fix: 修复了导致增量分析报告发送失败的几个核心问题:

修复报告分发器初始化错误:
在 AutoScheduler 中,初始化 ReportDispatcher 时漏传了 report_generator,导致在尝试生成报告(图片或文本)时出现 'NoneType' object has no attribute 'generate_image_report' 错误。
已更新 AutoScheduler 的构造函数并正确传递 report_generator。
修复 LLM 分析器在增量模式下的数据结构错误:UserTitleAnalyzer 报错:在增量模式下,用户活动数据被简化存储,导致 UserTitleAnalyzer 找不到 'hours' 键(Error: 'hours')。
补全数据维度:更新了 AnalysisApplicationService 的增量数据转换逻辑,保留了完整的每小时活跃分布 (hours) 和回复数 (reply_count)。
统一字段名:增量模式下使用了 nickname 字段以匹配分析器的预期(之前被错误转为了 name)。
增强容错性:优化了 UserTitleAnalyzer 的逻辑,现在它能安全地处理 hours 缺失或旧版本数据 schema,不再会因为找不到键而崩溃。
修复增量数据合并逻辑:
更新了 IncrementalMergeService,现在它能正确合并多个批次中的每小时发言统计和回复数统计,确保最终生成的报告数据准确。
This commit is contained in:
SXP-Simon
2026-02-11 12:54:02 +08:00
parent e292558965
commit ced38e6a9e
5 changed files with 81 additions and 57 deletions
+1
View File
@@ -78,6 +78,7 @@ class QQGroupDailyAnalysis(Star):
self.analysis_service, self.analysis_service,
self.bot_manager, self.bot_manager,
self.retry_manager, self.retry_manager,
self.report_generator,
self.html_render, self.html_render,
) )
@@ -250,9 +250,7 @@ class AnalysisApplicationService:
if last_analyzed_ts > 0: if last_analyzed_ts > 0:
unified_messages = [ unified_messages = [
msg msg for msg in unified_messages if msg.timestamp > last_analyzed_ts
for msg in unified_messages
if msg.timestamp > last_analyzed_ts
] ]
# 5. 检查最小消息阈值 # 5. 检查最小消息阈值
@@ -282,7 +280,7 @@ class AnalysisApplicationService:
# 7. LLM 增量分析(仅话题 + 金句) # 7. LLM 增量分析(仅话题 + 金句)
topics_per_batch = self.config_manager.get_incremental_topics_per_batch() topics_per_batch = self.config_manager.get_incremental_topics_per_batch()
quotes_per_batch = self.config_manager.get_incremental_quotes_per_batch() quotes_per_batch = self.config_manager.get_incremental_quotes_per_batch()
# 获取功能开关状态 # 获取功能开关状态
topic_enabled = self.config_manager.get_topic_analysis_enabled() topic_enabled = self.config_manager.get_topic_analysis_enabled()
golden_quote_enabled = self.config_manager.get_golden_quote_analysis_enabled() golden_quote_enabled = self.config_manager.get_golden_quote_analysis_enabled()
@@ -295,15 +293,17 @@ class AnalysisApplicationService:
f"{platform_id}:GroupMessage:{group_id}" if platform_id else group_id f"{platform_id}:GroupMessage:{group_id}" if platform_id else group_id
) )
topics, golden_quotes, token_usage = ( (
await self.llm_analyzer.analyze_incremental_concurrent( topics,
legacy_messages, golden_quotes,
umo=unified_msg_origin, token_usage,
topics_per_batch=topics_per_batch, ) = await self.llm_analyzer.analyze_incremental_concurrent(
quotes_per_batch=quotes_per_batch, legacy_messages,
topic_enabled=topic_enabled, umo=unified_msg_origin,
golden_quote_enabled=golden_quote_enabled, topics_per_batch=topics_per_batch,
) quotes_per_batch=quotes_per_batch,
topic_enabled=topic_enabled,
golden_quote_enabled=golden_quote_enabled,
) )
# 8. 构建 IncrementalBatch # 8. 构建 IncrementalBatch
@@ -441,9 +441,7 @@ class AnalysisApplicationService:
# 3. 检查批次有效性 # 3. 检查批次有效性
if not batches: if not batches:
logger.warning( logger.warning(f"{group_id} 滑动窗口内无增量分析数据,无法生成最终报告")
f"{group_id} 滑动窗口内无增量分析数据,无法生成最终报告"
)
return {"success": False, "reason": "no_incremental_data"} return {"success": False, "reason": "no_incremental_data"}
# 4. 合并批次为 IncrementalState # 4. 合并批次为 IncrementalState
@@ -466,19 +464,18 @@ class AnalysisApplicationService:
