Compare commits

..

3 Commits

Author SHA1 Message Date
srikanthccv
59329d3aaa feat: support semconv evolution in services 2026-08-07 01:20:49 +05:30
srikanthccv
7c58b6af49 feat: resolve semantic convention names in trace queries 2026-08-07 01:20:08 +05:30
srikanthccv
e730192885 feat: generate semantic convention families 2026-08-07 01:20:01 +05:30
33 changed files with 3257 additions and 157 deletions

View File

@@ -53,6 +53,21 @@ jobs:
with:
PRIMUS_REF: main
GO_VERSION: 1.24
semconv-generated:
if: |
github.event_name == 'merge_group' ||
(github.event_name == 'pull_request' && ! github.event.pull_request.head.repo.fork && github.event.pull_request.user.login != 'dependabot[bot]' && ! contains(github.event.pull_request.labels.*.name, 'safe-to-test')) ||
(github.event_name == 'pull_request_target' && contains(github.event.pull_request.labels.*.name, 'safe-to-test'))
runs-on: ubuntu-latest
steps:
- name: self-checkout
uses: actions/checkout@v4
- name: go-install
uses: actions/setup-go@v5
with:
go-version: "1.24"
- name: check-semconv-generated-files
run: go run ./scripts/semconv -check
build:
if: |
github.event_name == 'merge_group' ||

View File

@@ -233,6 +233,10 @@ py-clean: ## Clear all pycache and pytest cache from tests directory recursively
##############################################################
# generate commands
##############################################################
.PHONY: semconv-generate
semconv-generate: ## Regenerate semantic-convention families for Go and TypeScript
@go run ./scripts/semconv
.PHONY: gen-mocks
gen-mocks:
@echo ">> Generating mocks"

View File

@@ -0,0 +1,32 @@
// Code generated by scripts/semconv. DO NOT EDIT.
export type SemconvFamily = {
readonly current: string;
readonly old: readonly string[];
readonly kind: 'attribute' | 'metric';
readonly contexts: readonly string[];
readonly signals: readonly string[];
readonly applyToMetrics: readonly string[];
readonly valueMap: Readonly<Record<string, string>>;
};
export const SEMCONV_FAMILIES: readonly SemconvFamily[] = [
{
current: 'db.system.name',
old: ['db.system'],
kind: 'attribute',
contexts: [],
signals: [],
applyToMetrics: [],
valueMap: {},
},
{
current: 'deployment.environment.name',
old: ['deployment.environment'],
kind: 'attribute',
contexts: [],
signals: [],
applyToMetrics: [],
valueMap: {},
},
] as const;

View File

@@ -6,6 +6,8 @@ import (
v3 "github.com/SigNoz/signoz/pkg/query-service/model/v3"
"github.com/SigNoz/signoz/pkg/query-service/utils"
"github.com/SigNoz/signoz/pkg/semconv"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
)
var resourceLogOperators = map[v3.FilterOperator]string{
@@ -29,13 +31,61 @@ var resourceLogOperators = map[v3.FilterOperator]string{
v3.FilterOperatorNotILike: "NOT ILIKE",
}
func resourceSemconvMembers(key string) []string {
return semconv.Members(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: key,
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
})
}
func resourceValueExpression(key string) string {
members := resourceSemconvMembers(key)
if len(members) == 1 {
return fmt.Sprintf("simpleJSONExtractString(labels, '%s')", key)
}
values := make([]string, 0, len(members))
for _, member := range members {
values = append(values, fmt.Sprintf("NULLIF(simpleJSONExtractString(labels, '%s'), '')", member))
}
return "COALESCE(" + strings.Join(values, ", ") + ")"
}
func resourcePresenceExpression(key string, exists bool) string {
members := resourceSemconvMembers(key)
if len(members) == 1 {
if exists {
return fmt.Sprintf("simpleJSONHas(labels, '%s')", key)
}
return fmt.Sprintf("not simpleJSONHas(labels, '%s')", key)
}
conditions := make([]string, 0, len(members))
for _, member := range members {
if exists {
conditions = append(conditions, fmt.Sprintf("simpleJSONHas(labels, '%s')", member))
} else {
conditions = append(conditions, fmt.Sprintf("not simpleJSONHas(labels, '%s')", member))
}
}
separator := " OR "
if !exists {
separator = " AND "
}
return "(" + strings.Join(conditions, separator) + ")"
}
// buildResourceFilter builds a clickhouse filter string for resource labels
func buildResourceFilter(logsOp string, key string, op v3.FilterOperator, value interface{}) string {
// for all operators except contains and like
searchKey := fmt.Sprintf("simpleJSONExtractString(labels, '%s')", key)
searchKey := resourceValueExpression(key)
// for contains and like it will be case insensitive
lowerSearchKey := fmt.Sprintf("simpleJSONExtractString(lower(labels), '%s')", key)
if len(resourceSemconvMembers(key)) > 1 {
lowerSearchKey = "lower(" + searchKey + ")"
}
chFmtVal := utils.ClickHouseFormattedValue(value)
@@ -43,9 +93,9 @@ func buildResourceFilter(logsOp string, key string, op v3.FilterOperator, value
switch op {
case v3.FilterOperatorExists:
return fmt.Sprintf("simpleJSONHas(labels, '%s')", key)
return resourcePresenceExpression(key, true)
case v3.FilterOperatorNotExists:
return fmt.Sprintf("not simpleJSONHas(labels, '%s')", key)
return resourcePresenceExpression(key, false)
case v3.FilterOperatorRegex, v3.FilterOperatorNotRegex:
return fmt.Sprintf(logsOp, searchKey, chFmtVal)
case v3.FilterOperatorContains, v3.FilterOperatorNotContains:
@@ -110,6 +160,38 @@ func buildIndexFilterForInOperator(key string, op v3.FilterOperator, value inter
// we can use lower index for =, in etc but it's difficult to do it for !=, NIN etc
// if as x != "ABC" we cannot predict something like "not lower(labels) like '%%x%%abc%%'". It has it be "not lower(labels) like '%%x%%ABC%%'"
func buildResourceIndexFilter(key string, op v3.FilterOperator, value interface{}) string {
return buildResourceIndexFilterForKey(key, op, value, true)
}
func buildResourceIndexFilterForKey(key string, op v3.FilterOperator, value interface{}, resolveFamily bool) string {
members := []string{key}
if resolveFamily {
members = resourceSemconvMembers(key)
}
if len(members) > 1 {
switch op {
case v3.FilterOperatorNotEqual,
v3.FilterOperatorNotLike,
v3.FilterOperatorNotILike,
v3.FilterOperatorNotContains,
v3.FilterOperatorNotExists,
v3.FilterOperatorNotRegex,
v3.FilterOperatorNotIn:
return ""
}
conditions := make([]string, 0, len(members))
for _, member := range members {
if condition := buildResourceIndexFilterForKey(member, op, value, false); condition != "" {
conditions = append(conditions, condition)
}
}
if len(conditions) == 0 {
return ""
}
return "(" + strings.Join(conditions, " OR ") + ")"
}
// not using clickhouseFormattedValue as we don't wan't the quotes
strVal := fmt.Sprintf("%s", value)
fmtValEscapedForContains := utils.QuoteEscapedStringForContains(strVal, true)
@@ -206,14 +288,31 @@ func buildResourceFiltersFromGroupBy(groupBy []v3.AttributeKey) []string {
if attr.Type != v3.AttributeKeyTypeResource {
continue
}
conditions = append(conditions, fmt.Sprintf("(simpleJSONHas(labels, '%s') AND labels like '%%%s%%')", attr.Key, attr.Key))
members := resourceSemconvMembers(attr.Key)
if len(members) == 1 {
conditions = append(conditions, fmt.Sprintf("(simpleJSONHas(labels, '%s') AND labels like '%%%s%%')", attr.Key, attr.Key))
continue
}
indexConditions := make([]string, 0, len(members))
for _, member := range members {
indexConditions = append(indexConditions, fmt.Sprintf("labels like '%%%s%%'", member))
}
conditions = append(conditions, fmt.Sprintf("(%s AND (%s))", resourcePresenceExpression(attr.Key, true), strings.Join(indexConditions, " OR ")))
}
return conditions
}
func buildResourceFiltersFromAggregateAttribute(aggregateAttribute v3.AttributeKey) string {
if aggregateAttribute.Key != "" && aggregateAttribute.Type == v3.AttributeKeyTypeResource {
return fmt.Sprintf("(simpleJSONHas(labels, '%s') AND labels like '%%%s%%')", aggregateAttribute.Key, aggregateAttribute.Key)
members := resourceSemconvMembers(aggregateAttribute.Key)
if len(members) == 1 {
return fmt.Sprintf("(simpleJSONHas(labels, '%s') AND labels like '%%%s%%')", aggregateAttribute.Key, aggregateAttribute.Key)
}
indexConditions := make([]string, 0, len(members))
for _, member := range members {
indexConditions = append(indexConditions, fmt.Sprintf("labels like '%%%s%%'", member))
}
return fmt.Sprintf("(%s AND (%s))", resourcePresenceExpression(aggregateAttribute.Key, true), strings.Join(indexConditions, " OR "))
}
return ""

View File

@@ -5,6 +5,8 @@ import (
"testing"
v3 "github.com/SigNoz/signoz/pkg/query-service/model/v3"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func Test_buildResourceFilter(t *testing.T) {
@@ -552,3 +554,38 @@ func Test_buildResourceSubQuery(t *testing.T) {
})
}
}
func TestSemanticConventionResourceFamily(t *testing.T) {
const resolvedValue = "COALESCE(NULLIF(simpleJSONExtractString(labels, 'deployment.environment.name'), ''), NULLIF(simpleJSONExtractString(labels, 'deployment.environment'), ''))"
for _, requestedName := range []string{"deployment.environment.name", "deployment.environment"} {
t.Run(requestedName, func(t *testing.T) {
assert.Equal(t, resolvedValue+" = 'production'", buildResourceFilter("=", requestedName, v3.FilterOperatorEqual, "production"))
assert.Equal(t, "(simpleJSONHas(labels, 'deployment.environment.name') OR simpleJSONHas(labels, 'deployment.environment'))", buildResourceFilter("", requestedName, v3.FilterOperatorExists, nil))
assert.Equal(t, "(not simpleJSONHas(labels, 'deployment.environment.name') AND not simpleJSONHas(labels, 'deployment.environment'))", buildResourceFilter("", requestedName, v3.FilterOperatorNotExists, nil))
assert.Equal(t, "(labels like '%deployment.environment.name\":\"production%' OR labels like '%deployment.environment\":\"production%')", buildResourceIndexFilter(requestedName, v3.FilterOperatorEqual, "production"))
assert.Empty(t, buildResourceIndexFilter(requestedName, v3.FilterOperatorNotEqual, "production"), "negative family filter must not use a rejecting index hint")
})
}
filters, err := buildResourceFiltersFromFilterItems(&v3.FilterSet{Items: []v3.FilterItem{{
Key: v3.AttributeKey{
Key: "deployment.environment.name",
DataType: v3.AttributeKeyDataTypeString,
Type: v3.AttributeKeyTypeResource,
},
Operator: v3.FilterOperatorEqual,
Value: "production",
}}})
require.NoError(t, err, "family filter items must build before their output is inspected")
wantFilters := []string{
resolvedValue + " = 'production'",
"(labels like '%deployment.environment.name\":\"production%' OR labels like '%deployment.environment\":\"production%')",
}
assert.Equal(t, wantFilters, filters)
wantPresence := "((simpleJSONHas(labels, 'deployment.environment.name') OR simpleJSONHas(labels, 'deployment.environment')) AND (labels like '%deployment.environment.name%' OR labels like '%deployment.environment%'))"
groupBy := buildResourceFiltersFromGroupBy([]v3.AttributeKey{{Key: "deployment.environment", Type: v3.AttributeKeyTypeResource}})
assert.Equal(t, []string{wantPresence}, groupBy)
assert.Equal(t, wantPresence, buildResourceFiltersFromAggregateAttribute(v3.AttributeKey{Key: "deployment.environment.name", Type: v3.AttributeKeyTypeResource}))
}

View File

@@ -6,16 +6,32 @@ import (
"github.com/ClickHouse/clickhouse-go/v2"
"github.com/SigNoz/signoz/pkg/query-service/model"
"github.com/SigNoz/signoz/pkg/semconv"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
)
var (
columns = map[string]struct{}{
"deployment_environment": {},
"k8s_cluster_name": {},
"k8s_namespace_name": {},
}
columns = serviceMapColumns()
)
func serviceMapColumns() map[string]string {
columns := map[string]string{
"k8s_cluster_name": "k8s_cluster_name",
"k8s_namespace_name": "k8s_namespace_name",
}
// Dependency-graph rows keep their historical physical column name. Both
// semantic-convention request spellings target that same derived column.
for _, member := range semconv.Members(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: "deployment.environment.name",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
}) {
columns[strings.ReplaceAll(member, ".", "_")] = "deployment_environment"
}
return columns
}
func BuildServiceMapQuery(tags []model.TagQuery) (string, []interface{}) {
var filterQuery string
var namedArgs []interface{}
@@ -24,39 +40,40 @@ func BuildServiceMapQuery(tags []model.TagQuery) (string, []interface{}) {
operator := tag.GetOperator()
value := tag.GetValues()
if _, ok := columns[key]; !ok {
column, ok := columns[key]
if !ok {
continue
}
switch operator {
case model.InOperator:
filterQuery += fmt.Sprintf(" AND %s IN @%s", key, key)
filterQuery += fmt.Sprintf(" AND %s IN @%s", column, key)
namedArgs = append(namedArgs, clickhouse.Named(key, value))
case model.NotInOperator:
filterQuery += fmt.Sprintf(" AND %s NOT IN @%s", key, key)
filterQuery += fmt.Sprintf(" AND %s NOT IN @%s", column, key)
namedArgs = append(namedArgs, clickhouse.Named(key, value))
case model.EqualOperator:
filterQuery += fmt.Sprintf(" AND %s = @%s", key, key)
filterQuery += fmt.Sprintf(" AND %s = @%s", column, key)
namedArgs = append(namedArgs, clickhouse.Named(key, value))
case model.NotEqualOperator:
filterQuery += fmt.Sprintf(" AND %s != @%s", key, key)
filterQuery += fmt.Sprintf(" AND %s != @%s", column, key)
namedArgs = append(namedArgs, clickhouse.Named(key, value))
case model.ContainsOperator:
filterQuery += fmt.Sprintf(" AND %s LIKE @%s", key, key)
filterQuery += fmt.Sprintf(" AND %s LIKE @%s", column, key)
namedArgs = append(namedArgs, clickhouse.Named(key, fmt.Sprintf("%%%s%%", value)))
case model.NotContainsOperator:
filterQuery += fmt.Sprintf(" AND %s NOT LIKE @%s", key, key)
filterQuery += fmt.Sprintf(" AND %s NOT LIKE @%s", column, key)
namedArgs = append(namedArgs, clickhouse.Named(key, fmt.Sprintf("%%%s%%", value)))
case model.StartsWithOperator:
filterQuery += fmt.Sprintf(" AND %s LIKE @%s", key, key)
filterQuery += fmt.Sprintf(" AND %s LIKE @%s", column, key)
namedArgs = append(namedArgs, clickhouse.Named(key, fmt.Sprintf("%s%%", value)))
case model.NotStartsWithOperator:
filterQuery += fmt.Sprintf(" AND %s NOT LIKE @%s", key, key)
filterQuery += fmt.Sprintf(" AND %s NOT LIKE @%s", column, key)
namedArgs = append(namedArgs, clickhouse.Named(key, fmt.Sprintf("%s%%", value)))
case model.ExistsOperator:
filterQuery += fmt.Sprintf(" AND %s IS NOT NULL", key)
filterQuery += fmt.Sprintf(" AND %s IS NOT NULL", column)
case model.NotExistsOperator:
filterQuery += fmt.Sprintf(" AND %s IS NULL", key)
filterQuery += fmt.Sprintf(" AND %s IS NULL", column)
}
}
return filterQuery, namedArgs

View File

@@ -0,0 +1,35 @@
package services
import (
"testing"
"github.com/ClickHouse/clickhouse-go/v2/lib/driver"
"github.com/SigNoz/signoz/pkg/query-service/model"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestBuildServiceMapQueryAcceptsEnvironmentFamily(t *testing.T) {
for _, requestedName := range []string{"deployment.environment.name", "deployment.environment"} {
t.Run(requestedName, func(t *testing.T) {
tags := []model.TagQuery{model.NewTagQueryString(model.TagQueryParam{
Key: requestedName,
StringValues: []string{"production"},
Operator: model.InOperator,
TagType: model.ResourceAttributeTagType,
})}
query, args := BuildServiceMapQuery(tags)
argName := "deployment_environment"
if requestedName == "deployment.environment.name" {
argName = "deployment_environment_name"
}
assert.Equal(t, " AND deployment_environment IN @"+argName, query)
require.Len(t, args, 1)
named, ok := args[0].(driver.NamedValue)
require.True(t, ok)
assert.Equal(t, argName, named.Name)
assert.Equal(t, []interface{}{"production"}, named.Value)
})
}
}

