start work on prometheus exporter

Author: czhang771

exporter/opentelemetry-exporter-prometheus/src/opentelemetry/exporter/prometheus/__init__.py

@@ -67,7 +67,7 @@
 from json import dumps
 from logging import getLogger
 from os import environ
-from typing import Deque, Dict, Iterable, Sequence, Tuple, Union
+from typing import Deque, Dict, Iterable, Sequence, Tuple, Union, Optional
 
 from prometheus_client import start_http_server
 from prometheus_client.core import (
@@ -107,6 +107,9 @@
     Sum,
 )
 from opentelemetry.util.types import Attributes
+from opentelemetry.sdk.metrics._internal import Exemplar
+from prometheus_client.samples import Exemplar as PrometheusExemplar
+
 
 _logger = getLogger(__name__)
 
@@ -115,17 +118,29 @@
 
 
 def _convert_buckets(
-    bucket_counts: Sequence[int], explicit_bounds: Sequence[float]
-) -> Sequence[Tuple[str, int]]:
+    bucket_counts: Sequence[int], explicit_bounds: Sequence[float], exemplars: Sequence[Optional[PrometheusExemplar]] = None
+) -> Sequence[Tuple[str, int, Optional[Exemplar]]]:
     buckets = []
     total_count = 0
+    previous_bound = float('-inf')
+
     for upper_bound, count in zip(
         chain(explicit_bounds, ["+Inf"]),
         bucket_counts,
     ):
         total_count += count
-        buckets.append((f"{upper_bound}", total_count))
-
+        buckets.append((f"{upper_bound}", total_count, None))
+    
+    # assigning exemplars to their corresponding values
+    if exemplars:
+        for i, (upper_bound, _, _) in enumerate(buckets):
+            for exemplar in exemplars:
+                if previous_bound <= exemplar.value < float(upper_bound):
+                        # Assign the exemplar to the current bucket if it's the first valid one found
+                        _, current_count, current_exemplar = buckets[i]
+                        if current_exemplar is None:  # Only assign if no exemplar has been assigned yet
+                            buckets[i] = (upper_bound, current_count, exemplar)
+            previous_bound = float(upper_bound)
     return buckets
 
 
@@ -251,6 +266,10 @@ def _translate_to_prometheus(
             for number_data_point in metric.data.data_points:
                 label_keys = []
                 label_values = []
+                exemplars = [
+                    self._convert_exemplar(ex) for ex in number_data_point.exemplars
+                ]
+
 
                 for key, value in sorted(number_data_point.attributes.items()):
                     label_keys.append(sanitize_attribute(key))
@@ -276,6 +295,7 @@ def _translate_to_prometheus(
                                 number_data_point.explicit_bounds
                             ),
                             "sum": number_data_point.sum,
+                            "exemplars": exemplars,
                         }
                     )
                 else:
@@ -350,7 +370,7 @@ def _translate_to_prometheus(
                         [pre_metric_family_id, HistogramMetricFamily.__name__]
                     )
 
-                    if (
+                    if ( 
                         metric_family_id
                         not in metric_family_id_metric_family.keys()
                     ):
@@ -359,15 +379,15 @@ def _translate_to_prometheus(
                                 name=metric_name,
                                 documentation=metric_description,
                                 labels=label_keys,
-                                unit=metric_unit,
+                               unit=metric_unit,
                             )
                         )
                     metric_family_id_metric_family[
                         metric_family_id
                     ].add_metric(
                         labels=label_values,
                         buckets=_convert_buckets(
-                            value["bucket_counts"], value["explicit_bounds"]
+                            value["bucket_counts"], value["explicit_bounds"], value["exemplars"]
                         ),
                         sum_value=value["sum"],
                     )
@@ -395,7 +415,29 @@ def _create_info_metric(
         info = InfoMetricFamily(name, description, labels=attributes)
         info.add_metric(labels=list(attributes.keys()), value=attributes)
         return info
+    
+    def _convert_exemplar(self, exemplar_data: Exemplar) -> PrometheusExemplar:
+        """
+        Converts the SDK exemplar into a Prometheus Exemplar, including proper time conversion.
 
+        Parameters:
+        - value (float): The value associated with the exemplar.
+        - exemplar_data (ExemplarData): An OpenTelemetry exemplar data object containing attributes and timing information.
+
+        Returns:
+        - Exemplar: A Prometheus Exemplar object with correct labeling and timing.
+        """
+        labels = {self._sanitize_label(key): str(value) for key, value in exemplar_data.filtered_attributes.items()}
+    
+        # Add trace_id and span_id to labels only if they are valid and not None
+        if exemplar_data.trace_id and exemplar_data.span_id:
+            labels['trace_id'] = exemplar_data.trace_id
+            labels['span_id'] = exemplar_data.span_id
+    
+        # Convert time from nanoseconds to seconds
+        timestamp_seconds = exemplar_data.time_unix_nano / 1e9
+        prom_exemplar = PrometheusExemplar(labels, exemplar_data.value, timestamp_seconds)
+        return prom_exemplar
 
 class _AutoPrometheusMetricReader(PrometheusMetricReader):
     """Thin wrapper around PrometheusMetricReader used for the opentelemetry_metrics_exporter entry point.