top_users = state.get_user_activity_ranking(max_user_titles) top_users = state.get_user_activity_ranking(max_user_titles)
unified_msg_origin = ( unified_msg_origin = (
f"{platform_id}:GroupMessage:{group_id}" f"{platform_id}:GroupMessage:{group_id}" if platform_id else group_id
if platform_id
else group_id
) )
try: try:
user_titles_result, title_token_usage = ( (
await self.llm_analyzer.analyze_user_titles( user_titles_result,
messages=[], # 增量模式下不传原始消息 title_token_usage,
user_analysis=state.user_activities, ) = await self.llm_analyzer.analyze_user_titles(
umo=unified_msg_origin, messages=[], # 增量模式下不传原始消息
top_users=top_users, user_analysis=state.user_activities,
) umo=unified_msg_origin,
top_users=top_users,
) )
user_titles = user_titles_result user_titles = user_titles_result
@@ -579,11 +576,14 @@ class AnalysisApplicationService:
result: dict[str, dict] = {} result: dict[str, dict] = {}
for user_id, stats in user_activity.items(): for user_id, stats in user_activity.items():
result[user_id] = { result[user_id] = {
"name": stats.get("nickname", user_id), "nickname": stats.get("nickname", user_id),
"message_count": stats.get("message_count", 0), "message_count": stats.get("message_count", 0),
"char_count": stats.get("char_count", 0), "char_count": stats.get("char_count", 0),
"emoji_count": stats.get("emoji_count", 0), "emoji_count": stats.get("emoji_count", 0),
"active_hours": list(stats.get("hours", {}).keys()), "reply_count": stats.get("reply_count", 0),
"hours": dict(
stats.get("hours", {})
), # 这里的 hours 是 defaultdict(int),转为 dict
"last_message_time": user_last_time.get(user_id, 0), "last_message_time": user_last_time.get(user_id, 0),
} }
@@ -86,29 +86,46 @@ class IncrementalMergeService:
for user_id, stats in batch.user_stats.items(): for user_id, stats in batch.user_stats.items():
if user_id not in state.user_activities: if user_id not in state.user_activities:
state.user_activities[user_id] = { state.user_activities[user_id] = {
"name": stats.get("name", user_id), "nickname": stats.get("nickname", stats.get("name", user_id)),
"message_count": 0, "message_count": 0,
"char_count": 0, "char_count": 0,
"emoji_count": 0, "emoji_count": 0,
"active_hours": [], "reply_count": 0,
"hours": {},
"last_message_time": 0, "last_message_time": 0,
} }
existing = state.user_activities[user_id] existing = state.user_activities[user_id]
existing["message_count"] += stats.get("message_count", 0) existing["message_count"] += stats.get("message_count", 0)
existing["char_count"] += stats.get("char_count", 0) existing["char_count"] += stats.get("char_count", 0)
existing["emoji_count"] += stats.get("emoji_count", 0) existing["emoji_count"] += stats.get("emoji_count", 0)
# 合并活跃小时(去重) existing["reply_count"] += stats.get("reply_count", 0)
existing_hours = set(existing.get("active_hours", []))
existing_hours.update(stats.get("active_hours", [])) # 合并每小时统计
existing["active_hours"] = list(existing_hours) # 兼容旧版本 (active_hours 是 list) 和新版本 (hours 是 dict)
batch_hours = stats.get("hours", {})
if isinstance(batch_hours, dict):
# 现代 schema: hours 是 dict {hour: count}
for h_str, h_count in batch_hours.items():
h_int = int(h_str)
existing["hours"][h_int] = (
existing["hours"].get(h_int, 0) + h_count
)
else:
# 兼容旧 schema: 只有 active_hours (list)
active_hours = stats.get("active_hours", [])
for h in active_hours:
h_int = int(h)
existing["hours"][h_int] = existing["hours"].get(h_int, 0) + 1
