User collector cardinality, #288

Author: akpw

mktxp/collector/user_collector.py

@@ -11,7 +11,7 @@
 ## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 ## GNU General Public License for more details.
 
-
+from datetime import datetime
 from mktxp.collector.base_collector import BaseCollector
 from mktxp.datasource.user_ds import UserMetricsDataSource
 
@@ -24,7 +24,19 @@ def collect(router_entry):
 
         user_labels = ['name', 'when', 'address', 'via', 'group']
         user_records = UserMetricsDataSource.metric_records(router_entry, metric_labels=user_labels)        
+        
         if user_records:
-            # Auto de-duplicate the records from the scrapes via API silently
-            user_metrics = BaseCollector.info_collector('active_users', 'Active Users', user_records, user_labels, verbose_reporting = False)
+            for record in user_records:
+                if 'when' in record:
+                    try:
+                        dt = datetime.strptime(record['when'], "%Y-%m-%d %H:%M:%S")
+                        record['when'] = int(dt.timestamp())
+                    except ValueError:
+                        # Use current timestamp as fallback if parsing fails
+                        record['when'] = int(datetime.now().timestamp())        
+            
+            user_labels.remove('when')
+
+            # Auto de-duplicate the records from the scrapes via API silently            
+            user_metrics = BaseCollector.gauge_collector('active_users_info', 'Active Users', user_records, 'when', user_labels, verbose_reporting = False)
             yield user_metrics

tests/collector/test_user_collector.py

@@ -15,25 +15,26 @@
 from unittest.mock import Mock, patch
 from mktxp.collector.user_collector import UserCollector
 
+# MikroTik when format: YYYY-MM-DD HH:MM:SS
 # Case 1: Records with duplicates
 records_with_duplicates = [
-    {'name': 'user1', 'when': 't1', 'address': 'a1', 'via': 'v1', 'group': 'g1'},
-    {'name': 'user2', 'when': 't2', 'address': 'a2', 'via': 'v2', 'group': 'g2'},
-    {'name': 'user1', 'when': 't1', 'address': 'a1', 'via': 'v1', 'group': 'g1'},  # Duplicate
+    {'name': 'user1', 'when': '2024-12-18 10:54:02', 'address': 'a1', 'via': 'v1', 'group': 'g1'},
+    {'name': 'user2', 'when': '2024-12-18 10:55:02', 'address': 'a2', 'via': 'v2', 'group': 'g2'},
+    {'name': 'user1', 'when': '2024-12-18 10:54:02', 'address': 'a1', 'via': 'v1', 'group': 'g1'},  # Duplicate
 ]
 expected_names_1 = {'user1', 'user2'}
 
 # Case 2: No duplicates
 records_without_duplicates = [
-    {'name': 'user1', 'when': 't1', 'address': 'a1', 'via': 'v1', 'group': 'g1'},
-    {'name': 'user2', 'when': 't2', 'address': 'a2', 'via': 'v2', 'group': 'g2'},
+    {'name': 'user1', 'when': '2024-12-18 10:54:02', 'address': 'a1', 'via': 'v1', 'group': 'g1'},
+    {'name': 'user2', 'when': '2024-12-18 10:55:02', 'address': 'a2', 'via': 'v2', 'group': 'g2'},
 ]
 expected_names_2 = {'user1', 'user2'}
 
 # Case 3: All duplicates
 records_all_duplicates = [
-    {'name': 'user1', 'when': 't1', 'address': 'a1', 'via': 'v1', 'group': 'g1'},
-    {'name': 'user1', 'when': 't1', 'address': 'a1', 'via': 'v1', 'group': 'g1'},
+    {'name': 'user1', 'when': '2024-12-18 10:54:02', 'address': 'a1', 'via': 'v1', 'group': 'g1'},
+    {'name': 'user1', 'when': '2024-12-18 10:54:02', 'address': 'a1', 'via': 'v1', 'group': 'g1'},
 ]
 expected_names_3 = {'user1'}
 
@@ -61,7 +62,7 @@ def test_user_collector_deduplicates_records(mock_metric_records, input_records,
     assert len(metrics) == 1
 
     user_metric = metrics[0]
-    assert user_metric.name == 'mktxp_active_users'
+    assert user_metric.name == 'mktxp_active_users_info'
     assert len(user_metric.samples) == expected_sample_count
 
     # Check that the correct samples are present