View File

@@ -10,6 +10,7 @@ import (
"github.com/SigNoz/signoz/pkg/errors"
grammar "github.com/SigNoz/signoz/pkg/parser/filterquery/grammar"
"github.com/SigNoz/signoz/pkg/semconv"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
@@ -982,25 +983,78 @@ func assignIfEmpty(s *string, value string) {
// MatchingFieldKeys returns the field keys from the map that match the given key,
// honoring any context/data type the user specified.
func MatchingFieldKeys(field *telemetrytypes.TelemetryFieldKey, fieldKeys map[string][]*telemetrytypes.TelemetryFieldKey) []*telemetrytypes.TelemetryFieldKey {
fieldKeysForName := []*telemetrytypes.TelemetryFieldKey{}
selector := telemetrytypes.FieldKeySelector{
Name: field.Name,
Signal: field.Signal,
FieldContext: field.FieldContext,
}
members := semconv.Members(semconv.KindAttribute, selector)
isFamily := len(members) > 1
fieldKeysForName := make([]*telemetrytypes.TelemetryFieldKey, 0)
indexByIdentity := make(map[string]int)
// match by name; keep items whose context and data type match (unspecified matches any)
for _, item := range fieldKeys[field.Name] {
if (field.FieldContext == telemetrytypes.FieldContextUnspecified || field.FieldContext == item.FieldContext) &&
(field.FieldDataType == telemetrytypes.FieldDataTypeUnspecified || field.FieldDataType == item.FieldDataType) {
fieldKeysForName = append(fieldKeysForName, item)
appendMatches := func(lookupName string, memberName string, contextAlreadyMatched bool) {
for _, item := range fieldKeys[lookupName] {
if !contextAlreadyMatched && field.FieldContext != telemetrytypes.FieldContextUnspecified && field.FieldContext != item.FieldContext {
continue
}
if field.FieldDataType != telemetrytypes.FieldDataTypeUnspecified && field.FieldDataType != item.FieldDataType {
continue
}
// A wildcard lookup may have found a same-named field in a scope where
// this family does not apply. Keep exact names, but reject cross-member
// matches outside the generated family scope.
if memberName != field.Name {
itemSelector := telemetrytypes.FieldKeySelector{
Name: field.Name,
Signal: item.Signal,
FieldContext: item.FieldContext,
}
if !slices.Contains(semconv.Members(semconv.KindAttribute, itemSelector), memberName) {
continue
}
}
physicalMembers := item.SemconvMembers
if len(physicalMembers) == 0 {
physicalMembers = []string{memberName}
}
identity := item.Signal.StringValue() + ";" + item.FieldContext.StringValue() + ";" + item.FieldDataType.StringValue()
if isFamily {
if index, found := indexByIdentity[identity]; found {
for _, physicalMember := range physicalMembers {
if !slices.Contains(fieldKeysForName[index].SemconvMembers, physicalMember) {
fieldKeysForName[index].SemconvMembers = append(fieldKeysForName[index].SemconvMembers, physicalMember)
}
}
continue
}
indexByIdentity[identity] = len(fieldKeysForName)
}
resolved := *item
// The requested spelling is the response identity. Field mappers use
// it to resolve the available family members current-first.
if isFamily {
resolved.Name = field.Name
resolved.SemconvMembers = slices.Clone(physicalMembers)
}
fieldKeysForName = append(fieldKeysForName, &resolved)
}
}
// A context may have been split off a name that legitimately contained it (e.g.
// `attribute.key`); also look up the context-prefixed name so both readings resolve.
// Members are current-first, so metadata from the current key wins when
// both spellings describe the same signal/context/type.
for _, member := range members {
appendMatches(member, member, false)
}
// A context may have been split off a name that legitimately contained it
// (e.g. `attribute.key`); preserve that historical alternate reading for
// every family member.
if field.FieldContext != telemetrytypes.FieldContextUnspecified {
contextPrefixedFieldName := fmt.Sprintf("%s.%s", field.FieldContext.StringValue(), field.Name)
for _, item := range fieldKeys[contextPrefixedFieldName] {
// Context already matched via the lookup key; only data type needs checking.
if field.FieldDataType == telemetrytypes.FieldDataTypeUnspecified || item.FieldDataType == field.FieldDataType {
fieldKeysForName = append(fieldKeysForName, item)
}
for _, member := range members {
appendMatches(fmt.Sprintf("%s.%s", field.FieldContext.StringValue(), member), member, true)
}
}

View File

@@ -14,6 +14,7 @@ import (
"github.com/antlr4-go/antlr/v4"
sqlbuilder "github.com/huandu/go-sqlbuilder"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// TestPrepareWhereClause_EmptyVariableList ensures PrepareWhereClause errors when a variable has an empty list value.
@@ -685,6 +686,53 @@ func TestVisitKey(t *testing.T) {
}
}
func TestMatchingFieldKeysResolvesSemconvFamily(t *testing.T) {
current := &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment.name",
Description: "current metadata",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
old := &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment",
Description: "old metadata",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
fieldKeys := map[string][]*telemetrytypes.TelemetryFieldKey{
current.Name: {current},
old.Name: {old},
}
for _, requestedName := range []string{current.Name, old.Name} {
requested := telemetrytypes.NewTelemetryFieldKey(
requestedName,
telemetrytypes.FieldContextResource,
telemetrytypes.FieldDataTypeString,
)
matches := MatchingFieldKeys(requested, fieldKeys)
require.Len(t, matches, 1, "family lookup must return one field before its metadata is inspected")
assert.Equal(t, requestedName, matches[0].Name)
assert.Equal(t, "current metadata", matches[0].Description)
assert.Equal(t, []string{current.Name, old.Name}, matches[0].SemconvMembers)
}
// A current-name query still resolves when metadata has seen only the old
// spelling. The returned name remains the request identity.
requested := telemetrytypes.NewTelemetryFieldKey(
current.Name,
telemetrytypes.FieldContextResource,
telemetrytypes.FieldDataTypeString,
)
matches := MatchingFieldKeys(requested, map[string][]*telemetrytypes.TelemetryFieldKey{old.Name: {old}})
require.Len(t, matches, 1, "family lookup must return one field before its metadata is inspected")
assert.Equal(t, current.Name, matches[0].Name)
assert.Equal(t, "old metadata", matches[0].Description)
assert.Equal(t, []string{old.Name}, matches[0].SemconvMembers)
}
// ---------------------------------------------------------------------------
// TestVisitComparison
// ---------------------------------------------------------------------------

3
pkg/semconv/doc.go Normal file
View File

@@ -0,0 +1,3 @@
// Package semconv resolves historical OpenTelemetry semantic-convention names
// into logical field families.
package semconv

View File

@@ -0,0 +1,22 @@
// Code generated by scripts/semconv. DO NOT EDIT.
package semconv
var families = []Family{
{
Current: "db.system.name",
Old: []string{"db.system"},
Kind: KindAttribute,
Contexts: nil,
Signals: nil,
ApplyToMetrics: nil,
},
{
Current: "deployment.environment.name",
Old: []string{"deployment.environment"},
Kind: KindAttribute,
Contexts: nil,
Signals: nil,
ApplyToMetrics: nil,
},
}

146
pkg/semconv/semconv.go Normal file
View File

@@ -0,0 +1,146 @@
package semconv
import (
"slices"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
)
//go:generate go run ../../scripts/semconv
// Kind identifies whether a family describes an attribute or a metric name.
type Kind struct {
valuer.String
}
// Family is one logical telemetry field. Old is ordered from the most recent
// predecessor to the oldest one and therefore also defines fallback order.
type Family struct {
Current string
Old []string
Kind Kind
Contexts []telemetrytypes.FieldContext
Signals []telemetrytypes.Signal
ApplyToMetrics []string
ValueMap map[string]string
}
var (
KindAttribute = Kind{String: valuer.NewString("attribute")}
KindMetric = Kind{String: valuer.NewString("metric")}
)
var memberToFamilies, familyMembers = buildIndexes()
// Enum returns the acceptable values for Kind.
func (Kind) Enum() []any {
return []any{KindAttribute, KindMetric}
}
// Lookup returns the enabled family containing selector.Name for kind. The
// slices and map in the returned value reference immutable generated data and
// must be treated as read-only. Use All when a mutable copy is required.
func Lookup(kind Kind, selector telemetrytypes.FieldKeySelector) (Family, bool) {
idx, ok := lookupIndex(kind, selector)
if !ok {
return Family{}, false
}
return families[idx], true
}
// Members returns the current name first, followed by historical names in
// fallback order. For enabled families the returned precomputed slice is
// read-only. A name outside an enabled family is returned unchanged.
func Members(kind Kind, selector telemetrytypes.FieldKeySelector) []string {
idx, ok := lookupIndex(kind, selector)
if !ok {
return []string{selector.Name}
}
return familyMembers[idx]
}
// Current returns the current name for selector.Name, or the input name when
// it does not belong to an enabled family.
func Current(kind Kind, selector telemetrytypes.FieldKeySelector) string {
idx, ok := lookupIndex(kind, selector)
if !ok {
return selector.Name
}
return families[idx].Current
}
// All returns a defensive copy of every enabled family.
func All() []Family {
result := make([]Family, len(families))
for i := range families {
result[i] = cloneFamily(families[i])
}
return result
}
func buildIndexes() (map[string][]int, [][]string) {
index := make(map[string][]int)
members := make([][]string, len(families))
for i, family := range families {
members[i] = make([]string, 0, len(family.Old)+1)
members[i] = append(members[i], family.Current)
members[i] = append(members[i], family.Old...)
index[family.Current] = append(index[family.Current], i)
for _, old := range family.Old {
index[old] = append(index[old], i)
}
}
return index, members
}
func lookupIndex(kind Kind, selector telemetrytypes.FieldKeySelector) (int, bool) {
for _, idx := range memberToFamilies[selector.Name] {
if matchesSelector(families[idx], kind, selector) {
return idx, true
}
}
return 0, false
}
func matchesSelector(family Family, kind Kind, selector telemetrytypes.FieldKeySelector) bool {
if family.Kind != kind {
return false
}
if selector.Signal != telemetrytypes.SignalUnspecified && len(family.Signals) > 0 {
if !slices.Contains(family.Signals, selector.Signal) {
return false
}
}
if selector.FieldContext != telemetrytypes.FieldContextUnspecified && len(family.Contexts) > 0 {
if !slices.Contains(family.Contexts, selector.FieldContext) {
return false
}
}
if selector.Signal == telemetrytypes.SignalMetrics && len(family.ApplyToMetrics) > 0 {
if selector.MetricContext == nil {
return false
}
return slices.Contains(family.ApplyToMetrics, selector.MetricContext.MetricName)
}
return true
}
func cloneFamily(family Family) Family {
valueMap := family.ValueMap
family.Old = slices.Clone(family.Old)
family.Contexts = slices.Clone(family.Contexts)
family.Signals = slices.Clone(family.Signals)
family.ApplyToMetrics = slices.Clone(family.ApplyToMetrics)
if valueMap != nil {
family.ValueMap = make(map[string]string, len(valueMap))
for old, current := range valueMap {
family.ValueMap[old] = current
}
}
return family
}

View File

@@ -0,0 +1,59 @@
package semconv
import (
"testing"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestMembersReturnsCurrentBeforeHistoricalName(t *testing.T) {
selector := telemetrytypes.FieldKeySelector{
Name: "deployment.environment",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
}
assert.Equal(t,
[]string{"deployment.environment.name", "deployment.environment"},
Members(KindAttribute, selector),
"members should use current-first fallback order",
)
}
func TestCurrentReturnsCanonicalName(t *testing.T) {
selector := telemetrytypes.FieldKeySelector{
Name: "deployment.environment",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
}
assert.Equal(t,
"deployment.environment.name",
Current(KindAttribute, selector),
"historical name should resolve to the current family name",
)
}
func TestMembersReturnsInputWhenKindDoesNotMatch(t *testing.T) {
selector := telemetrytypes.FieldKeySelector{
Name: "deployment.environment",
Signal: telemetrytypes.SignalTraces,
}
assert.Equal(t,
[]string{"deployment.environment"},
Members(KindMetric, selector),
"an attribute family must not match a metric-name lookup",
)
}
func TestAllReturnsDefensiveCopies(t *testing.T) {
first := All()
require.NotEmpty(t, first, "generated families must not be empty")
first[0].Old[0] = "mutated"
second := All()
assert.NotEqual(t, "mutated", second[0].Old[0], "All must not expose mutable generated data")
}

View File

@@ -235,6 +235,7 @@ func NewSQLMigrationProviderFactories(
sqlmigration.NewFillDashboardSpecCollectionsFactory(sqlstore, dashboardStore),
sqlmigration.NewScrubEmailChannelTransportFactory(sqlstore),
sqlmigration.NewAddDashboardTuplesFactory(sqlstore),
sqlmigration.NewMigrateDeploymentEnvironmentQuickFilterFactory(),
)
}

View File

@@ -0,0 +1,127 @@
package sqlmigration
import (
"context"
"encoding/json"
"log/slog"
"time"
"github.com/SigNoz/signoz/pkg/factory"
"github.com/SigNoz/signoz/pkg/semconv"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/uptrace/bun"
"github.com/uptrace/bun/migrate"
)
const deploymentEnvironmentCurrent = "deployment.environment.name"
type migrateDeploymentEnvironmentQuickFilter struct {
logger *slog.Logger
}
type semconvQuickFilterRow struct {
bun.BaseModel `bun:"table:quick_filter"`
ID string `bun:"id"`
Filter string `bun:"filter"`
}
func NewMigrateDeploymentEnvironmentQuickFilterFactory() factory.ProviderFactory[SQLMigration, Config] {
return factory.NewProviderFactory(
factory.MustNewName("migrate_semconv_quick_filter"),
func(_ context.Context, settings factory.ProviderSettings, _ Config) (SQLMigration, error) {
return &migrateDeploymentEnvironmentQuickFilter{logger: settings.Logger}, nil
},
)
}
func (migration *migrateDeploymentEnvironmentQuickFilter) Register(migrations *migrate.Migrations) error {
return migrations.Register(migration.Up, migration.Down)
}
func deploymentEnvironmentOld() string {
members := semconv.Members(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: deploymentEnvironmentCurrent,
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
})
if len(members) < 2 {
return deploymentEnvironmentCurrent
}
return members[1]
}
func rewriteQuickFilterSemconv(filterJSON, from, to string) (string, bool, error) {
var filters []map[string]any
if err := json.Unmarshal([]byte(filterJSON), &filters); err != nil {
return "", false, err
}
changed := false
for _, filter := range filters {
if key, ok := filter["key"].(string); ok && key == from {
filter["key"] = to
changed = true
}
}
if !changed {
return filterJSON, false, nil
}
rewritten, err := json.Marshal(filters)
if err != nil {
return "", false, err
}
return string(rewritten), true, nil
}
func (migration *migrateDeploymentEnvironmentQuickFilter) migrate(ctx context.Context, db *bun.DB, from, to string) error {
tx, err := db.BeginTx(ctx, nil)
if err != nil {
return err
}
defer func() { _ = tx.Rollback() }()
rows := make([]*semconvQuickFilterRow, 0)
if err := tx.NewSelect().
Model(&rows).
Where("signal IN (?)", bun.In([]string{"traces", "api_monitoring", "exceptions"})).
Scan(ctx); err != nil {
return err
}
for _, row := range rows {
rewritten, changed, err := rewriteQuickFilterSemconv(row.Filter, from, to)
if err != nil {
// Quick filters are user-editable. One malformed legacy row must not
// prevent the application from starting or block every other org's
// migration.
if migration.logger != nil {
migration.logger.WarnContext(ctx, "skipping quick filter with unreadable filter JSON",
slog.String("quick_filter_id", row.ID), slog.Any("error", err))
}
continue
}
if !changed {
continue
}
if _, err := tx.NewUpdate().
Model((*semconvQuickFilterRow)(nil)).
Set("filter = ?", rewritten).
Set("updated_at = ?", time.Now()).
Where("id = ?", row.ID).
Exec(ctx); err != nil {
return err
}
}
return tx.Commit()
}
func (migration *migrateDeploymentEnvironmentQuickFilter) Up(ctx context.Context, db *bun.DB) error {
return migration.migrate(ctx, db, deploymentEnvironmentOld(), deploymentEnvironmentCurrent)
}
func (migration *migrateDeploymentEnvironmentQuickFilter) Down(ctx context.Context, db *bun.DB) error {
return migration.migrate(ctx, db, deploymentEnvironmentCurrent, deploymentEnvironmentOld())
}

View File

@@ -0,0 +1,38 @@
package sqlmigration
import (
"encoding/json"
"testing"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestRewriteQuickFilterSemconv(t *testing.T) {
oldName := deploymentEnvironmentOld()
input := `[{"key":"service.name","dataType":"string","type":"resource"},{"key":"` + oldName + `","dataType":"string","type":"resource","custom":true}]`
rewritten, changed, err := rewriteQuickFilterSemconv(input, oldName, deploymentEnvironmentCurrent)
require.NoError(t, err)
assert.True(t, changed)
var filters []map[string]any
require.NoError(t, json.Unmarshal([]byte(rewritten), &filters))
assert.Equal(t, "service.name", filters[0]["key"])
assert.Equal(t, deploymentEnvironmentCurrent, filters[1]["key"])
assert.Equal(t, true, filters[1]["custom"], "unknown filter properties must be preserved")
restored, changed, err := rewriteQuickFilterSemconv(rewritten, deploymentEnvironmentCurrent, oldName)
require.NoError(t, err)
assert.True(t, changed)
require.NoError(t, json.Unmarshal([]byte(restored), &filters))
assert.Equal(t, oldName, filters[1]["key"])
}
func TestRewriteQuickFilterSemconvNoop(t *testing.T) {
input := `[{"key":"service.name","dataType":"string","type":"resource"}]`
rewritten, changed, err := rewriteQuickFilterSemconv(input, deploymentEnvironmentOld(), deploymentEnvironmentCurrent)
require.NoError(t, err)
assert.False(t, changed)
assert.Equal(t, input, rewritten)
}

