wsk/deerflow-code
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: run leaderboard stats in background

Browse Source
This commit is contained in:
吉鑫升 committed 2026-06-05 21:15:20 +08:00
1 parent f4ccef8c70
commit db1b32612b
4 files changed
+269 -1
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

frontend-web/src/core/admin/api.ts
+3
View File
@@ -104,6 +104,9 @@ export interface AdminLeaderboardResponse {
range_days: number;
since: string;
until: string;
status?: "ready" | "running" | "refreshing";
generated_at?: string;
message?: string;
overview: AdminLeaderboardOverview;
users_by_activity: AdminLeaderboardUser[];
users_by_questions: AdminLeaderboardUser[];
frontend-web/src/core/admin/hooks.ts
+4
View File
@@ -33,6 +33,10 @@ export function useUserLeaderboard(
limit,
],
queryFn: () => getUserLeaderboard(range, limit),
refetchInterval: (query) => {
const status = query.state.data?.status;
return status === "running" || status === "refreshing" ? 3000 : false;
},
});
}
frontend-web/src/pages/AdminUserLeaderboardPage.tsx
+25 -1
View File
@@ -1087,6 +1087,9 @@ export default function AdminUserLeaderboardPage() {
}, [data, query, tab]);
const overview = data?.overview;
const leaderboardStatus = data?.status;
const leaderboardPending = leaderboardStatus === "running" || leaderboardStatus === "refreshing";
const leaderboardHasRows = rankedUsers.length > 0 || (overview?.total_runs ?? 0) > 0 || (overview?.total_users ?? 0) > 0;
const topUsers = rankedUsers.slice(0, 3);
const userPageCount = Math.max(1, Math.ceil(rankedUsers.length / LEADERBOARD_PAGE_SIZE));
const visibleUsers = useMemo(() => {
@@ -1446,8 +1449,29 @@ export default function AdminUserLeaderboardPage() {
</div>
)}
{!isLoading && data && overview && (
{!isLoading && !error && leaderboardPending && !leaderboardHasRows && (
<div className="flex flex-col items-center justify-center gap-3 rounded-lg border border-[#F1D6D6] bg-white px-6 py-16 text-center text-sm text-[#5F6368] shadow-sm dark:border-[#462123] dark:bg-[#230D0F]">
<div className="flex items-center gap-2 font-medium text-[#202124] dark:text-white">
<Loader2 className="size-4 animate-spin text-[#C1121F]" />
<span>排行榜正在后台统计</span>
</div>
<div className="max-w-lg text-xs leading-5 text-[#7A5A5A] dark:text-[#D8B8B8]">
{data?.message || "统计任务已放到后台执行，页面会自动刷新结果。这样排行榜慢慢算，不会长时间占住当前接口请求。"}
</div>
</div>
)}
{!isLoading && data && overview && (!leaderboardPending || leaderboardHasRows) && (
<>
{leaderboardPending && leaderboardHasRows && (
<div className="flex flex-wrap items-center justify-between gap-3 rounded-lg border border-[#FDE68A] bg-[#FFFBEB] px-4 py-3 text-sm text-[#92400E] dark:border-amber-200/25 dark:bg-amber-300/10 dark:text-amber-100">
<div className="flex items-center gap-2">
<Loader2 className="size-4 animate-spin" />
<span>{data.message || "正在后台刷新排行榜，当前展示缓存结果。"}</span>
</div>
{data.generated_at && <span className="text-xs">缓存生成于 {fmtTime(data.generated_at)}</span>}
</div>
)}
<div className="grid gap-3 md:grid-cols-2 xl:grid-cols-4">
<StatCard icon={Users} accent="crimson" label="总用户" value={fmtNumber(overview.total_users)} note={`${fmtNumber(overview.active_users)} 位在当前时间段活跃`} />
<StatCard icon={MousePointerClick} accent="teal" label="使用次数" value={fmtNumber(overview.total_runs)} note={`${fmtNumber(overview.total_threads)} 个会话被创建`} />
offline-backend-20260512/backend/app/gateway/routers/admin_users.py
+237
View File
@@ -18,6 +18,7 @@ import json
import logging
import os
import re
import time
from collections.abc import AsyncIterator
from difflib import SequenceMatcher
from hashlib import sha256
@@ -67,7 +68,14 @@ _LEADERBOARD_ACQUIRE_TIMEOUT_SECONDS = max(
0.1,
float(os.getenv("ADMIN_LEADERBOARD_ACQUIRE_TIMEOUT_SECONDS", "0.5")),
)
_LEADERBOARD_CACHE_TTL_SECONDS = max(
5.0,
float(os.getenv("ADMIN_LEADERBOARD_CACHE_TTL_SECONDS", "300")),
