feat(ai): 每个数据页面都有 AI 解读,靠一条带优先级的生产者/消费者队列
原来只有今日页有晨报、指标详情页有归因,其余页面一片空白。现在除设置外 的 10 个页面都有:健康、睡眠、运动、趋势、每日、身体成分、成绩预测、 身体年龄、挑战赛、运动详情。 不是给每个页面写一套,而是一个通用管线: - services/scopes.py:一个页面一个 context builder,返回同一个信封。 context["highlights"] 是已经算好的白话事实——模型负责解读它们,模型不 可用时规则引擎原样渲染。两者引用同一批数字,所以降级读起来不像换了个 App。 没数据的页面返回 None,宁可不出卡片,也不让模型对着空表格发挥。 - coach.scope_messages / parse_scope_insight:一套提示词吃所有页面,页面 的差异全在 context 里,加页面 = 加一个 builder。 - 前端 <AiPanel scope="…">:一个组件渲染所有页面,轮询逻辑抽成 lib/insight.ts 的 usePolledInsight,晨报卡也改用它。 ## 队列 一次生成 40 秒到 4.5 分钟,所以什么都不能在请求里生成。页面只负责入队, worker 负责消费(services/jobs.py)。 优先级才是用队列而不是后台线程的理由:同步完成后 prefetch 把所有页面按 背景优先级排进去,可能要跑半小时;而用户一打开某个页面,那个页面的任务 立刻提到队首、下一个就跑。你在看什么,队列就在算什么。 队列放在数据库而不是内存里,因为 gunicorn 有两个 worker:任务带 holder 声明后回读确认,和 scheduler.py 抢 tick 是同一套做法。id 由 user+kind+subject 推导,所以每几秒一次的轮询是幂等的入队,不会每几秒堆一 个任务。 ## 网关中断时踩到的两个坑(当场修了) 写完正好赶上 oracle 那台机器不通,于是看到: - 三次失败后任务被永久标 failed,网关恢复了也不会重试——一次瞬时中断就把 那个页面的解读判了死刑,直到它的数据碰巧变化。加了冷却期,过期后重置 尝试次数再排一次。 - 队列已经放弃了,页面还在 pending 转圈,要转满 8 分钟才停。meta.pending 现在跟着队列状态走,并把失败原因带给卡片。 顺带把 BAND_SOURCES 从 routes/settings.py 下沉到 services/insights.py: 教练要拿它做参照,而 services 不该反向依赖 routes。 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
@@ -7,6 +7,8 @@ from auth import require_auth
|
||||
from services import analysis as analysis_svc
|
||||
from services import ai as ai_svc
|
||||
from services import insights
|
||||
from services import jobs as ai_jobs
|
||||
from services import scopes
|
||||
|
||||
bp = Blueprint("analysis", __name__)
|
||||
|
||||
@@ -135,3 +137,39 @@ def copilot():
|
||||
# until it completes, which would undo the point of streaming it.
|
||||
headers={"Cache-Control": "no-cache", "X-Accel-Buffering": "no"},
|
||||
)
|
||||
|
||||
|
||||
@bp.route("/insight", methods=["GET"])
|
||||
@require_auth
|
||||
def insight():
|
||||
"""One screen's AI reading. `?scope=` names the screen.
|
||||
|
||||
Answers immediately, like the briefing: the computed highlights come back
|
||||
with `meta.pending` while the model's version is generated. Opening a
|
||||
screen queues it at interactive priority, ahead of any backfill still
|
||||
running, so what you are looking at is what the queue works on next.
|
||||
|
||||
`?subject=` identifies the item for per-item screens (`activity` needs an
|
||||
activity id, `daily` takes a date).
|
||||
"""
|
||||
scope = request.args.get("scope")
|
||||
if scope not in scopes.SCOPES:
|
||||
return jsonify({
|
||||
"error": f"不支持的页面: {scope}",
|
||||
"supported": sorted(scopes.SCOPES),
|
||||
}), 400
|
||||
return jsonify(analysis_svc.get_scope_insight(
|
||||
g.user_id, scope,
|
||||
subject=request.args.get("subject") or None,
|
||||
refresh=_flag("refresh"),
|
||||
))
|
||||
|
||||
|
||||
@bp.route("/insight/queue", methods=["GET"])
|
||||
@require_auth
|
||||
def insight_queue():
|
||||
"""What the coach still has to generate — for a progress indicator."""
|
||||
return jsonify({
|
||||
"pending": ai_jobs.pending_count(g.user_id),
|
||||
"enabled": ai_jobs.ENABLED,
|
||||
})
|
||||
|
||||
Reference in New Issue
Block a user