Compare commits

...

2 Commits

Author SHA1 Message Date
srikanthccv
303d6a0363 feat: add semconv migration product surfaces 2026-08-07 01:21:14 +05:30
srikanthccv
6b4bb82efb feat: resolve semconv families across logs and metrics 2026-08-07 01:20:58 +05:30
82 changed files with 1943 additions and 197 deletions

View File

@@ -224,6 +224,14 @@ py-test: ## Runs integration tests
py-test-semconv-phase1: py-test-setup ## Rebuild the shared stack and run the semantic-convention Phase 1 matrix
@cd tests && uv run pytest --basetemp=./tmp/ -vv --reuse --capture=no integration/tests/queriertraces/13_semconv_evolution.py
.PHONY: py-test-semconv-phase2
py-test-semconv-phase2: py-test-setup ## Rebuild the shared stack and run the Phase 1-2 cross-signal matrices
@cd tests && uv run pytest --basetemp=./tmp/ -vv --reuse --capture=no integration/tests/queriertraces/13_semconv_evolution.py integration/tests/queriersemconv/02_cross_signal.py
.PHONY: py-test-semconv-phase3
py-test-semconv-phase3: py-test-setup ## Rebuild the shared stack and run the Phase 1-3 compatibility and migration-report matrices
@cd tests && uv run pytest --basetemp=./tmp/ -vv --reuse --capture=no integration/tests/queriertraces/13_semconv_evolution.py integration/tests/queriersemconv/02_cross_signal.py
.PHONY: py-clean
py-clean: ## Clear all pycache and pytest cache from tests directory recursively
@echo ">> cleaning python cache files from tests directory"

View File

@@ -0,0 +1,9 @@
import axios from 'api';
import { SemconvMigrationReport } from 'types/api/semconvMigration';
async function getSemconvMigrationReport(): Promise<SemconvMigrationReport> {
const response = await axios.get('/fields/semconv-migration');
return response.data.data;
}
export default getSemconvMigrationReport;

View File

@@ -5,7 +5,7 @@ import { ArrowUpRight } from '@signozhq/icons';
const QUICK_FILTER_DOC_PATHS: Record<string, string> = {
severity_text: 'severity-text',
'deployment.environment': 'environment',
'deployment.environment.name': 'environment',
'service.name': 'service-name',
'host.name': 'hostname',
'k8s.cluster.name': 'k8s-cluster-name',

View File

@@ -0,0 +1,32 @@
import { Alert } from 'antd';
import { findOldSemconvNames } from 'utils/semconv';
interface SemconvEditorWarningProps {
value: unknown;
editor: string;
}
function SemconvEditorWarning({
value,
editor,
}: SemconvEditorWarningProps): JSX.Element | null {
const text = typeof value === 'string' ? value : JSON.stringify(value ?? '');
const renames = findOldSemconvNames(text);
if (renames.length === 0) {
return null;
}
return (
<Alert
type="warning"
showIcon
data-testid="semconv-editor-warning"
message={`${editor} contains renamed OpenTelemetry fields`}
description={renames
.map(({ old, current }) => `${old}${current}`)
.join(', ')}
/>
);
}
export default SemconvEditorWarning;

View File

@@ -0,0 +1,23 @@
import { Badge } from '@signozhq/ui/badge';
import { getSemconvRename } from 'utils/semconv';
interface SemconvOldNameBadgeProps {
name: string;
}
function SemconvOldNameBadge({
name,
}: SemconvOldNameBadgeProps): JSX.Element | null {
const rename = getSemconvRename(name);
if (!rename || rename.family.kind !== 'attribute') {
return null;
}
return (
<Badge color="amber" variant="outline" data-testid="semconv-old-name-badge">
old name, renamed to {rename.current}
</Badge>
);
}
export default SemconvOldNameBadge;

View File

@@ -0,0 +1,39 @@
import { render, screen } from '@testing-library/react';
import SemconvEditorWarning from '../SemconvEditorWarning';
import SemconvOldNameBadge from '../SemconvOldNameBadge';
describe('semantic convention product hints', () => {
it('badges an old raw attribute with its current name', () => {
render(<SemconvOldNameBadge name="deployment.environment" />);
expect(screen.getByTestId('semconv-old-name-badge')).toHaveTextContent(
'old name, renamed to deployment.environment.name',
);
});
it('does not badge a current raw attribute', () => {
render(<SemconvOldNameBadge name="deployment.environment.name" />);
expect(
screen.queryByTestId('semconv-old-name-badge'),
).not.toBeInTheDocument();
});
it('shows an informational editor warning without disabling the editor', () => {
render(
<>
<input aria-label="query" defaultValue="db.system = 'postgresql'" />
<SemconvEditorWarning
value="db.system = 'postgresql'"
editor="ClickHouse SQL"
/>
</>,
);
expect(screen.getByLabelText('query')).not.toBeDisabled();
expect(screen.getByTestId('semconv-editor-warning')).toHaveTextContent(
'db.system → db.system.name',
);
});
});

View File

@@ -0,0 +1,2 @@
export { default as SemconvEditorWarning } from './SemconvEditorWarning';
export { default as SemconvOldNameBadge } from './SemconvOldNameBadge';