View File

@@ -44,6 +44,73 @@ func keyIndexFilter(key *telemetrytypes.TelemetryFieldKey) any {
return fmt.Sprintf(`%%%s%%`, key.Name)
}
func memberKey(key *telemetrytypes.TelemetryFieldKey, name string) *telemetrytypes.TelemetryFieldKey {
member := *key
member.Name = name
return &member
}
func keyIndexCondition(sb *sqlbuilder.SelectBuilder, column string, key *telemetrytypes.TelemetryFieldKey, members []string) string {
conditions := make([]string, 0, len(members))
for _, member := range members {
conditions = append(conditions, sb.Like(column, keyIndexFilter(memberKey(key, member))))
}
if len(conditions) == 1 {
return conditions[0]
}
return sb.Or(conditions...)
}
func valueIndexCondition(
sb *sqlbuilder.SelectBuilder,
column string,
key *telemetrytypes.TelemetryFieldKey,
members []string,
op qbtypes.FilterOperator,
value any,
caseInsensitive bool,
) string {
conditions := make([]string, 0, len(members))
for _, member := range members {
patterns := valueForIndexFilter(op, memberKey(key, member), value)
switch values := patterns.(type) {
case []string:
for _, pattern := range values {
conditions = append(conditions, sb.Like(column, pattern))
}
default:
if caseInsensitive {
conditions = append(conditions, sb.ILike(column, values))
} else {
conditions = append(conditions, sb.Like(column, values))
}
}
}
if len(conditions) == 1 {
return conditions[0]
}
return sb.Or(conditions...)
}
func memberPresenceCondition(sb *sqlbuilder.SelectBuilder, column string, members []string, exists bool) string {
conditions := make([]string, 0, len(members))
for _, member := range members {
field := fmt.Sprintf("simpleJSONHas(%s, '%s')", column, member)
if exists {
conditions = append(conditions, sb.E(field, true))
} else {
conditions = append(conditions, sb.NE(field, true))
}
}
if exists {
if len(conditions) == 1 {
return conditions[0]
}
return sb.Or(conditions...)
}
return sb.And(conditions...)
}
// SkipResourceFilter is not applicable here: the fingerprint table only stores resource attributes.
func (b *defaultConditionBuilder) ConditionFor(
ctx context.Context,
@@ -115,8 +182,10 @@ func (b *defaultConditionBuilder) conditionForKey(
// as we have not changed the resource column in the resource fingerprint table.
column := columns[0]
keyIdxFilter := sb.Like(column.Name, keyIndexFilter(key))
valueForIndexFilter := valueForIndexFilter(op, key, value)
members := resourceSemconvMembers(key)
isFamily := len(members) > 1
keyIdxFilter := keyIndexCondition(sb, column.Name, key, members)
singleValueIndexFilter := valueForIndexFilter(op, memberKey(key, members[0]), value)
fieldName, err := b.fm.FieldFor(ctx, valuer.UUID{}, startNs, endNs, key)
if err != nil {
@@ -128,12 +197,15 @@ func (b *defaultConditionBuilder) conditionForKey(
return sb.And(
sb.E(fieldName, formattedValue),
keyIdxFilter,
sb.Like(column.Name, valueForIndexFilter),
valueIndexCondition(sb, column.Name, key, members, op, value, false),
), nil
case qbtypes.FilterOperatorNotEqual:
if isFamily {
return sb.NE(fieldName, formattedValue), nil
}
return sb.And(
sb.NE(fieldName, formattedValue),
sb.NotLike(column.Name, valueForIndexFilter),
sb.NotLike(column.Name, singleValueIndexFilter),
), nil
case qbtypes.FilterOperatorGreaterThan:
return sb.And(sb.GT(fieldName, formattedValue), keyIdxFilter), nil
@@ -148,7 +220,7 @@ func (b *defaultConditionBuilder) conditionForKey(
return sb.And(
sb.ILike(fieldName, formattedValue),
keyIdxFilter,
sb.ILike(column.Name, valueForIndexFilter),
valueIndexCondition(sb, column.Name, key, members, op, value, true),
), nil
case qbtypes.FilterOperatorNotLike, qbtypes.FilterOperatorNotILike:
// no index filter: as cannot apply `not contains x%y` as y can be somewhere else
@@ -185,13 +257,11 @@ func (b *defaultConditionBuilder) conditionForKey(
inConditions = append(inConditions, sb.E(fieldName, querybuilder.FormatValueForContains(v)))
}
mainCondition := sb.Or(inConditions...)
valConditions := make([]string, 0, len(values))
if valuesForIndexFilter, ok := valueForIndexFilter.([]string); ok {
for _, v := range valuesForIndexFilter {
valConditions = append(valConditions, sb.Like(column.Name, v))
}
}
mainCondition = sb.And(mainCondition, keyIdxFilter, sb.Or(valConditions...))
mainCondition = sb.And(
mainCondition,
keyIdxFilter,
valueIndexCondition(sb, column.Name, key, members, op, value, false),
)
return mainCondition, nil
case qbtypes.FilterOperatorNotIn:
@@ -204,8 +274,11 @@ func (b *defaultConditionBuilder) conditionForKey(
notInConditions = append(notInConditions, sb.NE(fieldName, querybuilder.FormatValueForContains(v)))
}
mainCondition := sb.And(notInConditions...)
if isFamily {
return mainCondition, nil
}
valConditions := make([]string, 0, len(values))
if valuesForIndexFilter, ok := valueForIndexFilter.([]string); ok {
if valuesForIndexFilter, ok := singleValueIndexFilter.([]string); ok {
for _, v := range valuesForIndexFilter {
valConditions = append(valConditions, sb.NotLike(column.Name, v))
}
@@ -215,13 +288,11 @@ func (b *defaultConditionBuilder) conditionForKey(
case qbtypes.FilterOperatorExists:
return sb.And(
sb.E(fmt.Sprintf("simpleJSONHas(%s, '%s')", column.Name, key.Name), true),
memberPresenceCondition(sb, column.Name, members, true),
keyIdxFilter,
), nil
case qbtypes.FilterOperatorNotExists:
return sb.And(
sb.NE(fmt.Sprintf("simpleJSONHas(%s, '%s')", column.Name, key.Name), true),
), nil
return memberPresenceCondition(sb, column.Name, members, false), nil
case qbtypes.FilterOperatorRegexp:
return sb.And(
@@ -237,7 +308,7 @@ func (b *defaultConditionBuilder) conditionForKey(
return sb.And(
sb.ILike(fieldName, fmt.Sprintf(`%%%s%%`, formattedValue)),
keyIdxFilter,
sb.ILike(column.Name, valueForIndexFilter),
valueIndexCondition(sb, column.Name, key, members, op, value, true),
), nil
case qbtypes.FilterOperatorNotContains:
// no index filter: as cannot apply `not contains x%y` as y can be somewhere else

View File

@@ -198,6 +198,75 @@ func TestConditionBuilder(t *testing.T) {
expected: "match(simpleJSONExtractString(labels, 'k8s.namespace.name'), ?) AND labels LIKE ?",
expectedArgs: []any{"ban.*", "%k8s.namespace.name%"},
},
{
name: "semantic convention family equality uses current-first fallback",
key: &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment.name",
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
SemconvMembers: []string{"deployment.environment.name", "deployment.environment"},
},
op: qbtypes.FilterOperatorEqual,
value: "production",
expected: "COALESCE(NULLIF(simpleJSONExtractString(labels, 'deployment.environment.name'), ''), NULLIF(simpleJSONExtractString(labels, 'deployment.environment'), '')) = ? AND (labels LIKE ? OR labels LIKE ?) AND (labels LIKE ? OR labels LIKE ?)",
expectedArgs: []any{
"production",
"%deployment.environment.name%",
"%deployment.environment%",
`%deployment.environment.name":"production%`,
`%deployment.environment":"production%`,
},
},
{
name: "old semantic convention request uses only current metadata member",
key: &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment",
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
SemconvMembers: []string{"deployment.environment.name"},
},
op: qbtypes.FilterOperatorEqual,
value: "production",
expected: "simpleJSONExtractString(labels, 'deployment.environment.name') = ? AND labels LIKE ? AND labels LIKE ?",
expectedArgs: []any{"production", "%deployment.environment.name%", `%deployment.environment.name":"production%`},
},
{
name: "semantic convention family negative filter does not reject fallback rows",
key: &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment",
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
SemconvMembers: []string{"deployment.environment.name", "deployment.environment"},
},
op: qbtypes.FilterOperatorNotEqual,
value: "staging",
expected: "COALESCE(NULLIF(simpleJSONExtractString(labels, 'deployment.environment.name'), ''), NULLIF(simpleJSONExtractString(labels, 'deployment.environment'), '')) <> ?",
expectedArgs: []any{"staging"},
},
{
name: "semantic convention family exists checks every member",
key: &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment.name",
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
SemconvMembers: []string{"deployment.environment.name", "deployment.environment"},
},
op: qbtypes.FilterOperatorExists,
expected: "(simpleJSONHas(labels, 'deployment.environment.name') = ? OR simpleJSONHas(labels, 'deployment.environment') = ?) AND (labels LIKE ? OR labels LIKE ?)",
expectedArgs: []any{true, true, "%deployment.environment.name%", "%deployment.environment%"},
},
{
name: "semantic convention family not exists checks every member",
key: &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment",
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
SemconvMembers: []string{"deployment.environment.name", "deployment.environment"},
},
op: qbtypes.FilterOperatorNotExists,
expected: "(simpleJSONHas(labels, 'deployment.environment.name') <> ? AND simpleJSONHas(labels, 'deployment.environment') <> ?)",
expectedArgs: []any{true, true},
},
}
fm := NewFieldMapper()

View File

@@ -3,8 +3,10 @@ package resourcefilter
import (
"context"
"fmt"
"strings"
schema "github.com/SigNoz/signoz-otel-collector/cmd/signozschemamigrator/schema_migrator"
"github.com/SigNoz/signoz/pkg/semconv"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
@@ -32,6 +34,20 @@ func NewFieldMapper() *defaultFieldMapper {
return &defaultFieldMapper{}
}
func resourceSemconvMembers(key *telemetrytypes.TelemetryFieldKey) []string {
if key.FieldContext != telemetrytypes.FieldContextResource {
return []string{key.Name}
}
if len(key.SemconvMembers) > 0 {
return key.SemconvMembers
}
return semconv.Members(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: key.Name,
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
})
}
func (m *defaultFieldMapper) getColumn(
_ context.Context,
_, _ uint64,
@@ -66,7 +82,15 @@ func (m *defaultFieldMapper) FieldFor(
return "", err
}
if key.FieldContext == telemetrytypes.FieldContextResource {
return fmt.Sprintf("simpleJSONExtractString(%s, '%s')", columns[0].Name, key.Name), nil
members := resourceSemconvMembers(key)
if len(members) > 1 {
values := make([]string, 0, len(members))
for _, member := range members {
values = append(values, fmt.Sprintf("NULLIF(simpleJSONExtractString(%s, '%s'), '')", columns[0].Name, member))
}
return "COALESCE(" + strings.Join(values, ", ") + ")", nil
}
return fmt.Sprintf("simpleJSONExtractString(%s, '%s')", columns[0].Name, members[0]), nil
}
return columns[0].Name, nil
}

View File

@@ -1,35 +0,0 @@
package telemetrymetadata
import "github.com/SigNoz/signoz/pkg/types/telemetrytypes"
type BackwardCompatibleKeyMap map[string]string
var (
TracesBackwardCompatKeys = BackwardCompatibleKeyMap{
"net.peer.name": "server.address",
"server.address": "net.peer.name",
"http.url": "url.full",
"url.full": "http.url",
}
// LogsBackwardCompatKeys contains bidirectional mappings for logs.
// Currently empty, can be extended in the future.
LogsBackwardCompatKeys = BackwardCompatibleKeyMap{}
// MetricsBackwardCompatKeys contains bidirectional mappings for metrics.
// Currently empty, can be extended in the future.
MetricsBackwardCompatKeys = BackwardCompatibleKeyMap{}
)
func GetBackwardCompatKeysForSignal(signal telemetrytypes.Signal) BackwardCompatibleKeyMap {
switch signal {
case telemetrytypes.SignalTraces:
return TracesBackwardCompatKeys
case telemetrytypes.SignalLogs:
return LogsBackwardCompatKeys
case telemetrytypes.SignalMetrics:
return MetricsBackwardCompatKeys
default:
return BackwardCompatibleKeyMap{}
}
}

