refactor(prompt): 优化任务并发执行逻辑，使用 as_completed 提升性能和错误处理

2025-11-12 21:19:19 +08:00
parent 61d86875ad
commit 17abfc74ae
1 changed files with 217 additions and 85 deletions
--- a/src/chat/utils/prompt.py
+++ b/src/chat/utils/prompt.py
@@ -24,7 +24,6 @@ install(extra_lines=3)
 logger = get_logger("unified_prompt")


-
 class PromptContext:
    """提示词上下文管理器.

@@ -41,7 +40,9 @@ class PromptContext:

        # _current_context_var: 使用contextvars来存储当前协程的上下文ID。
        # 这确保了在并发执行的异步任务中，每个任务都能访问到正确的上下文ID。
-        self._current_context_var = contextvars.ContextVar("current_context", default=None)
+        self._current_context_var = contextvars.ContextVar(
+            "current_context", default=None
+        )

        # _context_lock: 一个异步锁，用于保护对共享资源_context_prompts的并发访问。
        self._context_lock = asyncio.Lock()
@@ -120,7 +121,9 @@ class PromptContext:
                return self._context_prompts[current_context][name]
            return None

-    async def register_async(self, prompt: "Prompt", context_id: str | None = None) -> None:
+    async def register_async(
+        self, prompt: "Prompt", context_id: str | None = None
+    ) -> None:
        """异步、安全地将提示模板注册到指定的作用域.

        如果未指定context_id，则注册到当前协程的上下文中。
@@ -130,7 +133,9 @@ class PromptContext:
            if target_context := context_id or self._current_context:
                if prompt.name:
                    # 使用setdefault确保目标上下文的字典存在，然后注册prompt
-                    self._context_prompts.setdefault(target_context, {})[prompt.name] = prompt
+                    self._context_prompts.setdefault(target_context, {})[
+                        prompt.name
+                    ] = prompt


 class PromptManager:
@@ -158,7 +163,9 @@ class PromptManager:
        async with self._context.async_scope(message_id):
            yield self

-    async def get_prompt_async(self, name: str, parameters: PromptParameters | None = None) -> "Prompt":
+    async def get_prompt_async(
+        self, name: str, parameters: PromptParameters | None = None
+    ) -> "Prompt":
        """异步获取提示模板，并动态地将插件内容注入其中.

        获取提示词的优先级顺序为：
@@ -309,12 +316,16 @@ class Prompt:
        elif not isinstance(template, str):
            template = str(template)

-        return template.replace("\\{", Prompt._TEMP_LEFT_BRACE).replace("\\}", Prompt._TEMP_RIGHT_BRACE)
+        return template.replace("\\{", Prompt._TEMP_LEFT_BRACE).replace(
+            "\\}", Prompt._TEMP_RIGHT_BRACE
+        )

    @staticmethod
    def _restore_escaped_braces(template: str) -> str:
        """在格式化完成后，将临时标记还原为实际的花括号字符 `{` 和 `}`."""
-        return template.replace(Prompt._TEMP_LEFT_BRACE, "{").replace(Prompt._TEMP_RIGHT_BRACE, "}")
+        return template.replace(Prompt._TEMP_LEFT_BRACE, "{").replace(
+            Prompt._TEMP_RIGHT_BRACE, "}"
+        )

    def _parse_template_args(self, template: str) -> list[str]:
        """从模板字符串中解析出所有占位符（例如 "{user_name}" -> "user_name"）."""
@@ -365,7 +376,9 @@ class Prompt:
            result = main_formatted_prompt

            total_time = time.time() - start_time
-            logger.debug(f"Prompt构建完成，模式: {self.parameters.prompt_mode}, 耗时: {total_time:.2f}s")
+            logger.debug(
+                f"Prompt构建完成，模式: {self.parameters.prompt_mode}, 耗时: {total_time:.2f}s"
+            )