View File

@@ -11,12 +11,21 @@ export type SemconvFamily = {
};
export const SEMCONV_FAMILIES: readonly SemconvFamily[] = [
{
current: 'container.cpu.usage',
old: ['container.cpu.utilization'],
kind: 'metric',
contexts: ['metric'],
signals: ['metrics'],
applyToMetrics: [],
valueMap: {},
},
{
current: 'db.system.name',
old: ['db.system'],
kind: 'attribute',
contexts: [],
signals: [],
contexts: ['attribute', 'resource'],
signals: ['logs', 'metrics', 'traces'],
applyToMetrics: [],
valueMap: {},
},
@@ -24,8 +33,26 @@ export const SEMCONV_FAMILIES: readonly SemconvFamily[] = [
current: 'deployment.environment.name',
old: ['deployment.environment'],
kind: 'attribute',
contexts: [],
signals: [],
contexts: ['attribute', 'resource'],
signals: ['logs', 'metrics', 'traces'],
applyToMetrics: [],
valueMap: {},
},
{
current: 'k8s.node.cpu.usage',
old: ['k8s.node.cpu.utilization'],
kind: 'metric',
contexts: ['metric'],
signals: ['metrics'],
applyToMetrics: [],
valueMap: {},
},
{
current: 'k8s.pod.cpu.usage',
old: ['k8s.pod.cpu.utilization'],
kind: 'metric',
contexts: ['metric'],
signals: ['metrics'],
applyToMetrics: [],
valueMap: {},
},

View File

@@ -155,7 +155,7 @@ function DomainList(): JSX.Element {
dataSource={DataSource.TRACES}
queryData={query}
onChange={handleSearchChange}
placeholder="Enter your filter query (e.g., deployment.environment = 'otel-demo' AND service.name = 'frontend')"
placeholder="Enter your filter query (e.g., deployment.environment.name = 'otel-demo' AND service.name = 'frontend')"
hardcodedAttributeKeys={ApiMonitoringHardcodedAttributeKeys}
/>
</div>

View File

@@ -6,9 +6,9 @@ import { SPAN_ATTRIBUTES } from './Explorer/Domains/DomainDetails/constants';
export const ApiMonitoringHardcodedAttributeKeys: QueryKeyDataSuggestionsProps[] =
[
{
label: 'deployment.environment',
label: 'deployment.environment.name',
type: 'resource',
name: 'deployment.environment',
name: 'deployment.environment.name',
signal: 'traces',
fieldDataType: QUERY_BUILDER_KEY_TYPES.STRING,
},

View File

@@ -87,7 +87,7 @@ export const ApiMonitoringQuickFiltersConfig: IQuickFiltersConfig[] = [
title: 'Environment',
attributeKey: {
key: 'deployment.environment',
key: 'deployment.environment.name',
dataType: DataTypes.String,
type: 'resource',
},

View File

@@ -112,7 +112,7 @@ export const INFRA_MONITORING_ATTR_KEYS = {
K8S_OBJECT_NAME: 'k8s.object.name',
// Environment
DEPLOYMENT_ENVIRONMENT: 'deployment.environment',
DEPLOYMENT_ENVIRONMENT: 'deployment.environment.name',
// Host System
OS_TYPE: 'os.type',
@@ -733,7 +733,7 @@ export const ENTITY_FILTER_PLACEHOLDERS: Record<InfraMonitoringEntity, string> =
[InfraMonitoringEntity.NAMESPACES]:
"Enter your filter query (e.g., k8s.namespace.name = 'production' AND k8s.cluster.name = 'prod-cluster')",
[InfraMonitoringEntity.CLUSTERS]:
"Enter your filter query (e.g., k8s.cluster.name = 'prod-cluster' AND deployment.environment = 'production')",
"Enter your filter query (e.g., k8s.cluster.name = 'prod-cluster' AND deployment.environment.name = 'production')",
[InfraMonitoringEntity.DEPLOYMENTS]:
"Enter your filter query (e.g., k8s.deployment.name = 'api-server' AND k8s.namespace.name = 'production')",
[InfraMonitoringEntity.STATEFULSETS]:

View File

@@ -2,6 +2,17 @@
color: white;
}
.semconv-migration-report {
margin-top: 32px;
display: flex;
flex-direction: column;
gap: 12px;
.ant-table-wrapper {
margin-top: 4px;
}
}
.ingestion-key-container {
margin-top: 24px;
display: flex;

View File

@@ -5,6 +5,8 @@ import getIngestionData from 'api/settings/getIngestionData';
import { useAppContext } from 'providers/App/App';
import { IngestionDataType } from 'types/api/settings/ingestion';
import SemconvMigrationReport from './SemconvMigrationReport';
import './IngestionSettings.styles.scss';
export default function IngestionSettings(): JSX.Element {
@@ -84,6 +86,7 @@ export default function IngestionSettings(): JSX.Element {
dataSource={data}
bordered
/>
<SemconvMigrationReport />
</div>
);
}

View File

@@ -83,6 +83,8 @@ import { MeterAggregateOperator } from 'types/common/queryBuilder';
import { USER_ROLES } from 'types/roles';
import { getDaysUntilExpiry } from 'utils/timeUtils';
import SemconvMigrationReport from './SemconvMigrationReport';
import './IngestionSettings.styles.scss';
const { Option } = Select;
@@ -1705,6 +1707,7 @@ function MultiIngestionSettings(): JSX.Element {
}}
className="ingestion-keys-table"
/>
<SemconvMigrationReport />
</div>
{/* Delete Key Modal */}