View File

@@ -4,6 +4,7 @@ import (
"context"
"fmt"
"log/slog"
"slices"
"strings"
"time"
@@ -14,6 +15,7 @@ import (
"github.com/SigNoz/signoz/pkg/factory"
"github.com/SigNoz/signoz/pkg/flagger"
"github.com/SigNoz/signoz/pkg/querybuilder"
"github.com/SigNoz/signoz/pkg/semconv"
"github.com/SigNoz/signoz/pkg/telemetryschema/audittelemetryschema"
"github.com/SigNoz/signoz/pkg/telemetryschema/logstelemetryschema"
"github.com/SigNoz/signoz/pkg/telemetryschema/metertelemetryschema"
@@ -151,6 +153,102 @@ func (t *telemetryMetaStore) tracesTblStatementToFieldKeys(ctx context.Context)
return materialisedKeys, nil
}
func traceSemconvMembers(name string, fieldContext telemetrytypes.FieldContext) []string {
return semconv.Members(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: name,
Signal: telemetrytypes.SignalTraces,
FieldContext: fieldContext,
})
}
func traceSemconvDuplicateFactor() int {
factor := 1
for _, family := range semconv.All() {
if family.Kind != semconv.KindAttribute {
continue
}
if _, ok := semconv.Lookup(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: family.Current,
Signal: telemetrytypes.SignalTraces,
}); ok {
factor = max(factor, len(family.Old)+1)
}
}
return factor
}
// canonicalizeTraceSemconvKeys presents one current-name key for each family.
// If metadata contains both spellings, metadata attached to the current name
// wins; otherwise the old entry is copied under the current response name.
func canonicalizeTraceSemconvKeys(keys []*telemetrytypes.TelemetryFieldKey) []*telemetrytypes.TelemetryFieldKey {
result := make([]*telemetrytypes.TelemetryFieldKey, 0, len(keys))
indexByIdentity := make(map[string]int)
currentSourceByIdentity := make(map[string]bool)
for _, key := range keys {
if key.Signal != telemetrytypes.SignalTraces {
result = append(result, key)
continue
}
family, ok := semconv.Lookup(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: key.Name,
Signal: telemetrytypes.SignalTraces,
FieldContext: key.FieldContext,
})
if !ok {
result = append(result, key)
continue
}
resolved := *key
resolved.Name = family.Current
resolved.SemconvMembers = []string{key.Name}
identity := resolved.Name + ";" + resolved.Signal.StringValue() + ";" + resolved.FieldContext.StringValue() + ";" + resolved.FieldDataType.StringValue()
fromCurrent := key.Name == family.Current
if index, found := indexByIdentity[identity]; found {
physicalMembers := result[index].SemconvMembers
if fromCurrent && !currentSourceByIdentity[identity] {
result[index] = &resolved
currentSourceByIdentity[identity] = true
}
for _, member := range physicalMembers {
if !slices.Contains(result[index].SemconvMembers, member) {
result[index].SemconvMembers = append(result[index].SemconvMembers, member)
}
}
if !slices.Contains(result[index].SemconvMembers, key.Name) {
result[index].SemconvMembers = append(result[index].SemconvMembers, key.Name)
}
continue
}
indexByIdentity[identity] = len(result)
currentSourceByIdentity[identity] = fromCurrent
result = append(result, &resolved)
}
for _, key := range result {
if len(key.SemconvMembers) < 2 {
continue
}
present := make(map[string]bool, len(key.SemconvMembers))
for _, member := range key.SemconvMembers {
present[member] = true
}
ordered := make([]string, 0, len(key.SemconvMembers))
for _, member := range traceSemconvMembers(key.Name, key.FieldContext) {
if present[member] {
ordered = append(ordered, member)
}
}
key.SemconvMembers = ordered
}
return result
}
// getTracesKeys returns the keys from the spans that match the field selection criteria.
func (t *telemetryMetaStore) getTracesKeys(ctx context.Context, fieldKeySelectors []*telemetrytypes.FieldKeySelector) ([]*telemetrytypes.TelemetryFieldKey, bool, error) {
ctx = ctxtypes.NewContextWithCommentVals(ctx, map[string]string{
@@ -202,10 +300,23 @@ func (t *telemetryMetaStore) getTracesKeys(ctx context.Context, fieldKeySelector
// key part of the selector
fieldKeyConds := []string{}
members := traceSemconvMembers(fieldKeySelector.Name, fieldKeySelector.FieldContext)
if fieldKeySelector.SelectorMatchType == telemetrytypes.FieldSelectorMatchTypeExact {
fieldKeyConds = append(fieldKeyConds, sb.E("tagKey", fieldKeySelector.Name))
if len(members) == 1 {
fieldKeyConds = append(fieldKeyConds, sb.E("tagKey", members[0]))
} else {
memberValues := make([]any, 0, len(members))
for _, member := range members {
memberValues = append(memberValues, member)
}
fieldKeyConds = append(fieldKeyConds, sb.In("tagKey", memberValues...))
}
} else {
fieldKeyConds = append(fieldKeyConds, sb.ILike("tagKey", "%"+escapeForLike(fieldKeySelector.Name)+"%"))
memberConditions := make([]string, 0, len(members))
for _, member := range members {
memberConditions = append(memberConditions, sb.ILike("tagKey", "%"+escapeForLike(member)+"%"))
}
fieldKeyConds = append(fieldKeyConds, sb.Or(memberConditions...))
}
searchTexts = append(searchTexts, fieldKeySelector.Name)
@@ -238,8 +349,10 @@ func (t *telemetryMetaStore) getTracesKeys(ctx context.Context, fieldKeySelector
mainSb.From(mainSb.BuilderAs(sb, "sub_query"))
mainSb.GroupBy("tag_key", "tag_type", "tag_data_type")
mainSb.OrderBy("priority")
// query one extra to check if we hit the limit
mainSb.Limit(limit + 1)
// Family members collapse after the database query. In the worst case each
// logical key occupies one row per family member, so fetch enough physical
// rows to return the requested logical page, plus one to detect truncation.
mainSb.Limit(limit*traceSemconvDuplicateFactor() + 1)
query, args := mainSb.BuildWithFlavor(sqlbuilder.ClickHouse)
@@ -249,14 +362,7 @@ func (t *telemetryMetaStore) getTracesKeys(ctx context.Context, fieldKeySelector
}
defer rows.Close()
keys := []*telemetrytypes.TelemetryFieldKey{}
rowCount := 0
for rows.Next() {
rowCount++
// reached the limit, we know there are more results
if rowCount > limit {
break
}
var name string
var fieldContext telemetrytypes.FieldContext
var fieldDataType telemetrytypes.FieldDataType
@@ -285,8 +391,11 @@ func (t *telemetryMetaStore) getTracesKeys(ctx context.Context, fieldKeySelector
return nil, false, errors.Wrap(rows.Err(), errors.TypeInternal, errors.CodeInternal, ErrFailedToGetTracesKeys.Error())
}
// hit the limit? (only counting DB results)
complete := rowCount <= limit
keys = canonicalizeTraceSemconvKeys(keys)
complete := len(keys) <= limit
if !complete {
keys = keys[:limit]
}
staticKeys := []string{"isRoot", "isEntryPoint"}
staticKeys = append(staticKeys, maps.Keys(tracestelemetryschema.IntrinsicFields)...)
@@ -1108,40 +1217,6 @@ func (t *telemetryMetaStore) getMeterSourceMetricKeys(ctx context.Context, field
}
// applyBackwardCompatibleKeys adds backward compatible key aliases to the map.
func applyBackwardCompatibleKeys(mapOfKeys map[string][]*telemetrytypes.TelemetryFieldKey) {
// Get backward compatible keys for all signals
backwardCompatKeysBySignal := map[telemetrytypes.Signal]BackwardCompatibleKeyMap{
telemetrytypes.SignalTraces: GetBackwardCompatKeysForSignal(telemetrytypes.SignalTraces),
telemetrytypes.SignalLogs: GetBackwardCompatKeysForSignal(telemetrytypes.SignalLogs),
telemetrytypes.SignalMetrics: GetBackwardCompatKeysForSignal(telemetrytypes.SignalMetrics),
}
// Iterate over existing keys and add aliases if they exist in backward compat mapping
for srcKey, srcKeys := range mapOfKeys {
for _, srcKeyEntry := range srcKeys {
backwardCompatKeys := backwardCompatKeysBySignal[srcKeyEntry.Signal]
if backwardCompatKeys == nil {
continue
}
if aliasKey, ok := backwardCompatKeys[srcKey]; ok {
if _, aliasExists := mapOfKeys[aliasKey]; !aliasExists {
aliasKeyEntry := &telemetrytypes.TelemetryFieldKey{
Name: aliasKey,
Signal: srcKeyEntry.Signal,
FieldContext: srcKeyEntry.FieldContext,
FieldDataType: srcKeyEntry.FieldDataType,
}
mapOfKeys[aliasKey] = []*telemetrytypes.TelemetryFieldKey{aliasKeyEntry}
}
// Found the alias for this signal, no need to check other entries
break
}
}
}
}
func enrichWithIntrinsicMetricKeys(keys map[string][]*telemetrytypes.TelemetryFieldKey, selectors []*telemetrytypes.FieldKeySelector) map[string][]*telemetrytypes.TelemetryFieldKey {
if len(selectors) == 0 {
return keys
@@ -1272,7 +1347,6 @@ func (t *telemetryMetaStore) GetKeys(ctx context.Context, orgID valuer.UUID, fie
mapOfKeys[key.Name] = append(mapOfKeys[key.Name], key)
}
applyBackwardCompatibleKeys(mapOfKeys)
mapOfKeys = enrichWithIntrinsicMetricKeys(mapOfKeys, selectors)
if t.fl.BooleanOrEmpty(ctx, flagger.FeatureEnableAIObservability, featuretypes.NewFlaggerEvaluationContext(orgID)) {
mapOfKeys = enrichWithGenAIKeys(mapOfKeys, selectors)
@@ -1353,7 +1427,6 @@ func (t *telemetryMetaStore) GetKeysMulti(ctx context.Context, orgID valuer.UUID
mapOfKeys[key.Name] = append(mapOfKeys[key.Name], key)
}
applyBackwardCompatibleKeys(mapOfKeys)
mapOfKeys = enrichWithIntrinsicMetricKeys(mapOfKeys, fieldKeySelectors)
if t.fl.BooleanOrEmpty(ctx, flagger.FeatureEnableAIObservability, featuretypes.NewFlaggerEvaluationContext(orgID)) {
mapOfKeys = enrichWithGenAIKeys(mapOfKeys, fieldKeySelectors)
@@ -1367,7 +1440,18 @@ func (t *telemetryMetaStore) GetKey(ctx context.Context, orgID valuer.UUID, fiel
if err != nil {
return nil, err
}
return keys[fieldKeySelector.Name], nil
members := semconv.Members(semconv.KindAttribute, *fieldKeySelector)
resolved := make([]*telemetrytypes.TelemetryFieldKey, 0)
seen := make(map[*telemetrytypes.TelemetryFieldKey]bool)
for _, member := range members {
for _, key := range keys[member] {
if !seen[key] {
resolved = append(resolved, key)
seen[key] = true
}
}
}
return resolved, nil
}
func (t *telemetryMetaStore) getRelatedValues(ctx context.Context, orgID valuer.UUID, fieldValueSelector *telemetrytypes.FieldValueSelector) ([]string, bool, error) {
@@ -1542,7 +1626,16 @@ func (t *telemetryMetaStore) getSpanFieldValues(ctx context.Context, fieldValueS
sb := sqlbuilder.Select("DISTINCT string_value, number_value").From(t.tracesDBName + "." + t.tracesFieldsTblName)
if fieldValueSelector.Name != "" {
sb.Where(sb.E("tag_key", fieldValueSelector.Name))
members := traceSemconvMembers(fieldValueSelector.Name, fieldValueSelector.FieldContext)
if len(members) == 1 {
sb.Where(sb.E("tag_key", members[0]))
} else {
memberValues := make([]any, 0, len(members))
for _, member := range members {
memberValues = append(memberValues, member)
}
sb.Where(sb.In("tag_key", memberValues...))
}
}
// now look at the field context

View File

@@ -12,6 +12,7 @@ import (
"github.com/SigNoz/signoz/pkg/telemetrystore"
"github.com/SigNoz/signoz/pkg/telemetrystore/telemetrystoretest"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
@@ -83,3 +84,83 @@ func TestGetFirstSeenFromMetricMetadata(t *testing.T) {
t.Errorf("there were unfulfilled expectations: %s", err)
}
}
func TestCanonicalizeTraceSemconvKeys(t *testing.T) {
oldResource := &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment",
Description: "old resource metadata",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
currentResource := &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment.name",
Description: "current resource metadata",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
oldAttribute := &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
oldLogResource := &telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment",
Signal: telemetrytypes.SignalLogs,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
result := canonicalizeTraceSemconvKeys([]*telemetrytypes.TelemetryFieldKey{
oldResource,
currentResource,
oldAttribute,
oldLogResource,
})
require.Len(t, result, 3)
assert.Equal(t, "deployment.environment.name", result[0].Name)
assert.Equal(t, "current resource metadata", result[0].Description)
assert.Equal(t, []string{"deployment.environment.name", "deployment.environment"}, result[0].SemconvMembers)
assert.Equal(t, "deployment.environment.name", result[1].Name)
assert.Equal(t, telemetrytypes.FieldContextAttribute, result[1].FieldContext)
assert.Equal(t, []string{"deployment.environment"}, result[1].SemconvMembers)
assert.Equal(t, "deployment.environment", result[2].Name, "phase 1 must not rewrite raw log metadata")
}
func TestGetSpanFieldValuesMergesSemconvFamily(t *testing.T) {
mockTelemetryStore := telemetrystoretest.New(telemetrystore.Config{}, &regexMatcher{})
mock := mockTelemetryStore.Mock()
metadata := NewTelemetryMetaStore(
instrumentationtest.New().ToProviderSettings(),
mockTelemetryStore,
flaggertest.New(t),
)
mock.ExpectQuery(`SELECT DISTINCT string_value, number_value FROM signoz_traces\.distributed_tag_attributes_v2 WHERE tag_key IN \(\?, \?\) AND tag_type = \? AND tag_data_type = \? LIMIT \?`).
WithArgs("deployment.environment.name", "deployment.environment", "resource", "string", 51).
WillReturnRows(cmock.NewRows([]cmock.ColumnType{
{Name: "string_value", Type: "String"},
{Name: "number_value", Type: "Float64"},
}, [][]any{
{"production", float64(0)},
{"staging", float64(0)},
{"production", float64(0)},
}))
values, complete, err := metadata.GetAllValues(context.Background(), valuer.UUID{}, &telemetrytypes.FieldValueSelector{
FieldKeySelector: &telemetrytypes.FieldKeySelector{
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
Name: "deployment.environment",
},
})
require.NoError(t, err)
assert.True(t, complete)
assert.Equal(t, []string{"production", "staging"}, values.StringValues)
assert.NoError(t, mock.ExpectationsWereMet(), "all expected metadata queries should be executed")
}

View File

@@ -154,6 +154,15 @@ func (c *conditionBuilder) conditionFor(
// in the query builder, `exists` and `not exists` are used for
// key membership checks, so depending on the column type, the condition changes
case qbtypes.FilterOperatorExists, qbtypes.FilterOperatorNotExists:
// A semantic-convention family is represented by one current-first value
// expression, but presence still has to inspect every physical member. In
// particular, using ExistsExpression below with the requested key would add
// a mapContains check for only that spelling and reject fallback-only rows.
if isTraceSemconvFamily(key) {
if fm, ok := c.fm.(*fieldMapper); ok {
return fm.existsExpressionFor(ctx, orgID, startNs, endNs, key, operator == qbtypes.FilterOperatorExists)
}
}
columns, err := c.fm.ColumnFor(ctx, orgID, startNs, endNs, key)
if err != nil {
return "", err

View File

@@ -210,6 +210,32 @@ func TestConditionFor(t *testing.T) {
expectedSQL: "NOT mapContains(attributes_string, 'user.id')",
expectedError: nil,
},
{
name: "Equal operator - semantic convention family",
key: telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment.name",
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
SemconvMembers: []string{"deployment.environment.name", "deployment.environment"},
},
operator: qbtypes.FilterOperatorEqual,
value: "production",
expectedSQL: "(COALESCE(NULLIF(attributes_string['deployment.environment.name'], ''), NULLIF(attributes_string['deployment.environment'], '')) = ? AND ((mapContains(attributes_string, 'deployment.environment.name') OR mapContains(attributes_string, 'deployment.environment'))))",
expectedArgs: []any{"production"},
expectedError: nil,
},
{
name: "Not Exists operator - semantic convention family",
key: telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment",
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
SemconvMembers: []string{"deployment.environment.name", "deployment.environment"},
},
operator: qbtypes.FilterOperatorNotExists,
expectedSQL: "NOT (((mapContains(attributes_string, 'deployment.environment.name') OR mapContains(attributes_string, 'deployment.environment'))))",
expectedError: nil,
},
{
name: "Exists operator - json field",
key: telemetrytypes.TelemetryFieldKey{

View File

@@ -8,6 +8,7 @@ import (
schema "github.com/SigNoz/signoz-otel-collector/cmd/signozschemamigrator/schema_migrator"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/querybuilder"
"github.com/SigNoz/signoz/pkg/semconv"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
@@ -167,6 +168,32 @@ func NewFieldMapper() *fieldMapper {
return &fieldMapper{}
}
func traceSemconvMembers(key *telemetrytypes.TelemetryFieldKey) []string {
if key.FieldContext != telemetrytypes.FieldContextResource && key.FieldContext != telemetrytypes.FieldContextAttribute {
return []string{key.Name}
}
if len(key.SemconvMembers) > 0 {
return key.SemconvMembers
}
return semconv.Members(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: key.Name,
Signal: telemetrytypes.SignalTraces,
FieldContext: key.FieldContext,
})
}
func isTraceSemconvFamily(key *telemetrytypes.TelemetryFieldKey) bool {
if key.FieldContext != telemetrytypes.FieldContextResource && key.FieldContext != telemetrytypes.FieldContextAttribute {
return false
}
_, ok := semconv.Lookup(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: key.Name,
Signal: telemetrytypes.SignalTraces,
FieldContext: key.FieldContext,
})
return ok
}
func (m *fieldMapper) getColumn(
_ context.Context,
_, _ uint64,
@@ -291,10 +318,25 @@ func (m *fieldMapper) resolveColumnExprs(
if key.FieldContext != telemetrytypes.FieldContextResource {
return nil, nil, nil, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "only resource context fields are supported for json columns, got %s", key.FieldContext.String)
}
// have to add ::string as clickHouse throws an error :- data types Variant/Dynamic are not allowed in GROUP BY
// once clickHouse dependency is updated, we need to check if we can remove it.
exprs = append(exprs, fmt.Sprintf("%s.`%s`::String", columnName, key.Name))
existExprs = append(existExprs, fmt.Sprintf("%s.`%s` IS NOT NULL", columnName, key.Name))
members := traceSemconvMembers(key)
if len(members) > 1 {
values := make([]string, 0, len(members))
guards := make([]string, 0, len(members))
for _, member := range members {
// The String cast is required because ClickHouse does not allow
// Variant/Dynamic values in GROUP BY.
value := fmt.Sprintf("%s.`%s`::String", columnName, member)
values = append(values, fmt.Sprintf("NULLIF(%s, '')", value))
guards = append(guards, fmt.Sprintf("%s.`%s` IS NOT NULL", columnName, member))
}
exprs = append(exprs, "COALESCE("+strings.Join(values, ", ")+")")
existExprs = append(existExprs, "("+strings.Join(guards, " OR ")+")")
} else {
// have to add ::string as clickHouse throws an error :- data types Variant/Dynamic are not allowed in GROUP BY
// once ClickHouse is updated, check whether this cast can be removed.
exprs = append(exprs, fmt.Sprintf("%s.`%s`::String", columnName, members[0]))
existExprs = append(existExprs, fmt.Sprintf("%s.`%s` IS NOT NULL", columnName, members[0]))
}
case schema.ColumnTypeEnumString,
schema.ColumnTypeEnumUInt64,
schema.ColumnTypeEnumUInt32,
@@ -319,13 +361,35 @@ func (m *fieldMapper) resolveColumnExprs(
switch valueType := column.Type.(schema.MapColumnType).ValueType; valueType.GetType() {
case schema.ColumnTypeEnumString, schema.ColumnTypeEnumFloat64, schema.ColumnTypeEnumBool:
// a key could have been materialized, if so return the materialized column name
if key.Materialized {
exprs = append(exprs, telemetrytypes.FieldKeyToMaterializedColumnName(key))
existExprs = append(existExprs, telemetrytypes.FieldKeyToMaterializedColumnNameForExists(key))
members := traceSemconvMembers(key)
if len(members) > 1 {
guards := make([]string, 0, len(members))
for _, member := range members {
guards = append(guards, fmt.Sprintf("mapContains(%s, '%s')", columnName, member))
}
if valueType.GetType() == schema.ColumnTypeEnumString {
values := make([]string, 0, len(members))
for _, member := range members {
values = append(values, fmt.Sprintf("NULLIF(%s['%s'], '')", columnName, member))
}
exprs = append(exprs, "COALESCE("+strings.Join(values, ", ")+")")
} else {
branches := make([]string, 0, len(members)*2)
for i, member := range members {
branches = append(branches, guards[i], fmt.Sprintf("%s['%s']", columnName, member))
}
exprs = append(exprs, "multiIf("+strings.Join(branches, ", ")+", NULL)")
}
existExprs = append(existExprs, "("+strings.Join(guards, " OR ")+")")
} else if key.Materialized {
// a key could have been materialized, if so return the materialized column name
physicalKey := *key
physicalKey.Name = members[0]
exprs = append(exprs, telemetrytypes.FieldKeyToMaterializedColumnName(&physicalKey))
existExprs = append(existExprs, telemetrytypes.FieldKeyToMaterializedColumnNameForExists(&physicalKey))
} else {
exprs = append(exprs, fmt.Sprintf("%s['%s']", columnName, key.Name))
existExprs = append(existExprs, fmt.Sprintf("mapContains(%s, '%s')", columnName, key.Name))
exprs = append(exprs, fmt.Sprintf("%s['%s']", columnName, members[0]))
existExprs = append(existExprs, fmt.Sprintf("mapContains(%s, '%s')", columnName, members[0]))
}
default:
return nil, nil, nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "value type %s is not supported for map column type %s", valueType, column.Type)
@@ -529,6 +593,25 @@ func (m *fieldMapper) existsExpressionFor(
key *telemetrytypes.TelemetryFieldKey,
exists bool,
) (string, error) {
if isTraceSemconvFamily(key) {
_, existExprs, _, err := m.resolveColumnExprs(ctx, tsStart, tsEnd, key)
if err != nil {
return "", err
}
if len(existExprs) == 0 {
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "no existence expression found for field %s", key.Name)
}
parts := make([]string, 0, len(existExprs))
for _, expression := range existExprs {
parts = append(parts, "("+expression+")")
}
combined := strings.Join(parts, " OR ")
if exists {
return combined, nil
}
return "NOT (" + combined + ")", nil
}
columns, err := m.getColumn(ctx, tsStart, tsEnd, key)
if err != nil {
return "", err

