logging

monitor.py

@@ -8,7 +8,7 @@ import html
 import threading
 import time
 import csv
-from datetime import datetime
+from datetime import datetime, timedelta
 from pathlib import Path
 
 import requests
@@ -58,6 +58,65 @@ APPLICATIONS_FILE = DATA_DIR / "applications.json"
 
 # WGcompany specific files
 WGCOMPANY_LISTINGS_FILE = DATA_DIR / "wgcompany_listings.json"
+
+
+def _cleanup_old_files(png_hours: int = 24, log_days: int = 7):
+    """Remove PNG files older than `png_hours` and prune log lines older than `log_days` days.
+
+    Runs best-effort and logs exceptions to the logger.
+    """
+    try:
+        now = datetime.utcnow()
+
+        # Remove old PNGs in DATA_DIR
+        png_cutoff = now - timedelta(hours=png_hours)
+        removed_pngs = 0
+        for p in DATA_DIR.glob("*.png"):
+            try:
+                mtime = datetime.fromtimestamp(p.stat().st_mtime)
+                if mtime < png_cutoff:
+                    p.unlink()
+                    removed_pngs += 1
+            except Exception:
+                logger.exception(f"Error while checking/removing PNG: {p}")
+        if removed_pngs:
+            logger.info(f"Removed {removed_pngs} PNG(s) older than {png_hours} hours")
+
+        # Prune logfile lines older than log_days
+        if LOG_FILE.exists():
+            cutoff_log = now - timedelta(days=log_days)
+            kept_lines = []
+            try:
+                with open(LOG_FILE, "r", encoding="utf-8", errors="ignore") as f:
+                    for line in f:
+                        # Expect logging lines starting with 'YYYY-MM-DD HH:MM:SS,ms - '
+                        m = re.match(r"^(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2},\d+)\s+-\s+", line)
+                        if m:
+                            try:
+                                ts = datetime.strptime(m.group(1), "%Y-%m-%d %H:%M:%S,%f")
+                                if ts >= cutoff_log:
+                                    kept_lines.append(line)
+                            except Exception:
+                                # If parsing fails, keep the line
+                                kept_lines.append(line)
+                        else:
+                            # Keep non-standard lines
+                            kept_lines.append(line)
+                # Atomically replace the logfile with kept lines
+                if kept_lines:
+                    tmp = LOG_FILE.with_suffix(".tmp")
+                    with open(tmp, "w", encoding="utf-8") as f:
+                        f.writelines(kept_lines)
+                    tmp.replace(LOG_FILE)
+                else:
+                    # No recent lines; truncate the file
+                    with open(LOG_FILE, "w", encoding="utf-8") as f:
+                        f.truncate(0)
+                logger.info(f"Pruned logfile, kept {len(kept_lines)} lines from last {log_days} days")
+            except Exception:
+                logger.exception("Error while pruning logfile")
+    except Exception:
+        logger.exception("Unexpected error in cleanup task")
 WGCOMPANY_TIMING_FILE = DATA_DIR / "wgcompany_times.csv"
 
 # Setup logging
@@ -2082,6 +2141,9 @@ def main():
     logger.info(f"InBerlin Autopilot: {'ENABLED' if inberlin_monitor.is_autopilot_enabled() else 'DISABLED'}")
     logger.info(f"WGcompany: {'ENABLED' if WGCOMPANY_ENABLED else 'DISABLED'}")
 
+    # Run periodic cleanup hourly
+    last_cleanup = 0
+
     while True:
         # Check InBerlinWohnen
         try:
@@ -2089,6 +2151,15 @@ def main():
         except Exception as e:
             logger.error(f"InBerlin check failed: {e}")
 
+        # Periodic cleanup: remove PNGs older than 24h and prune logs older than 7 days
+        try:
+            if time.time() - last_cleanup > 3600:  # every hour
+                logger.info("Running periodic cleanup (old PNGs, prune logs)")
+                _cleanup_old_files(png_hours=24, log_days=7)
+                last_cleanup = time.time()
+        except Exception:
+            logger.exception("Cleanup failed")
+
         # Check WGcompany
         if wgcompany_monitor:
             try: