1033 lines
45 KiB
Python
1033 lines
45 KiB
Python
from flask import Blueprint, render_template, jsonify, request, current_app, redirect
|
||
from .auth import login_required
|
||
from loguru import logger
|
||
from utils.robot_cmd.robot_command import GroupBotManager, Feature, PermissionStatus
|
||
from datetime import datetime, timedelta
|
||
|
||
from db.base import BaseDBOperator
|
||
from plugins.ai_auto_response.memory.group_memory_profile import GroupMemoryService
|
||
|
||
# 创建群管理兼容蓝图（旧入口保留做跳转，实际功能已并入通讯录）
|
||
robot_bp = Blueprint('robot', __name__, url_prefix='/robot')
|
||
LOG = logger
|
||
|
||
|
||
def _serialize_login_qr_state(server) -> dict:
|
||
"""把 Robot 中的二维码登录态整理成 Dashboard 接口输出。"""
|
||
robot = getattr(server, "robot", None)
|
||
if robot is None or not hasattr(robot, "get_ipad_login_qr_state"):
|
||
return {
|
||
"logged_in": False,
|
||
"active": False,
|
||
"status": "unavailable",
|
||
"provider_name": "",
|
||
"provider_stage": "status_unavailable",
|
||
"connection_ready": False,
|
||
"login_required": False,
|
||
"status_text": "机器人运行态暂不可用",
|
||
"current": {},
|
||
"history": [],
|
||
"server_now": datetime.now().timestamp(),
|
||
}
|
||
|
||
state = robot.get_ipad_login_qr_state() or {}
|
||
return {
|
||
"logged_in": bool(state.get("logged_in", False)),
|
||
"active": bool(state.get("active", False)),
|
||
"status": str(state.get("status", "idle") or "idle"),
|
||
"provider_name": str(state.get("provider_name", "") or ""),
|
||
"provider_stage": str(state.get("provider_stage", "bootstrap") or "bootstrap"),
|
||
"connection_ready": bool(state.get("connection_ready", False)),
|
||
"login_required": bool(state.get("login_required", False)),
|
||
"status_text": str(state.get("status_text", "尚未进入扫码登录流程") or "尚未进入扫码登录流程"),
|
||
"runtime_running": bool(state.get("runtime_running", False)),
|
||
"wxid": str(state.get("wxid", "") or ""),
|
||
"nickname": str(state.get("nickname", "") or ""),
|
||
"updated_at": float(state.get("updated_at", 0) or 0),
|
||
"server_now": float(state.get("server_now", datetime.now().timestamp()) or datetime.now().timestamp()),
|
||
"current": dict(state.get("current", {}) or {}),
|
||
"history": [dict(item or {}) for item in (state.get("history", []) or [])],
|
||
}
|
||
|
||
|
||
def _build_group_ops_profile(server, group_id: str, group_name: str, peak_hours: list, plugin_stats: list) -> dict:
|
||
"""构建群运营 2.0 所需的群画像摘要。
|
||
|
||
设计说明：
|
||
1. 这里优先复用现有群画像快照与消息总结能力，不重新发明一套分析链路；
|
||
2. 若历史画像数据不足，也必须返回可展示的兜底结构，避免前端出现大片空白；
|
||
3. 输出结构专门面向后台运营视图，强调“群是什么群、最近聊什么、适合怎么运营”。
|
||
"""
|
||
profile_service = GroupMemoryService(server.db_manager, {})
|
||
profile = profile_service.build_group_memory_profile(group_id, group_name=group_name) or {}
|
||
|
||
style_profile = profile.get("style_profile", {}) or {}
|
||
focus_topics = [str(item).strip() for item in (profile.get("focus_topics") or []) if str(item).strip()]
|
||
structured_summary = profile.get("structured_summary", {}) or {}
|
||
recent_key_points = [
|
||
str(item).strip() for item in (structured_summary.get("recent_key_points") or []) if str(item).strip()
|
||
][:4]
|
||
unresolved_points = [
|
||
str(item).strip() for item in (structured_summary.get("unresolved_points") or []) if str(item).strip()
|
||
][:3]
|
||
stable_topics = [
|
||
str(item).strip() for item in (structured_summary.get("stable_topics") or []) if str(item).strip()
|
||
][:4]
|
||
|
||
peak_hour_labels = [str(item.get("label") or "").strip() for item in peak_hours if str(item.get("label") or "").strip()]
|
||
dominant_plugin_names = []
|
||
for item in plugin_stats[:3]:
|
||
plugin_name = str(item.get("plugin_name") or "").strip()
|
||
if plugin_name and plugin_name not in dominant_plugin_names:
|
||
dominant_plugin_names.append(plugin_name)
|
||
|
||
# 群定位摘要优先使用群画像服务的领域判断，再结合真实主题给出一段更像“运营结论”的描述。
|
||
inferred_domain = str(profile.get("inferred_domain") or "general").strip()
|
||
domain_label_map = {
|
||
"openclaw": "工作流 / 编排讨论型群",
|
||
"robotics": "机器人 / 自动化协作型群",
|
||
"dota": "Dota / 游戏讨论型群",
|
||
"tech": "技术排障 / 工具协作型群",
|
||
"casual": "熟人闲聊 / 日常水群型群",
|
||
"general": "综合交流型群",
|
||
}
|
||
group_identity = domain_label_map.get(inferred_domain, "综合交流型群")
|
||
|
||
profile_tags = [
|
||
tag for tag in [
|
||
group_identity,
|
||
str(style_profile.get("interaction_tone") or "").strip(),
|
||
str(style_profile.get("humor_style") or "").strip(),
|
||
str(style_profile.get("expressiveness_style") or "").strip(),
|
||
] if tag
|
||
]
|
||
|
||
summary_lines = []
|
||
if focus_topics:
|
||
summary_lines.append(f"近期高频主题集中在：{'、'.join(focus_topics[:4])}")
|
||
if peak_hour_labels:
|
||
summary_lines.append(f"活跃时段主要落在：{'、'.join(peak_hour_labels[:2])}")
|
||
if dominant_plugin_names:
|
||
summary_lines.append(f"当前最容易被接受的机器人能力偏向：{'、'.join(dominant_plugin_names[:2])}")
|
||
if not summary_lines:
|
||
summary_lines.append("当前历史数据仍偏少，建议先继续积累群消息、总结与插件使用记录。")
|
||
|
||
return {
|
||
"group_identity": group_identity,
|
||
"profile_tags": profile_tags,
|
||
"focus_topics": focus_topics[:6],
|
||
"stable_topics": stable_topics,
|
||
"recent_key_points": recent_key_points,
|
||
"unresolved_points": unresolved_points,
|
||
"summary_text": "；".join(summary_lines),
|
||
"interaction_tone": str(style_profile.get("interaction_tone") or "").strip(),
|
||
"humor_style": str(style_profile.get("humor_style") or "").strip(),
|
||
"expressiveness_style": str(style_profile.get("expressiveness_style") or "").strip(),
|
||
"peak_hours": peak_hour_labels[:3],
|
||
"dominant_plugins": dominant_plugin_names[:3],
|
||
"cache_status": str(profile.get("cache_status") or "").strip(),
|
||
"last_generated_at": str(profile.get("last_generated_at") or "").strip(),
|
||
}
|
||
|
||
|
||
def _load_recent_group_summaries(server, group_id: str, limit: int = 3) -> list:
|
||
"""读取最近几条群总结，供运营面板直接展示。
|
||
|
||
设计说明：
|
||
1. 群画像适合给“结论”，但运营者往往还想看到最近几期总结到底说了什么；
|
||
2. 默认卡片仍使用短摘要，避免详情页被大段文本挤占；
|
||
3. 同时补充完整总结正文，供前端弹窗按需展开查看；
|
||
4. 若没有总结记录，返回空列表，由前端优雅降级。
|
||
"""
|
||
summary_db = getattr(server, "message_summary_db", None)
|
||
if summary_db is None:
|
||
return []
|
||
|
||
rows = summary_db.execute_query(
|
||
"""
|
||
SELECT period_key, summary_type, source_message_count, summary_text, last_generated_at
|
||
FROM t_message_summary
|
||
WHERE chatroom_id = %s
|
||
ORDER BY period_end DESC, update_time DESC
|
||
LIMIT %s
|
||
""",
|
||
(group_id, int(limit)),
|
||
) or []
|
||
|
||
result = []
|
||
for row in rows:
|
||
raw_summary = str(row.get("summary_text") or "").strip()
|
||
normalized_summary = " ".join(raw_summary.split())
|
||
result.append({
|
||
"period_key": str(row.get("period_key") or "").strip(),
|
||
"summary_type": str(row.get("summary_type") or "").strip(),
|
||
"source_message_count": int(row.get("source_message_count") or 0),
|
||
"last_generated_at": str(row.get("last_generated_at") or "").strip(),
|
||
# 详情弹窗需要查看完整总结正文；这里保留原文，不做截断。
|
||
"summary_text": raw_summary,
|
||
# 卡片区域继续展示压缩后的摘要，避免详情首屏信息密度过高。
|
||
"summary_excerpt": normalized_summary[:180] + ("..." if len(normalized_summary) > 180 else ""),
|
||
})
|
||
return result
|
||
|
||
|
||
def _load_value_rank_top_members(server, group_id: str, limit: int = 5) -> list:
|
||
"""读取群内最近一期高价值成员快照。
|
||
|
||
说明：
|
||
1. 该榜单直接复用已落库的身价快照，而不是运行时重新算分；
|
||
2. 若群里尚未开启或尚未产出这类数据，则静默返回空列表；
|
||
3. 返回结构会补齐成员展示名，避免前端出现一串 wxid。
|
||
"""
|
||
db = BaseDBOperator(server.db_manager)
|
||
latest_row = db.execute_query(
|
||
"""
|
||
SELECT MAX(stat_date) AS latest_stat_date
|
||
FROM t_value_rank_snapshot
|
||
WHERE group_id = %s
|
||
""",
|
||
(group_id,),
|
||
fetch_one=True,
|
||
) or {}
|
||
latest_stat_date = latest_row.get("latest_stat_date")
|
||
if not latest_stat_date:
|
||
return []
|
||
|
||
rows = db.execute_query(
|
||
"""
|
||
SELECT user_id, score, rank_no, title, points_total, msg_count_7d, active_days_30, inactive_days
|
||
FROM t_value_rank_snapshot
|
||
WHERE group_id = %s AND stat_date = %s
|
||
ORDER BY rank_no ASC
|
||
LIMIT %s
|
||
""",
|
||
(group_id, latest_stat_date, int(limit)),
|
||
) or []
|
||
|
||
result = []
|
||
for row in rows:
|
||
user_id = str(row.get("user_id") or "").strip()
|
||
result.append({
|
||
"wxid": user_id,
|
||
"display_name": server.contact_manager.get_group_name(group_id, user_id) or user_id,
|
||
"score": float(row.get("score") or 0.0),
|
||
"rank_no": int(row.get("rank_no") or 0),
|
||
"title": str(row.get("title") or "").strip(),
|
||
"points_total": int(row.get("points_total") or 0),
|
||
"msg_count_7d": int(row.get("msg_count_7d") or 0),
|
||
"active_days_30": int(row.get("active_days_30") or 0),
|
||
"inactive_days": int(row.get("inactive_days") or 0),
|
||
})
|
||
return result
|
||
|
||
|
||
def _load_activation_candidates(server, group_id: str, limit: int = 6) -> list:
|
||
"""识别“待激活成员”。
|
||
|
||
口径选择：
|
||
1. 不直接拿最沉默的人，因为那类成员很多已经处于长期静默，激活收益未必最高；
|
||
2. 这里优先找“7~30 天未发言，但仍不是完全流失”的边缘成员，更适合做召回动作；
|
||
3. 这类成员通常对低门槛互动、固定时段提醒更敏感，运营价值更高。
|
||
"""
|
||
db = BaseDBOperator(server.db_manager)
|
||
rows = db.execute_query(
|
||
"""
|
||
SELECT
|
||
wxid,
|
||
COALESCE(NULLIF(display_name, ''), nick_name, wxid) AS display_name,
|
||
latest_active_time,
|
||