View File

@@ -80,7 +80,7 @@ func TestGetFieldKeyName(t *testing.T) {
Materialized: true,
Evolutions: mockEvolution,
},
expectedResult: "multiIf(resource.`deployment.environment` IS NOT NULL, resource.`deployment.environment`::String, `resource_string_deployment$$environment_exists`, `resource_string_deployment$$environment`, NULL)",
expectedResult: "multiIf((resource.`deployment.environment.name` IS NOT NULL OR resource.`deployment.environment` IS NOT NULL), COALESCE(NULLIF(resource.`deployment.environment.name`::String, ''), NULLIF(resource.`deployment.environment`::String, '')), (mapContains(resources_string, 'deployment.environment.name') OR mapContains(resources_string, 'deployment.environment')), COALESCE(NULLIF(resources_string['deployment.environment.name'], ''), NULLIF(resources_string['deployment.environment'], '')), NULL)",
expectedError: nil,
},
{
@@ -120,6 +120,51 @@ func TestGetFieldKeyName(t *testing.T) {
}
}
func TestFieldForSemconvFamily(t *testing.T) {
ctx := context.Background()
fm := NewFieldMapper()
start := uint64(time.Date(2024, 6, 1, 0, 0, 0, 0, time.UTC).UnixNano())
end := uint64(time.Date(2024, 6, 5, 0, 0, 0, 0, time.UTC).UnixNano())
for _, requestedName := range []string{"deployment.environment.name", "deployment.environment"} {
attributeKey := telemetrytypes.TelemetryFieldKey{
Name: requestedName,
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
attributeExpression, err := fm.FieldFor(ctx, valuer.UUID{}, start, end, &attributeKey)
require.NoError(t, err)
assert.Equal(t,
"COALESCE(NULLIF(attributes_string['deployment.environment.name'], ''), NULLIF(attributes_string['deployment.environment'], ''))",
attributeExpression,
)
resourceKey := telemetrytypes.TelemetryFieldKey{
Name: requestedName,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
Materialized: true,
Evolutions: MockEvolutionData(time.Date(2024, 6, 2, 0, 0, 0, 0, time.UTC)),
}
resourceExpression, err := fm.FieldFor(ctx, valuer.UUID{}, start, end, &resourceKey)
require.NoError(t, err)
assert.Equal(t,
"multiIf((resource.`deployment.environment.name` IS NOT NULL OR resource.`deployment.environment` IS NOT NULL), COALESCE(NULLIF(resource.`deployment.environment.name`::String, ''), NULLIF(resource.`deployment.environment`::String, '')), (mapContains(resources_string, 'deployment.environment.name') OR mapContains(resources_string, 'deployment.environment')), COALESCE(NULLIF(resources_string['deployment.environment.name'], ''), NULLIF(resources_string['deployment.environment'], '')), NULL)",
resourceExpression,
)
}
oldRequestWithCurrentOnlyMetadata := telemetrytypes.TelemetryFieldKey{
Name: "deployment.environment",
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
SemconvMembers: []string{"deployment.environment.name"},
}
expression, err := fm.FieldFor(ctx, valuer.UUID{}, start, end, &oldRequestWithCurrentOnlyMetadata)
require.NoError(t, err)
assert.Equal(t, "attributes_string['deployment.environment.name']", expression)
}
func TestFieldForResourceWithEvolution(t *testing.T) {
ctx := context.Background()
releaseTime := time.Date(2025, 1, 1, 0, 0, 0, 0, time.UTC)
@@ -176,7 +221,7 @@ func TestFieldForResourceWithEvolution(t *testing.T) {
},
tsStart: uint64(time.Date(2025, 6, 1, 0, 0, 0, 0, time.UTC).UnixNano()),
tsEnd: uint64(time.Date(2025, 7, 1, 0, 0, 0, 0, time.UTC).UnixNano()),
expectedResult: "resource.`deployment.environment`::String",
expectedResult: "COALESCE(NULLIF(resource.`deployment.environment.name`::String, ''), NULLIF(resource.`deployment.environment`::String, ''))",
},
{
name: "Window straddles release - materialized resource",
@@ -189,7 +234,7 @@ func TestFieldForResourceWithEvolution(t *testing.T) {
},
tsStart: uint64(time.Date(2024, 6, 1, 0, 0, 0, 0, time.UTC).UnixNano()),
tsEnd: uint64(time.Date(2025, 6, 1, 0, 0, 0, 0, time.UTC).UnixNano()),
expectedResult: "multiIf(resource.`deployment.environment` IS NOT NULL, resource.`deployment.environment`::String, `resource_string_deployment$$environment_exists`, `resource_string_deployment$$environment`, NULL)",
expectedResult: "multiIf((resource.`deployment.environment.name` IS NOT NULL OR resource.`deployment.environment` IS NOT NULL), COALESCE(NULLIF(resource.`deployment.environment.name`::String, ''), NULLIF(resource.`deployment.environment`::String, '')), (mapContains(resources_string, 'deployment.environment.name') OR mapContains(resources_string, 'deployment.environment')), COALESCE(NULLIF(resources_string['deployment.environment.name'], ''), NULLIF(resources_string['deployment.environment'], '')), NULL)",
},
}

View File

@@ -141,7 +141,7 @@ func NewSignalFilterFromStorableQuickFilter(storableQuickFilter *StorableQuickFi
func NewDefaultQuickFilter(orgID valuer.UUID) ([]*StorableQuickFilter, error) {
tracesFilters := []map[string]interface{}{
{"key": "duration_nano", "dataType": "float64", "type": "tag"},
{"key": "deployment.environment", "dataType": "string", "type": "resource"},
{"key": "deployment.environment.name", "dataType": "string", "type": "resource"},
{"key": "hasError", "dataType": "bool", "type": "tag"},
{"key": "service.name", "dataType": "string", "type": "resource"},
{"key": "name", "dataType": "string", "type": "tag"},
@@ -166,13 +166,13 @@ func NewDefaultQuickFilter(orgID valuer.UUID) ([]*StorableQuickFilter, error) {
}
apiMonitoringFilters := []map[string]interface{}{
{"key": "deployment.environment", "dataType": "string", "type": "resource"},
{"key": "deployment.environment.name", "dataType": "string", "type": "resource"},
{"key": "service.name", "dataType": "string", "type": "resource"},
{"key": "rpc.method", "dataType": "string", "type": "tag"},
}
exceptionsFilters := []map[string]interface{}{
{"key": "deployment.environment", "dataType": "string", "type": "resource"},
{"key": "deployment.environment.name", "dataType": "string", "type": "resource"},
{"key": "service.name", "dataType": "string", "type": "resource"},
{"key": "host.name", "dataType": "string", "type": "resource"},
{"key": "k8s.cluster.name", "dataType": "string", "type": "resource"},

View File

@@ -0,0 +1,37 @@
package quickfiltertypes
import (
"encoding/json"
"testing"
v3 "github.com/SigNoz/signoz/pkg/query-service/model/v3"
"github.com/SigNoz/signoz/pkg/valuer"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestDefaultTraceQuickFiltersUseCurrentEnvironmentName(t *testing.T) {
filters, err := NewDefaultQuickFilter(valuer.GenerateUUID())
require.NoError(t, err)
traceSignals := map[string]bool{
SignalTraces.StringValue(): true,
SignalApiMonitoring.StringValue(): true,
SignalExceptions.StringValue(): true,
}
for _, filter := range filters {
if !traceSignals[filter.Signal.StringValue()] {
continue
}
var keys []v3.AttributeKey
require.NoError(t, json.Unmarshal([]byte(filter.Filter), &keys))
found := false
for _, key := range keys {
if key.Key == "deployment.environment.name" {
found = true
}
assert.NotEqual(t, "deployment.environment", key.Key)
}
assert.True(t, found, "missing environment quick filter for %s", filter.Signal.StringValue())
}
}

View File

@@ -47,7 +47,8 @@ type TelemetryFieldKey struct {
Indexes []TelemetryFieldKeySkipIndex `json:"-"`
Materialized bool `json:"-"` // refers to promoted in case of body.... fields
Evolutions []*EvolutionEntry `json:"-"`
Evolutions []*EvolutionEntry `json:"-"`
SemconvMembers []string `json:"-"`
}
func (f *TelemetryFieldKey) KeyNameContainsArray() bool {
@@ -128,6 +129,7 @@ func (f *TelemetryFieldKey) OverrideMetadataFrom(src *TelemetryFieldKey) {
f.Materialized = src.Materialized
f.JSONPlan = src.JSONPlan
f.Evolutions = src.Evolutions
f.SemconvMembers = src.SemconvMembers
}
func (f *TelemetryFieldKey) Equal(key *TelemetryFieldKey) bool {

