Add test to Validate if duplicate metrics are intentional (#26687)

**Description:**
This test alerts a user if a metric is duplicated across receivers, and
instructs the user to add said metric to an allow list if this is
intentional.

If semantic conventions on the duplicate metric exist, validating is a
matter of ensuring they both fulfill semantic intent. For validating
duplicate metrics for which there is no semantic conventions, additional
discussion is needed with the semantic conventions WG as mentioned in
the issue. For the time being, it will be best judgment.

**Link to tracking Issue:** #26499

cmd/mdatagen/validate_test.go

@@ -4,8 +4,12 @@
 package main
 
 import (
+ "fmt"
+ "io/fs"
+ "path/filepath"
  "testing"
 
+ "github.com/stretchr/testify/assert"
  "github.com/stretchr/testify/require"
 )
 
@@ -101,3 +105,49 @@ func TestValidate(t *testing.T) {
  })
  }
 }
+
+func TestValidateMetricDuplicates(t *testing.T) {
+ allowedMetrics := map[string][]string{
+ "container.cpu.utilization": {"docker_stats", "kubeletstats"},
+ "container.memory.rss": {"docker_stats", "kubeletstats"},
+ "container.uptime": {"docker_stats", "kubeletstats"},
+ }
+ allMetrics := map[string][]string{}
+ err := filepath.Walk("../../receiver", func(path string, info fs.FileInfo, err error) error {
+ if info.Name() == "metadata.yaml" {
+ md, err := loadMetadata(path)
+ require.NoError(t, err)
+ if len(md.Metrics) > 0 {
+ for metricName := range md.Metrics {
+ allMetrics[md.Type] = append(allMetrics[md.Type], string(metricName))
+ }
+ }
+ }
+ return nil
+ })
+ require.NoError(t, err)
+
+ seen := make(map[string]string)
+ for receiver, metrics := range allMetrics {
+ for _, metricName := range metrics {
+ if val, exists := seen[metricName]; exists {
+ receivers, allowed := allowedMetrics[metricName]
+ assert.True(
+ t,
+ allowed && contains(receiver, receivers) && contains(val, receivers),
+ fmt.Sprintf("Duplicate metric %v in receivers %v and %v. Please validate that this is intentional by adding the metric name and receiver types in the allowedMetrics map in this test\n", metricName, receiver, val),
+ )
+ }
+ seen[metricName] = receiver
+ }
+ }
+}
+
+func contains(r string, rs []string) bool {
+ for _, s := range rs {
+ if s == r {
+ return true
+ }
+ }
+ return false
+}