TIMESTAMPDIFF(DAY, latest_active_time, NOW()) AS inactivity_days
|
||
FROM t_chatroom_member
|
||
WHERE chatroom_id = %s
|
||
AND status = 1
|
||
AND latest_active_time IS NOT NULL
|
||
AND latest_active_time < DATE_SUB(NOW(), INTERVAL 7 DAY)
|
||
AND latest_active_time >= DATE_SUB(NOW(), INTERVAL 30 DAY)
|
||
ORDER BY latest_active_time DESC
|
||
LIMIT %s
|
||
""",
|
||
(group_id, int(limit)),
|
||
) or []
|
||
|
||
result = []
|
||
for row in rows:
|
||
result.append({
|
||
"wxid": str(row.get("wxid") or "").strip(),
|
||
"display_name": str(row.get("display_name") or "").strip(),
|
||
"latest_active_time": str(row.get("latest_active_time") or "").strip(),
|
||
"inactivity_days": int(row.get("inactivity_days") or 0),
|
||
})
|
||
return result
|
||
|
||
|
||
def _build_ops_member_segments(server, group_id: str, speaker_ranking: list, inactive_members: list) -> dict:
|
||
"""构建群成员分层结果。
|
||
|
||
设计目标：
|
||
1. 保留原有“沉默成员 / 发言排行”表格；
|
||
2. 在此基础上额外产出“核心成员 / 高价值成员 / 待激活成员”三类运营视角；
|
||
3. 前端展示时以卡片为主，让运营者更快找到“该看谁”。
|
||
"""
|
||
value_rank_members = _load_value_rank_top_members(server, group_id, limit=5)
|
||
activation_candidates = _load_activation_candidates(server, group_id, limit=6)
|
||
member_context_map = {
|
||
str(item.get("wxid") or "").strip(): item
|
||
for item in (server.member_context_db.list_group_member_contexts(group_id) or [])
|
||
}
|
||
|
||
core_members = []
|
||
for item in speaker_ranking[:5]:
|
||
wxid = str(item.get("sender") or "").strip()
|
||
context = member_context_map.get(wxid, {}) or {}
|
||
core_members.append({
|
||
"wxid": wxid,
|
||
"display_name": str(item.get("display_name") or wxid).strip(),
|
||
"message_count": int(item.get("message_count") or 0),
|
||
"last_message_time": str(item.get("last_message_time") or "").strip(),
|
||
"activity_level": str(context.get("activity_level") or "").strip(),
|
||
"summary_text": str(context.get("summary_text") or "").strip(),
|
||
})
|
||
|
||
dormant_members = []
|
||
for item in inactive_members[:5]:
|
||
wxid = str(item.get("wxid") or "").strip()
|
||
context = member_context_map.get(wxid, {}) or {}
|
||
dormant_members.append({
|
||
"wxid": wxid,
|
||
"display_name": str(item.get("display_name") or wxid).strip(),
|
||
"inactivity_days": int(item.get("inactivity_days") or 0),
|
||
"latest_active_time": str(item.get("latest_active_time") or "").strip(),
|
||
"response_style_hint": str(context.get("response_style_hint") or "").strip(),
|
||
})
|
||
|
||
return {
|
||
"core_members": core_members,
|
||
"value_rank_members": value_rank_members,
|
||
"activation_candidates": activation_candidates,
|
||
"dormant_members": dormant_members,
|
||
}
|
||
|
||
|
||
def _build_ops_action_cards(group_name: str, ops_profile: dict, ops_members: dict, plugin_stats: list) -> list:
|
||
"""生成更偏“可执行动作”的运营建议卡片。
|
||
|
||
与旧版 `operation_suggestions` 的区别：
|
||
1. 旧版更像诊断提示；新版更强调“下一步怎么做”；
|
||
2. 仍然采用规则组合，避免引入高成本、不可控的 LLM 依赖；
|
||
3. 结构尽量稳定，方便前端做卡片展示与后续继续增强。
|
||
"""
|
||
action_cards = []
|
||
peak_hours = ops_profile.get("peak_hours") or []
|
||
focus_topics = ops_profile.get("focus_topics") or []
|
||
dominant_plugins = ops_profile.get("dominant_plugins") or []
|
||
activation_candidates = ops_members.get("activation_candidates") or []
|
||
core_members = ops_members.get("core_members") or []
|
||
|
||
if peak_hours:
|
||
action_cards.append({
|
||
"type": "time",
|
||
"title": "建议把互动动作放到高峰时段",
|
||
"summary": f"优先考虑在 {peak_hours[0]} 发起签到、话题或菜单引导。",
|
||
"detail": f"{group_name} 最近高峰主要集中在 {'、'.join(peak_hours[:2])}，在这个时间段发起动作更容易拿到首轮反馈。",
|
||
})
|
||
|
||
if dominant_plugins:
|
||
action_cards.append({
|
||
"type": "feature",
|
||
"title": "建议优先强化已被接受的功能入口",
|
||
"summary": f"先围绕 {dominant_plugins[0]} 做菜单、公告或欢迎语曝光。",
|
||
"detail": f"当前群里真实被用起来的功能更偏向 {'、'.join(dominant_plugins[:3])}，说明这类能力更容易被接受，适合作为下一阶段主推入口。",
|
||
})
|
||
|
||
if activation_candidates:
|
||
candidate_names = [str(item.get("display_name") or "").strip() for item in activation_candidates[:3] if str(item.get("display_name") or "").strip()]
|
||
action_cards.append({
|
||
"type": "activation",
|
||
"title": "建议优先激活边缘活跃成员",
|
||
"summary": f"可以先针对 {'、'.join(candidate_names) if candidate_names else '近 7~30 天边缘成员'} 做轻量召回。",
|
||
"detail": "这批成员不是长期沉默用户，仍有回流可能，更适合通过低门槛互动、固定提醒或轻福利重新拉回讨论。",
|
||
})
|
||
|
||
if focus_topics:
|
||
action_cards.append({
|
||
"type": "topic",
|
||
"title": "建议围绕当前高频主题设计话题入口",
|
||
"summary": f"优先从 {focus_topics[0]}、{focus_topics[1] if len(focus_topics) > 1 else focus_topics[0]} 这类主题切入。",
|
||
"detail": "与其硬推通用活动，不如顺着群里原本就在讨论的内容做问题、投票、接龙或功能引导，转化更自然。",
|
||
})
|
||
|
||
if core_members:
|
||
core_names = [str(item.get("display_name") or "").strip() for item in core_members[:3] if str(item.get("display_name") or "").strip()]
|
||
action_cards.append({
|
||
"type": "core",
|
||
"title": "建议借核心成员带动扩散",
|
||
"summary": f"可以优先让 {'、'.join(core_names) if core_names else '核心活跃成员'} 参与首轮互动。",
|
||
"detail": "核心成员往往能决定首轮响应速度和讨论氛围，先让他们接住话题，比单纯增加曝光更有效。",
|
||
})
|
||
|
||
if not action_cards:
|
||
action_cards.append({
|
||
"type": "fallback",
|
||
"title": "建议先继续积累群画像与使用数据",
|
||
"summary": "当前数据不足以给出更细的动作建议。",
|
||
"detail": "可以先保持基础运营动作，等群总结、成员画像和插件使用数据更完整后，再做更细的策略推荐。",
|
||
})
|
||
|
||
return action_cards[:4]
|
||
|
||
|
||
def _build_group_list_ops_preview(server, group_id: str, group_name: str) -> dict:
|
||
"""构建群列表层可展示的轻量运营速览。
|
||
|
||
设计原则：
|
||
1. 列表页只放“筛选信号”，不放完整成员明细，避免信息密度过高；
|
||
2. 详情页里已经有完整诊断，因此这里优先输出健康度、核心成员、待激活成员等摘要；
|
||
3. 为了控制接口成本，这里不复用重型画像构建，只做轻量聚合和短摘要生成。
|
||
"""
|
||
now = datetime.now()
|
||
last_30d = now - timedelta(days=30)
|
||
|
||
member_summary = server.contact_db.get_group_member_summary(group_id, inactive_days=30)
|
||
message_stats_30d = server.message_storage.get_message_stats_by_date_range(group_id, last_30d, now)
|
||
plugin_summary = server.stats_db.get_group_plugin_summary(group_id, days=30)
|
||
speaker_ranking = server.message_storage.get_group_member_message_ranking(group_id, last_30d, now, limit=3)
|
||
activation_candidates = _load_activation_candidates(server, group_id, limit=3)
|
||
recent_summaries = _load_recent_group_summaries(server, group_id, limit=1)
|
||
|
||
total_members = member_summary.get("in_group_members") or member_summary.get("total_members") or 0
|
||
active_30d = int(member_summary.get("active_30d_members") or 0)
|
||
zombie_count = int(member_summary.get("inactive_members") or 0)
|
||
msg_30d_total = int(message_stats_30d.get("total_count") or 0)
|
||
plugin_total_calls = int(plugin_summary.get("total_calls") or 0)
|
||
|
||
# 列表页也复用和详情页一致的健康度口径：
|
||
# 1. 这样用户从列表点进详情时，数值语义不会跳变；
|
||
# 2. 逻辑保持一套，后续调分只改一处思路更稳；
|
||
# 3. 列表上只展示分值，不额外塞入复杂推导过程。
|
||
health_score = 100
|
||
if total_members > 0:
|
||
health_score -= min(35, round((zombie_count / total_members) * 35))
|
||
health_score -= min(25, round((1 - min(active_30d / total_members, 1)) * 25))
|
||
if msg_30d_total == 0:
|
||
health_score -= 25
|
||
elif msg_30d_total < 50:
|
||
health_score -= 12
|
||
if plugin_total_calls == 0:
|
||
health_score -= 10
|
||
health_score = max(0, min(100, health_score))
|
||
|
||
core_member_names = []
|
||
for item in speaker_ranking:
|
||
sender = str(item.get("sender") or "").strip()
|
||
display_name = server.contact_manager.get_group_name(group_id, sender) or sender
|
||
if display_name:
|
||
core_member_names.append(display_name)
|
||
|
||
activation_names = [
|
||
str(item.get("display_name") or "").strip()
|
||
for item in activation_candidates
|
||
if str(item.get("display_name") or "").strip()
|
||
]
|
||
|
||
summary_excerpt = ""
|
||
if recent_summaries:
|
||
summary_excerpt = str(recent_summaries[0].get("summary_excerpt") or "").strip()
|
||
if not summary_excerpt:
|
||
if core_member_names:
|
||
summary_excerpt = f"核心讨论主要由 {'、'.join(core_member_names[:2])} 带动。"
|
||
elif plugin_total_calls > 0:
|
||
summary_excerpt = f"近30天已有 {plugin_total_calls} 次插件调用，可继续做功能渗透。"
|
||
else:
|
||
summary_excerpt = "当前运营数据较少，建议先观察活跃与功能渗透。"
|
||
|
||
return {
|
||
"group_id": group_id,
|
||
"group_name": group_name,
|
||
"health_score": int(health_score),
|
||
"active_member_count_30d": active_30d,
|
||
"zombie_member_count": zombie_count,
|
||
"core_member_names": core_member_names[:3],
|
||
"activation_candidate_count": len(activation_names),
|
||
"activation_candidate_names": activation_names[:3],
|
||
"plugin_call_count_30d": plugin_total_calls,
|
||
"summary_excerpt": summary_excerpt[:120] + ("..." if len(summary_excerpt) > 120 else ""),
|
||
}
|
||
|
||
|
||
# 旧群权限页入口兼容跳转
|
||
@robot_bp.route('/')
|
||
@login_required
|
||
def robot_management():
|
||
return redirect('/contacts')
|
||
|
||
|
||
@robot_bp.route('/api/login_qr_status', methods=['GET'])
|
||
@login_required
|
||
def api_login_qr_status():
|
||
"""返回当前二维码登录状态，供 Dashboard 首页弹窗轮询。"""
|
||
try:
|
||
server = current_app.dashboard_server
|
||
return jsonify({
|
||
"success": True,
|
||
"data": _serialize_login_qr_state(server),
|
||
})
|
||
except Exception as e:
|
||
LOG.error(f"获取登录二维码状态失败: {e}")
|
||
return jsonify({"success": False, "error": str(e)}), 500
|
||
|
||
|
||
# API路由
|
||
@robot_bp.route('/api/groups')
|
||
@login_required
|
||
def api_robot_groups():
