From c84fe4f8ed40bc19030e5fa542edf7f33edb07b3 Mon Sep 17 00:00:00 2001 From: MuWinds Date: Thu, 13 Mar 2025 23:55:09 +0800 Subject: [PATCH] =?UTF-8?q?Update:=E8=8E=B7=E5=8F=96=E8=AE=B0=E5=BF=86?= =?UTF-8?q?=E6=A0=B7=E6=9C=AC=E5=A2=9E=E5=8A=A0=E5=8E=BB=E9=87=8D=E6=9C=BA?= =?UTF-8?q?=E5=88=B6?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- src/plugins/chat/utils.py | 2 +- src/plugins/memory_system/memory.py | 34 +++++++++++++++++++++++++---- 2 files changed, 31 insertions(+), 5 deletions(-) diff --git a/src/plugins/chat/utils.py b/src/plugins/chat/utils.py index f28d0e19..fa6a7649 100644 --- a/src/plugins/chat/utils.py +++ b/src/plugins/chat/utils.py @@ -105,11 +105,11 @@ def get_closest_chat_from_db(length: int, timestamp: str): formatted_records = [] for record in chat_records: formatted_records.append({ + 'message_id': record['_id'], 'time': record["time"], 'chat_id': record["chat_id"], 'detailed_plain_text': record.get("detailed_plain_text", "") # 添加文本内容 }) - return formatted_records return [] diff --git a/src/plugins/memory_system/memory.py b/src/plugins/memory_system/memory.py index f5c7181b..7354e29a 100644 --- a/src/plugins/memory_system/memory.py +++ b/src/plugins/memory_system/memory.py @@ -192,26 +192,52 @@ class Hippocampus: """ current_timestamp = datetime.datetime.now().timestamp() chat_samples = [] - + added_ids = set() # 全局已添加消息ID集合，用于去重 # 短期：1h 中期：4h 长期：24h for _ in range(time_frequency.get('near')): random_time = current_timestamp - random.randint(1, 3600) messages = get_closest_chat_from_db(length=chat_size, timestamp=random_time) if messages: - chat_samples.append(messages) + # 过滤已存在的消息 + filtered = [] + for msg in messages: + msg_id = msg.get("message_id") # 假设消息中存在唯一标识字段 + if msg_id not in added_ids: + filtered.append(msg) + added_ids.add(msg_id) + + if filtered: + chat_samples.append(filtered) for _ in range(time_frequency.get('mid')): random_time = current_timestamp - random.randint(3600, 3600 * 4) messages = get_closest_chat_from_db(length=chat_size, timestamp=random_time) if messages: - chat_samples.append(messages) + # 过滤已存在的消息 + filtered = [] + for msg in messages: + msg_id = msg.get("message_id") # 假设消息中存在唯一标识字段 + if msg_id not in added_ids: + filtered.append(msg) + added_ids.add(msg_id) + + if filtered: + chat_samples.append(filtered) for _ in range(time_frequency.get('far')): random_time = current_timestamp - random.randint(3600 * 4, 3600 * 24) messages = get_closest_chat_from_db(length=chat_size, timestamp=random_time) if messages: - chat_samples.append(messages) + # 过滤已存在的消息 + filtered = [] + for msg in messages: + msg_id = msg.get("message_id") # 假设消息中存在唯一标识字段 + if msg_id not in added_ids: + filtered.append(msg) + added_ids.add(msg_id) + if filtered: + chat_samples.append(filtered) return chat_samples async def memory_compress(self, messages: list, compress_rate=0.1):