            # 缓存结果
            self._formatted_result = result
@@ -410,9 +423,13 @@ class Prompt:
            # 如果参数对象中已经包含了某些block，说明它们是外部预构建的，
            # 我们将它们存起来，并跳过对应的实时构建任务。
            if self.parameters.expression_habits_block:
-                pre_built_params["expression_habits_block"] = self.parameters.expression_habits_block
+                pre_built_params["expression_habits_block"] = (
+                    self.parameters.expression_habits_block
+                )
            if self.parameters.relation_info_block:
-                pre_built_params["relation_info_block"] = self.parameters.relation_info_block
+                pre_built_params["relation_info_block"] = (
+                    self.parameters.relation_info_block
+                )
            if self.parameters.memory_block:
                pre_built_params["memory_block"] = self.parameters.memory_block
                logger.debug("使用预构建的memory_block，跳过实时构建")
@@ -421,12 +438,16 @@ class Prompt:
            if self.parameters.knowledge_prompt:
                pre_built_params["knowledge_prompt"] = self.parameters.knowledge_prompt
            if self.parameters.cross_context_block:
-                pre_built_params["cross_context_block"] = self.parameters.cross_context_block
+                pre_built_params["cross_context_block"] = (
+                    self.parameters.cross_context_block
+                )
            if self.parameters.notice_block:
                pre_built_params["notice_block"] = self.parameters.notice_block

            # --- 步骤 1.2: 根据参数和预构建情况，决定需要实时运行的任务 ---
-            if self.parameters.enable_expression and not pre_built_params.get("expression_habits_block"):
+            if self.parameters.enable_expression and not pre_built_params.get(
+                "expression_habits_block"
+            ):
                tasks.append(self._build_expression_habits())
                task_names.append("expression_habits")

@@ -434,19 +455,27 @@ class Prompt:
            # 使用新的记忆图系统，不再在 prompt.py 中构建记忆
            # 如果需要记忆，必须通过 pre_built_params 传入

-            if self.parameters.enable_relation and not pre_built_params.get("relation_info_block"):
+            if self.parameters.enable_relation and not pre_built_params.get(
+                "relation_info_block"
+            ):
                tasks.append(self._build_relation_info())
                task_names.append("relation_info")

-            if self.parameters.enable_tool and not pre_built_params.get("tool_info_block"):
+            if self.parameters.enable_tool and not pre_built_params.get(
+                "tool_info_block"
+            ):
                tasks.append(self._build_tool_info())
                task_names.append("tool_info")

-            if self.parameters.enable_knowledge and not pre_built_params.get("knowledge_prompt"):
+            if self.parameters.enable_knowledge and not pre_built_params.get(
+                "knowledge_prompt"
+            ):
                tasks.append(self._build_knowledge_info())
                task_names.append("knowledge_info")

-            if self.parameters.enable_cross_context and not pre_built_params.get("cross_context_block"):
+            if self.parameters.enable_cross_context and not pre_built_params.get(
+                "cross_context_block"
+            ):
                tasks.append(self._build_cross_context())
                task_names.append("cross_context")

@@ -463,32 +492,61 @@ class Prompt:
                "expression_habits": 10.0,
            }

-            results = []
-            # 循环等待每个任务，而不是使用`asyncio.gather`，
-            # 这样可以为每个任务应用独立的超时控制。
+            # 使用 as_completed 并发执行任务，提供更好的性能和错误处理
+            results = [None] * len(tasks)  # 预分配结果列表，保持任务顺序
+            task_with_meta = []
+
+            # 准备任务和元数据
            for i, task in enumerate(tasks):
                task_name = task_names[i] if i < len(task_names) else f"task_{i}"
-                task_timeout = task_timeouts.get(task_name, 2.0)  # 未指定超时的任务默认为2秒
+                task_timeout = task_timeouts.get(
+                    task_name, 2.0
+                )  # 未指定超时的任务默认为2秒
+
+                # 检查任务是否为协程，非协程任务直接使用默认值
+                if asyncio.iscoroutine(task):
+                    task_with_meta.append(
+                        (
+                            asyncio.wait_for(task, timeout=task_timeout),
+                            task_name,
+                            i,
+                            task_timeout,
+                        )
+                    )
+                else:
+                    logger.warning(
+                        f"任务{task_name}不是协程对象，类型: {type(task)}，跳过处理"
+                    )
+                    results[i] = self._get_default_result_for_task(task_name)  # type: ignore
+
+            # 并发执行任务，使用 as_completed 获得更好的性能
+            for future in asyncio.as_completed(
+                [task_meta[0] for task_meta in task_with_meta]
+            ):
+                # 找到对应的任务元数据
+                task_index = None
+                task_name = None
+                task_timeout = None
+
+                for idx, (task, name, index, timeout) in enumerate(task_with_meta):
+                    if task == future:
+                        task_index = index
+                        task_name = name
+                        task_timeout = timeout
+                        break

                try:
-                    # 确保任务是可等待的协程
-                    if asyncio.iscoroutine(task):
-                        # 使用 `asyncio.wait_for` 来执行带超时的任务
-                        result = await asyncio.wait_for(task, timeout=task_timeout)
-                        results.append(result)
-                        logger.debug(f"构建任务{task_name}完成 ({task_timeout}s)")
-                    else:
-                        # 如果任务不是协程，记录警告并使用默认值
-                        logger.warning(f"任务{task_name}不是协程对象，类型: {type(task)}，跳过处理")
-                        results.append(self._get_default_result_for_task(task_name))
+                    result = await future
+                    results[task_index] = result  # type: ignore
+                    logger.debug(f"构建任务{task_name}完成 ({task_timeout}s)")
                except asyncio.TimeoutError:
-                    # 如果任务超时，记录警告并使用默认值
-                    logger.warning(f"构建任务{task_name}超时 ({task_timeout}s)，使用默认值")
-                    results.append(self._get_default_result_for_task(task_name))
+                    logger.warning(
+                        f"构建任务{task_name}超时 ({task_timeout}s)，使用默认值"
+                    )
+                    results[task_index] = self._get_default_result_for_task(task_name)  # type: ignore
                except Exception as e:
-                    # 如果任务执行出错，记录错误并使用默认值
                    logger.error(f"构建任务{task_name}失败: {e!s}")
-                    results.append(self._get_default_result_for_task(task_name))
+                    results[task_index] = self._get_default_result_for_task(task_name)  # type: ignore

            # --- 步骤 3: 合并所有结果 ---
            context_data = {}
@@ -501,7 +559,9 @@ class Prompt:
                    context_data.update(result)

            # 合并预构建的参数，这会覆盖任何同名的实时构建结果
-            context_data.update({key: value for key, value in pre_built_params.items() if value})
+            context_data.update(
+                {key: value for key, value in pre_built_params.items() if value}
+            )

        except asyncio.TimeoutError:
            # 这是一个不太可能发生的、总体的构建超时，作为最后的保障
@@ -522,7 +582,8 @@ class Prompt:
            {
                "keywords_reaction_prompt": self.parameters.keywords_reaction_prompt,
                "extra_info_block": self.parameters.extra_info_block,
-                "time_block": self.parameters.time_block or f"当前时间：{time.strftime('%Y-%m-%d %H:%M:%S')}",
+                "time_block": self.parameters.time_block
+                or f"当前时间：{time.strftime('%Y-%m-%d %H:%M:%S')}",
                "identity": self.parameters.identity_block,
                "schedule_block": self.parameters.schedule_block,
                "moderation_prompt": self.parameters.moderation_prompt_block,
@@ -549,20 +610,25 @@ class Prompt:
            target_user_id = self.parameters.target_user_info.get("user_id") or ""

        # 调用核心构建逻辑
-        read_history_prompt, unread_history_prompt = await self._build_s4u_chat_history_prompts(
-            self.parameters.message_list_before_now_long,
-            target_user_id,
-            self.parameters.sender,
-            self.parameters.chat_id,
+        read_history_prompt, unread_history_prompt = (
+            await self._build_s4u_chat_history_prompts(
+                self.parameters.message_list_before_now_long,
+                target_user_id,
+                self.parameters.sender,
+                self.parameters.chat_id,
+            )
        )