|
||
try:
|
||
server = current_app.dashboard_server
|
||
# 获取所有群组列表
|
||
groups = GroupBotManager.get_group_list()
|
||
|
||
# 如果方法返回None或发生异常，使用本地缓存
|
||
if groups is None and hasattr(GroupBotManager, "local_cache"):
|
||
groups = GroupBotManager.local_cache.get("group_list", set())
|
||
|
||
# 如果仍然为None，则初始化为空集合
|
||
if groups is None:
|
||
groups = set()
|
||
|
||
LOG.info(f"获取到 {len(groups)} 个群组")
|
||
group_data = []
|
||
|
||
for group_id in groups:
|
||
try:
|
||
# 获取群名称，如果失败则使用默认值
|
||
group_name = server.contact_manager.get_nickname(group_id)
|
||
if not group_name:
|
||
group_name = f"未知群组({group_id})"
|
||
|
||
# 获取机器人状态，如果失败则使用默认值
|
||
try:
|
||
robot_status = GroupBotManager.get_group_permission(group_id, Feature.ROBOT)
|
||
except:
|
||
robot_status = PermissionStatus.DISABLED
|
||
|
||
group_data.append({
|
||
"group_id": group_id,
|
||
"group_name": group_name,
|
||
"robot_status": robot_status.value if robot_status else "unknown"
|
||
})
|
||
except Exception as e:
|
||
LOG.warning(f"处理群组 {group_id} 信息时出错: {e}")
|
||
# 添加基本信息，避免单个群组错误影响整个列表
|
||
group_data.append({
|
||
"group_id": group_id,
|
||
"group_name": "获取失败",
|
||
"robot_status": "unknown"
|
||
})
|
||
|
||
return jsonify({"success": True, "data": group_data})
|
||
except Exception as e:
|
||
LOG.error(f"获取群组列表失败: {e}")
|
||
return jsonify({"success": False, "error": str(e)}), 500
|
||
|
||
|
||
@robot_bp.route('/api/group/<group_id>/permissions')
|
||
@login_required
|
||
def api_robot_group_permissions(group_id):
|
||
try:
|
||
permissions = GroupBotManager.list_group_permissions(group_id)
|
||
permission_data = []
|
||
|
||
for feature, status in permissions.items():
|
||
permission_data.append({
|
||
"feature_id": feature.value,
|
||
"feature_name": feature.name,
|
||
"feature_description": feature.description,
|
||
"status": status.value
|
||
})
|
||
|
||
return jsonify({"success": True, "data": permission_data})
|
||
except Exception as e:
|
||
LOG.error(f"获取群组权限失败: {e}")
|
||
return jsonify({"success": False, "error": str(e)}), 500
|
||
|
||
|
||
@robot_bp.route('/api/group/<group_id>/permissions', methods=['POST'])
|
||
@login_required
|
||
def api_update_robot_permissions(group_id):
|
||
# 更新群组功能权限
|
||
server = current_app.dashboard_server
|
||
data = request.json
|
||
feature_id = data.get('feature_id')
|
||
status = data.get('status')
|
||
|
||
try:
|
||
feature = Feature(int(feature_id))
|
||
new_status = PermissionStatus(status)
|
||
|
||
# 特殊处理ROBOT功能
|
||
if feature == Feature.ROBOT:
|
||
r = server.db_manager.get_redis_connection()
|
||
if new_status == PermissionStatus.ENABLED:
|
||
GroupBotManager.local_cache["group_list"].add(group_id)
|
||
r.sadd("group:list", group_id)
|
||
else:
|
||
GroupBotManager.local_cache["group_list"].remove(group_id)
|
||
r.srem("group:list", group_id)
|
||
|
||
GroupBotManager.set_group_permission(group_id, feature, new_status)
|
||
return jsonify({"success": True})
|
||
except Exception as e:
|
||
LOG.error(f"更新群组权限失败: {e}")
|
||
return jsonify({"success": False, "error": str(e)}), 400
|
||
|
||
|
||
@robot_bp.route('/api/batch_operation', methods=['POST'])
|
||
@login_required
|
||
def api_robot_batch_operation():
|
||
# 批量操作接口
|
||
data = request.json
|
||
operation = data.get('operation')
|
||
group_ids = data.get('group_ids', [])
|
||
|
||
results = {}
|
||
|
||
try:
|
||
if operation == 'remove_groups':
|
||
for group_id in group_ids:
|
||
result = GroupBotManager.remove_group(group_id)
|
||
results[group_id] = result
|
||
|
||
return jsonify({"success": True, "results": results})
|
||
else:
|
||
return jsonify({"success": False, "error": "不支持的操作类型"}), 400
|
||
except Exception as e:
|
||
LOG.error(f"批量操作失败: {e}")
|
||
return jsonify({"success": False, "error": str(e)}), 400
|
||
|
||
|
||
@robot_bp.route('/api/add_group', methods=['POST'])
|
||
@login_required
|
||
def api_add_group():
|
||
try:
|
||
server = current_app.dashboard_server
|
||
data = request.json
|
||
group_id = data.get('group_id')
|
||
|
||
if not group_id or not group_id.strip():
|
||
return jsonify({"success": False, "error": "群组ID不能为空"}), 400
|
||
|
||
group_id = group_id.strip()
|
||
# 如果group_id 不是@chatroom 结尾，这提示错误
|
||
if not group_id.endswith("@chatroom"):
|
||
return jsonify({"success": False, "error": "群组ID必须以 @chatroom 结尾"}), 400
|
||
|
||
# 检查群组是否已存在
|
||
if group_id in GroupBotManager.local_cache["group_list"]:
|
||
return jsonify({"success": False, "error": "该群组已存在"}), 400
|
||
|
||
# 添加群组到列表并启用机器人功能
|
||
GroupBotManager.local_cache["group_list"].add(group_id)
|
||