View File

@@ -0,0 +1,72 @@
import { useQuery } from 'react-query';
import { Alert, Table, TableColumnsType } from 'antd';
import { Typography } from '@signozhq/ui/typography';
import getSemconvMigrationReport from 'api/semconv/getMigrationReport';
import dayjs from 'dayjs';
import { SemconvMigrationReportEntry } from 'types/api/semconvMigration';
function SemconvMigrationReport(): JSX.Element {
const { data, isLoading, isError } = useQuery({
queryKey: ['semconv-migration-report'],
queryFn: getSemconvMigrationReport,
});
const columns: TableColumnsType<SemconvMigrationReportEntry> = [
{
title: 'Old name',
dataIndex: 'old',
key: 'old',
},
{
title: 'Current name',
dataIndex: 'current',
key: 'current',
},
{
title: 'Signal',
dataIndex: 'signal',
key: 'signal',
},
{
title: 'Services still sending only the old name',
dataIndex: 'services',
key: 'services',
render: (services: string[]): string => services.join(', '),
},
{
title: 'Last seen',
dataIndex: 'lastSeenUnixMilli',
key: 'lastSeenUnixMilli',
render: (value: number): string =>
dayjs(value).format('YYYY-MM-DD HH:mm:ss'),
},
];
return (
<section className="semconv-migration-report">
<Typography.Title level={4}>Semantic convention migration</Typography.Title>
<Typography.Text>
Services in this report sent an old OpenTelemetry field during the last 24
hours without sending its current replacement. Update their SDK or
instrumentation when practical; SigNoz queries remain backward compatible.
</Typography.Text>
{isError && (
<Alert
type="error"
showIcon
message="Could not load the semantic convention migration report"
/>
)}
<Table
loading={isLoading}
columns={columns}
dataSource={data?.entries ?? []}
rowKey={(entry): string => `${entry.current}-${entry.old}-${entry.signal}`}
pagination={false}
locale={{ emptyText: 'No old-only services found in the last 24 hours' }}
/>
</section>
);
}
export default SemconvMigrationReport;

View File

@@ -15,7 +15,7 @@ export const SAMPLE_SPAN_JSON = `{
},
"resource": {
"service.name": "llm-gateway",
"deployment.environment": "production"
"deployment.environment.name": "production"
}
}`;

View File

@@ -1120,7 +1120,7 @@
"plugin": {
"kind": "signoz/QueryVariable",
"spec": {
"queryValue": "SELECT DISTINCT resources_string['deployment.environment'] AS environment FROM signoz_traces.distributed_signoz_index_v3 WHERE mapContains(resources_string, 'deployment.environment') AND timestamp >= now() - INTERVAL 1 DAY"
"queryValue": "SELECT DISTINCT resources_string['deployment.environment.name'] AS environment FROM signoz_traces.distributed_signoz_index_v3 WHERE mapContains(resources_string, 'deployment.environment.name') AND timestamp >= now() - INTERVAL 1 DAY"
}
}
}

View File

@@ -1,6 +1,7 @@
import { Divider } from '@signozhq/ui/divider';
import { TooltipSimple } from '@signozhq/ui/tooltip';
import { Typography } from '@signozhq/ui/typography';
import { SemconvOldNameBadge } from 'components/Semconv';
import { TagContainer, TagLabel, TagValue } from './FieldRenderer.styles';
import { FieldRendererProps } from './LogDetailedView.types';
@@ -28,6 +29,7 @@ function FieldRenderer({ field }: FieldRendererProps): JSX.Element {
<Typography.Text truncate={1} className="label">
{newField}{' '}
</Typography.Text>
<SemconvOldNameBadge name={newField} />
</TooltipSimple>
<div className="tags">
@@ -47,7 +49,10 @@ function FieldRenderer({ field }: FieldRendererProps): JSX.Element {
</div>
</>
) : (
<span className="label">{field}</span>
<>
<span className="label">{field}</span>
<SemconvOldNameBadge name={field} />
</>
)}
</span>
);

View File

@@ -164,7 +164,9 @@ describe('useInitialQuery - Priority-Based Resource Filtering', () => {
value: 'frontend-service',
}),
expect.objectContaining({
key: expect.objectContaining({ key: 'deployment.environment' }),
key: expect.objectContaining({
key: 'deployment.environment.name',
}),
value: 'production',
}),
expect.objectContaining({
@@ -286,7 +288,9 @@ describe('useInitialQuery - Priority-Based Resource Filtering', () => {
value: 'legacy-app',
}),
expect.objectContaining({
key: expect.objectContaining({ key: 'deployment.environment' }),
key: expect.objectContaining({
key: 'deployment.environment.name',
}),
value: 'production',
}),
expect.objectContaining({