721
scripts/semconv/generate.go Normal file
View File

@@ -0,0 +1,721 @@
package main
import (
"bytes"
"errors"
"flag"
"fmt"
"go/format"
"os"
"path/filepath"
"sort"
"strconv"
"strings"
"gopkg.in/yaml.v3"
)
const (
kindAttribute = "attribute"
kindMetric = "metric"
)
type stringListFlag []string
func (f *stringListFlag) String() string { return strings.Join(*f, ",") }
func (f *stringListFlag) Set(value string) error {
*f = append(*f, value)
return nil
}
type schemaFile struct {
FileFormat string `yaml:"file_format"`
SchemaURL string `yaml:"schema_url"`
Versions map[string]schemaVersion `yaml:"versions"`
}
type schemaVersion struct {
All changeSection `yaml:"all"`
Resources changeSection `yaml:"resources"`
Spans changeSection `yaml:"spans"`
Logs changeSection `yaml:"logs"`
Metrics changeSection `yaml:"metrics"`
}
type changeSection struct {
Changes []schemaChange `yaml:"changes"`
}
type schemaChange struct {
RenameAttributes *attributeRename `yaml:"rename_attributes"`
RenameMetrics map[string]string `yaml:"rename_metrics"`
}
type attributeRename struct {
AttributeMap map[string]string `yaml:"attribute_map"`
ApplyToMetrics []string `yaml:"apply_to_metrics"`
}
type overlayFile struct {
DefaultEnabled bool `yaml:"default_enabled"`
// Families is keyed only by current name. One name cannot carry separate
// policies for attribute and metric families; set kind explicitly whenever
// a metric-name family is configured.
Families map[string]overlayFamily `yaml:"families"`
}
type overlayFamily struct {
Enabled *bool `yaml:"enabled"`
Kind string `yaml:"kind"`
Old []string `yaml:"old"`
AddOld []string `yaml:"add_old"`
ExcludeOld []string `yaml:"exclude_old"`
Contexts []string `yaml:"contexts"`
Signals []string `yaml:"signals"`
AddContexts []string `yaml:"add_contexts"`
AddSignals []string `yaml:"add_signals"`
ApplyToMetrics []string `yaml:"apply_to_metrics"`
AddApplyToMetrics []string `yaml:"add_apply_to_metrics"`
ValueMap map[string]string `yaml:"value_map"`
}
type edge struct {
old string
current string
kind string
contexts []string
signals []string
allContexts bool
allSignals bool
applyToMetrics []string
}
type graphKey struct{ kind, name string }
type generatedFamily struct {
Current string
Old []string
Kind string
Contexts []string
Signals []string
ApplyToMetrics []string
ValueMap map[string]string
}
func main() {
root, err := findRepoRoot()
if err != nil {
fatal(err)
}
var schemaPaths stringListFlag
flag.Var(&schemaPaths, "schema", "schema source (repeatable)")
overlayPath := flag.String("overlay", filepath.Join(root, "scripts/semconv/overlay.yaml"), "SigNoz overlay")
goOutput := flag.String("go-out", filepath.Join(root, "pkg/semconv/families_gen.go"), "generated Go output")
tsOutput := flag.String("ts-out", filepath.Join(root, "frontend/src/constants/generated/semconvFamilies.gen.ts"), "generated TypeScript output")
check := flag.Bool("check", false, "fail if generated files are stale")
flag.Parse()
if len(schemaPaths) == 0 {
schemaPaths = append(schemaPaths, filepath.Join(root, "scripts/semconv/schema-1.42.0.yaml"))
}
families, err := generate(schemaPaths, *overlayPath)
if err != nil {
fatal(err)
}
goBytes, err := renderGo(families)
if err != nil {
fatal(err)
}
tsBytes := renderTypeScript(families)
if *check {
if err := checkFile(*goOutput, goBytes); err != nil {
fatal(err)
}
if err := checkFile(*tsOutput, tsBytes); err != nil {
fatal(err)
}
return
}
if err := os.WriteFile(*goOutput, goBytes, 0o644); err != nil {
fatal(err)
}
if err := os.WriteFile(*tsOutput, tsBytes, 0o644); err != nil {
fatal(err)
}
}
func fatal(err error) {
fmt.Fprintln(os.Stderr, err)
os.Exit(1)
}
func findRepoRoot() (string, error) {
dir, err := os.Getwd()
if err != nil {
return "", err
}
for {
if _, err := os.Stat(filepath.Join(dir, "go.mod")); err == nil {
return dir, nil
}
parent := filepath.Dir(dir)
if parent == dir {
return "", errors.New("could not find repository root")
}
dir = parent
}
}
func generate(schemaPaths []string, overlayPath string) ([]generatedFamily, error) {
var schemas []schemaFile
for _, path := range schemaPaths {
data, err := os.ReadFile(path)
if err != nil {
return nil, fmt.Errorf("read schema %s: %w", path, err)
}
var schema schemaFile
if err := decodeKnownFields(data, &schema); err != nil {
return nil, fmt.Errorf("parse schema %s: %w", path, err)
}
schemas = append(schemas, schema)
}
overlayData, err := os.ReadFile(overlayPath)
if err != nil {
return nil, fmt.Errorf("read overlay: %w", err)
}
var overlay overlayFile
if err := decodeKnownFields(overlayData, &overlay); err != nil {
return nil, fmt.Errorf("parse overlay: %w", err)
}
return buildFamilies(schemas, overlay)
}
func decodeKnownFields(data []byte, target any) error {
decoder := yaml.NewDecoder(bytes.NewReader(data))
decoder.KnownFields(true)
return decoder.Decode(target)
}
func collectEdges(schemas []schemaFile) ([]edge, error) {
var edges []edge
for _, schema := range schemas {
versions := make([]string, 0, len(schema.Versions))
versionParts := make(map[string][3]int, len(schema.Versions))
for version := range schema.Versions {
parts, err := parseSchemaVersion(version)
if err != nil {
return nil, err
}
versions = append(versions, version)
versionParts[version] = parts
}
sort.Slice(versions, func(i, j int) bool {
return compareVersionParts(versionParts[versions[i]], versionParts[versions[j]]) < 0
})
for _, versionName := range versions {
version := schema.Versions[versionName]
var versionEdges []edge
sections := []struct {
name string
section changeSection
}{
{name: "all", section: version.All},
{name: "resources", section: version.Resources},
{name: "spans", section: version.Spans},
{name: "logs", section: version.Logs},
{name: "metrics", section: version.Metrics},
}
for _, scoped := range sections {
contexts, signals, allContexts, allSignals, err := scopeForSection(scoped.name)
if err != nil {
return nil, err
}
for _, change := range scoped.section.Changes {
if change.RenameAttributes != nil {
for _, old := range sortedMapKeys(change.RenameAttributes.AttributeMap) {
versionEdges = append(versionEdges, edge{
old: old, current: change.RenameAttributes.AttributeMap[old], kind: kindAttribute,
contexts: contexts, signals: signals,
allContexts: allContexts, allSignals: allSignals,
applyToMetrics: change.RenameAttributes.ApplyToMetrics,
})
}
}
for _, old := range sortedMapKeys(change.RenameMetrics) {
versionEdges = append(versionEdges, edge{
old: old, current: change.RenameMetrics[old], kind: kindMetric,
contexts: []string{"metric"}, signals: []string{"metrics"},
})
}
}
}
if err := rejectSameVersionChains(versionName, versionEdges); err != nil {
return nil, err
}
edges = append(edges, versionEdges...)
}
}
return edges, nil
}
func rejectSameVersionChains(version string, edges []edge) error {
oldNames := make(map[graphKey]struct{}, len(edges))
for _, item := range edges {
oldNames[graphKey{kind: item.kind, name: item.old}] = struct{}{}
}
for _, item := range edges {
if _, ok := oldNames[graphKey{kind: item.kind, name: item.current}]; ok {
return fmt.Errorf(
"schema version %q contains a same-version %s rename chain through %q",
version,
item.kind,
item.current,
)
}
}
return nil
}
func parseSchemaVersion(version string) ([3]int, error) {
parts := strings.Split(version, ".")
if len(parts) != 3 {
return [3]int{}, fmt.Errorf("schema version %q must contain major, minor, and patch numbers", version)
}
var parsed [3]int
for i, part := range parts {
value, err := strconv.Atoi(part)
if err != nil || value < 0 {
return [3]int{}, fmt.Errorf("schema version %q contains invalid numeric component %q", version, part)
}
parsed[i] = value
}
return parsed, nil
}
func compareVersionParts(left, right [3]int) int {
for i := range left {
if left[i] < right[i] {
return -1
}
if left[i] > right[i] {
return 1
}
}
return 0
}
func scopeForSection(section string) (contexts, signals []string, allContexts, allSignals bool, err error) {
switch section {
case "all":
return nil, nil, true, true, nil
case "resources":
return []string{"resource"}, nil, false, true, nil
case "spans":
return []string{"attribute"}, []string{"traces"}, false, false, nil
case "logs":
return []string{"attribute"}, []string{"logs"}, false, false, nil
case "metrics":
return []string{"attribute"}, []string{"metrics"}, false, false, nil
default:
return nil, nil, false, false, fmt.Errorf("unsupported schema section %q", section)
}
}
func buildFamilies(schemas []schemaFile, overlay overlayFile) ([]generatedFamily, error) {
edges, err := collectEdges(schemas)
if err != nil {
return nil, err
}
next := make(map[graphKey]string)
for _, item := range edges {
key := graphKey{kind: item.kind, name: item.old}
if existing, ok := next[key]; ok && existing == item.current {
// Repeated entries are common in chained schema histories. Treat an
// identical edge as a no-op so it cannot sever a later edge in the
// same chain (A -> B, B -> C, then a repeated A -> B).
continue
}
// Schema history occasionally repeats an old name with a newer direct
// destination or rolls a rename back. Edges are collected
// oldest-to-newest, so the latest published current name must be a root.
delete(next, graphKey{kind: item.kind, name: item.current})
next[key] = item.current
}
type familyState struct {
family generatedFamily
distance map[string]int
allContexts bool
allSignals bool
}
states := map[graphKey]*familyState{}
for _, item := range edges {
root, distance, err := rootFor(next, item.kind, item.old)
if err != nil {
return nil, err
}
key := graphKey{kind: item.kind, name: root}
state := states[key]
if state == nil {
state = &familyState{
family: generatedFamily{Current: root, Kind: item.kind},
distance: map[string]int{},
}
states[key] = state
}
if prior, ok := state.distance[item.old]; !ok || distance < prior {
state.distance[item.old] = distance
}
state.allContexts = state.allContexts || item.allContexts
state.allSignals = state.allSignals || item.allSignals
state.family.Contexts = appendUnique(state.family.Contexts, item.contexts...)
state.family.Signals = appendUnique(state.family.Signals, item.signals...)
state.family.ApplyToMetrics = appendUnique(state.family.ApplyToMetrics, item.applyToMetrics...)
}
for _, state := range states {
for old := range state.distance {
if old != state.family.Current {
state.family.Old = append(state.family.Old, old)
}
}
sort.Slice(state.family.Old, func(i, j int) bool {
left, right := state.family.Old[i], state.family.Old[j]
if state.distance[left] != state.distance[right] {
return state.distance[left] < state.distance[right]
}
return left < right
})
if state.allContexts {
state.family.Contexts = nil
} else {
sort.Strings(state.family.Contexts)
}
if state.allSignals {
state.family.Signals = nil
} else {
sort.Strings(state.family.Signals)
}
sort.Strings(state.family.ApplyToMetrics)
}
for _, current := range sortedMapKeys(overlay.Families) {
policy := overlay.Families[current]
kind, err := normalizedOverlayKind(current, policy)
if err != nil {
return nil, err
}
policy.Kind = kind
overlay.Families[current] = policy
key := graphKey{kind: kind, name: current}
state := states[key]
if state == nil {
if len(policy.Old) == 0 {
return nil, fmt.Errorf(
"overlay family %q with kind %q is absent from schemas and has no old members",
current,
kind,
)
}
state = &familyState{
family: generatedFamily{Current: current, Kind: kind, Old: append([]string(nil), policy.Old...)},
distance: map[string]int{},
}
states[key] = state
}
applyOverlay(&state.family, policy)
}
var result []generatedFamily
for key, state := range states {
policy, hasPolicy := overlay.Families[key.name]
enabled := overlay.DefaultEnabled
if hasPolicy && policy.Kind != key.kind {
hasPolicy = false
}
if hasPolicy && policy.Enabled != nil {
enabled = *policy.Enabled
}
if !enabled {
continue
}
if len(state.family.Old) == 0 {
return nil, fmt.Errorf(
"enabled family %q with kind %q has no old members",
state.family.Current,
state.family.Kind,
)
}
sort.Strings(state.family.Contexts)
sort.Strings(state.family.Signals)
sort.Strings(state.family.ApplyToMetrics)
result = append(result, state.family)
}
sort.Slice(result, func(i, j int) bool {
if result[i].Current != result[j].Current {
return result[i].Current < result[j].Current
}
return result[i].Kind < result[j].Kind
})
return result, nil
}
func rootFor(next map[graphKey]string, kind, name string) (string, int, error) {
seen := map[string]bool{}
distance := 0
for {
if seen[name] {
return "", 0, fmt.Errorf("rename cycle for %s %q", kind, name)
}
seen[name] = true
current, ok := next[graphKey{kind: kind, name: name}]
if !ok {
return name, distance, nil
}
name = current
distance++
}
}
func normalizedOverlayKind(current string, policy overlayFamily) (string, error) {
kind := policy.Kind
if kind == "" {
kind = kindAttribute
}
if kind != kindAttribute && kind != kindMetric {
return "", fmt.Errorf("overlay family %q has unsupported kind %q", current, kind)
}
return kind, nil
}
func applyOverlay(family *generatedFamily, policy overlayFamily) {
if policy.Kind != "" {
family.Kind = policy.Kind
}
if policy.Old != nil {
family.Old = append([]string(nil), policy.Old...)
}
family.Old = appendUnique(family.Old, policy.AddOld...)
if len(policy.ExcludeOld) > 0 {
excluded := make(map[string]bool, len(policy.ExcludeOld))
for _, old := range policy.ExcludeOld {
excluded[old] = true
}
family.Old = deleteMatching(family.Old, excluded)
}
if policy.Contexts != nil {
family.Contexts = append([]string(nil), policy.Contexts...)
}
if policy.Signals != nil {
family.Signals = append([]string(nil), policy.Signals...)
}
family.Contexts = appendUnique(family.Contexts, policy.AddContexts...)
family.Signals = appendUnique(family.Signals, policy.AddSignals...)
if policy.ApplyToMetrics != nil {
family.ApplyToMetrics = append([]string(nil), policy.ApplyToMetrics...)
}
family.ApplyToMetrics = appendUnique(family.ApplyToMetrics, policy.AddApplyToMetrics...)
if policy.ValueMap != nil {
family.ValueMap = make(map[string]string, len(policy.ValueMap))
for old, current := range policy.ValueMap {
family.ValueMap[old] = current
}
}
}
func appendUnique(values []string, additions ...string) []string {
seen := make(map[string]bool, len(values)+len(additions))
for _, value := range values {
seen[value] = true
}
for _, value := range additions {
if value == "" || seen[value] {
continue
}
seen[value] = true
values = append(values, value)
}
return values
}
func deleteMatching(values []string, excluded map[string]bool) []string {
result := values[:0]
for _, value := range values {
if !excluded[value] {
result = append(result, value)
}
}
return result
}
func renderGo(families []generatedFamily) ([]byte, error) {
var out bytes.Buffer
out.WriteString("// Code generated by scripts/semconv. DO NOT EDIT.\n\n")
out.WriteString("package semconv\n\n")
needsTelemetryTypes := false
for _, family := range families {
if len(family.Contexts) > 0 || len(family.Signals) > 0 {
needsTelemetryTypes = true
break
}
}
if needsTelemetryTypes {
out.WriteString("import \"github.com/SigNoz/signoz/pkg/types/telemetrytypes\"\n\n")
}
out.WriteString("var families = []Family{\n")
for _, family := range families {
contexts, err := goFieldContextSlice(family.Contexts)
if err != nil {
return nil, fmt.Errorf("render family %q: %w", family.Current, err)
}
signals, err := goSignalSlice(family.Signals)
if err != nil {
return nil, fmt.Errorf("render family %q: %w", family.Current, err)
}
out.WriteString("\t{\n")
fmt.Fprintf(&out, "\t\tCurrent: %s,\n", strconv.Quote(family.Current))
fmt.Fprintf(&out, "\t\tOld: %s,\n", goStringSlice(family.Old))
if family.Kind == kindMetric {
out.WriteString("\t\tKind: KindMetric,\n")
} else {
out.WriteString("\t\tKind: KindAttribute,\n")
}
fmt.Fprintf(&out, "\t\tContexts: %s,\n", contexts)
fmt.Fprintf(&out, "\t\tSignals: %s,\n", signals)
fmt.Fprintf(&out, "\t\tApplyToMetrics: %s,\n", goStringSlice(family.ApplyToMetrics))
if len(family.ValueMap) > 0 {
out.WriteString("\t\tValueMap: map[string]string{\n")
keys := sortedMapKeys(family.ValueMap)
for _, key := range keys {
fmt.Fprintf(&out, "\t\t\t%s: %s,\n", strconv.Quote(key), strconv.Quote(family.ValueMap[key]))
}
out.WriteString("\t\t},\n")
}
out.WriteString("\t},\n")
}
out.WriteString("}\n")
return format.Source(out.Bytes())
}
func goStringSlice(values []string) string {
if len(values) == 0 {
return "nil"
}
quoted := make([]string, len(values))
for i, value := range values {
quoted[i] = strconv.Quote(value)
}
return "[]string{" + strings.Join(quoted, ", ") + "}"
}
func goFieldContextSlice(values []string) (string, error) {
if len(values) == 0 {
return "nil", nil
}
constants := make([]string, len(values))
for i, value := range values {
switch value {
case "metric":
constants[i] = "telemetrytypes.FieldContextMetric"
case "resource":
constants[i] = "telemetrytypes.FieldContextResource"
case "attribute":
constants[i] = "telemetrytypes.FieldContextAttribute"
default:
return "", fmt.Errorf("unsupported field context %q", value)
}
}
return "[]telemetrytypes.FieldContext{" + strings.Join(constants, ", ") + "}", nil
}
func goSignalSlice(values []string) (string, error) {
if len(values) == 0 {
return "nil", nil
}
constants := make([]string, len(values))
for i, value := range values {
switch value {
case "traces":
constants[i] = "telemetrytypes.SignalTraces"
case "logs":
constants[i] = "telemetrytypes.SignalLogs"
case "metrics":
constants[i] = "telemetrytypes.SignalMetrics"
default:
return "", fmt.Errorf("unsupported signal %q", value)
}
}
return "[]telemetrytypes.Signal{" + strings.Join(constants, ", ") + "}", nil
}
func renderTypeScript(families []generatedFamily) []byte {
var out bytes.Buffer
out.WriteString("// Code generated by scripts/semconv. DO NOT EDIT.\n\n")
out.WriteString("export type SemconvFamily = {\n")
out.WriteString("\treadonly current: string;\n\treadonly old: readonly string[];\n")
out.WriteString("\treadonly kind: 'attribute' | 'metric';\n")
out.WriteString("\treadonly contexts: readonly string[];\n\treadonly signals: readonly string[];\n")
out.WriteString("\treadonly applyToMetrics: readonly string[];\n")
out.WriteString("\treadonly valueMap: Readonly<Record<string, string>>;\n};\n\n")
out.WriteString("export const SEMCONV_FAMILIES: readonly SemconvFamily[] = [\n")
for _, family := range families {
out.WriteString("\t{\n")
fmt.Fprintf(&out, "\t\tcurrent: %s,\n", tsString(family.Current))
fmt.Fprintf(&out, "\t\told: %s,\n", tsStringSlice(family.Old))
fmt.Fprintf(&out, "\t\tkind: %s,\n", tsString(family.Kind))
fmt.Fprintf(&out, "\t\tcontexts: %s,\n", tsStringSlice(family.Contexts))
fmt.Fprintf(&out, "\t\tsignals: %s,\n", tsStringSlice(family.Signals))
fmt.Fprintf(&out, "\t\tapplyToMetrics: %s,\n", tsStringSlice(family.ApplyToMetrics))
out.WriteString("\t\tvalueMap: {")
keys := sortedMapKeys(family.ValueMap)
for i, key := range keys {
if i > 0 {
out.WriteString(", ")
}
fmt.Fprintf(&out, "%s: %s", tsString(key), tsString(family.ValueMap[key]))
}
out.WriteString("},\n\t},\n")
}
out.WriteString("] as const;\n")
return out.Bytes()
}
func tsString(value string) string {
quoted := strconv.Quote(value)
return "'" + strings.ReplaceAll(quoted[1:len(quoted)-1], "'", `\'`) + "'"
}
func tsStringSlice(values []string) string {
quoted := make([]string, len(values))
for i, value := range values {
quoted[i] = tsString(value)
}
return "[" + strings.Join(quoted, ", ") + "]"
}
func sortedMapKeys[T any](values map[string]T) []string {
keys := make([]string, 0, len(values))
for key := range values {
keys = append(keys, key)
}
sort.Strings(keys)
return keys
}
func checkFile(path string, expected []byte) error {
actual, err := os.ReadFile(path)
if err != nil {
return fmt.Errorf("generated file %s is missing: run go run ./scripts/semconv", path)
}
if !bytes.Equal(actual, expected) {
return fmt.Errorf("generated file %s is stale: run go run ./scripts/semconv", path)
}
return nil
}

