pkg/translator/prometheus: Allow not translating UTF8-characters

Signed-off-by: Arthur Silva Sens <[email protected]>

exporter/prometheusexporter/README.md

@@ -60,6 +60,10 @@ Given the example, metrics will be available at `https://1.2.3.4:1234/metrics`.
 
 OpenTelemetry metric names and attributes are normalized to be compliant with Prometheus naming rules. [Details on this normalization process are described in the Prometheus translator module](../../pkg/translator/prometheus/).
 
+Prometheus 2.55.0 introduced support for UTF-8 characters behind the feature-flag `utf8-names`. Prometheus 3.0.0 and later accept UTF-8 by default. This means that name and attribute normalization is not required if you're using those versions. To allow UTF-8 characters to be exposed without normalization, start the collector with the feature gate: `--feature-gates=pkg.translator.prometheus.allow_utf8`.
+
+The scraper must include `scaping=allow-utf-8` in the `Accept` header for UTF-8 characters to be exposed.
+
 ## Setting resource attributes as metric labels
 
 By default, resource attributes are added to a special metric called `target_info`. To select and group by metrics by resource attributes, you [need to do join on `target_info`](https://prometheus.io/docs/prometheus/latest/querying/operators/#many-to-one-and-one-to-many-vector-matches). For example, to select metrics with `k8s_namespace_name` attribute equal to `my-namespace`:

exporter/prometheusremotewriteexporter/README.md

@@ -111,6 +111,8 @@ To enable it run collector with enabled feature gate `exporter.prometheusremotew
 
 OpenTelemetry metric names and attributes are normalized to be compliant with Prometheus naming rules. [Details on this normalization process are described in the Prometheus translator module](../../pkg/translator/prometheus/).
 
+Prometheus 2.55.0 introduced support for UTF-8 characters behind the feature-flag `utf8-names`. Prometheus 3.0.0 and later accept UTF-8 by default. This means that name and attribute normalization is not required if you're using those versions. To allow UTF-8 characters to be exposed without normalization, start the collector with the feature gate: `--feature-gates=pkg.translator.prometheus.allow_utf8`.
+
 ## Setting resource attributes as metric labels
 
 By default, resource attributes are added to a special metric called `target_info`. To select and group by metrics by resource attributes, you [need to do join on `target_info`](https://prometheus.io/docs/prometheus/latest/querying/operators/#many-to-one-and-one-to-many-vector-matches). For example, to select metrics with `k8s_namespace_name` attribute equal to `my-namespace`:

pkg/translator/prometheus/feature_gates.go

@@ -0,0 +1,39 @@
+// Copyright The OpenTelemetry Authors
+// SPDX-License-Identifier: Apache-2.0
+
+package prometheus // import "github.com/open-telemetry/opentelemetry-collector-contrib/pkg/translator/prometheus"
+
+import (
+	"github.com/prometheus/common/model"
+	"go.opentelemetry.io/collector/featuregate"
+)
+
+var (
+	dropSanitizationGate = featuregate.GlobalRegistry().MustRegister(
+		"pkg.translator.prometheus.PermissiveLabelSanitization",
+		featuregate.StageAlpha,
+		featuregate.WithRegisterDescription("Controls whether to change labels starting with '_' to 'key_'."),
+		featuregate.WithRegisterReferenceURL("https://github.com/open-telemetry/opentelemetry-collector-contrib/issues/8950"),
+	)
+
+	normalizeNameGate = featuregate.GlobalRegistry().MustRegister(
+		"pkg.translator.prometheus.NormalizeName",
+		featuregate.StageBeta,
+		featuregate.WithRegisterDescription("Controls whether metrics names are automatically normalized to follow Prometheus naming convention. Attention: if 'pkg.translator.prometheus.allowUTF8' is enabled, UTF-8 characters will not be normalized."),
+		featuregate.WithRegisterReferenceURL("https://github.com/open-telemetry/opentelemetry-collector-contrib/issues/8950"),
+	)
+
+	allowUTF8FeatureGate = featuregate.GlobalRegistry().MustRegister(
+		"pkg.translator.prometheus.allowUTF8",
+		featuregate.StageAlpha,
+		featuregate.WithRegisterDescription("When enabled, UTF-8 characters will not be translated to underscores."),
+		featuregate.WithRegisterReferenceURL("https://github.com/open-telemetry/opentelemetry-collector-contrib/issues/35459"),
+		featuregate.WithRegisterFromVersion("v0.110.0"),
+	)
+)
+
+func init() {
+	if allowUTF8FeatureGate.IsEnabled() {
+		model.NameValidationScheme = model.UTF8Validation
+	}
+}

