重构了一下

function_calls/llm.py

@@ -49,10 +49,12 @@ class FunctionCallLLM:
             self.logger.error("无可用的AI模型")
             return LLMRunResult(handled=False, error="no_model")
 
+        if not hasattr(chat_model, "call_with_functions"):
+            self.logger.error("当前模型不支持函数调用接口，请配置支持 function calling 的模型")
+            return LLMRunResult(handled=False, error="no_function_call_support")
+
         try:
-            if hasattr(chat_model, "call_with_functions"):
-                return self._run_native_loop(ctx, chat_model, functions, executor, formatter)
-            return self._run_prompt_loop(ctx, chat_model, functions, executor)
+            return self._run_native_loop(ctx, chat_model, functions, executor, formatter)
         except Exception as exc:  # pragma: no cover - safeguard
             self.logger.error(f"LLM 调用失败: {exc}")
             return LLMRunResult(handled=False, error=str(exc))
@@ -133,62 +135,6 @@ class FunctionCallLLM:
         self.logger.warning("达到最大函数调用轮数，未得到最终回答")
         return LLMRunResult(handled=False, error="max_rounds")
 
-    # ---------------------------------------------------------------------
-    # Prompt-based fallback workflow
-    # ---------------------------------------------------------------------
-
-    def _run_prompt_loop(
-        self,
-        ctx: MessageContext,
-        chat_model: Any,
-        functions: Dict[str, FunctionSpec],
-        executor: Callable[[FunctionSpec, Dict[str, Any]], FunctionResult],
-    ) -> LLMRunResult:
-        system_prompt = self._build_prompt_system_text(functions)
-        user_input = f"用户输入：{ctx.text}"
-
-        ai_response = chat_model.get_answer(
-            user_input,
-            wxid=ctx.get_receiver(),
-            system_prompt_override=system_prompt,
-        )
-
-        json_match = re.search(r"\{.*\}", ai_response, re.DOTALL)
-        if not json_match:
-            self.logger.warning(f"提示词模式下无法解析JSON: {ai_response}")
-            return LLMRunResult(handled=False)
-
-        try:
-            decision = json.loads(json_match.group(0))
-        except json.JSONDecodeError as exc:
-            self.logger.error(f"提示词模式 JSON 解析失败: {exc}")
-            return LLMRunResult(handled=False)
-
-        action_type = decision.get("action_type")
-        if action_type == "chat":
-            # 提示词模式下无法获得模型最终回答，交给上层兜底
-            return LLMRunResult(handled=False)
-
-        if action_type != "function":
-            self.logger.warning(f"未知的action_type: {action_type}")
-            return LLMRunResult(handled=False)
-
-        function_name = decision.get("function_name")
-        if function_name not in functions:
-            self.logger.warning(f"未知的功能名 - {function_name}")
-            return LLMRunResult(handled=False)
-
-        arguments = decision.get("arguments", {})
-        result = executor(functions[function_name], arguments)
-        if not result.handled:
-            return LLMRunResult(handled=False)
-
-        return LLMRunResult(handled=True, final_response="\n".join(result.messages))
-
-    # ------------------------------------------------------------------
-    # Helpers
-    # ------------------------------------------------------------------
-
     @staticmethod
     def _convert_assistant_message(message: Any) -> Dict[str, Any]:
         entry: Dict[str, Any] = {
@@ -224,16 +170,6 @@ class FunctionCallLLM:
             )
         return openai_functions
 
-    @staticmethod
-    def _build_prompt_system_text(functions: Dict[str, FunctionSpec]) -> str:
-        prompt = """你是一个智能路由助手。根据用户输入判断是否需要调用以下函数之一。"""
-        for spec in functions.values():
-            prompt += f"\n- {spec.name}: {spec.description}"
-        prompt += """
-请严格输出JSON：{"action_type": "chat"} 或 {"action_type": "function", "function_name": "...", "arguments": {...}}
-"""
-        return prompt
-
     def validate_arguments(self, arguments: Dict[str, Any], schema: Dict[str, Any]) -> bool:
         try:
             required_fields = schema.get("required", [])

function_calls/services/__init__.py

@@ -3,6 +3,7 @@
 from .reminder import create_reminder, list_reminders, delete_reminder
 from .group_tools import summarize_messages
 from .perplexity import run_perplexity
+from .chat import run_chat_fallback
 