View File

@@ -0,0 +1,366 @@
package main
import (
"os"
"path/filepath"
"testing"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestSchemaDecoderRejectsUnsupportedSection(t *testing.T) {
var schema schemaFile
err := decodeKnownFields([]byte(`
versions:
1.0.0:
span_events:
changes:
- rename_events:
event_map:
old: current
`), &schema)
assert.ErrorContains(t, err, "field span_events not found", "unsupported schema sections must fail generation")
}
func TestBuildFamiliesRejectsMalformedSchemaVersion(t *testing.T) {
var schema schemaFile
require.NoError(t, decodeKnownFields([]byte(`
versions:
latest:
spans:
changes: []
`), &schema), "test schema must decode")
_, err := buildFamilies([]schemaFile{schema}, overlayFile{})
assert.ErrorContains(t, err, `schema version "latest"`, "malformed versions must not be silently reordered")
}
func TestBuildFamiliesResolvesRenameChain(t *testing.T) {
var schema schemaFile
require.NoError(t, decodeKnownFields([]byte(`
versions:
4.0.0:
spans:
changes:
- rename_attributes:
attribute_map:
a: b
3.0.0:
spans:
changes:
- rename_attributes:
attribute_map:
b: c
x: c
2.0.0:
spans:
changes:
- rename_attributes:
attribute_map:
a: b
- rename_attributes:
attribute_map:
a: b
`), &schema), "test schema must decode")
enabled := true
families, err := buildFamilies([]schemaFile{schema}, overlayFile{Families: map[string]overlayFamily{
"c": {Enabled: &enabled},
}})
require.NoError(t, err)
assert.Equal(t, []generatedFamily{{
Current: "c",
Old: []string{"b", "x", "a"},
Kind: kindAttribute,
Contexts: []string{"attribute"},
Signals: []string{"traces"},
}}, families, "predecessors should be ordered by distance and then name")
}
func TestBuildFamiliesMapsSchemaSectionsToScopes(t *testing.T) {
var schema schemaFile
require.NoError(t, decodeKnownFields([]byte(`
versions:
1.0.0:
resources:
changes:
- rename_attributes:
attribute_map:
resource.old: resource.current
logs:
changes:
- rename_attributes:
attribute_map:
log.old: log.current
metrics:
changes:
- rename_attributes:
attribute_map:
state: cpu.mode
apply_to_metrics: [system.cpu.time]
- rename_metrics:
old.metric: current.metric
`), &schema), "test schema must decode")
enabled := true
families, err := buildFamilies([]schemaFile{schema}, overlayFile{Families: map[string]overlayFamily{
"resource.current": {Enabled: &enabled},
"log.current": {Enabled: &enabled},
"cpu.mode": {Enabled: &enabled},
"current.metric": {Enabled: &enabled, Kind: kindMetric},
}})
require.NoError(t, err)
assert.Equal(t, []generatedFamily{
{
Current: "cpu.mode", Old: []string{"state"}, Kind: kindAttribute,
Contexts: []string{"attribute"}, Signals: []string{"metrics"},
ApplyToMetrics: []string{"system.cpu.time"},
},
{
Current: "current.metric", Old: []string{"old.metric"}, Kind: kindMetric,
Contexts: []string{"metric"}, Signals: []string{"metrics"},
},
{
Current: "log.current", Old: []string{"log.old"}, Kind: kindAttribute,
Contexts: []string{"attribute"}, Signals: []string{"logs"},
},
{
Current: "resource.current", Old: []string{"resource.old"}, Kind: kindAttribute,
Contexts: []string{"resource"},
},
}, families, "schema sections should produce their documented signal and context scopes")
}
func TestOverlayAddsFamilyWithoutSchemaHistory(t *testing.T) {
enabled := true
families, err := buildFamilies(nil, overlayFile{Families: map[string]overlayFamily{
"added.current": {
Enabled: &enabled,
Old: []string{"added.old"},
Contexts: []string{"resource"},
Signals: []string{"traces"},
},
}})
require.NoError(t, err)
assert.Equal(t, []generatedFamily{{
Current: "added.current",
Old: []string{"added.old"},
Kind: kindAttribute,
Contexts: []string{"resource"},
Signals: []string{"traces"},
}}, families, "an explicit overlay family should not require schema history")
}
func TestOverlayOverridesGeneratedFamily(t *testing.T) {
var schema schemaFile
require.NoError(t, decodeKnownFields([]byte(`
versions:
1.0.0:
spans:
changes:
- rename_attributes:
attribute_map:
old: current
`), &schema), "test schema must decode")
enabled := true
families, err := buildFamilies([]schemaFile{schema}, overlayFile{Families: map[string]overlayFamily{
"current": {
Enabled: &enabled,
AddOld: []string{"older"},
ExcludeOld: []string{"old"},
AddContexts: []string{"resource"},
AddSignals: []string{"logs"},
ValueMap: map[string]string{"legacy": "current"},
},
}})
require.NoError(t, err)
assert.Equal(t, []generatedFamily{{
Current: "current",
Old: []string{"older"},
Kind: kindAttribute,
Contexts: []string{"attribute", "resource"},
Signals: []string{"logs", "traces"},
ValueMap: map[string]string{"legacy": "current"},
}}, families, "overlay additions and exclusions should be applied to the generated family")
}
func TestOverlayDisablesFamilyWhenDefaultIsEnabled(t *testing.T) {
var schema schemaFile
require.NoError(t, decodeKnownFields([]byte(`
versions:
1.0.0:
spans:
changes:
- rename_attributes:
attribute_map:
old: current
`), &schema), "test schema must decode")
disabled := false
families, err := buildFamilies([]schemaFile{schema}, overlayFile{
DefaultEnabled: true,
Families: map[string]overlayFamily{
"current": {Enabled: &disabled},
},
})
require.NoError(t, err)
assert.Empty(t, families, "an explicitly disabled family must override default_enabled")
}
func TestRenderGoIsDeterministic(t *testing.T) {
families := []generatedFamily{{
Current: "current", Old: []string{"old"}, Kind: kindAttribute,
ValueMap: map[string]string{"b": "2", "a": "1"},
}}
first, err := renderGo(families)
require.NoError(t, err)
second, err := renderGo(families)
require.NoError(t, err)
assert.Equal(t, first, second, "Go generation must not depend on map iteration order")
}
func TestRenderGoUsesCanonicalTelemetryTypes(t *testing.T) {
families := []generatedFamily{{
Current: "current", Old: []string{"old"}, Kind: kindAttribute,
Contexts: []string{"resource"}, Signals: []string{"traces"},
}}
output, err := renderGo(families)
require.NoError(t, err)
assert.Contains(t, string(output), "telemetrytypes.FieldContextResource", "generated contexts should use telemetrytypes")
assert.Contains(t, string(output), "telemetrytypes.SignalTraces", "generated signals should use telemetrytypes")
}
func TestRenderTypeScriptIsDeterministic(t *testing.T) {
families := []generatedFamily{{
Current: "current", Old: []string{"old"}, Kind: kindAttribute,
ValueMap: map[string]string{"b": "2", "a": "1"},
}}
assert.Equal(t, renderTypeScript(families), renderTypeScript(families), "TypeScript generation must not depend on map iteration order")
}
func TestBuildFamiliesHandlesRenameRollback(t *testing.T) {
var schema schemaFile
require.NoError(t, decodeKnownFields([]byte(`
versions:
2.0.0:
metrics:
changes:
- rename_metrics:
temporary: original
1.0.0:
metrics:
changes:
- rename_metrics:
original: temporary
`), &schema), "test schema must decode")
enabled := true
families, err := buildFamilies([]schemaFile{schema}, overlayFile{Families: map[string]overlayFamily{
"original": {Enabled: &enabled, Kind: kindMetric},
}})
require.NoError(t, err)
assert.Equal(t, []generatedFamily{{
Current: "original",
Old: []string{"temporary"},
Kind: kindMetric,
Contexts: []string{"metric"},
Signals: []string{"metrics"},
}}, families, "the latest rollback destination should remain the family root")
}
func TestBuildFamiliesRejectsSameVersionRenameChain(t *testing.T) {
var schema schemaFile
require.NoError(t, decodeKnownFields([]byte(`
versions:
1.0.0:
spans:
changes:
- rename_attributes:
attribute_map:
x: y
y: z
`), &schema), "test schema must decode")
_, err := buildFamilies([]schemaFile{schema}, overlayFile{})
assert.ErrorContains(t, err, `same-version attribute rename chain through "y"`, "order-sensitive same-version chains must be rejected")
}
func TestBuildFamiliesRejectsOverlayFamilyWithoutHistory(t *testing.T) {
enabled := true
_, err := buildFamilies(nil, overlayFile{Families: map[string]overlayFamily{
"missing": {Enabled: &enabled},
}})
assert.ErrorContains(t, err, `overlay family "missing" with kind "attribute" is absent`, "an overlay cannot invent a family without old members")
}
func TestBuildFamiliesRejectsEnabledFamilyWithoutOldMembers(t *testing.T) {
var schema schemaFile
require.NoError(t, decodeKnownFields([]byte(`
versions:
1.0.0:
spans:
changes:
- rename_attributes:
attribute_map:
old: current
`), &schema), "test schema must decode")
enabled := true
_, err := buildFamilies([]schemaFile{schema}, overlayFile{Families: map[string]overlayFamily{
"current": {Enabled: &enabled, ExcludeOld: []string{"old"}},
}})
assert.ErrorContains(t, err, `enabled family "current" with kind "attribute" has no old members`, "exclude_old cannot empty an enabled family")
}
func TestOverlayKindDefaultsToAttribute(t *testing.T) {
var schema schemaFile
require.NoError(t, decodeKnownFields([]byte(`
versions:
1.0.0:
spans:
changes:
- rename_attributes:
attribute_map:
attribute.old: shared.current
metrics:
changes:
- rename_metrics:
metric.old: shared.current
`), &schema), "test schema must decode")
enabled := true
families, err := buildFamilies([]schemaFile{schema}, overlayFile{Families: map[string]overlayFamily{
"shared.current": {Enabled: &enabled},
}})
require.NoError(t, err)
assert.Equal(t, []generatedFamily{{
Current: "shared.current",
Old: []string{"attribute.old"},
Kind: kindAttribute,
Contexts: []string{"attribute"},
Signals: []string{"traces"},
}}, families, "a kind-less overlay policy should affect only the attribute family")
}
func TestCheckFileReportsStaleOutput(t *testing.T) {
path := filepath.Join(t.TempDir(), "generated.go")
require.NoError(t, os.WriteFile(path, []byte("old"), 0o600), "test output must be writable")
assert.ErrorContains(t, checkFile(path, []byte("new")), "is stale", "check mode must reject stale generated output")
}
func TestTypeScriptStringEscapesControlCharacters(t *testing.T) {
assert.Equal(t, `'line\n\t\x01\'\\end'`, tsString("line\n\t\x01'\\end"), "generated TypeScript strings must remain valid literals")
}

View File

@@ -0,0 +1,11 @@
# SigNoz semantic-convention rollout policy.
#
# Families are keyed by their current OpenTelemetry name. Schema-derived
# families are disabled by default so rollout remains explicit and reversible.
default_enabled: false
families:
deployment.environment.name:
enabled: true
db.system.name:
enabled: true

