Update wehago matching logic and exclude reports

scripts/rebuild_hanmac_aggregate_cache_once.py

@@ -0,0 +1,86 @@
+from __future__ import annotations
+
+import argparse
+import json
+import sqlite3
+import sys
+from datetime import date
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).resolve().parents[1]))
+import main
+
+
+def load_latest_payloads(years: list[int]) -> dict[int, dict[str, object]]:
+    conn = sqlite3.connect(main.DB_PATH)
+    conn.row_factory = sqlite3.Row
+    rows = conn.execute(
+        """
+        SELECT params_json
+        FROM system_jobs
+        WHERE page_key = 'hanmac_browser'
+          AND job_type = 'hanmac_aggregate_cache'
+          AND status = 'done'
+        ORDER BY updated_at DESC
+        """
+    ).fetchall()
+    latest: dict[int, dict[str, object]] = {}
+    fallback: dict[str, object] = {}
+    for row in rows:
+        try:
+            payload = json.loads(row["params_json"] or "{}")
+        except Exception:
+            continue
+        if not isinstance(payload, dict):
+            continue
+        if not fallback and payload.get("host") and payload.get("user") and payload.get("password"):
+            fallback = payload
+        start_text = str(payload.get("start_date") or "")
+        for year in years:
+            if year not in latest and start_text.startswith(str(year)):
+                latest[year] = payload
+    return {year: dict(latest.get(year) or fallback) for year in years}
+
+
+def rebuild_year(year: int, base_payload: dict[str, object], end_today: bool = False) -> dict[str, object]:
+    if not base_payload:
+        raise RuntimeError(f"{year}년 캐시를 만들 접속정보가 없습니다.")
+    end_date = date.today().isoformat() if end_today and year == date.today().year else f"{year}-12-31"
+    payload = {
+        **base_payload,
+        "database": main.normalize_text(base_payload.get("database")) or main.HANMAC_PRIMARY_MANHOUR_SCHEMA,
+        "view": "member",
+        "employment": main.normalize_text(base_payload.get("employment") or "all"),
+        "start_date": f"{year}-01-01",
+        "end_date": end_date,
+    }
+    result = main.get_hanmac_aggregate_summary(payload)
+    cache_key = main._hanmac_aggregate_cache_key(payload)
+    main._store_hanmac_aggregate_cache(cache_key, result)
+    diagnostics = result.get("source_diagnostics") or {}
+    return {
+        "year": year,
+        "start_date": result.get("start_date"),
+        "end_date": result.get("end_date"),
+        "row_count": len(result.get("rows") or []),
+        "cache_key": cache_key,
+        "joint_assignment_records": diagnostics.get("joint_assignment_records", 0),
+        "joint_assignment_regular_rows": diagnostics.get("joint_assignment_regular_rows", 0),
+        "joint_assignment_overtime_rows": diagnostics.get("joint_assignment_overtime_rows", 0),
+    }
+
+
+def main_cli() -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--years", nargs="+", type=int, default=[2022, 2023, 2024, 2025, 2026])
+    parser.add_argument("--end-today", action="store_true")
+    args = parser.parse_args()
+
+    payloads = load_latest_payloads(args.years)
+    for year in args.years:
+        summary = rebuild_year(year, payloads.get(year) or {}, end_today=args.end_today)
+        print(json.dumps(summary, ensure_ascii=False, sort_keys=True))
+
+
+if __name__ == "__main__":
+    main_cli()