# 取最后消息时间的较大值 # 取最后消息时间的较大值
batch_last = stats.get("last_message_time", 0) batch_last = stats.get("last_message_time", 0)
if batch_last > existing.get("last_message_time", 0): if batch_last > existing.get("last_message_time", 0):
existing["last_message_time"] = batch_last existing["last_message_time"] = batch_last
# 更新昵称(使用最新批次的昵称) # 更新昵称(使用最新批次的昵称)
name = stats.get("name", "") nickname = stats.get("nickname", stats.get("name", ""))
if name: if nickname:
existing["name"] = name existing["nickname"] = nickname
# 合并表情统计(按键累加) # 合并表情统计(按键累加)
for emoji_key, count in batch.emoji_stats.items(): for emoji_key, count in batch.emoji_stats.items():
@@ -187,32 +187,36 @@ class UserTitleAnalyzer(BaseAnalyzer):
continue continue
# 分析用户特征 (此处已基于已清理的 stats) # 分析用户特征 (此处已基于已清理的 stats)
night_messages = sum(stats["hours"][h] for h in range(6)) # 兼容性处理:优先使用 hours (dict),如果没有则尝试从消息推断或使用空
avg_chars = ( hours_data = stats.get("hours")
stats["char_count"] / stats["message_count"] if hours_data is None:
if stats["message_count"] > 0 # 尝试兼容旧 schema 或简化版
else 0 active_hours = stats.get("active_hours", [])
) hours_data = dict.fromkeys(active_hours, 1)
# 安全计算夜间发言数
night_messages = sum(hours_data.get(h, 0) for h in range(6))
message_count = stats.get("message_count", 0)
if message_count <= 0:
continue
avg_chars = stats.get("char_count", 0) / message_count
# 称号所需维度
user_summaries.append( user_summaries.append(
{ {
"name": stats["nickname"], "name": stats.get("nickname", stats.get("name", user_id_str)),
"user_id": user_id_str, "user_id": user_id_str,
"message_count": stats["message_count"], "message_count": message_count,
"avg_chars": round(avg_chars, 1), "avg_chars": round(avg_chars, 1),
"emoji_ratio": round( "emoji_ratio": round(
stats["emoji_count"] / stats["message_count"], 2 stats.get("emoji_count", 0) / message_count, 2
) ),
if stats["message_count"] > 0 "night_ratio": round(night_messages / message_count, 2),
else 0,
"night_ratio": round(night_messages / stats["message_count"], 2)
if stats["message_count"] > 0
else 0,
"reply_ratio": round( "reply_ratio": round(
stats["reply_count"] / stats["message_count"], 2 stats.get("reply_count", 0) / message_count, 2
) ),
if stats["message_count"] > 0
else 0,
} }
) )
@@ -25,18 +25,20 @@ class AutoScheduler:
analysis_service, analysis_service,
bot_manager, bot_manager,
retry_manager, retry_manager,
report_generator=None,
html_render_func=None, html_render_func=None,
): ):
self.config_manager = config_manager self.config_manager = config_manager
self.analysis_service = analysis_service self.analysis_service = analysis_service
self.bot_manager = bot_manager self.bot_manager = bot_manager
self.retry_manager = retry_manager self.retry_manager = retry_manager
self.report_generator = report_generator
self.html_render_func = html_render_func self.html_render_func = html_render_func
# 初始化核心组件 # 初始化核心组件
self.message_sender = MessageSender(bot_manager, config_manager, retry_manager) self.message_sender = MessageSender(bot_manager, config_manager, retry_manager)
self.report_dispatcher = ReportDispatcher( self.report_dispatcher = ReportDispatcher(
config_manager, None, self.message_sender, retry_manager config_manager, report_generator, self.message_sender, retry_manager
) )
if html_render_func: if html_render_func:
self.report_dispatcher.set_html_render(html_render_func) self.report_dispatcher.set_html_render(html_render_func)