r = server.db_manager.get_redis_connection()
|
||
r.sadd("group:list", group_id)
|
||
|
||
# 设置ROBOT功能为启用状态
|
||
GroupBotManager.set_group_permission(group_id, Feature.ROBOT, PermissionStatus.ENABLED)
|
||
|
||
# 获取群组名称（如果可能）
|
||
group_name = server.contact_manager.get_nickname(group_id)
|
||
|
||
return jsonify({
|
||
"success": True,
|
||
"message": f"群组 {group_id} 已成功添加",
|
||
"group": {
|
||
"group_id": group_id,
|
||
"group_name": group_name,
|
||
"robot_status": "enabled"
|
||
}
|
||
})
|
||
except Exception as e:
|
||
LOG.error(f"添加群组失败: {e}")
|
||
return jsonify({"success": False, "error": str(e)}), 500
|
||
|
||
|
||
@robot_bp.route('/api/group/<group_id>/message_trend')
|
||
@login_required
|
||
def api_group_message_trend(group_id):
|
||
try:
|
||
server = current_app.dashboard_server
|
||
days = request.args.get('days', 7, type=int)
|
||
trend_data = server.message_storage.get_message_trend(group_id, days)
|
||
|
||
# 格式化数据为前端需要的格式
|
||
dates = []
|
||
counts = []
|
||
for item in trend_data:
|
||
# 将日期转换为字符串
|
||
if isinstance(item['date'], datetime):
|
||
date_str = item['date'].strftime('%Y-%m-%d')
|
||
else:
|
||
date_str = str(item['date'])
|
||
|
||
dates.append(date_str)
|
||
counts.append(item['message_count'])
|
||
|
||
return jsonify({
|
||
'success': True,
|
||
'data': {
|
||
'dates': dates,
|
||
'counts': counts
|
||
}
|
||
})
|
||
except Exception as e:
|
||
LOG.error(f"获取群组消息趋势数据出错: {e}")
|
||
return jsonify({'success': False, 'error': str(e)}), 500
|
||
|
||
|
||
@robot_bp.route('/api/groups/ops_preview', methods=['POST'])
|
||
@login_required
|
||
def api_groups_ops_preview():
|
||
"""批量获取群列表层的运营速览摘要。"""
|
||
try:
|
||
server = current_app.dashboard_server
|
||
data = request.json or {}
|
||
group_ids = data.get("group_ids") or []
|
||
# 列表批量接口只接受有限数量群组：
|
||
# 1. 避免一次把全量群的详情聚合全算一遍；
|
||
# 2. 通讯录列表本身就有分页/筛选，首版以“当前批次可见群”为主；
|
||
# 3. 这里做硬上限，防止误调用把后台拖慢。
|
||
cleaned_group_ids = []
|
||
for item in group_ids:
|
||
group_id = str(item or "").strip()
|
||
if group_id and group_id not in cleaned_group_ids:
|
||
cleaned_group_ids.append(group_id)
|
||
cleaned_group_ids = cleaned_group_ids[:30]
|
||
|
||
preview_map = {}
|
||
for group_id in cleaned_group_ids:
|
||
group_name = server.contact_manager.get_nickname(group_id) or group_id
|
||
preview_map[group_id] = _build_group_list_ops_preview(server, group_id, group_name)
|
||
|
||
return jsonify({
|
||
"success": True,
|
||
"data": {
|
||
"preview_map": preview_map,
|
||
}
|
||
})
|
||
except Exception as e:
|
||
LOG.error(f"批量获取群运营速览失败: {e}")
|
||
return jsonify({"success": False, "error": str(e)}), 500
|
||
|
||
|
||
@robot_bp.route('/api/group/<group_id>/detail')
|
||
@login_required
|
||
def api_group_detail(group_id):
|
||
"""群组详情聚合接口"""
|
||
try:
|
||
server = current_app.dashboard_server
|
||
now = datetime.now()
|
||
last_24h = now - timedelta(days=1)
|
||
last_7d = now - timedelta(days=7)
|
||
last_30d = now - timedelta(days=30)
|
||
|
||
group_name = server.contact_manager.get_nickname(group_id) or group_id
|
||
permission_map = GroupBotManager.list_group_permissions(group_id)
|
||
enabled_features = [
|
||
feature.description
|
||
for feature, status in permission_map.items()
|
||
if status == PermissionStatus.ENABLED
|
||
]
|
||
|
||
member_summary = server.contact_db.get_group_member_summary(group_id, inactive_days=30)
|
||
inactive_members = server.contact_db.get_inactive_members_rank(group_id, days=30, limit=8)
|
||
message_stats_24h = server.message_storage.get_message_stats_by_date_range(group_id, last_24h, now)
|
||
message_stats_7d = server.message_storage.get_message_stats_by_date_range(group_id, last_7d, now)
|
||
message_stats_30d = server.message_storage.get_message_stats_by_date_range(group_id, last_30d, now)
|
||
speaker_ranking = server.message_storage.get_group_member_message_ranking(group_id, last_30d, now, limit=10)
|
||
plugin_summary = server.stats_db.get_group_plugin_summary(group_id, days=30)
|
||
plugin_stats = server.stats_db.get_group_plugin_stats(group_id, days=30, limit=8)
|
||
last_message = server.message_storage.get_group_last_message(group_id)
|
||
trend_data = server.message_storage.get_message_trend(group_id, 14)
|
||
hourly_distribution = server.message_storage.get_group_hourly_distribution(group_id, 30)
|
||
|
||
for item in inactive_members:
|
||
wxid = item.get("wxid")
|
||
item["display_name"] = server.contact_manager.get_group_name(group_id, wxid) or item.get("nick_name") or wxid
|
||
|
||
for item in speaker_ranking:
|
||
sender = item.get("sender")
|
||