View File

@@ -6,13 +6,14 @@ import {
TagFilterItem,
} from 'types/api/queryBuilder/queryBuilderData';
import { v4 as uuid } from 'uuid';
import { getSemconvRename } from 'utils/semconv';
const FALLBACK_STARTS_WITH_REGEX = /^(k8s|cloud|host|deployment)/; // regex to filter out resources that start with the specified keywords
const FALLBACK_CONTAINS_REGEX = /(env|service|file|container|tenant)/; // regex to filter out resources that contains the specified keywords
// Priority categories for filter selection
// Strategy:
// - Always include: service.name, deployment.environment, env, environment
// - Always include: service.name, deployment.environment.name, env, environment
// - Select ONE category only: stops at the first category with a matching attribute
// - Within category: picks the first available attribute by order
// - Order (highest to lowest priority): Kubernetes > Cloud > Host > Container
@@ -26,27 +27,36 @@ const PRIORITY_CATEGORIES = [
const SERVICE_AND_ENVIRONMENT_KEYS = [
'service.name',
'deployment.environment',
'deployment.environment.name',
'env',
'environment',
];
export const getFiltersFromResources = (
resources: ILog['resources_string'],
): TagFilterItem[] =>
Object.keys(resources).map((key: string) => {
): TagFilterItem[] => {
const items = new Map<string, TagFilterItem>();
Object.keys(resources).forEach((key: string) => {
const currentKey = getSemconvRename(key)?.current ?? key;
const resourceValue = resources[key] as string;
return {
const item = {
id: uuid(),
key: {
key,
key: currentKey,
dataType: DataTypes.String,
type: 'resource',
},
op: OPERATORS['='],
value: resourceValue,
};
// If raw data contains both names, retain the current value just like the
// backend's current-first resolver.
if (!items.has(currentKey) || key === currentKey) {
items.set(currentKey, item);
}
});
return Array.from(items.values());
};
export const isServiceOrEnvironmentAttribute = (key: string): boolean =>
SERVICE_AND_ENVIRONMENT_KEYS.includes(key);

View File

@@ -94,7 +94,7 @@ function DBCall(): JSX.Element {
featureFlags?.find((flag) => flag.name === FeatureKeys.DOT_METRICS_ENABLED)
?.active || false;
const legend = dotMetricsEnabled ? '{{db.system}}' : '{{db_system}}';
const legend = dotMetricsEnabled ? '{{db.system.name}}' : '{{db_system_name}}';
const databaseCallsRPSWidget = useMemo(
() =>

View File

@@ -28,7 +28,7 @@ import { v4 as uuid } from 'uuid';
export const dbSystemTags: Tags[] = [
{
Key: 'db.system.(string)',
Key: 'db.system.name.(string)',
StringValues: [''],
NumberValues: [],
BoolValues: [],

View File

@@ -103,7 +103,7 @@ export enum WidgetKeys {
SignozExternalCallLatencySum = 'signoz_external_call_latency_sum',
Signoz_latency_bucket_norm = 'signoz_latency_bucket',
Signoz_latency_bucket = 'signoz_latency.bucket',
Db_system = 'db.system',
Db_system = 'db.system.name',
Db_system_norm = 'db_system',
}

View File

@@ -2,6 +2,7 @@ import { ChangeEvent, useCallback } from 'react';
import MEditor, { Monaco } from '@monaco-editor/react';
import { Color } from '@signozhq/design-tokens';
import { Input } from 'antd';
import { SemconvEditorWarning } from 'components/Semconv';
import { LEGEND } from 'constants/global';
import { useQueryBuilder } from 'hooks/queryBuilder/useQueryBuilder';
import { useIsDarkMode } from 'hooks/useDarkMode';
@@ -118,6 +119,7 @@ function ClickHouseQueryBuilder({
theme={isDarkMode ? 'my-theme' : 'light'}
beforeMount={setEditorTheme}
/>
<SemconvEditorWarning value={queryData?.query} editor="ClickHouse SQL" />
<Input
onChange={handleUpdateInput}
name="legend"

View File

@@ -1,6 +1,7 @@
import { ChangeEvent, useCallback } from 'react';
import { Input } from 'antd';
import { LEGEND } from 'constants/global';
import { SemconvEditorWarning } from 'components/Semconv';
import { useQueryBuilder } from 'hooks/queryBuilder/useQueryBuilder';
import { IPromQLQuery } from 'types/api/queryBuilder/queryBuilderData';
import { EQueryType } from 'types/common/dashboard';
@@ -66,6 +67,7 @@ function PromQLQueryBuilder({
style={{ marginBottom: '0.5rem' }}
data-testid="promql-query-input"
/>
<SemconvEditorWarning value={queryData?.query} editor="PromQL" />
<Input
onChange={handleUpdateQuery}

View File

@@ -1,6 +1,7 @@
import { useTranslation } from 'react-i18next';
import { Form } from 'antd';
import { initialQueryBuilderFormValuesMap } from 'constants/queryBuilder';
import { SemconvEditorWarning } from 'components/Semconv';
import QueryBuilderSearchV2 from 'container/QueryBuilder/filters/QueryBuilderSearchV2/QueryBuilderSearchV2';
import isEqual from 'lodash-es/isEqual';
import { TagFilter } from 'types/api/queryBuilder/queryBuilderData';
@@ -55,6 +56,7 @@ function TagFilterInputWithLogsResultPreview({
value={value}
onChange={onChange}
/>
<SemconvEditorWarning value={value} editor="Pipeline filter" />
<div className="pipeline-filter-input-preview-container">
<LogsFilterPreview filter={value} />
</div>