)
_leaderboard_semaphore = asyncio.Semaphore(_LEADERBOARD_MAX_CONCURRENT)
_leaderboard_jobs: dict[str, asyncio.Task] = {}
_leaderboard_cache: dict[str, dict] = {}
_leaderboard_jobs_lock = asyncio.Lock()
async def _require_admin(request: Request) -> None:
@@ -154,6 +162,9 @@ class AdminLeaderboardResponse(BaseModel):
range_days: int
since: str
until: str
status: str = "ready"
generated_at: str = ""
message: str = ""
overview: AdminLeaderboardOverview
users_by_activity: list[AdminLeaderboardUser] = Field(default_factory=list)
users_by_questions: list[AdminLeaderboardUser] = Field(default_factory=list)
@@ -981,6 +992,169 @@ async def _query_overview(
)
def _empty_leaderboard_response(
since_dt: datetime,
until_dt: datetime,
range_days: int,
*,
status: str,
message: str,
) -> AdminLeaderboardResponse:
return AdminLeaderboardResponse(
range_days=range_days,
since=_iso(since_dt),
until=_iso(until_dt),
status=status,
message=message,
overview=AdminLeaderboardOverview(),
)
def _leaderboard_cache_key(
*,
since_dt: datetime,
until_dt: datetime,
limit: int,
excluded: set[str],
include_scheduled: bool,
include_admins: bool,
include_failed: bool,
cleanup_system_questions: bool,
) -> str:
payload = {
"version": 2,
"since": _iso(since_dt),
"until": _iso(until_dt),
"limit": limit,
"excluded": sorted(excluded),
"include_scheduled": include_scheduled,
"include_admins": include_admins,
"include_failed": include_failed,
"cleanup_system_questions": cleanup_system_questions,
}
return sha256(json.dumps(payload, sort_keys=True, ensure_ascii=False).encode("utf-8")).hexdigest()
async def _build_user_leaderboard_response(
session_factory,
settings,
since_dt: datetime,
until_dt: datetime,
range_days: int,
limit: int,
excluded: set[str],
include_scheduled: bool,
include_admins: bool,
include_failed: bool,
) -> AdminLeaderboardResponse:
async with session_factory() as session:
common = {
"exclude_user_ids": excluded,
"include_scheduled": include_scheduled,
"include_admins": include_admins,
"include_failed": include_failed,
}
tool_common = {
"exclude_user_ids": excluded,
"include_scheduled": include_scheduled,
"include_admins": include_admins,
}
question_common = {
**common,
"cleanup_system_questions": settings.cleanup_system_questions,
}
overview = await _query_overview(session, since_dt, until_dt, **common)
users = await _query_user_leaderboard(session, since_dt, until_dt, **common)
skills = await _query_call_rankings(session, since_dt, until_dt, user_id=None, field="skill", limit=limit, **tool_common)
tools = await _query_call_rankings(session, since_dt, until_dt, user_id=None, field="tool", limit=limit, **tool_common)
models = await _query_model_rankings(session, since_dt, until_dt, user_id=None, limit=limit, **common)
if not models:
models = await _query_run_dimensions(session, since_dt, until_dt, user_id=None, field="model", limit=limit, **common)
agents = await _query_run_dimensions(session, since_dt, until_dt, user_id=None, field="agent", limit=limit, **common)
run_statuses = await _query_run_dimensions(session, since_dt, until_dt, user_id=None, field="status", limit=limit, **common)
recent_questions = await _query_recent_questions(session, since_dt, until_dt, user_id=None, limit=limit, **question_common)
frequent_questions = await _query_frequent_questions(session, since_dt, until_dt, limit=limit, **question_common)
trends = await _query_trends(session, since_dt, until_dt, **common)
return AdminLeaderboardResponse(
range_days=range_days,
since=_iso(since_dt),
until=_iso(until_dt),
status="ready",
generated_at=_iso(datetime.utcnow()),
overview=overview,
users_by_activity=sorted(users, key=lambda u: (u.run_count + u.tool_call_count, u.total_tokens), reverse=True)[:limit],
users_by_questions=sorted(users, key=lambda u: (u.qa_count, u.run_count), reverse=True)[:limit],