item["display_name"] = server.contact_manager.get_group_name(group_id, sender) or sender
|
||
|
||
total_members = member_summary.get("in_group_members") or member_summary.get("total_members") or 0
|
||
active_30d = member_summary.get("active_30d_members", 0)
|
||
zombie_count = member_summary.get("inactive_members", 0)
|
||
msg_30d_total = message_stats_30d.get("total_count", 0)
|
||
plugin_total_calls = plugin_summary.get("total_calls", 0)
|
||
|
||
health_score = 100
|
||
if total_members > 0:
|
||
health_score -= min(35, round((zombie_count / total_members) * 35))
|
||
health_score -= min(25, round((1 - min(active_30d / total_members, 1)) * 25))
|
||
if msg_30d_total == 0:
|
||
health_score -= 25
|
||
elif msg_30d_total < 50:
|
||
health_score -= 12
|
||
if plugin_total_calls == 0:
|
||
health_score -= 10
|
||
health_score = max(0, min(100, health_score))
|
||
|
||
trend_dates = []
|
||
trend_counts = []
|
||
for item in trend_data:
|
||
date_val = item.get("date")
|
||
if isinstance(date_val, datetime):
|
||
date_str = date_val.strftime('%Y-%m-%d')
|
||
else:
|
||
date_str = str(date_val)
|
||
trend_dates.append(date_str)
|
||
trend_counts.append(int(item.get("message_count") or 0))
|
||
|
||
hourly_distribution_sorted = sorted(
|
||
hourly_distribution,
|
||
key=lambda x: x.get("message_count", 0),
|
||
reverse=True
|
||
)
|
||
peak_hours = [
|
||
{
|
||
"hour": item.get("hour", 0),
|
||
"label": f"{int(item.get('hour', 0)):02d}:00-{(int(item.get('hour', 0)) + 1) % 24:02d}:00",
|
||
"message_count": item.get("message_count", 0),
|
||
}
|
||
for item in hourly_distribution_sorted[:3]
|
||
if item.get("message_count", 0) > 0
|
||
]
|
||
|
||
plugin_success_rate = (
|
||
round(plugin_summary.get("success_calls", 0) / plugin_total_calls * 100, 1)
|
||
if plugin_total_calls else 0
|
||
)
|
||
zombie_ratio = round((zombie_count / total_members) * 100, 1) if total_members else 0
|
||
active_ratio = round((active_30d / total_members) * 100, 1) if total_members else 0
|
||
|
||
operation_suggestions = []
|
||
if zombie_ratio >= 50:
|
||
operation_suggestions.append({
|
||
"level": "danger",
|
||
"title": "僵尸成员偏多，建议分层清理或召回",
|
||
"desc": f"当前僵尸成员占比 {zombie_ratio}% ，建议优先筛查 30 天未发言成员，结合群定位决定清理、激活或转移。"
|
||
})
|
||
elif zombie_ratio >= 30:
|
||
operation_suggestions.append({
|
||
"level": "warning",
|
||
"title": "沉默成员较多，建议做激活动作",
|
||
"desc": f"当前僵尸成员占比 {zombie_ratio}% ，可以通过签到、抽奖、任务或话题互动做一次激活。"
|
||
})
|
||
|
||
if active_ratio < 20:
|
||
operation_suggestions.append({
|
||
"level": "danger",
|
||
"title": "活跃覆盖率偏低，群内容可能只被少数人消费",
|
||
"desc": f"近30天活跃覆盖率仅 {active_ratio}% ，建议增加低门槛互动，如投票、签到、接龙、关键词触发。"
|
||
})
|
||
elif active_ratio < 40:
|
||
operation_suggestions.append({
|
||
"level": "warning",
|
||
"title": "活跃度中等，可继续扩大参与面",
|
||
"desc": f"近30天活跃覆盖率为 {active_ratio}% ，建议在高峰时段安排固定互动，提高更多成员参与率。"
|
||
})
|
||
|
||
if plugin_total_calls == 0:
|
||
operation_suggestions.append({
|
||
"level": "info",
|
||
"title": "插件渗透率为 0，建议补一次功能引导",
|
||
"desc": "这个群近30天没有插件调用，建议置顶菜单、欢迎语或群公告里给出可用指令示例。"
|
||
})
|
||
elif plugin_summary.get("plugin_count", 0) <= 2:
|
||
operation_suggestions.append({
|
||
"level": "info",
|
||
"title": "插件使用面偏窄，可以扩展场景",
|
||
"desc": f"近30天仅使用了 {plugin_summary.get('plugin_count', 0)} 个插件，建议根据群定位补充娱乐、工具或运营型插件。"
|
||
})
|
||
|
||
if plugin_total_calls > 0 and plugin_success_rate < 85:
|
||
operation_suggestions.append({
|
||
"level": "warning",
|
||
"title": "插件成功率偏低，建议检查指令体验",
|
||
"desc": f"近30天插件成功率 {plugin_success_rate}% ，可排查失败插件、异常命令和提示文案是否清晰。"
|
||
})
|
||
|
||
if msg_30d_total < 30:
|
||
operation_suggestions.append({
|
||
"level": "info",
|
||
"title": "群消息基数偏低，建议降低互动门槛",
|
||
"desc": f"近30天消息量仅 {msg_30d_total} 条，更适合短反馈、轻任务、固定时间提醒等低成本互动。"
|
||
})
|
||
|
||
if not operation_suggestions:
|
||
operation_suggestions.append({
|
||
"level": "success",
|
||
"title": "群整体状态健康，可持续做精细化运营",
|
||
"desc": "当前活跃、消息量、插件调用都比较均衡，建议继续围绕高峰时段和高频插件做强化。"
|
||
})
|
||
|
||
# 群运营分析 2.0 采用“增量增强”策略：
|
||
# 1. 现有 overview / diagnosis / operation_suggestions 等结构继续保留，避免影响旧页面；
|
||
# 2. 新增 ops_profile / ops_members / ops_actions 三类字段，供现有详情页逐步增强使用；
|
||
# 3. 即使某些增强数据暂时为空，也不应影响老内容展示。
|
||
ops_profile = _build_group_ops_profile(
|
||
server=server,
|
||
group_id=group_id,
|
||
group_name=group_name,
|
||
peak_hours=peak_hours,
|
||
plugin_stats=plugin_stats,
|
||
)
|
||
ops_members = _build_ops_member_segments(
|
||
server=server,
|
||
group_id=group_id,
|
||
speaker_ranking=speaker_ranking,
|
||
inactive_members=inactive_members,
|
||
)
|
||
recent_summaries = _load_recent_group_summaries(server, group_id, limit=3)
|
||
ops_actions = _build_ops_action_cards(