View File

@@ -424,7 +424,7 @@ describe('ResourceProvider', () => {
await waitFor(() => {
expect(result.current.queries).toHaveLength(1);
expect(result.current.queries[0]).toMatchObject({
tagKey: 'resource_deployment_environment',
tagKey: 'resource_deployment_environment_name',
operator: 'IN',
tagValue: ['production'],
});
@@ -435,7 +435,7 @@ describe('ResourceProvider', () => {
const seeded = [
{
id: 'env',
tagKey: 'resource_deployment_environment',
tagKey: 'resource_deployment_environment_name',
operator: 'IN',
tagValue: ['production'],
},
@@ -459,7 +459,7 @@ describe('ResourceProvider', () => {
await waitFor(() => {
const tagKeys = result.current.queries.map((q) => q.tagKey);
expect(tagKeys).not.toContain('resource_deployment_environment');
expect(tagKeys).not.toContain('resource_deployment_environment_name');
expect(tagKeys).toContain('resource_service_name');
});
});
@@ -468,7 +468,7 @@ describe('ResourceProvider', () => {
const seeded = [
{
id: 'env',
tagKey: 'resource_deployment_environment',
tagKey: 'resource_deployment_environment_name',
operator: 'IN',
tagValue: ['production'],
},
@@ -486,7 +486,7 @@ describe('ResourceProvider', () => {
await waitFor(() => {
const envQueries = result.current.queries.filter(
(q) => q.tagKey === 'resource_deployment_environment',
(q) => q.tagKey === 'resource_deployment_environment_name',
);
expect(envQueries).toHaveLength(1);
expect(envQueries[0].tagValue).toStrictEqual(['staging']);
@@ -518,7 +518,7 @@ describe('ResourceProvider', () => {
await waitFor(() => {
expect(result.current.queries[0].tagKey).toBe(
'resource_deployment.environment',
'resource_deployment.environment.name',
);
});
});

View File

@@ -6,13 +6,13 @@ import { mappingWithRoutesAndKeys } from '../utils';
describe('useResourceAttribute config', () => {
describe('whilelistedKeys', () => {
it('should include underscore-notation keys (DOT_METRICS_ENABLED=false)', () => {
expect(whilelistedKeys).toContain('resource_deployment_environment');
expect(whilelistedKeys).toContain('resource_deployment_environment_name');
expect(whilelistedKeys).toContain('resource_k8s_cluster_name');
expect(whilelistedKeys).toContain('resource_k8s_cluster_namespace');
});
it('should include dot-notation keys (DOT_METRICS_ENABLED=true)', () => {
expect(whilelistedKeys).toContain('resource_deployment.environment');
expect(whilelistedKeys).toContain('resource_deployment.environment.name');
expect(whilelistedKeys).toContain('resource_k8s.cluster.name');
expect(whilelistedKeys).toContain('resource_k8s.cluster.namespace');
});
@@ -21,8 +21,8 @@ describe('useResourceAttribute config', () => {
describe('mappingWithRoutesAndKeys', () => {
const dotNotationFilters = [
{
label: 'deployment.environment',
value: 'resource_deployment.environment',
label: 'deployment.environment.name',
value: 'resource_deployment.environment.name',
},
{ label: 'k8s.cluster.name', value: 'resource_k8s.cluster.name' },
{ label: 'k8s.cluster.namespace', value: 'resource_k8s.cluster.namespace' },
@@ -30,8 +30,8 @@ describe('useResourceAttribute config', () => {
const underscoreNotationFilters = [
{
label: 'deployment.environment',
value: 'resource_deployment_environment',
label: 'deployment.environment.name',
value: 'resource_deployment_environment_name',
},
{ label: 'k8s.cluster.name', value: 'resource_k8s_cluster_name' },
{ label: 'k8s.cluster.namespace', value: 'resource_k8s_cluster_namespace' },

View File

@@ -1,6 +1,6 @@
export const whilelistedKeys = [
'resource_deployment_environment',
'resource_deployment.environment',
'resource_deployment_environment_name',
'resource_deployment.environment.name',
'resource_k8s_cluster_name',
'resource_k8s.cluster.name',
'resource_k8s_cluster_namespace',

View File

@@ -148,9 +148,9 @@ export const getResourceDeploymentKeys = (
dotMetricsEnabled: boolean,
): string => {
if (dotMetricsEnabled) {
return 'resource_deployment.environment';
return 'resource_deployment.environment.name';
}
return 'resource_deployment_environment';
return 'resource_deployment_environment_name';
};
export const GetTagKeys = async (

View File

@@ -40,7 +40,7 @@ export const LogsQuickFiltersConfig: IQuickFiltersConfig[] = [
type: FiltersType.CHECKBOX,
title: 'Environment',
attributeKey: {
key: 'deployment.environment',
key: 'deployment.environment.name',
dataType: DataTypes.String,
type: 'resource',
},