        # 将构建好的prompt添加到上下文数据中
        context_data["read_history_prompt"] = read_history_prompt
        context_data["unread_history_prompt"] = unread_history_prompt

-
    async def _build_s4u_chat_history_prompts(
-        self, message_list_before_now: list[dict[str, Any]], target_user_id: str, sender: str, chat_id: str
+        self,
+        message_list_before_now: list[dict[str, Any]],
+        target_user_id: str,
+        sender: str,
+        chat_id: str,
    ) -> tuple[str, str]:
        """构建S4U风格的已读/未读历史消息prompt.

@@ -574,7 +640,9 @@ class Prompt:
            from src.plugin_system.apis.generator_api import get_replyer

            # 获取一个临时的生成器实例来访问其方法
-            temp_generator = await get_replyer(None, chat_id, request_type="prompt_building")
+            temp_generator = await get_replyer(
+                None, chat_id, request_type="prompt_building"
+            )
            if temp_generator:
                # 调用实际的构建方法
                return await temp_generator.build_s4u_chat_history_prompts(
@@ -588,7 +656,9 @@ class Prompt:
    async def _build_expression_habits(self) -> dict[str, Any]:
        """构建表达习惯（如表情、口癖）的上下文块."""
        # 检查当前聊天是否启用了表达习惯功能
-        use_expression, _, _ = global_config.expression.get_expression_config_for_chat(self.parameters.chat_id)
+        use_expression, _, _ = global_config.expression.get_expression_config_for_chat(
+            self.parameters.chat_id
+        )
        if not use_expression:
            return {"expression_habits_block": ""}

@@ -601,15 +671,20 @@ class Prompt:
            if self.parameters.message_list_before_now_long:
                recent_messages = self.parameters.message_list_before_now_long[-10:]
                chat_history = await build_readable_messages(
-                    recent_messages, replace_bot_name=True, timestamp_mode="normal", truncate=True
+                    recent_messages,
+                    replace_bot_name=True,
+                    timestamp_mode="normal",
+                    truncate=True,
                )

            # 使用统一的表达方式选择入口（支持classic和exp_model模式）
            expression_selector = ExpressionSelector(self.parameters.chat_id)
-            selected_expressions = await expression_selector.select_suitable_expressions(
-                chat_id=self.parameters.chat_id,
-                chat_history=chat_history,
-                target_message=self.parameters.target,
+            selected_expressions = (
+                await expression_selector.select_suitable_expressions(
+                    chat_id=self.parameters.chat_id,
+                    chat_history=chat_history,
+                    target_message=self.parameters.target,
+                )
            )

            # 将选择的表达习惯格式化为提示词的一部分
@@ -641,7 +716,9 @@ class Prompt:
        """构建与对话目标相关的关系信息."""
        try:
            # 调用静态方法来执行实际的构建逻辑
-            relation_info = await Prompt.build_relation_info(self.parameters.chat_id, self.parameters.reply_to)
+            relation_info = await Prompt.build_relation_info(
+                self.parameters.chat_id, self.parameters.reply_to
+            )
            return {"relation_info_block": relation_info}
        except Exception as e:
            logger.error(f"构建关系信息失败: {e}")
@@ -660,7 +737,10 @@ class Prompt:
            if self.parameters.message_list_before_now_long:
                recent_messages = self.parameters.message_list_before_now_long[-15:]
                chat_history = await build_readable_messages(
-                    recent_messages, replace_bot_name=True, timestamp_mode="normal", truncate=True
+                    recent_messages,
+                    replace_bot_name=True,
+                    timestamp_mode="normal",
+                    truncate=True,
                )

            # 决定是否调用工具并执行
@@ -682,7 +762,9 @@ class Prompt:

                    tool_info_parts.append(f"- 【{tool_name}】{result_type}: {content}")

