#!/usr/bin/env python3
# =============================================================================
# QWHPI — Quebec Weekly Housing Price Index
# Author  : Simon-Pierre Boucher
# Contact : contact@spboucher.ai
# File    : ops/scheduler/refresh.py
# Purpose : Weekly refresh job — re-run the incremental pipeline every Monday
#           06:00, write a run manifest, load the DB, log structured events.
# =============================================================================
"""Weekly refresh scheduler.

Every Monday 06:00 (container time): ingest new rows → re-clean (append-only)
→ re-estimate monthly indexes → new data_vintage → canonical rebuild → DB load.
A JSON run manifest (input hash, row counts, timings) is written to
``ops/scheduler/runs/``. The API's parquet cache invalidates automatically on
file mtime; ETag values include the data vintage, so HTTP caches roll over.
"""

from __future__ import annotations

import json
import logging
import subprocess
import sys
import time
from datetime import datetime, timezone
from pathlib import Path

import schedule

WORKSPACE = Path("/workspace")
RUNS_DIR = WORKSPACE / "ops" / "scheduler" / "runs"

logging.basicConfig(
    level=logging.INFO,
    format='{"ts":"%(asctime)s","level":"%(levelname)s","msg":"%(message)s"}',
)
log = logging.getLogger("qwhpi.scheduler")

STEPS = [
    "engine/scripts/03_clean.py",
    "engine/scripts/09_monthly.py",
    "engine/scripts/11_monthly_canonical.py",
]


def run_refresh() -> None:
    started = time.time()
    manifest: dict = {"started_at": datetime.now(timezone.utc).isoformat(),
                      "steps": []}
    ok = True
    for step in STEPS:
        t0 = time.time()
        proc = subprocess.run([sys.executable, str(WORKSPACE / step)],
                              cwd=WORKSPACE, capture_output=True, text=True)
        manifest["steps"].append({
            "step": step,
            "seconds": round(time.time() - t0, 1),
            "returncode": proc.returncode,
        })
        if proc.returncode != 0:
            log.error("step %s failed: %s", step, proc.stderr[-2000:])
            ok = False
            break
        log.info("step %s ok (%.0fs)", step, time.time() - t0)

    if ok:
        sys.path.insert(0, str(WORKSPACE / "engine" / "src"))
        try:
            from qwhpi.export import input_hash, load_database
            manifest["input_hash"] = input_hash()
            manifest["db_loaded"] = load_database()
        except Exception as exc:  # noqa: BLE001
            log.error("db load failed: %s", exc)
            manifest["db_loaded"] = False

    manifest["ok"] = ok
    manifest["total_seconds"] = round(time.time() - started, 1)
    RUNS_DIR.mkdir(parents=True, exist_ok=True)
    stamp = datetime.now(timezone.utc).strftime("%Y%m%dT%H%M%SZ")
    (RUNS_DIR / f"run_{stamp}.json").write_text(json.dumps(manifest, indent=2))
    log.info("refresh finished ok=%s in %.0fs", ok, manifest["total_seconds"])


if __name__ == "__main__":
    log.info("scheduler up — weekly refresh Mondays 06:00")
    schedule.every().monday.at("06:00").do(run_refresh)
    while True:
        schedule.run_pending()
        time.sleep(60)