+        except Exception as e:
+            print(f"创建GUI输入框失败: {e}")
+            return None
+
+    def _create_simple_gui_entry(self, pos=(0, 0, 0), placeholder="输入文本...", size=0.08):
+        """创建简单的GUI输入框，不依赖QT树形控件"""
+        try:
+            from direct.gui.DirectGui import DirectEntry
+
+            gui_pos = (pos[0] * 0.1, 0, pos[2] * 0.1)
+            font = self._get_chinese_font()
+
+            if isinstance(size, (list, tuple)) and len(size) >= 2:
+                scale_value = size[0]
+            else:
+                scale_value = size
+
+            entry = DirectEntry(
+                text=placeholder,
+                pos=gui_pos,
+                scale=scale_value,
+                text_font=font,
+                width=20,
+                numLines=1,
+                focus=1
+            )
+
+            entry_wrapper = type("GUIElement", (), {})()
+            entry_wrapper.node = entry
+            entry_wrapper.name = placeholder
+            entry_wrapper.gui_type = "GUI_ENTRY"
+            entry_wrapper.position = pos
+            entry_wrapper.size = size
+
+            self.gui_manager.gui_elements.append(entry_wrapper)
+
+            print(f"✓ GUI输入框创建成功: {placeholder}")
+            return entry_wrapper
+        except Exception as e:
+            print(f"✗ 创建简单GUI输入框失败: {e}")
+            return None
+
+    def createGUIImage(self, pos=(0, 0, 0), image_path=None, size=2):
+        """创建2D GUI图片"""
+        try:
+            if hasattr(self, "gui_manager") and self.gui_manager:
+                return self._create_simple_gui_image(pos, image_path, size)
+            return None
+        except Exception as e:
+            print(f"创建GUI图片失败: {e}")
+            return None
+
+    def _create_simple_gui_image(self, pos=(0, 0, 0), image_path=None, size=2):
+        """创建简单的GUI图片，不依赖QT树形控件"""
+        try:
+            from direct.gui.DirectGui import DirectFrame
+            from panda3d.core import Filename
+
+            gui_pos = (pos[0] * 0.1, 0, pos[2] * 0.1)
+
+            if isinstance(size, (list, tuple)) and len(size) >= 2:
+                scale_value = size[0]
+            else:
+                scale_value = size
+
+            if image_path and os.path.exists(image_path):
+                tex = self.loader.loadTexture(Filename.fromOsSpecific(image_path))
+                image = DirectFrame(
+                    image=tex,
+                    pos=gui_pos,
+                    scale=scale_value
+                )
+                image_name = os.path.basename(image_path)
+            else:
+                image = DirectFrame(
+                    frameColor=(0.5, 0.5, 0.5, 1.0),
+                    frameSize=(-scale_value, scale_value, -scale_value, scale_value),
+                    pos=gui_pos
+                )
+                image_name = "占位符图片"
+
+            image_wrapper = type("GUIElement", (), {})()
+            image_wrapper.node = image
+            image_wrapper.name = image_name
+            image_wrapper.gui_type = "GUI_IMAGE"
+            image_wrapper.position = pos
+            image_wrapper.size = size
+            image_wrapper.image_path = image_path
+
+            self.gui_manager.gui_elements.append(image_wrapper)
+
+            print(f"✓ GUI图片创建成功: {image_name}")
+            return image_wrapper
+        except Exception as e:
+            print(f"✗ 创建简单GUI图片失败: {e}")
+            return None
+
+    def createVideoScreen(self, pos=(0, 0, 0), size=1, video_path=None):
+        """创建视频屏幕"""
+        try:
+            if hasattr(self, "gui_manager") and self.gui_manager:
+                return self.gui_manager.createVideoScreen(pos, size, video_path)
+            return None
+        except Exception as e:
+            print(f"创建视频屏幕失败: {e}")
+            return None
+
+    def create2DVideoScreen(self, pos=(0, 0, 0), size=0.2, video_path=None):
+        """创建2D视频屏幕"""
+        try:
+            if hasattr(self, "gui_manager") and self.gui_manager:
+                return self.gui_manager.createGUI2DVideoScreen(pos, size, video_path)
+            return None
+        except Exception as e:
+            print(f"创建2D视频屏幕失败: {e}")
+            return None
+
+    def createSphericalVideo(self, pos=(0, 0, 0), radius=5.0, video_path=None):
+        """创建360度视频"""
+        try:
+            if hasattr(self, "gui_manager") and self.gui_manager:
+                return self.gui_manager.createSphericalVideo(pos, radius, video_path)
+            return None
+        except Exception as e:
+            print(f"创建球形视频失败: {e}")
+            return None
+
+    def createVirtualScreen(self, pos=(0, 0, 0), size=(2, 1), text="虚拟屏幕"):
+        """创建虚拟屏幕"""
+        try:
+            if hasattr(self, "gui_manager") and self.gui_manager:
+                return self.gui_manager.createGUIVirtualScreen(pos, size, text)
+            return None
+        except Exception as e:
+            print(f"创建虚拟屏幕失败: {e}")
+            return None
+
+    def createSpotLight(self, pos=(0, 0, 5)):
+        """创建聚光灯"""
+        try:
+            if hasattr(self, "scene_manager") and self.scene_manager:
+                return self.scene_manager.createSpotLight(pos)
+            return None
+        except Exception as e:
+            print(f"创建聚光灯失败: {e}")
+            return None
+
+    def createPointLight(self, pos=(0, 0, 5)):
+        """创建点光源"""
+        try:
+            if hasattr(self, "scene_manager") and self.scene_manager:
+                return self.scene_manager.createPointLight(pos)
+            return None
+        except Exception as e:
+            print(f"创建点光源失败: {e}")
+            return None
+
+    def createFlatTerrain(self, size=(10, 10), resolution=129):
+        """创建平面地形"""
+        try:
+            if hasattr(self, "terrain_manager") and self.terrain_manager:
+                return self.terrain_manager.createFlatTerrain(size, resolution)
+            return None
+        except Exception as e:
+            print(f"创建平面地形失败: {e}")
+            return None
+
+    def createTerrainFromHeightMap(self, heightmap_path, scale=(1.0, 1.0, 10.0)):
+        """从高度图创建地形"""
+        try:
+            if hasattr(self, "terrain_manager") and self.terrain_manager:
+                return self.terrain_manager.createTerrainFromHeightMap(heightmap_path, scale)
+            return None
+        except Exception as e:
+            print(f"创建高度图地形失败: {e}")
+            return None
+
+    def createScript(self, script_name, template="basic"):
+        """创建脚本"""
+        try:
+            if hasattr(self, "script_manager") and self.script_manager:
+                return self.script_manager.createScript(script_name, template)
+            return None
+        except Exception as e:
+            print(f"创建脚本失败: {e}")
+            return None
+
+    def loadScript(self, script_path):
+        """加载脚本"""
+        try:
+            if hasattr(self, "script_manager") and self.script_manager:
+                return self.script_manager.loadScript(script_path)
+            return None
+        except Exception as e:
+            print(f"加载脚本失败: {e}")
+            return None
+