View File

@@ -11,6 +11,7 @@ import { Skeleton } from 'antd';
import { DetailsHeader, DetailsPanelDrawer } from 'components/DetailsPanel';
import { HeaderAction } from 'components/DetailsPanel/DetailsHeader/DetailsHeader';
import { DetailsPanelState } from 'components/DetailsPanel/types';
import { SemconvOldNameBadge } from 'components/Semconv';
import { QueryParams } from 'constants/query';
import {
initialQueryBuilderFormValuesMap,
@@ -108,6 +109,12 @@ function SpanDetailsContent({
() => getSpanDisplayData(selectedSpan),
[selectedSpan],
);
const semconvLabelSuffix = useCallback(
(fieldKey: string): React.ReactNode => (
<SemconvOldNameBadge name={fieldKey} />
),
[],
);
// Map span attribute actions to PrettyView actions format.
// Use the last key in fieldKeyPath (the actual attribute key), not the full display path.
@@ -329,6 +336,7 @@ function SpanDetailsContent({
visibleActions: VISIBLE_ACTIONS,
pinnedFieldsValue,
onPinnedFieldsChange,
labelSuffixRenderer: semconvLabelSuffix,
}}
/>
</TabsContent>

View File

@@ -29,7 +29,7 @@ export const KEY_ATTRIBUTE_KEYS: Record<string, string[]> = {
traces: [
'service.name',
'service.namespace',
'deployment.environment',
'deployment.environment.name',
'timestamp',
'duration_nano',
'kind_string',

View File

@@ -22,7 +22,7 @@ export const SPAN_CATEGORIES: readonly SpanCategory[] = [
// Map each category to the attribute key it filters on
const CATEGORY_KEYS: Record<Exclude<SpanCategory, 'All'>, string> = {
Database: 'db.system',
Database: 'db.system.name',
HTTP: 'http.method',
Functions: 'kind_string',
Jobs: 'messaging.system',
@@ -34,7 +34,7 @@ const ALL_CATEGORY_KEYS = Object.values(CATEGORY_KEYS);
// The expression clause to add for each category
const CATEGORY_EXPRESSIONS: Record<Exclude<SpanCategory, 'All'>, string> = {
Database: 'db.system exists',
Database: 'db.system.name exists',
HTTP: 'http.method exists',
Functions: "kind_string = 'Internal'",
Jobs: 'messaging.system exists',

View File

@@ -38,7 +38,7 @@ export function Section(props: SectionProps): JSX.Element {
'hasError',
'durationNano',
'serviceName',
'deployment.environment',
'deployment.environment.name',
]),
),
[selectedFilters],

View File

@@ -14,7 +14,7 @@ export const AllTraceFilterKeyValue: Record<string, string> = {
durationNano: 'Duration',
duration_nano: 'Duration',
durationNanoMax: 'Duration',
'deployment.environment': 'Environment',
'deployment.environment.name': 'Environment',
hasError: 'Status',
has_error: 'Status',
serviceName: 'Service Name',
@@ -208,11 +208,11 @@ export const traceFilterKeys: Record<AllTraceFilterKeys, BaseAutocompleteData> =
id: 'serviceName--string--tag--true',
},
'deployment.environment': {
key: 'deployment.environment',
'deployment.environment.name': {
key: 'deployment.environment.name',
dataType: DataTypes.String,
type: 'resource',
id: 'deployment.environment--string--resource--false',
id: 'deployment.environment.name--string--resource--false',
},
name: {
key: 'name',

View File

@@ -223,6 +223,12 @@
padding-left: 6px !important;
}
&__label {
display: inline-flex;
align-items: baseline;
gap: 6px;
}
&__pinned-icon {
flex-shrink: 0;
color: var(--text-robin-400);

View File

@@ -67,6 +67,7 @@ export interface PrettyViewProps {
*/
pinnedFieldsValue?: string[];
onPinnedFieldsChange?: (next: string[]) => void;
labelSuffixRenderer?: (fieldKey: string) => React.ReactNode;
}
function PrettyView({
@@ -78,6 +79,7 @@ function PrettyView({
drawerKey = 'default',
pinnedFieldsValue,
onPinnedFieldsChange,
labelSuffixRenderer,
}: PrettyViewProps): JSX.Element {
const isDarkMode = useIsDarkMode();
const [, setCopy] = useCopyToClipboard();
@@ -305,10 +307,24 @@ function PrettyView({
}}
/>
<span>{displayKey}</span>
{labelSuffixRenderer?.(displayKey)}
</span>
);
},
[togglePin, pinnedEntries],
[togglePin, pinnedEntries, labelSuffixRenderer],
);
const labelRenderer = useCallback(
(keyPath: KeyPath): React.ReactNode => {
const displayKey = String(keyPath[0]);
return (
<span className="pretty-view__label">
<span>{displayKey}</span>
{labelSuffixRenderer?.(displayKey)}
</span>
);
},
[labelSuffixRenderer],
);
return (
@@ -351,6 +367,7 @@ function PrettyView({
shouldExpandNodeInitially={shouldExpandNodeInitially}
valueRenderer={valueRenderer}
getItemString={getItemString}
labelRenderer={labelRenderer}
/>
</div>
);

View File

@@ -0,0 +1,14 @@
export interface SemconvMigrationReportEntry {
current: string;
old: string;
signal: string;
services: string[];
resourceSets: number;
lastSeenUnixMilli: number;
}
export interface SemconvMigrationReport {
startUnixMilli: number;
endUnixMilli: number;
entries: SemconvMigrationReportEntry[];
}