 __all__ = [
     "create_reminder",
@@ -10,4 +11,5 @@ __all__ = [
     "delete_reminder",
     "summarize_messages",
     "run_perplexity",
+    "run_chat_fallback",
 ]

function_calls/services/chat.py

@@ -0,0 +1,130 @@
+"""Chat fallback utilities for Function Call routing."""
+from __future__ import annotations
+
+import time
+from typing import Optional
+
+from commands.context import MessageContext
+
+
+def run_chat_fallback(ctx: MessageContext) -> bool:
+    """Send a conversational reply using the active chat model.
+
+    This is used when no Function Call handler processes the message.
+    Returns True if a reply was sent successfully.
+    """
+    chat_model = getattr(ctx, "chat", None) or getattr(ctx.robot, "chat", None)
+    if not chat_model:
+        if ctx.logger:
+            ctx.logger.error("聊天兜底失败：没有可用的 chat 模型")
+        ctx.send_text("抱歉，我现在无法进行对话。")
+        return False
+
+    specific_max_history: Optional[int] = getattr(ctx, "specific_max_history", None)
+
+    if getattr(ctx, "is_quoted_image", False):
+        if not _handle_quoted_image(ctx, chat_model):
+            return False
+        return True
+
+    prompt = _build_prompt(ctx)
+    if ctx.logger:
+        ctx.logger.info(f"闲聊兜底发送给 AI 的内容:\n{prompt}")
+
+    try:
+        answer = chat_model.get_answer(
+            question=prompt,
+            wxid=ctx.get_receiver(),
+            specific_max_history=specific_max_history,
+        )
+    except Exception as exc:  # pragma: no cover - safety net
+        if ctx.logger:
+            ctx.logger.error(f"闲聊兜底调用模型失败: {exc}")
+        return False
+
+    if not answer:
+        if ctx.logger:
+            ctx.logger.warning("闲聊兜底返回空响应")
+        return False
+
+    at_list = ctx.msg.sender if ctx.is_group else ""
+    ctx.send_text(answer, at_list)
+    return True
+
+
+def _build_prompt(ctx: MessageContext) -> str:
+    sender_name = ctx.sender_name
+    content = ctx.text or ""
+
+    if ctx.robot and hasattr(ctx.robot, "xml_processor"):
+        if ctx.is_group:
+            msg_data = ctx.robot.xml_processor.extract_quoted_message(ctx.msg)
+            formatted = ctx.robot.xml_processor.format_message_for_ai(msg_data, sender_name)
+        else:
+            msg_data = ctx.robot.xml_processor.extract_private_quoted_message(ctx.msg)
+            formatted = ctx.robot.xml_processor.format_message_for_ai(msg_data, sender_name)
+
+        if formatted:
+            return formatted
+
+    current_time = time.strftime("%H:%M", time.localtime())
+    return f"[{current_time}] {sender_name}: {content or '[空内容]'}"
+
+
+def _handle_quoted_image(ctx: MessageContext, chat_model) -> bool:
+    if ctx.logger:
+        ctx.logger.info("检测到引用图片，尝试走模型图片理解能力")
+
+    from ai_providers.ai_chatgpt import ChatGPT  # 避免循环导入
+
+    support_vision = False
+    if isinstance(chat_model, ChatGPT):
+        support_vision = getattr(chat_model, "support_vision", False)
+        if not support_vision and hasattr(chat_model, "model"):
+            model_name = getattr(chat_model, "model", "")
+            support_vision = model_name in {"gpt-4.1-mini", "gpt-4o"} or "-vision" in model_name
+
+    if not support_vision:
+        ctx.send_text("当前模型不支持图片理解，请联系管理员配置支持视觉的模型。")
+        return True
+
+    import os
+
+    temp_dir = "temp/image_cache"
+    os.makedirs(temp_dir, exist_ok=True)
+
+    try:
+        image_path = ctx.wcf.download_image(
+            id=ctx.quoted_msg_id,
+            extra=ctx.quoted_image_extra,
+            dir=temp_dir,
+            timeout=30,
+        )
+    except Exception as exc:  # pragma: no cover - IO 失败
+        if ctx.logger:
+            ctx.logger.error(f"图片下载失败: {exc}")
+        ctx.send_text("抱歉，无法下载图片进行分析。")
+        return True
+
+    if not image_path or not os.path.exists(image_path):
+        ctx.send_text("抱歉，无法下载图片进行分析。")
+        return True
+
+    prompt = ctx.text.strip() or "请详细描述这张图片"
+
+    try:
+        response = chat_model.get_image_description(image_path, prompt)
+        ctx.send_text(response)
+    except Exception as exc:  # pragma: no cover - 模型异常
+        if ctx.logger:
+            ctx.logger.error(f"图片分析失败: {exc}")
+        ctx.send_text(f"分析图片时出错: {exc}")
+    finally:
+        try:
+            if os.path.exists(image_path):
+                os.remove(image_path)
+        except OSError:
+            if ctx.logger:
+                ctx.logger.warning(f"清理临时图片失败: {image_path}")
+
+    return True