initial commit — dropLake

backend/utils/cleanup.py

@@ -0,0 +1,63 @@
+import logging
+from datetime import datetime
+from pathlib import Path
+
+from apscheduler.schedulers.background import BackgroundScheduler
+from sqlalchemy.orm import Session
+
+from config import settings
+from database import SecurityEvent, SessionLocal, Share
+
+log = logging.getLogger("droplake.cleanup")
+
+
+def purge_expired():
+    db: Session = SessionLocal()
+    try:
+        now = datetime.utcnow()
+
+        # Expired shares
+        expired = db.query(Share).filter(Share.expires_at < now).all()
+
+        # Burn-after-read shares that have been downloaded
+        burnt = (
+            db.query(Share)
+            .filter(Share.burn_after_read.is_(True), Share.downloaded.is_(True))
+            .all()
+        )
+
+        # Deduplicate by id
+        seen = {}
+        for s in expired + burnt:
+            seen[s.id] = s
+        to_remove = list(seen.values())
+
+        for share in to_remove:
+            blob = Path(settings.storage_path) / share.storage_path
+            blob.unlink(missing_ok=True)
+            db.add(SecurityEvent(event_type="share_cleaned", share_id=share.id))
+            db.delete(share)
+
+        if to_remove:
+            db.commit()
+            log.info("Cleanup: removed %d share(s)", len(to_remove))
+    except Exception as exc:
+        log.error("Cleanup error: %s", exc)
+        db.rollback()
+    finally:
+        db.close()
+
+
+def start_scheduler() -> BackgroundScheduler:
+    scheduler = BackgroundScheduler(daemon=True)
+    scheduler.add_job(
+        purge_expired,
+        "interval",
+        minutes=settings.cleanup_interval_minutes,
+        id="cleanup",
+        replace_existing=True,
+        misfire_grace_time=60,
+    )
+    scheduler.start()
+    log.info("Cleanup scheduler started (every %dm)", settings.cleanup_interval_minutes)
+    return scheduler