users_by_tokens=sorted(users, key=lambda u: u.total_tokens, reverse=True)[:limit],
skills=skills,
tools=tools,
models=models,
agents=agents,
run_statuses=run_statuses,
recent_questions=recent_questions,
frequent_questions=frequent_questions,
trends=trends,
)
async def _refresh_leaderboard_cache(
cache_key: str,
*,
session_factory,
settings,
since_dt: datetime,
until_dt: datetime,
range_days: int,
limit: int,
excluded: set[str],
include_scheduled: bool,
include_admins: bool,
include_failed: bool,
) -> None:
acquired = False
try:
await _leaderboard_semaphore.acquire()
acquired = True
response = await asyncio.wait_for(
_build_user_leaderboard_response(
session_factory,
settings,
since_dt,
until_dt,
range_days,
limit,
excluded,
include_scheduled,
include_admins,
include_failed,
),
timeout=_LEADERBOARD_TIMEOUT_SECONDS,
)
_leaderboard_cache[cache_key] = {
"created_at": time.monotonic(),
"response": response,
}
except asyncio.TimeoutError:
logger.warning("Leaderboard refresh timed out for key=%s", cache_key)
except Exception:
logger.exception("Leaderboard refresh failed for key=%s", cache_key)
finally:
if acquired:
_leaderboard_semaphore.release()
async with _leaderboard_jobs_lock:
_leaderboard_jobs.pop(cache_key, None)
async def _ensure_leaderboard_refresh(cache_key: str, **kwargs) -> None:
async with _leaderboard_jobs_lock:
task = _leaderboard_jobs.get(cache_key)
if task and not task.done():
return
_leaderboard_jobs[cache_key] = asyncio.create_task(
_refresh_leaderboard_cache(cache_key, **kwargs)
)
@router.get("/leaderboard", response_model=AdminLeaderboardResponse)
async def get_user_leaderboard(
request: Request,
@@ -999,6 +1173,69 @@ async def get_user_leaderboard(
if session_factory is None:
raise HTTPException(status_code=503, detail="Analytics requires database persistence")
settings = load_system_settings().leaderboard
since_dt, until_dt, range_days = _resolve_range(days, since, until)
excluded = {uid for uid in settings.exclude_user_ids if uid}
excluded.update({uid for uid in (exclude_user_ids or []) if uid})
resolved_include_scheduled = settings.include_scheduled if include_scheduled is None else include_scheduled
resolved_include_admins = settings.include_admins if include_admins is None else include_admins
resolved_include_failed = settings.include_failed if include_failed is None else include_failed
cache_key = _leaderboard_cache_key(
since_dt=since_dt,
until_dt=until_dt,
limit=limit,
excluded=excluded,
include_scheduled=resolved_include_scheduled,
include_admins=resolved_include_admins,
include_failed=resolved_include_failed,
cleanup_system_questions=settings.cleanup_system_questions,
)
cached = _leaderboard_cache.get(cache_key)
if cached:
response = cached["response"].model_copy()
age = time.monotonic() - float(cached["created_at"])
if age <= _LEADERBOARD_CACHE_TTL_SECONDS:
response.status = "ready"
return response
response.status = "refreshing"
response.message = "排行榜缓存已返回，后台正在刷新最新统计。"
await _ensure_leaderboard_refresh(
cache_key,
session_factory=session_factory,
settings=settings,
since_dt=since_dt,
until_dt=until_dt,
range_days=range_days,
limit=limit,
excluded=excluded,
include_scheduled=resolved_include_scheduled,
include_admins=resolved_include_admins,
include_failed=resolved_include_failed,
)
return response
await _ensure_leaderboard_refresh(
cache_key,
session_factory=session_factory,
settings=settings,
since_dt=since_dt,
until_dt=until_dt,
range_days=range_days,
limit=limit,
excluded=excluded,
include_scheduled=resolved_include_scheduled,
include_admins=resolved_include_admins,
include_failed=resolved_include_failed,
)
return _empty_leaderboard_response(
since_dt,
until_dt,
range_days,
status="running",
message="排行榜统计已在后台运行。线上用户较多时会慢一些，页面会自动刷新结果。",
)
try:
await asyncio.wait_for(
_leaderboard_semaphore.acquire(),