|
||
group_name=group_name,
|
||
ops_profile=ops_profile,
|
||
ops_members=ops_members,
|
||
plugin_stats=plugin_stats,
|
||
)
|
||
|
||
return jsonify({
|
||
"success": True,
|
||
"data": {
|
||
"group_id": group_id,
|
||
"group_name": group_name,
|
||
"health_score": health_score,
|
||
"overview": {
|
||
"member_count": total_members,
|
||
"active_member_count_7d": member_summary.get("active_7d_members", 0),
|
||
"active_member_count_30d": active_30d,
|
||
"zombie_member_count": zombie_count,
|
||
"never_spoken_count": member_summary.get("never_spoken_members", 0),
|
||
"message_count_24h": message_stats_24h.get("total_count", 0),
|
||
"message_count_7d": message_stats_7d.get("total_count", 0),
|
||
"message_count_30d": msg_30d_total,
|
||
"plugin_call_count_30d": plugin_total_calls,
|
||
"plugin_count_30d": plugin_summary.get("plugin_count", 0),
|
||
},
|
||
"diagnosis": [
|
||
{
|
||
"title": "活跃覆盖率",
|
||
"value": f"{round((active_30d / total_members) * 100, 1) if total_members else 0}%",
|
||
"desc": "近30天发言成员占全群人数比例"
|
||
},
|
||
{
|
||
"title": "僵尸用户占比",
|
||
"value": f"{round((zombie_count / total_members) * 100, 1) if total_members else 0}%",
|
||
"desc": "30天未发言或从未发言成员占比"
|
||
},
|
||
{
|
||
"title": "人均消息量",
|
||
"value": round(msg_30d_total / active_30d, 1) if active_30d else 0,
|
||
"desc": "近30天活跃成员的人均发言量"
|
||
},
|
||
{
|
||
"title": "插件渗透度",
|
||
"value": f"{plugin_summary.get('plugin_count', 0)} 个插件",
|
||
"desc": "近30天在本群真实发生调用的插件数"
|
||
}
|
||
],
|
||
"permissions": {
|
||
"enabled_count": len(enabled_features),
|
||
"enabled_features": enabled_features,
|
||
},
|
||
"member_summary": member_summary,
|
||
"message_summary": {
|
||
"last_message": last_message,
|
||
"last_24h": message_stats_24h,
|
||
"last_7d": message_stats_7d,
|
||
"last_30d": message_stats_30d,
|
||
"type_mix_30d": [
|
||
{"label": "文本", "count": message_stats_30d.get("text_count", 0)},
|
||
{"label": "图片", "count": message_stats_30d.get("image_count", 0)},
|
||
{"label": "视频", "count": message_stats_30d.get("video_count", 0)},
|
||
{"label": "链接", "count": message_stats_30d.get("link_count", 0)},
|
||
{"label": "表情", "count": message_stats_30d.get("emoji_count", 0)},
|
||
],
|
||
"trend_14d": {
|
||
"dates": trend_dates,
|
||
"counts": trend_counts
|
||
},
|
||
"hourly_distribution_30d": hourly_distribution,
|
||
"peak_hours_30d": peak_hours,
|
||
},
|
||
"inactive_members": inactive_members,
|
||
"speaker_ranking": speaker_ranking,
|
||
"plugin_summary": plugin_summary,
|
||
"plugin_stats": plugin_stats,
|
||
"operation_suggestions": operation_suggestions,
|
||
"ops_profile": ops_profile,
|
||
"ops_members": ops_members,
|
||
"ops_actions": ops_actions,
|
||
"recent_summaries": recent_summaries,
|
||
}
|
||
})
|
||
except Exception as e:
|
||
LOG.error(f"获取群组详情失败: {e}")
|
||
return jsonify({"success": False, "error": str(e)}), 500
|
||
|
||
|
||
# 添加缺失的群组状态更新接口
|
||
@robot_bp.route('/api/group/<group_id>/status', methods=['POST'])
|
||
@login_required
|
||
def api_update_group_status(group_id):
|
||
try:
|
||
server = current_app.dashboard_server
|
||
data = request.json
|
||
status = data.get('status')
|
||
|
||
if status == 'disabled':
|
||
# 禁用该群组的所有功能
|
||
LOG.info(f"正在禁用群组 {group_id} 的所有功能")
|
||
|
||
# 获取所有功能并禁用
|
||
for feature in Feature:
|
||
GroupBotManager.set_group_permission(group_id, feature, PermissionStatus.DISABLED)
|
||
|
||
# 特殊处理ROBOT功能，从群组列表中移除
|
||
if group_id in GroupBotManager.local_cache["group_list"]:
|
||
GroupBotManager.local_cache["group_list"].remove(group_id)
|
||
|
||
# 从Redis中移除
|
||
r = server.db_manager.get_redis_connection()
|
||
r.srem("group:list", group_id)
|
||
|
||
return jsonify({
|
||
"success": True,
|
||
"message": f"群组 {group_id} 的所有功能已禁用"
|
||
})
|
||
elif status == 'enabled':
|
||
# 启用该群组的基本功能
|
||
LOG.info(f"正在启用群组 {group_id} 的基本功能")
|
||
|
||
# 添加到群组列表
|
||
if group_id not in GroupBotManager.local_cache["group_list"]:
|
||
GroupBotManager.local_cache["group_list"].add(group_id)
|
||
|
||
# 添加到Redis
|
||
r = server.db_manager.get_redis_connection()
|
||
r.sadd("group:list", group_id)
|
||
|
||
# 启用ROBOT基本功能
|
||
GroupBotManager.set_group_permission(group_id, Feature.ROBOT, PermissionStatus.ENABLED)
|
||
|
||
return jsonify({
|
||
"success": True,
|
||
"message": f"群组 {group_id} 的基本功能已启用"
|
||
})
|
||
else:
|
||
return jsonify({
|
||
"success": False,
|
||
"error": "不支持的状态值，只接受 'enabled' 或 'disabled'"
|
||
}), 400
|
||
|
||
except Exception as e:
|
||
LOG.error(f"更新群组状态失败: {e}")
|
||
return jsonify({"success": False, "error": str(e)}), 500
|