-                tool_info_parts.append("以上是你获取到的实时信息，请在回复时参考这些信息。")
+                tool_info_parts.append(
+                    "以上是你获取到的实时信息，请在回复时参考这些信息。"
+                )
                tool_info_block = "\n".join(tool_info_parts)
            else:
                tool_info_block = ""
@@ -710,7 +792,10 @@ class Prompt:

            # 将检索结果格式化为提示词
            if knowledge_results and knowledge_results.get("knowledge_items"):
-                knowledge_parts = ["## 知识库信息", "以下是与你当前对话相关的知识信息："]
+                knowledge_parts = [
+                    "## 知识库信息",
+                    "以下是与你当前对话相关的知识信息：",
+                ]

                for item in knowledge_results["knowledge_items"]:
                    content = item.get("content", "")
@@ -720,20 +805,27 @@ class Prompt:
                        # 过滤掉相关性低于阈值的知识
                        try:
                            relevance_float = float(relevance)
-                            if relevance_float < global_config.lpmm_knowledge.qa_paragraph_threshold:
+                            if (
+                                relevance_float
+                                < global_config.lpmm_knowledge.qa_paragraph_threshold
+                            ):
                                continue
                            relevance_str = f"{relevance_float:.2f}"
                        except (ValueError, TypeError):
                            relevance_str = str(relevance)

                        if source:
-                            knowledge_parts.append(f"- [{relevance_str}] {content} (来源: {source})")
+                            knowledge_parts.append(
+                                f"- [{relevance_str}] {content} (来源: {source})"
+                            )
                        else:
                            knowledge_parts.append(f"- [{relevance_str}] {content}")

                # 如果有总结，也一并加入
                if knowledge_results.get("summary"):
-                    knowledge_parts.append(f"\n知识总结: {knowledge_results['summary']}")
+                    knowledge_parts.append(
+                        f"\n知识总结: {knowledge_results['summary']}"
+                    )

                knowledge_prompt = "\n".join(knowledge_parts)
            else:
@@ -750,7 +842,9 @@ class Prompt:
        try:
            # 调用静态方法来执行实际的构建逻辑
            cross_context = await Prompt.build_cross_context(
-                self.parameters.chat_id, self.parameters.prompt_mode, self.parameters.target_user_info
+                self.parameters.chat_id,
+                self.parameters.prompt_mode,
+                self.parameters.target_user_info,
            )
            return {"cross_context_block": cross_context}
        except Exception as e:
@@ -769,7 +863,11 @@ class Prompt:
            params = self._prepare_default_params(context_data)

        # 如果prompt有名称，则通过全局管理器格式化（这样可以应用注入逻辑），否则直接格式化
-        return await global_prompt_manager.format_prompt(self.name, **params) if self.name else self.format(**params)
+        return (
+            await global_prompt_manager.format_prompt(self.name, **params)
+            if self.name
+            else self.format(**params)
+        )

