add metric object type

datadog/dogstatsd/metrics.py

@@ -0,0 +1,53 @@
+class MetricAggregator(object):
+    def __init__(self, name, tags, rate, timestamp=0):
+        self.name = name
+        self.tags = tags
+        self.rate = rate
+        self.timestamp = timestamp
+
+    def aggregate(self, value):
+        raise NotImplementedError("Subclasses should implement this method.")
+
+    # TODO: This may be implemented if flushing aggregated metrics is supported
+    def unsafe_flush(self):
+        pass
+
+
+class CountMetric(MetricAggregator):
+    def __init__(self, name, value, tags, rate, timestamp=0):
+        super(CountMetric, self).__init__(name, tags, rate, timestamp)
+        self.value = value
+
+    def aggregate(self, v):
+        self.value += v
+
+    # TODO: This may be implemented if flushing aggregated metrics is supported
+    def unsafe_flush(self):
+        pass
+
+
+class GaugeMetric(MetricAggregator):
+    def __init__(self, name, value, tags, rate, timestamp=0):
+        super(GaugeMetric, self).__init__(name, tags, rate, timestamp)
+        self.value = value
+
+    def aggregate(self, v):
+        self.value = v
+
+    # TODO: This may be implemented once flushing aggregated metrics is supported
+    def unsafe_flush(self):
+        pass
+
+
+class SetMetric(MetricAggregator):
+    def __init__(self, name, value, tags, rate, timestamp=0):
+        super(SetMetric, self).__init__(name, tags, rate, timestamp)
+        self.data = set()
+        self.data.add(value)
+
+    def aggregate(self, v):
+        self.data.add(v)
+
+    # TODO: This may be implemented once flushing aggregated metrics is supported
+    def unsafe_flush(self):
+        pass

tests/unit/dogstatsd/test_metrics.py

@@ -0,0 +1,80 @@
+import unittest
+
+from datadog.dogstatsd.metrics import CountMetric, GaugeMetric, SetMetric
+
+class TestMetrics(unittest.TestCase):
+    def test_new_count_metric(self):
+        c = CountMetric("test", 21, ["tag1", "tag2"], 1, 1713804588)
+        self.assertEqual(c.value, 21)
+        self.assertEqual(c.name, "test")
+        self.assertEqual(c.tags, ["tag1", "tag2"])
+        self.assertEqual(c.rate, 1.0)
+        self.assertEqual(c.timestamp, 1713804588)
+
+        c_default_timestamp = CountMetric("test", 21, ["tag1", "tag2"], 1)
+        self.assertEqual(c_default_timestamp.value, 21)
+        self.assertEqual(c_default_timestamp.name, "test")
+        self.assertEqual(c_default_timestamp.tags, ["tag1", "tag2"])
+        self.assertEqual(c_default_timestamp.rate, 1.0)
+        self.assertEqual(c_default_timestamp.timestamp, 0)
+
+    def test_count_metric_aggregate(self):
+        c = CountMetric("test", 10, ["tag1", "tag2"], 1, 1713804588)
+        c.aggregate(20)
+        self.assertEqual(c.value, 30)
+        self.assertEqual(c.name, "test")
+        self.assertEqual(c.tags, ["tag1", "tag2"])
+        self.assertEqual(c.rate, 1.0)
+        self.assertEqual(c.timestamp, 1713804588)
+
+    def test_new_gauge_metric(self):
+        g = GaugeMetric("test", 10, ["tag1", "tag2"], 1, 1713804588)
+        self.assertEqual(g.value, 10)
+        self.assertEqual(g.name, "test")
+        self.assertEqual(g.tags, ["tag1", "tag2"])
+        self.assertEqual(g.rate, 1)
+        self.assertEqual(g.timestamp, 1713804588)
+
+        g_default_timestamp = GaugeMetric("test", 10, ["tag1", "tag2"], 1)
+        self.assertEqual(g_default_timestamp.value, 10)
+        self.assertEqual(g_default_timestamp.name, "test")
+        self.assertEqual(g_default_timestamp.tags, ["tag1", "tag2"])
+        self.assertEqual(g_default_timestamp.rate, 1)
+        self.assertEqual(g_default_timestamp.timestamp, 0)
+
+    def test_gauge_metric_aggregate(self):
+        g = GaugeMetric("test", 10, ["tag1", "tag2"], 1, 1713804588)
+        g.aggregate(20)
+        self.assertEqual(g.value, 20)
+        self.assertEqual(g.name, "test")
+        self.assertEqual(g.tags, ["tag1", "tag2"])
+        self.assertEqual(g.rate, 1.0)
+        self.assertEqual(g.timestamp, 1713804588)
+
+    def test_new_set_metric(self):
+        s = SetMetric("test", "value1", ["tag1", "tag2"], 1, 1713804588)
+        self.assertEqual(s.data, {"value1"})
+        self.assertEqual(s.name, "test")
+        self.assertEqual(s.tags, ["tag1", "tag2"])
+        self.assertEqual(s.rate, 1)
+        self.assertEqual(s.timestamp, 1713804588)
+
+        s_default_timestamp = SetMetric("test", "value1", ["tag1", "tag2"], 1)
+        self.assertEqual(s_default_timestamp.data, {"value1"})
+        self.assertEqual(s_default_timestamp.name, "test")
+        self.assertEqual(s_default_timestamp.tags, ["tag1", "tag2"])
+        self.assertEqual(s_default_timestamp.rate, 1)
+        self.assertEqual(s_default_timestamp.timestamp, 0)
+
+    def test_set_metric_aggregate(self):
+        s = SetMetric("test", "value1", ["tag1", "tag2"], 1, 1713804588)
+        s.aggregate("value2")
+        s.aggregate("value2")
+        self.assertEqual(s.data, {"value1", "value2"})
+        self.assertEqual(s.name, "test")
+        self.assertEqual(s.tags, ["tag1", "tag2"])
+        self.assertEqual(s.rate, 1)
+        self.assertEqual(s.timestamp, 1713804588)
+
+if __name__ == '__main__':
+    unittest.main()