将项目文件整理到 refer 目录

2026-07-03 22:26:00 +08:00
parent 49880e785f
commit 9f57af4d37
1223 changed files with 98 additions and 0 deletions
@@ -0,0 +1,150 @@
+"""策略结果缓存 — 写入本地文件，供策略页面秒加载。
+
+缓存结构:
+  {
+    "as_of": "2024-01-15",
+    "results": { strategy_id: { total, as_of, rows } },
+    "today_ever_matched": { strategy_id: [symbol, ...] },    // 今日曾命中 symbol 并集
+    "today_ever_rows": { strategy_id: { symbol: row_data } },// 今日曾命中的完整行数据
+    "updated_at": 1705324800000  # Unix ms
+  }
+
+文件路径: data/user_data/strategy_cache.json
+"""
+from __future__ import annotations
+
+import json
+import logging
+import time
+from datetime import date, datetime
+from pathlib import Path
+from typing import Any
+
+
+def _json_default(obj: Any) -> Any:
+    """处理 date/datetime 等 JSON 不认识的类型。"""
+    if isinstance(obj, date):
+        return obj.isoformat()
+    if isinstance(obj, datetime):
+        return obj.isoformat()
+    raise TypeError(f"Object of type {type(obj).__name__} is not JSON serializable")
+
+
+logger = logging.getLogger(__name__)
+
+_CACHE_FILENAME = "strategy_cache.json"
+
+
+def _cache_path(data_dir: Path) -> Path:
+    return data_dir / "user_data" / _CACHE_FILENAME
+
+
+def _enriched_parquet_path(data_dir: Path, as_of: str) -> Path:
+    """返回 enriched parquet 文件路径。"""
+    return data_dir / "kline_daily_enriched" / f"date={as_of}" / "part.parquet"
+
+
+def _get_enriched_mtime(data_dir: Path, as_of: str) -> float | None:
+    """返回 enriched parquet 文件的 mtime (秒)。文件不存在返回 None。"""
+    p = _enriched_parquet_path(data_dir, as_of)
+    try:
+        return p.stat().st_mtime
+    except FileNotFoundError:
+        return None
+
+
+def read_cache(data_dir: Path) -> dict | None:
+    """读取策略缓存文件。返回 None 表示无缓存、读取失败或 enriched 数据已更新导致缓存过期。"""
+    path = _cache_path(data_dir)
+    if not path.exists():
+        return None
+    try:
+        text = path.read_text(encoding="utf-8")
+        if not text.strip():
+            return None
+        cached = json.loads(text)
+    except Exception as e:  # noqa: BLE001
+        logger.warning("读取策略缓存失败: %s", e)
+        return None
+
+    # 校验 enriched mtime: 数据文件变化 → 缓存过期
+    as_of = cached.get("as_of")
+    stored_mtime = cached.get("enriched_mtime")
+    if as_of and stored_mtime:
+        current_mtime = _get_enriched_mtime(data_dir, as_of)
+        if current_mtime is not None and current_mtime != stored_mtime:
+            logger.info("策略缓存过期: enriched 数据已更新 (as_of=%s)", as_of)
+            return None
+
+    return cached
+
+
+def _rows_to_symbol_map(rows: list[dict]) -> dict[str, dict]:
+    """将 rows 列表转为 {symbol: row_data} 映射。"""
+    result: dict[str, dict] = {}
+    for row in rows:
+        sym = row.get("symbol")
+        if sym:
+            result[sym] = row
+    return result
+
+
+def write_cache(
+    data_dir: Path,
+    as_of: str,
+    results: dict[str, Any],
+) -> None:
+    """将策略结果写入缓存文件，同时更新今日曾命中集合。
+
+    - 日期变更时重置 today_ever_matched 和 today_ever_rows
+    - 同一天内合并 (并集) 之前曾命中的 symbol，并用最新行数据更新
+    """
+    path = _cache_path(data_dir)
+    path.parent.mkdir(parents=True, exist_ok=True)
+
+    # 读取旧缓存
+    old = read_cache(data_dir)
+    old_as_of = old.get("as_of") if old else None
+    old_ever_rows: dict[str, dict[str, dict]] = old.get("today_ever_rows", {}) if old else {}
+
+    # 当前命中的行数据 → symbol 映射
+    current_row_maps: dict[str, dict[str, dict]] = {}
+    for sid, r in results.items():
+        current_row_maps[sid] = _rows_to_symbol_map(r.get("rows", []))
+
+    if old_as_of and old_as_of == as_of and old_ever_rows:
+        # 同一天: 合并 — 用当前行数据更新旧数据 (保持最新价格等)
+        merged_rows: dict[str, dict[str, dict]] = {}
+        all_keys = set(old_ever_rows.keys()) | set(current_row_maps.keys())
+        for sid in all_keys:
+            old_map = old_ever_rows.get(sid, {})
+            cur_map = current_row_maps.get(sid, {})
+            # 以旧数据为基础，用当前数据覆盖 (当前数据更新鲜)
+            combined = {**old_map, **cur_map}
+            merged_rows[sid] = combined
+        today_ever_rows = merged_rows
+    else:
+        # 新的一天或首次写入
+        today_ever_rows = current_row_maps
+
+    # 从 ever_rows 提取 symbol 列表 (用于快速计数)
+    today_ever_matched = {sid: sorted(maps.keys()) for sid, maps in today_ever_rows.items()}
+
+    # 记录 enriched parquet 文件的 mtime，用于后续校验缓存是否过期
+    enriched_mtime = _get_enriched_mtime(data_dir, as_of)
+
+    payload = {
+        "as_of": as_of,
+        "results": results,
+        "today_ever_matched": today_ever_matched,
+        "today_ever_rows": today_ever_rows,
+        "enriched_mtime": enriched_mtime,
+        "updated_at": int(time.time() * 1000),
+    }
+    try:
+        path.write_text(json.dumps(payload, ensure_ascii=False, default=_json_default), encoding="utf-8")
+        total_rows = sum(len(r.get("rows", [])) for r in results.values())
+        total_ever = sum(len(v) for v in today_ever_matched.values())
+        logger.info("策略缓存已写入: %s, %d 策略, %d 命中, %d 曾命中", as_of, len(results), total_rows, total_ever)
+    except Exception as e:  # noqa: BLE001
+        logger.warning("写入策略缓存失败: %s", e)