    def _prepare_s4u_params(self, context_data: dict[str, Any]) -> dict[str, Any]:
        """为S4U（Scene for You）模式准备最终用于格式化的参数字典."""
@@ -780,14 +878,20 @@ class Prompt:
            "knowledge_prompt": context_data.get("knowledge_prompt", ""),
            "memory_block": context_data.get("memory_block", ""),
            "relation_info_block": context_data.get("relation_info_block", ""),
-            "extra_info_block": self.parameters.extra_info_block or context_data.get("extra_info_block", ""),
+            "extra_info_block": self.parameters.extra_info_block
+            or context_data.get("extra_info_block", ""),
            "cross_context_block": context_data.get("cross_context_block", ""),
-            "notice_block": self.parameters.notice_block or context_data.get("notice_block", ""),
-            "identity": self.parameters.identity_block or context_data.get("identity", ""),
-            "action_descriptions": self.parameters.action_descriptions or context_data.get("action_descriptions", ""),
-            "schedule_block": self.parameters.schedule_block or context_data.get("schedule_block", ""),
+            "notice_block": self.parameters.notice_block
+            or context_data.get("notice_block", ""),
+            "identity": self.parameters.identity_block
+            or context_data.get("identity", ""),
+            "action_descriptions": self.parameters.action_descriptions
+            or context_data.get("action_descriptions", ""),
+            "schedule_block": self.parameters.schedule_block
+            or context_data.get("schedule_block", ""),
            "sender_name": self.parameters.sender or "未知用户",
-            "mood_state": self.parameters.mood_prompt or context_data.get("mood_state", ""),
+            "mood_state": self.parameters.mood_prompt
+            or context_data.get("mood_state", ""),
            "read_history_prompt": context_data.get("read_history_prompt", ""),
            "unread_history_prompt": context_data.get("unread_history_prompt", ""),
            "time_block": context_data.get("time_block", ""),
@@ -795,7 +899,8 @@ class Prompt:
            "reply_style": global_config.personality.reply_style,
            "keywords_reaction_prompt": self.parameters.keywords_reaction_prompt
            or context_data.get("keywords_reaction_prompt", ""),
-            "moderation_prompt": self.parameters.moderation_prompt_block or context_data.get("moderation_prompt", ""),
+            "moderation_prompt": self.parameters.moderation_prompt_block
+            or context_data.get("moderation_prompt", ""),
            "safety_guidelines_block": self.parameters.safety_guidelines_block
            or context_data.get("safety_guidelines_block", ""),
            "auth_role_prompt_block": self.parameters.auth_role_prompt_block
@@ -813,22 +918,29 @@ class Prompt:
            "knowledge_prompt": context_data.get("knowledge_prompt", ""),
            "memory_block": context_data.get("memory_block", ""),
            "relation_info_block": context_data.get("relation_info_block", ""),
-            "extra_info_block": self.parameters.extra_info_block or context_data.get("extra_info_block", ""),
+            "extra_info_block": self.parameters.extra_info_block
+            or context_data.get("extra_info_block", ""),
            "cross_context_block": context_data.get("cross_context_block", ""),
-            "notice_block": self.parameters.notice_block or context_data.get("notice_block", ""),
-            "identity": self.parameters.identity_block or context_data.get("identity", ""),
-            "action_descriptions": self.parameters.action_descriptions or context_data.get("action_descriptions", ""),
-            "schedule_block": self.parameters.schedule_block or context_data.get("schedule_block", ""),
+            "notice_block": self.parameters.notice_block
+            or context_data.get("notice_block", ""),
+            "identity": self.parameters.identity_block
+            or context_data.get("identity", ""),
+            "action_descriptions": self.parameters.action_descriptions
+            or context_data.get("action_descriptions", ""),
+            "schedule_block": self.parameters.schedule_block
+            or context_data.get("schedule_block", ""),
            "time_block": context_data.get("time_block", ""),
            "chat_info": context_data.get("chat_info", ""),
            "reply_target_block": context_data.get("reply_target_block", ""),
            "reply_style": global_config.personality.reply_style,
-            "mood_state": self.parameters.mood_prompt or context_data.get("mood_state", ""),
+            "mood_state": self.parameters.mood_prompt
+            or context_data.get("mood_state", ""),
            "read_history_prompt": context_data.get("read_history_prompt", ""),
            "unread_history_prompt": context_data.get("unread_history_prompt", ""),
            "keywords_reaction_prompt": self.parameters.keywords_reaction_prompt
            or context_data.get("keywords_reaction_prompt", ""),
-            "moderation_prompt": self.parameters.moderation_prompt_block or context_data.get("moderation_prompt", ""),
+            "moderation_prompt": self.parameters.moderation_prompt_block
+            or context_data.get("moderation_prompt", ""),
            "safety_guidelines_block": self.parameters.safety_guidelines_block
            or context_data.get("safety_guidelines_block", ""),
            "auth_role_prompt_block": self.parameters.auth_role_prompt_block
@@ -847,17 +959,21 @@ class Prompt:
            "chat_target": "",
            "time_block": context_data.get("time_block", ""),
            "chat_info": context_data.get("chat_info", ""),
-            "identity": self.parameters.identity_block or context_data.get("identity", ""),
-            "schedule_block": self.parameters.schedule_block or context_data.get("schedule_block", ""),
+            "identity": self.parameters.identity_block
+            or context_data.get("identity", ""),
+            "schedule_block": self.parameters.schedule_block
+            or context_data.get("schedule_block", ""),
            "chat_target_2": "",
            "reply_target_block": context_data.get("reply_target_block", ""),
            "raw_reply": self.parameters.target,
            "reason": "",
-            "mood_state": self.parameters.mood_prompt or context_data.get("mood_state", ""),
+            "mood_state": self.parameters.mood_prompt
+            or context_data.get("mood_state", ""),
            "reply_style": global_config.personality.reply_style,
            "keywords_reaction_prompt": self.parameters.keywords_reaction_prompt
            or context_data.get("keywords_reaction_prompt", ""),
-            "moderation_prompt": self.parameters.moderation_prompt_block or context_data.get("moderation_prompt", ""),
+            "moderation_prompt": self.parameters.moderation_prompt_block
+            or context_data.get("moderation_prompt", ""),
            "safety_guidelines_block": self.parameters.safety_guidelines_block
            or context_data.get("safety_guidelines_block", ""),
            "auth_role_prompt_block": self.parameters.auth_role_prompt_block
@@ -902,7 +1018,9 @@ class Prompt:
            return result
        except (IndexError, KeyError) as e:
            # 捕获格式化错误并抛出更具信息量的异常
-            raise ValueError(f"格式化模板失败: {self.template}, args={args}, kwargs={kwargs} {e!s}") from e
+            raise ValueError(
+                f"格式化模板失败: {self.template}, args={args}, kwargs={kwargs} {e!s}"
+            ) from e

    def __str__(self) -> str:
        """返回格式化后的结果，如果还未格式化，则返回原始模板."""
@@ -976,8 +1094,12 @@ class Prompt:
            return f"你完全不认识{sender}，不理解ta的相关信息。"

        # 使用关系提取器构建用户关系信息和聊天流印象
-        user_relation_info = await relationship_fetcher.build_relation_info(person_id, points_num=5)
-        stream_impression = await relationship_fetcher.build_chat_stream_impression(chat_id)
+        user_relation_info = await relationship_fetcher.build_relation_info(
+            person_id, points_num=5
+        )
+        stream_impression = await relationship_fetcher.build_chat_stream_impression(
+            chat_id
+        )

        # 组合两部分信息
        info_parts = []
@@ -1016,7 +1138,9 @@ class Prompt:
            return {}

    @staticmethod
-    async def build_cross_context(chat_id: str, prompt_mode: str, target_user_info: dict[str, Any] | None) -> str:
+    async def build_cross_context(
+        chat_id: str, prompt_mode: str, target_user_info: dict[str, Any] | None
+    ) -> str:
        """构建跨群聊的上下文信息.

        Args:
@@ -1039,7 +1163,9 @@ class Prompt:

        # 目前只为s4u模式构建跨群上下文
        if prompt_mode == "s4u":
-            return await cross_context_api.build_cross_context_s4u(chat_stream, target_user_info)
+            return await cross_context_api.build_cross_context_s4u(
+                chat_stream, target_user_info
+            )

        return ""

@@ -1073,7 +1199,10 @@ class Prompt:

 # 工厂函数
 def create_prompt(
-    template: str, name: str | None = None, parameters: PromptParameters | None = None, **kwargs
+    template: str,
+    name: str | None = None,
+    parameters: PromptParameters | None = None,
+    **kwargs,
 ) -> Prompt:
    """一个用于快速创建`Prompt`实例的工厂函数.

@@ -1094,7 +1223,10 @@ def create_prompt(


 async def create_prompt_async(
-    template: str, name: str | None = None, parameters: PromptParameters | None = None, **kwargs
+    template: str,
+    name: str | None = None,
+    parameters: PromptParameters | None = None,
+    **kwargs,
 ) -> Prompt:
    """异步创建`Prompt`实例，并自动处理插件内容的动态注入.