pkg/translator/prometheus/go.mod

@@ -4,6 +4,7 @@ go 1.22.0
 
 require (
 	github.com/open-telemetry/opentelemetry-collector-contrib/internal/common v0.110.0
+	github.com/prometheus/common v0.60.0
 	github.com/stretchr/testify v1.9.0
 	go.opentelemetry.io/collector/featuregate v1.16.0
 	go.opentelemetry.io/collector/pdata v1.16.0
@@ -19,10 +20,11 @@ require (
 	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
 	github.com/modern-go/reflect2 v1.0.2 // indirect
 	github.com/pmezard/go-difflib v1.0.0 // indirect
+	github.com/prometheus/client_model v0.6.1 // indirect
 	go.uber.org/multierr v1.11.0 // indirect
-	golang.org/x/net v0.26.0 // indirect
-	golang.org/x/sys v0.21.0 // indirect
-	golang.org/x/text v0.16.0 // indirect
+	golang.org/x/net v0.29.0 // indirect
+	golang.org/x/sys v0.25.0 // indirect
+	golang.org/x/text v0.18.0 // indirect
 	google.golang.org/genproto/googleapis/rpc v0.0.0-20240604185151-ef581f913117 // indirect
 	google.golang.org/grpc v1.66.2 // indirect
 	google.golang.org/protobuf v1.34.2 // indirect

pkg/translator/prometheus/normalize_label.go

@@ -6,15 +6,6 @@ package prometheus // import "github.com/open-telemetry/opentelemetry-collector-
 import (
 	"strings"
 	"unicode"
-
-	"go.opentelemetry.io/collector/featuregate"
-)
-
-var dropSanitizationGate = featuregate.GlobalRegistry().MustRegister(
-	"pkg.translator.prometheus.PermissiveLabelSanitization",
-	featuregate.StageAlpha,
-	featuregate.WithRegisterDescription("Controls whether to change labels starting with '_' to 'key_'."),
-	featuregate.WithRegisterReferenceURL("https://github.com/open-telemetry/opentelemetry-collector-contrib/issues/8950"),
 )
 
 // Normalizes the specified label to follow Prometheus label names standard
@@ -31,8 +22,10 @@ func NormalizeLabel(label string) string {
 		return label
 	}
 
-	// Replace all non-alphanumeric runes with underscores
-	label = strings.Map(sanitizeRune, label)
+	if !allowUTF8FeatureGate.IsEnabled() {
+		// Replace all non-alphanumeric runes with underscores
+		label = strings.Map(sanitizeRune, label)
+	}
 
 	// If label starts with a number, prepend with "key_"
 	if unicode.IsDigit(rune(label[0])) {

pkg/translator/prometheus/normalize_label_test.go

@@ -11,25 +11,68 @@ import (
 	"github.com/open-telemetry/opentelemetry-collector-contrib/internal/common/testutil"
 )
 
-func TestSanitize(t *testing.T) {
+func TestNormalizeLabel(t *testing.T) {
+	testCases := []struct {
+		label           string
+		utfAllowed      bool
+		dropSantization bool
+		expected        string
+	}{
+		{
+			label:    "",
+			expected: "",
+		},
+		{
+			label:    "test",
+			expected: "test",
+		},
+		{
+			label:    "__test",
+			expected: "__test",
+		},
+		{
+			label:    "_test",
+			expected: "key_test",
+		},
+		{
+			label:           "_test",
+			dropSantization: true,
+			expected:        "_test",
+		},
+		{
+			label:    "0test",
+			expected: "key_0test",
+		},
+		{
+			label:           "0test",
+			dropSantization: true,
+			expected:        "key_0test",
+		},
+		{
+			label:    "test_/",
+			expected: "test__",
+		},
+		{
+			label:      "test_/",
+			utfAllowed: true,
+			expected:   "test_/",
+		},
+		{
+			label:    "test.test",
+			expected: "test_test",
+		},
+		{
+			label:      "test.test",
+			utfAllowed: true,
+			expected:   "test.test",
+		},
+	}
 
-	defer testutil.SetFeatureGateForTest(t, dropSanitizationGate, false)()
-
-	require.Equal(t, "", NormalizeLabel(""), "")
-	require.Equal(t, "key_test", NormalizeLabel("_test"))
-	require.Equal(t, "key_0test", NormalizeLabel("0test"))
-	require.Equal(t, "test", NormalizeLabel("test"))
-	require.Equal(t, "test__", NormalizeLabel("test_/"))
-	require.Equal(t, "__test", NormalizeLabel("__test"))
-}
-
-func TestSanitizeDropSanitization(t *testing.T) {
-
-	defer testutil.SetFeatureGateForTest(t, dropSanitizationGate, true)()
-
-	require.Equal(t, "", NormalizeLabel(""))
-	require.Equal(t, "_test", NormalizeLabel("_test"))
-	require.Equal(t, "key_0test", NormalizeLabel("0test"))
-	require.Equal(t, "test", NormalizeLabel("test"))
-	require.Equal(t, "__test", NormalizeLabel("__test"))
+	for _, tc := range testCases {
+		t.Run(tc.label, func(t *testing.T) {
+			testutil.SetFeatureGateForTest(t, allowUTF8FeatureGate, tc.utfAllowed)
+			testutil.SetFeatureGateForTest(t, dropSanitizationGate, tc.dropSantization)
+			require.Equal(t, tc.expected, NormalizeLabel(tc.label))
+		})
+	}
 }