View File

@@ -0,0 +1,29 @@
import { findOldSemconvNames, getSemconvRename } from 'utils/semconv';
describe('semantic convention helpers', () => {
it('returns the current name for an old attribute', () => {
expect(getSemconvRename('deployment.environment')).toMatchObject({
old: 'deployment.environment',
current: 'deployment.environment.name',
});
});
it('finds old names in editor text without matching larger custom names', () => {
expect(
findOldSemconvNames(
"deployment.environment = 'prod' AND custom.db.system.value = 'x'",
),
).toStrictEqual([
expect.objectContaining({
old: 'deployment.environment',
current: 'deployment.environment.name',
}),
]);
});
it('does not warn for current names', () => {
expect(
findOldSemconvNames('deployment.environment.name = prod'),
).toStrictEqual([]);
});
});

View File

@@ -0,0 +1,50 @@
import {
SEMCONV_FAMILIES,
SemconvFamily,
} from 'constants/generated/semconvFamilies.gen';
export type SemconvRename = {
old: string;
current: string;
family: SemconvFamily;
};
const OLD_NAMES = SEMCONV_FAMILIES.flatMap((family) =>
family.old.map((old) => ({ old, current: family.current, family })),
);
const OLD_NAME_INDEX = new Map(OLD_NAMES.map((rename) => [rename.old, rename]));
export function getSemconvRename(name: string): SemconvRename | undefined {
return OLD_NAME_INDEX.get(name);
}
export function findOldSemconvNames(text: string): SemconvRename[] {
if (!text) {
return [];
}
return OLD_NAMES.filter(({ old }) => containsSemconvName(text, old));
}
function containsSemconvName(text: string, name: string): boolean {
let offset = 0;
while (offset < text.length) {
const index = text.indexOf(name, offset);
if (index === -1) {
return false;
}
const before = index === 0 ? '' : text[index - 1];
const afterIndex = index + name.length;
const after = afterIndex === text.length ? '' : text[afterIndex];
if (!isSemconvNameCharacter(before) && !isSemconvNameCharacter(after)) {
return true;
}
offset = index + 1;
}
return false;
}
function isSemconvNameCharacter(value: string): boolean {
return /[A-Za-z0-9_.-]/.test(value);
}

View File

@@ -46,5 +46,23 @@ func (provider *provider) addFieldsRoutes(router *mux.Router) error {
return err
}
if err := router.Handle("/api/v1/fields/semconv-migration", handler.New(provider.authzMiddleware.ViewAccess(provider.fieldsHandler.GetSemconvMigrationReport), handler.OpenAPIDef{
ID: "GetSemconvMigrationReport",
Tags: []string{"fields"},
Summary: "Get semantic-convention migration report",
Description: "Returns services that still emit old semantic-convention names without the current family name",
Request: nil,
RequestQuery: new(telemetrytypes.PostableSemconvMigrationReportParams),
RequestContentType: "",
Response: new(telemetrytypes.GettableSemconvMigrationReport),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{},
Deprecated: false,
SecuritySchemes: newSecuritySchemes(types.RoleViewer),
})).Methods(http.MethodGet).GetError(); err != nil {
return err
}
return nil
}

View File

@@ -8,4 +8,7 @@ type Handler interface {
// Gets the fields values for the given field value selector
GetFieldsValues(http.ResponseWriter, *http.Request)
// Gets services that still emit only historical semantic-convention names.
GetSemconvMigrationReport(http.ResponseWriter, *http.Request)
}

View File

@@ -2,6 +2,7 @@ package implfields
import (
"net/http"
"time"
"github.com/SigNoz/signoz/pkg/factory"
"github.com/SigNoz/signoz/pkg/http/binding"
@@ -16,6 +17,43 @@ type handler struct {
telemetryMetadataStore telemetrytypes.MetadataStore
}
func (handler *handler) GetSemconvMigrationReport(rw http.ResponseWriter, req *http.Request) {
ctx := req.Context()
var params telemetrytypes.PostableSemconvMigrationReportParams
if err := binding.Query.BindQuery(req.URL.Query(), &params); err != nil {
render.Error(rw, err)
return
}
now := time.Now()
if params.EndUnixMilli == 0 {
params.EndUnixMilli = now.UnixMilli()
}
if params.StartUnixMilli == 0 {
params.StartUnixMilli = now.Add(-24 * time.Hour).UnixMilli()
}
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
render.Error(rw, err)
return
}
report, err := handler.telemetryMetadataStore.GetSemconvMigrationReport(
ctx,
valuer.MustNewUUID(claims.OrgID),
params.StartUnixMilli,
params.EndUnixMilli,
)
if err != nil {
render.Error(rw, err)
return
}
render.Success(rw, http.StatusOK, report)
}
func NewHandler(settings factory.ProviderSettings, telemetryMetadataStore telemetrytypes.MetadataStore) fields.Handler {
return &handler{
telemetryMetadataStore: telemetryMetadataStore,

