From cd2024dfb5a0f97459231195ba24a1a27be8e708 Mon Sep 17 00:00:00 2001 From: liuwei Date: Fri, 24 Apr 2026 15:06:05 +0800 Subject: [PATCH] =?UTF-8?q?=E4=BC=98=E5=8C=96=20ai=5Fauto=5Fresponse=20?= =?UTF-8?q?=E4=BB=85=E5=9B=9E=E5=A4=8D=E5=90=8C=E7=BE=A4=E6=9C=80=E6=96=B0?= =?UTF-8?q?=E6=B6=88=E6=81=AF?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - 为同群消息引入递增序号，记录当前群的最新消息版本 - 在出队前跳过已被同群更新消息覆盖的旧消息 - 在发送前再次检查消息是否已被覆盖，避免 LLM 慢返回后补发旧回复 - 保持实现仅影响同一群内的消息时效，不干扰其他群的并行处理 --- plugins/ai_auto_response/main.py | 50 ++++++++++++++++++++++++++++++++ 1 file changed, 50 insertions(+) diff --git a/plugins/ai_auto_response/main.py b/plugins/ai_auto_response/main.py index 6b7a6b5..441cb84 100644 --- a/plugins/ai_auto_response/main.py +++ b/plugins/ai_auto_response/main.py @@ -102,6 +102,8 @@ class AIAutoResponsePlugin(MessagePluginInterface): self.reply_limits: Dict[str, Any] = {} self.prompt_compact_config: Dict[str, Any] = {} self.message_expire_sec = 0.0 + self.room_message_seq_counter = 0 + self.latest_room_message_seq: Dict[str, int] = {} def initialize(self, context: Dict[str, Any]) -> bool: self.LOG = logger @@ -231,6 +233,11 @@ class AIAutoResponsePlugin(MessagePluginInterface): # 记录入队时刻，供后续判断这条消息是否已经“聊过时”。 # 使用 monotonic 避免系统时间调整影响队列老化判断。 queued_message["_queued_at_mono"] = time.monotonic() + # 记录“同群最新消息版本号”： + # 1. 每来一条新消息，就给当前群分配一个更大的序号； + # 2. 后续旧消息即使已经排队甚至已经进模型，只要序号落后，就视为过时； + # 3. 这样可以保证群里只会优先回应最新现场，避免补发旧话。 + queued_message["_room_message_seq"] = self._next_room_message_seq(room_id) try: self.message_queue.put_nowait(queued_message) self._log_event( @@ -270,6 +277,16 @@ class AIAutoResponsePlugin(MessagePluginInterface): age_sec=round(stale_age_sec, 2), ) return False, "stale_queued_message" + if self._is_message_superseded(message): + self._log_event( + "skip", + room_id=room_id, + sender=sender, + reason="superseded_by_newer_message", + trigger_type="latest_only_guard", + reply_mode="drop", + ) + return False, "superseded_by_newer_message" message_key = self._build_message_key(message, content) dedup_expiry = int(self.cooldown_config.get("message_dedup_window_sec", 180)) if not self.dedup.begin_message_processing(message_key, dedup_expiry): @@ -630,6 +647,21 @@ class AIAutoResponsePlugin(MessagePluginInterface): age_sec=round(self._get_message_queue_age_sec(message), 2), ) return False, "stale_before_send" + # 第二次“只回最新消息”判断： + # 1. 旧消息可能已经进了 LLM，但这期间同群又来了更新内容； + # 2. 这时即使模型产出了结果，也不应该再把旧回复补发出去； + # 3. 直接丢弃旧结果，让群里只看到贴着最新现场的回复。 + if self._is_message_superseded(message): + self._log_event( + "skip", + room_id=room_id, + sender=sender, + reason="superseded_before_send", + trigger_type=trigger.trigger_type, + reply_mode=reply_mode, + topic=selected_topic, + ) + return False, "superseded_before_send" reply_dedup_expiry = int(self.cooldown_config.get("reply_dedup_window_sec", 90)) if not reply_chunks or self.dedup.should_skip_duplicate_reply( room_id=room_id, @@ -1280,6 +1312,24 @@ class AIAutoResponsePlugin(MessagePluginInterface): # 3. 这样实现最稳定，也最符合“超过多久就别回了”的产品语义。 return self._get_message_queue_age_sec(message) >= float(self.message_expire_sec) + def _next_room_message_seq(self, room_id: str) -> int: + self.room_message_seq_counter += 1 + seq = self.room_message_seq_counter + if room_id: + self.latest_room_message_seq[room_id] = seq + return seq + + def _is_message_superseded(self, message: Dict[str, Any]) -> bool: + room_id = str(message.get("roomid", "") or "") + if not room_id: + return False + current_seq = message.get("_room_message_seq") + latest_seq = self.latest_room_message_seq.get(room_id) + try: + return int(current_seq or 0) < int(latest_seq or 0) + except (TypeError, ValueError): + return False + def _normalize_content(self, message: Dict[str, Any]) -> str: msg_type = message.get("type") content = str(message.get("content", "")).strip()