binary-husky/gpt_academic · error · RuntimeError

用户取消了程序。

Error message

用户取消了程序。

What it means

Watchdog RuntimeError inside bridge_openrouter.predict_no_ui_long_connection: observe_window[1] holds the caller's last heartbeat timestamp, and if it is older than watch_dog_patience seconds the loop aborts with '用户取消了程序。' (user cancelled). It is a cooperative-cancellation mechanism - any caller that stops feeding the dog causes it, intentional or not.

Source

Thrown at request_llms/bridge_openrouter.py:199

                raise RuntimeError("OpenAI拒绝了请求：" + error_msg)
        if ('data: [DONE]' in chunk_decoded): break # api2d 正常完成
        # 提前读取一些信息 （用于判断异常）
        json_data = chunkjson['choices'][0]
        delta = json_data["delta"]
        if len(delta) == 0: break
        if (not has_content) and has_role: continue
        if (not has_content) and (not has_role): continue # raise RuntimeError("发现不标准的第三方接口："+delta)
        if has_content: # has_role = True/False
            result += delta["content"]
            if not console_silence: print(delta["content"], end='')
            if observe_window is not None:
                # 观测窗，把已经获取的数据显示出去
                if len(observe_window) >= 1:
                    observe_window[0] += delta["content"]
                # 看门狗，如果超过期限没有喂狗，则终止
                if len(observe_window) >= 2:
                    if (time.time()-observe_window[1]) > watch_dog_patience:
                        raise RuntimeError("用户取消了程序。")
        else: raise RuntimeError("意外Json结构："+delta)
    if json_data and json_data['finish_reason'] == 'content_filter':
        raise RuntimeError("由于提问含不合规内容被Azure过滤。")
    if json_data and json_data['finish_reason'] == 'length':
        raise ConnectionAbortedError("正常结束，但显示Token不足，导致输出不完整，请削减单次输入的文本量。")
    return result


def predict(inputs:str, llm_kwargs:dict, plugin_kwargs:dict, chatbot:ChatBotWithCookies,
            history:list=[], system_prompt:str='', stream:bool=True, additional_fn:str=None):
    """
    发送至chatGPT，流式获取输出。
    用于基础的对话功能。
    inputs 是本次问询的输入
    top_p, temperature是chatGPT的内部调优参数
    history 是之前的对话列表（注意无论是inputs还是history，内容太长了都会触发token数量溢出的错误）
    chatbot 为WebUI中显示的对话列表，修改它，然后yield出去，可以直接修改对话界面内容
    additional_fn代表点击的哪个按钮，按钮见functional.py

View on GitHub (pinned to d6bde0fa54)

Solutions

1. If the user cancelled, treat this as normal termination and discard partial state
2. Ensure the consumer thread updates observe_window[1] = time.time() each poll iteration
3. Increase watch_dog_patience for slow providers
4. Wrap the call in try/except RuntimeError and distinguish cancellation from provider errors before retrying

Example fix

# consumer contract
import threading, time

def feed(ow):
    while streaming[0]:
        ow[1] = time.time()
        time.sleep(1)

import time
assert len(observe_window) >= 2, 'watchdog requires observe_window[1] heartbeat'
ow[1] = time.time()  # prime the dog before first chunk

try:
    result = predict_no_ui_long_connection(...)
except RuntimeError as e:
    if str(e) == '用户取消了程序。':
        cancelled.set(); return ow[0]
    raise

When it happens

Trigger: Caller stops refreshing observe_window[1] (user pressed Stop, or the monitoring thread died); network stall so no delta arrives while the watchdog compares against an unfed timestamp; plugin using predict_no_ui_long_connection without a live feeder thread.

Common situations: Gradio Stop button during a long OpenRouter generation; plugin author forgetting the feeder-thread contract; slow upstream where chunks pause longer than the patience value.

Related errors

• 程序终止。
• 程序终止。
• 用户取消了程序。
• 用户取消了程序。
• 用户取消了程序。

AI-assisted analysis of binary-husky/gpt_academic@d6bde0fa54 (2026-08-14). Data as JSON: /api/errors/0fd2fd2940a61efd. Report an issue: GitHub.