View File

@@ -0,0 +1,760 @@
file_format: 1.1.0
schema_url: https://opentelemetry.io/schemas/1.42.0
versions:
1.42.0:
metrics:
changes:
- rename_metrics:
v8js.memory.heap.limit: v8js.memory.heap.space.size
1.41.1:
1.41.0:
metrics:
changes:
- rename_metrics:
k8s.container.cpu.limit: k8s.container.cpu.limit.desired
k8s.container.cpu.limit_utilization: k8s.container.cpu.limit.utilization
k8s.container.cpu.request: k8s.container.cpu.request.desired
k8s.container.cpu.request_utilization: k8s.container.cpu.request.utilization
k8s.container.memory.limit: k8s.container.memory.limit.desired
k8s.container.memory.request: k8s.container.memory.request.desired
1.40.0:
all:
changes:
- rename_attributes:
attribute_map:
feature_flag.evaluation.error.message: feature_flag.error.message
metrics:
changes:
- rename_metrics:
system.memory.shared: system.memory.linux.shared
1.39.0:
all:
changes:
- rename_attributes:
attribute_map:
linux.memory.slab.state: system.memory.linux.slab.state
peer.service: service.peer.name
rpc.connect_rpc.error_code: rpc.response.status_code
rpc.connect_rpc.request.metadata: rpc.request.metadata
rpc.connect_rpc.response.metadata: rpc.response.metadata
rpc.grpc.request.metadata: rpc.request.metadata
rpc.grpc.response.metadata: rpc.response.metadata
rpc.jsonrpc.request_id: jsonrpc.request.id
rpc.jsonrpc.version: jsonrpc.protocol.version
rpc.system: rpc.system.name
metrics:
changes:
- rename_metrics:
process.open_file_descriptor.count: process.unix.file_descriptor.count
system.linux.memory.available: system.memory.linux.available
system.linux.memory.slab.usage: system.memory.linux.slab.usage
1.38.0:
all:
changes:
- rename_attributes:
attribute_map:
process.context_switch_type: process.context_switch.type
process.paging.fault_type: system.paging.fault.type
system.cpu.logical_number: cpu.logical_number
system.paging.type: system.paging.fault.type
system.process.status: process.state
system.processes.status: process.state
metrics:
changes:
- rename_metrics:
k8s.cronjob.active_jobs: k8s.cronjob.job.active
k8s.daemonset.current_scheduled_nodes: k8s.daemonset.node.current_scheduled
k8s.daemonset.desired_scheduled_nodes: k8s.daemonset.node.desired_scheduled
k8s.daemonset.misscheduled_nodes: k8s.daemonset.node.misscheduled
k8s.daemonset.ready_nodes: k8s.daemonset.node.ready
k8s.deployment.available_pods: k8s.deployment.pod.available
k8s.deployment.desired_pods: k8s.deployment.pod.desired
k8s.hpa.current_pods: k8s.hpa.pod.current
k8s.hpa.desired_pods: k8s.hpa.pod.desired
k8s.hpa.max_pods: k8s.hpa.pod.max
k8s.hpa.min_pods: k8s.hpa.pod.min
k8s.job.active_pods: k8s.job.pod.active
k8s.job.desired_successful_pods: k8s.job.pod.desired_successful
k8s.job.failed_pods: k8s.job.pod.failed
k8s.job.max_parallel_pods: k8s.job.pod.max_parallel
k8s.job.successful_pods: k8s.job.pod.successful
k8s.node.allocatable.cpu: k8s.node.cpu.allocatable
k8s.node.allocatable.ephemeral_storage: k8s.node.ephemeral_storage.allocatable
k8s.node.allocatable.memory: k8s.node.memory.allocatable
k8s.node.allocatable.pods: k8s.node.pod.allocatable
k8s.replicaset.available_pods: k8s.replicaset.pod.available
k8s.replicaset.desired_pods: k8s.replicaset.pod.desired
k8s.replication_controller.available_pods: k8s.replicationcontroller.pod.available
k8s.replication_controller.desired_pods: k8s.replicationcontroller.pod.desired
k8s.replicationcontroller.available_pods: k8s.replicationcontroller.pod.available
k8s.replicationcontroller.desired_pods: k8s.replicationcontroller.pod.desired
k8s.statefulset.current_pods: k8s.statefulset.pod.current
k8s.statefulset.desired_pods: k8s.statefulset.pod.desired
k8s.statefulset.ready_pods: k8s.statefulset.pod.ready
k8s.statefulset.updated_pods: k8s.statefulset.pod.updated
v8js.heap.space.available_size: v8js.memory.heap.space.available_size
v8js.heap.space.physical_size: v8js.memory.heap.space.physical_size
1.37.0:
all:
changes:
- rename_attributes:
attribute_map:
android.state: android.app.state
container.runtime: container.runtime.name
enduser.role: user.roles
gen_ai.openai.request.service_tier: openai.request.service_tier
gen_ai.openai.response.service_tier: openai.response.service_tier
gen_ai.openai.response.system_fingerprint: openai.response.system_fingerprint
gen_ai.system: gen_ai.provider.name
ios.state: ios.app.state
1.36.0:
1.35.0:
all:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1698
- rename_attributes:
attribute_map:
az.namespace: azure.resource_provider.namespace
az.service_request_id: azure.service.request.id
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/issues/1800
- rename_metrics:
system.network.connections: system.network.connection.count
1.34.0:
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/2295
- rename_metrics:
cpu.time: system.cpu.time
cpu.utilization: system.cpu.utilization
cpu.frequency: system.cpu.frequency
1.33.0:
all:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1982
- rename_attributes:
attribute_map:
feature_flag.provider_name: feature_flag.provider.name
# https://github.com/open-telemetry/semantic-conventions/pull/1994
- rename_attributes:
attribute_map:
feature_flag.evaluation.error.message: error.message
1.32.0:
all:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1989
- rename_attributes:
attribute_map:
feature_flag.evaluation.reason: feature_flag.result.reason
feature_flag.variant: feature_flag.result.variant
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/2042
- rename_metrics:
otel.sdk.span.live.count: otel.sdk.span.live
otel.sdk.span.ended.count: otel.sdk.span.ended
otel.sdk.processor.span.processed.count: otel.sdk.processor.span.processed
otel.sdk.exporter.span.inflight.count: otel.sdk.exporter.span.inflight
otel.sdk.exporter.span.exported.count: otel.sdk.exporter.span.exported
1.31.0:
all:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1880
- rename_attributes:
attribute_map:
android.state: android.app.state
io.state: ios.app.state
metrics:
changes:
- rename_metrics:
k8s.replication_controller.desired_pods: k8s.replicationcontroller.desired_pods
k8s.replication_controller.available_pods: k8s.replicationcontroller.available_pods
# https://github.com/open-telemetry/semantic-conventions/pull/1896
- rename_metrics:
system.cpu.time: cpu.time
system.cpu.utilization: cpu.utilization
system.cpu.frequency: cpu.frequency
# https://github.com/open-telemetry/semantic-conventions/pull/1896
- rename_attributes:
attribute_map:
system.cpu.logical_number: cpu.logical_number
1.30.0:
all:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1632
- rename_attributes:
attribute_map:
gen_ai.openai.request.seed: gen_ai.request.seed
system.network.state: network.connection.state
# https://github.com/open-telemetry/semantic-conventions/pull/1624
- rename_attributes:
attribute_map:
code.function: code.function.name
code.filepath: code.file.path
code.lineno: code.line.number
code.column: code.column.number
# https://github.com/open-telemetry/semantic-conventions/pull/1734
- rename_attributes:
attribute_map:
db.system: db.system.name
db.cassandra.coordinator.dc: cassandra.coordinator.dc
db.cassandra.coordinator.id: cassandra.coordinator.id
db.cassandra.consistency_level: cassandra.consistency.level
db.cassandra.idempotence: cassandra.query.idempotent
db.cassandra.page_size: cassandra.page.size
db.cassandra.speculative_execution_count: cassandra.speculative_execution.count
db.cosmosdb.client_id: azure.client.id
db.cosmosdb.connection_mode: azure.cosmosdb.connection.mode
db.cosmosdb.consistency_level: azure.cosmosdb.consistency.level
db.cosmosdb.request_charge: azure.cosmosdb.operation.request_charge
db.cosmosdb.request_content_length: azure.cosmosdb.request.body.size
db.cosmosdb.regions_contacted: azure.cosmosdb.operation.contacted_regions
db.cosmosdb.sub_status_code: azure.cosmosdb.response.sub_status_code
db.elasticsearch.node.name: elasticsearch.node.name
# db.elasticsearch.path_parts is a template attribute, schema transformation
# does not support it, adding as a comment for consistency
# db.elasticsearch.path_parts.<key> -> db.operation.parameter.<key>
metrics:
changes:
- rename_metrics:
db.client.cosmosdb.operation.request_charge: azure.cosmosdb.client.operation.request_charge
db.client.cosmosdb.active_instance.count: azure.cosmosdb.client.active_instance.count
1.29.0:
all:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1520
- rename_attributes:
attribute_map:
process.executable.build_id.profiling: process.executable.build_id.htlhash
# https://github.com/open-telemetry/semantic-conventions/pull/1383
- rename_attributes:
attribute_map:
vcs.repository.change.id: vcs.change.id
vcs.repository.change.title: vcs.change.title
vcs.repository.ref.name: vcs.ref.head.name
vcs.repository.ref.revision: vcs.ref.head.revision
vcs.repository.ref.type: vcs.ref.head.type
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1492
- rename_attributes:
attribute_map:
system.device: network.interface.name
apply_to_metrics:
- container.network.io
- system.network.dropped
- system.network.errors
- system.network.io
- system.network.connections
1.28.0:
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1422
- rename_metrics:
messaging.client.published.messages: messaging.client.sent.messages
1.27.0:
all:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1216
- rename_attributes:
attribute_map:
tls.client.server_name: server.address
# https://github.com/open-telemetry/semantic-conventions/pull/1075
- rename_attributes:
attribute_map:
deployment.environment: deployment.environment.name
# https://github.com/open-telemetry/semantic-conventions/pull/1245
- rename_attributes:
attribute_map:
messaging.kafka.message.offset: messaging.kafka.offset
# https://github.com/open-telemetry/semantic-conventions/pull/815
- rename_attributes:
attribute_map:
messaging.kafka.consumer.group: messaging.consumer.group.name
messaging.rocketmq.client_group: messaging.consumer.group.name
messaging.eventhubs.consumer.group: messaging.consumer.group.name
messaging.servicebus.destination.subscription_name: messaging.destination.subscription.name
# https://github.com/open-telemetry/semantic-conventions/pull/1200
- rename_attributes:
attribute_map:
gen_ai.usage.completion_tokens: gen_ai.usage.output_tokens
gen_ai.usage.prompt_tokens: gen_ai.usage.input_tokens
spans:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1002
- rename_attributes:
attribute_map:
db.elasticsearch.cluster.name: db.namespace
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/1125
- rename_attributes:
attribute_map:
db.client.connections.state: db.client.connection.state
apply_to_metrics:
- db.client.connection.count
- rename_attributes:
attribute_map:
db.client.connections.pool.name: db.client.connection.pool.name
apply_to_metrics:
- db.client.connection.count
- db.client.connection.idle.max
- db.client.connection.idle.min
- db.client.connection.max
- db.client.connection.pending_requests
- db.client.connection.timeouts
- db.client.connection.create_time
- db.client.connection.wait_time
- db.client.connection.use_time
# https://github.com/open-telemetry/semantic-conventions/pull/1006
- rename_metrics:
messaging.publish.messages: messaging.client.published.messages
# https://github.com/open-telemetry/semantic-conventions/pull/1026
- rename_attributes:
attribute_map:
system.cpu.state: cpu.mode
process.cpu.state: cpu.mode
container.cpu.state: cpu.mode
apply_to_metrics:
- system.cpu.time
- system.cpu.utilization
- process.cpu.time
- process.cpu.utilization
- container.cpu.time
# https://github.com/open-telemetry/semantic-conventions/pull/1265
- rename_metrics:
jvm.buffer.memory.usage: jvm.buffer.memory.used
1.26.0:
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/966
- rename_metrics:
db.client.connections.usage: db.client.connection.count
db.client.connections.idle.max: db.client.connection.idle.max
db.client.connections.idle.min: db.client.connection.idle.min
db.client.connections.max: db.client.connection.max
db.client.connections.pending_requests: db.client.connection.pending_requests
db.client.connections.timeouts: db.client.connection.timeouts
# https://github.com/open-telemetry/semantic-conventions/pull/948
- rename_attributes:
attribute_map:
messaging.client_id: messaging.client.id
# https://github.com/open-telemetry/semantic-conventions/pull/909
- rename_attributes:
attribute_map:
state: db.client.connections.state
apply_to_metrics:
- db.client.connections.usage
- rename_attributes:
attribute_map:
pool.name: db.client.connections.pool.name
apply_to_metrics:
- db.client.connections.usage
- db.client.connections.idle.max
- db.client.connections.idle.min
- db.client.connections.max
- db.client.connections.pending_requests
- db.client.connections.timeouts
- db.client.connections.create_time
- db.client.connections.wait_time
- db.client.connections.use_time
all:
changes:
# https://github:com/open-telemetry/semantic-conventions/pull/731/
- rename_attributes:
attribute_map:
enduser.id: user.id
1.25.0:
spans:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/911
- rename_attributes:
attribute_map:
db.name: db.namespace
# https://github.com/open-telemetry/semantic-conventions/pull/870
- rename_attributes:
attribute_map:
db.sql.table: db.collection.name
db.mongodb.collection: db.collection.name
db.cosmosdb.container: db.collection.name
db.cassandra.table: db.collection.name
# https://github.com/open-telemetry/semantic-conventions/pull/798
- rename_attributes:
attribute_map:
messaging.kafka.destination.partition: messaging.destination.partition.id
# https://github.com/open-telemetry/semantic-conventions/pull/875
- rename_attributes:
attribute_map:
db.operation: db.operation.name
# https://github.com/open-telemetry/semantic-conventions/pull/913
- rename_attributes:
attribute_map:
messaging.operation: messaging.operation.type
# https://github.com/open-telemetry/semantic-conventions/pull/866
- rename_attributes:
attribute_map:
db.statement: db.query.text
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/484
- rename_attributes:
attribute_map:
system.processes.status: system.process.status
apply_to_metrics:
- system.processes.count
- rename_metrics:
system.processes.count: system.process.count
system.processes.created: system.process.created
# https://github.com/open-telemetry/semantic-conventions/pull/625
- rename_attributes:
attribute_map:
container.labels: container.label
k8s.pod.labels: k8s.pod.label
# https://github.com/open-telemetry/semantic-conventions/pull/330
- rename_metrics:
process.threads: process.thread.count
process.open_file_descriptors: process.open_file_descriptor.count
- rename_attributes:
attribute_map:
state: process.cpu.state
apply_to_metrics:
- process.cpu.time
- process.cpu.utilization
- rename_attributes:
attribute_map:
direction: disk.io.direction
apply_to_metrics:
- process.disk.io
- rename_attributes:
attribute_map:
type: process.context_switch_type
apply_to_metrics:
- process.context_switches
- rename_attributes:
attribute_map:
direction: network.io.direction
apply_to_metrics:
- process.network.io
- rename_attributes:
attribute_map:
type: process.paging.fault_type
apply_to_metrics:
- process.paging.faults
all:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/854
- rename_attributes:
attribute_map:
message.type: rpc.message.type
message.id: rpc.message.id
message.compressed_size: rpc.message.compressed_size
message.uncompressed_size: rpc.message.uncompressed_size
1.24.0:
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/536
- rename_metrics:
jvm.memory.usage: jvm.memory.used
jvm.memory.usage_after_last_gc: jvm.memory.used_after_last_gc
# https://github.com/open-telemetry/semantic-conventions/pull/530
- rename_attributes:
attribute_map:
system.network.io.direction: network.io.direction
system.disk.io.direction: disk.io.direction
1.23.1:
1.23.0:
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/20
- rename_attributes:
attribute_map:
thread.daemon: jvm.thread.daemon
apply_to_metrics:
- jvm.thread.count
1.22.0:
spans:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/229
- rename_attributes:
attribute_map:
messaging.message.payload_size_bytes: messaging.message.body.size
# https://github.com/open-telemetry/opentelemetry-specification/pull/374
- rename_attributes:
attribute_map:
http.resend_count: http.request.resend_count
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/224
- rename_metrics:
http.client.duration: http.client.request.duration
http.server.duration: http.server.request.duration
# https://github.com/open-telemetry/semantic-conventions/pull/241
- rename_metrics:
process.runtime.jvm.memory.usage: jvm.memory.usage
process.runtime.jvm.memory.committed: jvm.memory.committed
process.runtime.jvm.memory.limit: jvm.memory.limit
process.runtime.jvm.memory.usage_after_last_gc: jvm.memory.usage_after_last_gc
process.runtime.jvm.gc.duration: jvm.gc.duration
# also https://github.com/open-telemetry/semantic-conventions/pull/252
process.runtime.jvm.threads.count: jvm.thread.count
# also https://github.com/open-telemetry/semantic-conventions/pull/252
process.runtime.jvm.classes.loaded: jvm.class.loaded
# also https://github.com/open-telemetry/semantic-conventions/pull/252
process.runtime.jvm.classes.unloaded: jvm.class.unloaded
# also https://github.com/open-telemetry/semantic-conventions/pull/252
# and https://github.com/open-telemetry/semantic-conventions/pull/60
process.runtime.jvm.classes.current_loaded: jvm.class.count
process.runtime.jvm.cpu.time: jvm.cpu.time
process.runtime.jvm.cpu.recent_utilization: jvm.cpu.recent_utilization
process.runtime.jvm.memory.init: jvm.memory.init
process.runtime.jvm.system.cpu.utilization: jvm.system.cpu.utilization
process.runtime.jvm.system.cpu.load_1m: jvm.system.cpu.load_1m
# https://github.com/open-telemetry/semantic-conventions/pull/253
process.runtime.jvm.buffer.usage: jvm.buffer.memory.usage
# https://github.com/open-telemetry/semantic-conventions/pull/253
process.runtime.jvm.buffer.limit: jvm.buffer.memory.limit
process.runtime.jvm.buffer.count: jvm.buffer.count
# https://github.com/open-telemetry/semantic-conventions/pull/20
- rename_attributes:
attribute_map:
type: jvm.memory.type
pool: jvm.memory.pool.name
apply_to_metrics:
- jvm.memory.usage
- jvm.memory.committed
- jvm.memory.limit
- jvm.memory.usage_after_last_gc
- jvm.memory.init
- rename_attributes:
attribute_map:
name: jvm.gc.name
action: jvm.gc.action
apply_to_metrics:
- jvm.gc.duration
- rename_attributes:
attribute_map:
daemon: thread.daemon
apply_to_metrics:
- jvm.threads.count
- rename_attributes:
attribute_map:
pool: jvm.buffer.pool.name
apply_to_metrics:
- jvm.buffer.memory.usage
- jvm.buffer.memory.limit
- jvm.buffer.count
# https://github.com/open-telemetry/semantic-conventions/pull/89
- rename_attributes:
attribute_map:
state: system.cpu.state
cpu: system.cpu.logical_number
apply_to_metrics:
- system.cpu.time
- system.cpu.utilization
- rename_attributes:
attribute_map:
state: system.memory.state
apply_to_metrics:
- system.memory.usage
- system.memory.utilization
- rename_attributes:
attribute_map:
state: system.paging.state
apply_to_metrics:
- system.paging.usage
- system.paging.utilization
- rename_attributes:
attribute_map:
type: system.paging.type
direction: system.paging.direction
apply_to_metrics:
- system.paging.faults
- system.paging.operations
- rename_attributes:
attribute_map:
device: system.device
direction: system.disk.direction
apply_to_metrics:
- system.disk.io
- system.disk.operations
- system.disk.io_time
- system.disk.operation_time
- system.disk.merged
- rename_attributes:
attribute_map:
device: system.device
state: system.filesystem.state
type: system.filesystem.type
mode: system.filesystem.mode
mountpoint: system.filesystem.mountpoint
apply_to_metrics:
- system.filesystem.usage
- system.filesystem.utilization
- rename_attributes:
attribute_map:
device: system.device
direction: system.network.direction
protocol: network.protocol
state: system.network.state
apply_to_metrics:
- system.network.dropped
- system.network.packets
- system.network.errors
- system.network.io
- system.network.connections
- rename_attributes:
attribute_map:
status: system.processes.status
apply_to_metrics:
- system.processes.count
# https://github.com/open-telemetry/semantic-conventions/pull/247
- rename_metrics:
http.server.request.size: http.server.request.body.size
http.server.response.size: http.server.response.body.size
resources:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/178
- rename_attributes:
attribute_map:
telemetry.auto.version: telemetry.distro.version
1.21.0:
spans:
changes:
# https://github.com/open-telemetry/opentelemetry-specification/pull/3336
- rename_attributes:
attribute_map:
messaging.kafka.client_id: messaging.client_id
messaging.rocketmq.client_id: messaging.client_id
# https://github.com/open-telemetry/opentelemetry-specification/pull/3402
- rename_attributes:
attribute_map:
# net.peer.(name|port) attributes were usually populated on client side
# so they should be usually translated to server.(address|port)
# net.host.* attributes were only populated on server side
net.host.name: server.address
net.host.port: server.port
# was only populated on client side
net.sock.peer.name: server.socket.domain
# net.sock.peer.(addr|port) mapping is not possible
# since they applied to both client and server side
# were only populated on server side
net.sock.host.addr: server.socket.address
net.sock.host.port: server.socket.port
http.client_ip: client.address
# https://github.com/open-telemetry/opentelemetry-specification/pull/3426
- rename_attributes:
attribute_map:
net.protocol.name: network.protocol.name
net.protocol.version: network.protocol.version
net.host.connection.type: network.connection.type
net.host.connection.subtype: network.connection.subtype
net.host.carrier.name: network.carrier.name
net.host.carrier.mcc: network.carrier.mcc
net.host.carrier.mnc: network.carrier.mnc
net.host.carrier.icc: network.carrier.icc
# https://github.com/open-telemetry/opentelemetry-specification/pull/3355
- rename_attributes:
attribute_map:
http.method: http.request.method
http.status_code: http.response.status_code
http.scheme: url.scheme
http.url: url.full
http.request_content_length: http.request.body.size
http.response_content_length: http.response.body.size
metrics:
changes:
# https://github.com/open-telemetry/semantic-conventions/pull/53
- rename_metrics:
process.runtime.jvm.cpu.utilization: process.runtime.jvm.cpu.recent_utilization
1.20.0:
spans:
changes:
# https://github.com/open-telemetry/opentelemetry-specification/pull/3272
- rename_attributes:
attribute_map:
net.app.protocol.name: net.protocol.name
net.app.protocol.version: net.protocol.version
1.19.0:
spans:
changes:
# https://github.com/open-telemetry/opentelemetry-specification/pull/3209
- rename_attributes:
attribute_map:
faas.execution: faas.invocation_id
# https://github.com/open-telemetry/opentelemetry-specification/pull/3188
- rename_attributes:
attribute_map:
faas.id: cloud.resource_id
# https://github.com/open-telemetry/opentelemetry-specification/pull/3190
- rename_attributes:
attribute_map:
http.user_agent: user_agent.original
resources:
changes:
# https://github.com/open-telemetry/opentelemetry-specification/pull/3190
- rename_attributes:
attribute_map:
browser.user_agent: user_agent.original
1.18.0:
1.17.0:
spans:
changes:
# https://github.com/open-telemetry/opentelemetry-specification/pull/2957
- rename_attributes:
attribute_map:
messaging.consumer_id: messaging.consumer.id
messaging.protocol: net.app.protocol.name
messaging.protocol_version: net.app.protocol.version
messaging.destination: messaging.destination.name
messaging.temp_destination: messaging.destination.temporary
messaging.destination_kind: messaging.destination.kind
messaging.message_id: messaging.message.id
messaging.conversation_id: messaging.message.conversation_id
messaging.message_payload_size_bytes: messaging.message.payload_size_bytes
messaging.message_payload_compressed_size_bytes: messaging.message.payload_compressed_size_bytes
messaging.rabbitmq.routing_key: messaging.rabbitmq.destination.routing_key
messaging.kafka.message_key: messaging.kafka.message.key
messaging.kafka.partition: messaging.kafka.destination.partition
messaging.kafka.tombstone: messaging.kafka.message.tombstone
messaging.rocketmq.message_type: messaging.rocketmq.message.type
messaging.rocketmq.message_tag: messaging.rocketmq.message.tag
messaging.rocketmq.message_keys: messaging.rocketmq.message.keys
messaging.kafka.consumer_group: messaging.kafka.consumer.group
1.16.0:
1.15.0:
spans:
changes:
# https://github.com/open-telemetry/opentelemetry-specification/pull/2743
- rename_attributes:
attribute_map:
http.retry_count: http.resend_count
1.14.0:
1.13.0:
spans:
changes:
# https://github.com/open-telemetry/opentelemetry-specification/pull/2614
- rename_attributes:
attribute_map:
net.peer.ip: net.sock.peer.addr
net.host.ip: net.sock.host.addr
1.12.0:
1.11.0:
1.10.0:
1.9.0:
1.8.0:
spans:
changes:
- rename_attributes:
attribute_map:
db.cassandra.keyspace: db.name
db.hbase.namespace: db.name
1.7.0:
1.6.1:
1.5.0:
1.4.0: