From de12e8253a4df3703f9797c55426b1c31ba8c8a6 Mon Sep 17 00:00:00 2001 From: SXP-Simon Date: Tue, 16 Sep 2025 12:08:56 +0800 Subject: [PATCH] =?UTF-8?q?[fix]=20(LLM=20=E6=8F=90=E4=BE=9B=E5=95=86)=20?= =?UTF-8?q?=E4=BC=A0=E9=80=92=20unified=5Fmsg=5Forigin=20=E4=BB=A5?= =?UTF-8?q?=E8=8E=B7=E5=8F=96=E6=AD=A3=E7=A1=AE=E7=9A=84=20LLM=20=E6=8F=90?= =?UTF-8?q?=E4=BE=9B=E5=95=86?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- main.py | 4 ++-- src/analysis/llm_analyzer.py | 32 ++++++-------------------------- src/scheduler/auto_scheduler.py | 21 +++++++++++++++++++-- src/utils/helpers.py | 8 ++++---- 4 files changed, 31 insertions(+), 34 deletions(-) diff --git a/main.py b/main.py index 207d619..73be5e4 100644 --- a/main.py +++ b/main.py @@ -146,8 +146,8 @@ class QQGroupDailyAnalysis(Star): yield event.plain_result(f"📊 已获取{len(messages)}条消息，正在进行智能分析...") - # 进行分析 - analysis_result = await message_analyzer.analyze_messages(messages, group_id) + # 进行分析 - 传递 unified_msg_origin 以获取正确的 LLM 提供商 + analysis_result = await message_analyzer.analyze_messages(messages, group_id, event.unified_msg_origin) # 检查分析结果 if not analysis_result or not analysis_result.get("statistics"): diff --git a/src/analysis/llm_analyzer.py b/src/analysis/llm_analyzer.py index a3f9cf7..aada56f 100644 --- a/src/analysis/llm_analyzer.py +++ b/src/analysis/llm_analyzer.py @@ -81,8 +81,6 @@ class LLMAnalyzer: # 确保使用当前指定的模型 if provider is None: provider = self.context.get_using_provider(umo=umo) - - # 安全地获取 provider ID provider_id = 'unknown' if provider: try: @@ -90,10 +88,7 @@ class LLMAnalyzer: provider_id = meta.id except Exception as e: logger.debug(f"获取提供商ID失败: {e}") - logger.info(f"获取到的 provider ID: {provider_id}") - - # 如果获取失败，尝试使用全局默认提供商或者回退方案 if not provider or provider_id == 'unknown': logger.warning(f"获取的提供商不正确 (Provider ID: {provider_id})") @@ -118,7 +113,7 @@ class LLMAnalyzer: logger.error(f"LLM请求全部重试失败: {last_exc}") return None - async def analyze_topics(self, messages: List[Dict]) -> Tuple[List[SummaryTopic], TokenUsage]: + async def analyze_topics(self, messages: List[Dict], umo: str = None) -> Tuple[List[SummaryTopic], TokenUsage]: """使用LLM分析话题""" try: # 提取文本消息 @@ -203,12 +198,7 @@ class LLMAnalyzer: """ # 调用LLM - provider = self.context.get_using_provider() - if not provider: - logger.warning("未配置LLM提供商，跳过话题分析") - return [], TokenUsage() - - response = await self._call_provider_with_retry(provider, prompt, max_tokens=10000, temperature=0.6) + response = await self._call_provider_with_retry(None, prompt, max_tokens=10000, temperature=0.6, umo=umo) if response is None: logger.error("话题分析调用LLM失败: provider返回None（重试失败）") return [], TokenUsage() @@ -359,7 +349,7 @@ class LLMAnalyzer: logger.error(f"正则表达式提取失败: {e}") return [] - async def analyze_user_titles(self, messages: List[Dict], user_analysis: Dict) -> Tuple[List[UserTitle], TokenUsage]: + async def analyze_user_titles(self, messages: List[Dict], user_analysis: Dict, umo: str = None) -> Tuple[List[UserTitle], TokenUsage]: """使用LLM分析用户称号""" try: # 准备用户数据 @@ -430,12 +420,7 @@ class LLMAnalyzer: """ # 调用LLM - provider = self.context.get_using_provider() - if not provider: - logger.warning("未配置LLM提供商，跳过用户称号分析") - return [], TokenUsage() - - response = await self._call_provider_with_retry(provider, prompt, max_tokens=1500, temperature=0.5) + response = await self._call_provider_with_retry(None, prompt, max_tokens=1500, temperature=0.5, umo=umo) if response is None: logger.error("用户称号分析调用LLM失败: provider返回None（重试失败）") return [], TokenUsage() @@ -477,7 +462,7 @@ class LLMAnalyzer: logger.error(f"用户称号分析失败: {e}") return [], TokenUsage() - async def analyze_golden_quotes(self, messages: List[Dict]) -> Tuple[List[GoldenQuote], TokenUsage]: + async def analyze_golden_quotes(self, messages: List[Dict], umo: str = None) -> Tuple[List[GoldenQuote], TokenUsage]: """使用LLM分析群聊金句""" try: # 提取有趣的文本消息 @@ -538,12 +523,7 @@ class LLMAnalyzer: """ # 调用LLM - provider = self.context.get_using_provider() - if not provider: - logger.warning("未配置LLM提供商，跳过金句分析") - return [], TokenUsage() - - response = await self._call_provider_with_retry(provider, prompt, max_tokens=1500, temperature=0.7) + response = await self._call_provider_with_retry(None, prompt, max_tokens=1500, temperature=0.7, umo=umo) if response is None: logger.error("金句分析调用LLM失败: provider返回None（重试失败）") return [], TokenUsage() diff --git a/src/scheduler/auto_scheduler.py b/src/scheduler/auto_scheduler.py index b06d7b4..585e606 100644 --- a/src/scheduler/auto_scheduler.py +++ b/src/scheduler/auto_scheduler.py @@ -30,6 +30,21 @@ class AutoScheduler: """设置bot QQ号（保持向后兼容）""" self.bot_manager.set_bot_qq_id(bot_qq_id) + def _get_platform_id(self): + """获取平台ID""" + try: + if hasattr(self.bot_manager, '_context') and self.bot_manager._context: + context = self.bot_manager._context + if hasattr(context, 'platform_manager') and hasattr(context.platform_manager, 'platform_insts'): + platforms = context.platform_manager.platform_insts + for platform in platforms: + if hasattr(platform, 'metadata') and hasattr(platform.metadata, 'id'): + platform_id = platform.metadata.id + return platform_id + return "aiocqhttp" # 默认值 + except Exception as e: + return "aiocqhttp" # 默认值 + async def start_scheduler(self): """启动定时任务调度器""" if not self.config_manager.get_enable_auto_analysis(): @@ -144,8 +159,10 @@ class AutoScheduler: logger.info(f"群 {group_id} 获取到 {len(messages)} 条消息，开始分析") - # 进行分析 - analysis_result = await self.analyzer.analyze_messages(messages, group_id) + # 进行分析 - 构造正确的 unified_msg_origin + platform_id = self._get_platform_id() + umo = f"{platform_id}:group:{group_id}" if platform_id else None + analysis_result = await self.analyzer.analyze_messages(messages, group_id, umo) if not analysis_result: logger.error(f"群 {group_id} 分析失败") return diff --git a/src/utils/helpers.py b/src/utils/helpers.py index eba4272..0eefd9b 100644 --- a/src/utils/helpers.py +++ b/src/utils/helpers.py @@ -28,7 +28,7 @@ class MessageAnalyzer: else: await self.message_handler.set_bot_qq_id(bot_instance) - async def analyze_messages(self, messages: List[Dict], group_id: str) -> Dict: + async def analyze_messages(self, messages: List[Dict], group_id: str, unified_msg_origin: str = None) -> Dict: """完整的消息分析流程""" try: # 基础统计 @@ -45,20 +45,20 @@ class MessageAnalyzer: # 话题分析 if self.config_manager.get_topic_analysis_enabled(): - topics, topic_tokens = await self.llm_analyzer.analyze_topics(messages) + topics, topic_tokens = await self.llm_analyzer.analyze_topics(messages, umo=unified_msg_origin) total_token_usage.prompt_tokens += topic_tokens.prompt_tokens total_token_usage.completion_tokens += topic_tokens.completion_tokens total_token_usage.total_tokens += topic_tokens.total_tokens # 用户称号分析 if self.config_manager.get_user_title_analysis_enabled(): - user_titles, title_tokens = await self.llm_analyzer.analyze_user_titles(messages, user_analysis) + user_titles, title_tokens = await self.llm_analyzer.analyze_user_titles(messages, user_analysis, umo=unified_msg_origin) total_token_usage.prompt_tokens += title_tokens.prompt_tokens total_token_usage.completion_tokens += title_tokens.completion_tokens total_token_usage.total_tokens += title_tokens.total_tokens # 金句分析 - golden_quotes, quote_tokens = await self.llm_analyzer.analyze_golden_quotes(messages) + golden_quotes, quote_tokens = await self.llm_analyzer.analyze_golden_quotes(messages, umo=unified_msg_origin) total_token_usage.prompt_tokens += quote_tokens.prompt_tokens total_token_usage.completion_tokens += quote_tokens.completion_tokens total_token_usage.total_tokens += quote_tokens.total_tokens