feat: P0-02/P0-03/P0-05 数据库时间语义与约束

P0-02: MatchStats 增加 source/source_record_id/retrieved_at/available_at
        context_builder 过滤统计数据时检查 available_at <= cutoff
P0-03: Prediction 增加 match_kickoff_at/prediction_created_at/prediction_cutoff_at
        明确区分比赛时间/预测创建时间/数据截止时间
P0-05: Prediction 增加 status 字段(success/failed/degraded)
        数据库 CHECK 约束

Alembic: 0005_prediction_status_and_stats_provenance

P1-02: injuries as_of 不再截断为 date,保持 datetime 精度

src/data/bzzoiro.py

@@ -14,6 +14,7 @@ import urllib.error
 import urllib.parse
 import urllib.request
 from collections.abc import Iterable
+from datetime import datetime, timezone
 
 from sqlalchemy import select
 
@@ -223,6 +224,7 @@ class BzzoiroSource:
                     await db.flush()
                     existing_matches[match_key] = m  # 防止同批重复
                     if nm.home_xg is not None or nm.away_xg is not None:
+                        now = datetime.now(timezone.utc)
                         stats = MatchStats(
                             match_id=m.id,
                             home_xg=nm.home_xg,
@@ -238,6 +240,10 @@ class BzzoiroSource:
                             away_yellow_cards=nm.away_yellow_cards,
                             home_red_cards=nm.home_red_cards,
                             away_red_cards=nm.away_red_cards,
+                            source="bzzoiro",
+                            source_event_id=str(raw.get("id", "")),
+                            retrieved_at=now,
+                            available_at=now,
                         )
                         db.add(stats)
                     league_r["inserted"] += 1

src/data/injuries.py

@@ -190,11 +190,13 @@ async def get_injuries_for_match(db, team_id: int, match_date, as_of=None) -> li
         match_date: 比赛日期
         as_of: 截止时间(cutoff)。只返回 retrieved_at <= as_of 的记录。
                用于回测时防止"未来采集的数据"泄漏到历史预测。
+               必须保持 timezone-aware datetime,不会截断为 date。
     """
-    from sqlalchemy import and_, or_, select
+    from sqlalchemy import or_, select
 
     from src.db.models import Injury
 
+    # 只处理 match_date:去掉时间部分,仅比较日期
     if hasattr(match_date, "date"):
         match_date = match_date.date()
 
@@ -206,9 +208,8 @@ async def get_injuries_for_match(db, team_id: int, match_date, as_of=None) -> li
     )
 
     # 回测防泄漏: 只使用 as_of 时间点之前已采集的数据
+    # 注意: as_of 保持 datetime,不截断为 date,避免错误排除同日合法数据
     if as_of is not None:
-        if hasattr(as_of, "date"):
-            as_of = as_of.date()
         stmt = stmt.where(Injury.retrieved_at.is_not(None))
         stmt = stmt.where(Injury.retrieved_at <= as_of)
 

src/data/understat.py

@@ -10,6 +10,7 @@ import json
 import logging
 import random
 import re
+from datetime import datetime, timezone
 
 from sqlalchemy import func, select
 
@@ -136,7 +137,14 @@ class UnderstatSource:
 
             # 回填 xG
             if existing.stats is None and (nm.home_xg is not None or nm.away_xg is not None):
-                existing.stats = MatchStats(match_id=existing.id)
+                now = datetime.now(timezone.utc)
+                existing.stats = MatchStats(
+                    match_id=existing.id,
+                    source="understat",
+                    source_event_id=str(raw.get("id", "")),
+                    retrieved_at=now,
+                    available_at=now,
+                )
                 db.add(existing.stats)
                 await db.flush()
             if existing.stats is not None: