robust parser for timestamps

analytics.py

@@ -56,6 +56,40 @@ def get_device_type(user_agent):
     else:
         return 'Other'
 
+def parse_timestamp(ts_str):
+    try:
+        # Try the normal ISO parsing.
+        return datetime.fromisoformat(ts_str)
+    except ValueError as e:
+        if 'unconverted data remains' in str(e):
+            # Find where the timezone starts. Look for a '+' or '-' after the time.
+            for sign in ['+', '-']:
+                pos = ts_str.find(sign)
+                if pos != -1:
+                    # Assume the base part is up to pos and then the tz part
+                    base = ts_str[:pos]
+                    tz_part = ts_str[pos:]
+                    # Remove any colon from the tz part to help with parsing.
+                    tz_clean = tz_part.replace(':', '')
+                    # Try parsing the base part. It might or might not have fractional seconds.
+                    try:
+                        dt = datetime.fromisoformat(base)
+                    except ValueError:
+                        dt = datetime.strptime(base, '%Y-%m-%dT%H:%M:%S')
+                    # Extract hours and minutes from the tz portion.
+                    try:
+                        offset_hours = int(tz_clean[1:3])
+                        offset_minutes = int(tz_clean[3:5])
+                    except Exception:
+                        raise ValueError(f"Unable to parse timezone from {ts_str}")
+                    offset = timedelta(hours=offset_hours, minutes=offset_minutes)
+                    if tz_clean[0] == '-':
+                        offset = -offset
+                    # Return a timezone-aware datetime.
+                    return dt.replace(tzinfo=timezone(offset))
+        # If it's some other ValueError, re-raise it.
+        raise
+
 def log_file_access(rel_path, filesize, mime, ip_address, user_agent, device_id, cached):
     """Insert a file access record into the database and prune entries older than 10 minutes."""
     global file_access_temp
@@ -70,11 +104,11 @@ def log_file_access(rel_path, filesize, mime, ip_address, user_agent, device_id,
             VALUES (?, ?, ?, ?, ?, ?, ?, ?)
         ''', (iso_ts, rel_path, filesize, mime, ip_address, user_agent, device_id, cached))
     
-    # Remove entries older than 10 minutes
+    # Remove entries older than 10 minutes using our robust parser.
     cutoff_time = datetime.now(timezone.utc).astimezone() - timedelta(minutes=10)
     file_access_temp[:] = [
         entry for entry in file_access_temp
-        if datetime.fromisoformat(entry[0]) >= cutoff_time
+        if parse_timestamp(entry[0]) >= cutoff_time
     ]
     
     # Add the new entry at the beginning of the list

app.py

@@ -364,13 +364,13 @@ def query_recent_connections():
             rows = a.return_file_access()
             connections = [
                 {
-                    'timestamp': datetime.strptime(row[0], '%Y-%m-%dT%H:%M:%S.%f').strftime('%d.%m.%Y %H:%M:%S'),
+                    'timestamp': datetime.fromisoformat(row[0]).strftime('%d.%m.%Y %H:%M:%S'),
                     'full_path': row[1],
-                    'filesize' : row[2],
-                    'mime_typ' : row[3],
+                    'filesize': row[2],
+                    'mime_typ': row[3],
                     'ip_address': row[4],
                     'user_agent': row[5],
-                    'cached':  row[7]
+                    'cached': row[7]
                 }
                 for row in rows
             ]
@@ -384,6 +384,7 @@ def query_recent_connections():
         background_thread_running = False
         print("No clients connected; stopping query thread.")
 
+
 @socketio.on('connect')
 def handle_connect(auth=None):
     global clients_connected, background_thread_running
@@ -405,7 +406,7 @@ def handle_request_initial_data():
     rows = a.return_file_access()
     connections = [
         {
-            'timestamp': datetime.strptime(row[0], '%Y-%m-%dT%H:%M:%S.%f').strftime('%d.%m.%Y %H:%M:%S'),
+            'timestamp': datetime.fromisoformat(row[0]).strftime('%d.%m.%Y %H:%M:%S'),
             'full_path': row[1],
             'filesize' : row[2],
             'mime_typ' : row[3],