[Bug] StoryDiffusion_Apply 节点在 infer_mode="story" 时报错：UnboundLocalError: local variable 'edit_mode' referenced before assignment #181
Description
老板您好，我在使用您的 StoryDiffusion 节点时遇到了一个无法解决的运行时错误。
问题描述：
在使用时，每次运行到StoryDiffusion_Apply节点，会立刻报错 UnboundLocalError: local variable 'edit_mode' referenced before assignment。节点安装后在界面上显示为6节点版本，加载正常，但在执行时失败；我直接用git命令或者在manager中下载会直接拉取到以前的那两个节点而不是这6个节点，而下载code的zip安装包后再上传至服务器后则是这6个节点，但是会报我刚刚说的这个错误。
我的环境
服务器 (Server): 云服务器
GPU: Tesla V100
操作系统 (OS): Ubuntu (Linux)
Python 环境: Conda 3.10.11
ComfyUI 版本: 0.3.56
我已尝试过的解决步骤
- 修复网络问题：尝试通过修改
hosts文件和设置git全局代理来解决国内网络连接问题。 - 验证核心依赖：通过
git clone获取完整的IP-Adapter源码，并在服务器上使用pip install .进行标准安装，确保依赖完整且版本正确。 - 清理缓存：多次清除了所有
__pycache__文件夹，排除了缓存污染的可能。 - 创建全新Python环境：创建了一个全新的、隔离的 Conda 环境，并在这个干净的环境里从零开始重新安装了 PyTorch、ComfyUI 和所有相关节点及其依赖。
然而，在完成了以上所有步骤后，同样的错误依然稳定复现。**
日志与截图：
以下是 ComfyUI 中显示的错误截图和日志回溯，希望您能抽空查看一下看是否遇到过此类问题。
错误日志文本 :
ComfyUI Error Report
Error Details
- Node ID: 3
- Node Type: StoryDiffusion_Apply
- Exception Type: UnboundLocalError
- Exception Message: local variable 'edit_mode' referenced before assignment
Stack Trace
File "/root/ComfyUI/execution.py", line 496, in execute
output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
File "/root/ComfyUI/execution.py", line 315, in get_output_data
return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
File "/root/ComfyUI/execution.py", line 289, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/root/ComfyUI/execution.py", line 277, in process_inputs
result = f(**inputs)
File "/root/ComfyUI/custom_nodes/ComfyUI_StoryDiffusion/Storydiffusion_node.py", line 303, in main_apply
"vision_model_config_ar":vision_model_config_ar,"no_dif_quantization":no_dif_quantization,"edit_mode":edit_mode,
System Information
- ComfyUI Version: 0.3.56
- Arguments: main.py --listen --port=6889 --disable-auto-launch --preview-method auto
- OS: posix
- Python Version: 3.10.18 | packaged by conda-forge | (main, Jun 4 2025, 14:45:41) [GCC 13.3.0]
- Embedded Python: false
- PyTorch Version: 2.8.0+cu128
Devices
- Name: cuda:0 Tesla V100-SXM2-32GB : cudaMallocAsync
- Type: cuda
- VRAM Total: 34079637504
- VRAM Free: 33755889664
- Torch VRAM Total: 0
- Torch VRAM Free: 0
2025-09-01T13:46:23.500012 - got prompt
2025-09-01T13:46:23.505484 - infer model is story2025-09-01T13:46:23.505665 -
2025-09-01T13:46:23.505788 - total_vram is 32500.875,aggressive_offload is False,offload is True
2025-09-01T13:46:25.709970 - FETCH ComfyRegistry Data: 25/962025-09-01T13:46:25.710189 -
2025-09-01T13:46:34.851020 - FETCH ComfyRegistry Data: 30/962025-09-01T13:46:34.851239 -
2025-09-01T13:46:42.986003 - FETCH ComfyRegistry Data: 35/962025-09-01T13:46:42.986324 -
2025-09-01T13:46:43.918231 -
Loading pipeline components...: 50%|██████████████████████████████████████████████████████████████████████████ | 2/4 [00:00<00:00, 16.24it/s]2025-09-01T13:46:43.923287 -
Loading pipeline components...: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:00<00:00, 31.23it/s]2025-09-01T13:46:43.923510 -
2025-09-01T13:46:43.962409 - !!! Exception during processing !!! local variable 'edit_mode' referenced before assignment
2025-09-01T13:46:43.962858 - Traceback (most recent call last):
File "/root/ComfyUI/execution.py", line 496, in execute
output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
File "/root/ComfyUI/execution.py", line 315, in get_output_data
return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
File "/root/ComfyUI/execution.py", line 289, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/root/ComfyUI/execution.py", line 277, in process_inputs
result = f(**inputs)
File "/root/ComfyUI/custom_nodes/ComfyUI_StoryDiffusion/Storydiffusion_node.py", line 303, in main_apply
"vision_model_config_ar":vision_model_config_ar,"no_dif_quantization":no_dif_quantization,"edit_mode":edit_mode,
UnboundLocalError: local variable 'edit_mode' referenced before assignment
2025-09-01T13:46:43.964062 - Prompt executed in 20.46 seconds