View File

@@ -786,10 +786,11 @@ func (q *querier) run(
Results: maps.Values(processedResults),
},
Meta: qbtypes.ExecStats{
RowsScanned: stats.RowsScanned,
BytesScanned: stats.BytesScanned,
DurationMS: stats.DurationMS,
StepIntervals: stepIntervals,
RowsScanned: stats.RowsScanned,
BytesScanned: stats.BytesScanned,
DurationMS: stats.DurationMS,
StepIntervals: stepIntervals,
SemconvResolutions: semconvResolutionsForRequest(req),
},
}

View File

@@ -0,0 +1,128 @@
package querier
import (
"encoding/json"
"slices"
"strings"
"unicode"
"github.com/SigNoz/signoz/pkg/semconv"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
)
// semconvResolutionsForRequest reports only query-builder resolutions. Raw
// ClickHouse SQL and PromQL are deliberately excluded: SigNoz does not rewrite
// those languages and their editors surface non-blocking warnings instead.
func semconvResolutionsForRequest(req *qbtypes.QueryRangeRequest) []qbtypes.SemconvResolution {
if req == nil {
return nil
}
resolutions := make([]qbtypes.SemconvResolution, 0)
seen := make(map[string]struct{})
for _, envelope := range req.CompositeQuery.Queries {
signal, applies := semconvResolutionSignal(envelope.Spec)
if !applies {
continue
}
payload, err := json.Marshal(envelope.Spec)
if err != nil {
continue
}
text := string(payload)
for _, family := range semconv.All() {
if _, ok := semconv.Lookup(family.Kind, telemetrytypes.FieldKeySelector{
Name: family.Current,
Signal: signal,
}); !ok {
continue
}
for _, requested := range semconvRequestSpellings(family, signal) {
if !containsSemconvName(text, requested) {
continue
}
identity := family.Kind.StringValue() + "\x00" + requested + "\x00" + family.Current
if _, ok := seen[identity]; ok {
continue
}
seen[identity] = struct{}{}
resolutions = append(resolutions, qbtypes.SemconvResolution{
Requested: requested,
Current: family.Current,
Members: append([]string{family.Current}, family.Old...),
Kind: family.Kind.StringValue(),
})
}
}
}
if len(resolutions) == 0 {
return nil
}
return resolutions
}
func semconvResolutionSignal(spec any) (telemetrytypes.Signal, bool) {
switch spec.(type) {
case qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation], qbtypes.QueryBuilderTraceOperator:
return telemetrytypes.SignalTraces, true
case qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]:
return telemetrytypes.SignalLogs, true
case qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]:
return telemetrytypes.SignalMetrics, true
case qbtypes.QueryBuilderJoin:
// Joins can contain more than one signal. An unspecified signal keeps
// family signal scopes in force while allowing every applicable family.
return telemetrytypes.SignalUnspecified, true
default:
return telemetrytypes.SignalUnspecified, false
}
}
func semconvRequestSpellings(family semconv.Family, signal telemetrytypes.Signal) []string {
spellings := append([]string{family.Current}, family.Old...)
if signal != telemetrytypes.SignalMetrics {
return spellings
}
logical := slices.Clone(spellings)
for _, name := range logical {
normalized := strings.ReplaceAll(name, ".", "_")
if !slices.Contains(spellings, normalized) {
spellings = append(spellings, normalized)
}
if family.Kind == semconv.KindAttribute {
for _, resourceName := range []string{"resource_" + name, "resource_" + normalized} {
if !slices.Contains(spellings, resourceName) {
spellings = append(spellings, resourceName)
}
}
}
}
return spellings
}
func containsSemconvName(text, name string) bool {
for offset := 0; offset < len(text); {
index := strings.Index(text[offset:], name)
if index < 0 {
return false
}
start := offset + index
end := start + len(name)
if (start == 0 || !isSemconvNameRune(rune(text[start-1]))) &&
(end == len(text) || !isSemconvNameRune(rune(text[end]))) {
return true
}
offset = start + 1
}
return false
}
func isSemconvNameRune(r rune) bool {
return unicode.IsLetter(r) || unicode.IsDigit(r) || r == '_' || r == '.' || r == '-'
}

View File

@@ -0,0 +1,70 @@
package querier
import (
"testing"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/stretchr/testify/assert"
)
func TestSemconvResolutionsReportsOldTraceAttribute(t *testing.T) {
req := &qbtypes.QueryRangeRequest{
CompositeQuery: qbtypes.CompositeQuery{Queries: []qbtypes.QueryEnvelope{{
Spec: qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]{
Signal: telemetrytypes.SignalTraces,
Filter: &qbtypes.Filter{Expression: "deployment.environment = 'prod'"},
},
}}},
}
assert.Equal(t, []qbtypes.SemconvResolution{{
Requested: "deployment.environment",
Current: "deployment.environment.name",
Members: []string{"deployment.environment.name", "deployment.environment"},
Kind: "attribute",
}}, semconvResolutionsForRequest(req), "old trace attribute should be reported as a family resolution")
}
func TestSemconvResolutionsReportsCurrentLogAttribute(t *testing.T) {
req := &qbtypes.QueryRangeRequest{
CompositeQuery: qbtypes.CompositeQuery{Queries: []qbtypes.QueryEnvelope{{
Spec: qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]{
Signal: telemetrytypes.SignalLogs,
SelectFields: []telemetrytypes.TelemetryFieldKey{{
Name: "db.system.name",
}},
},
}}},
}
assert.Equal(t, []qbtypes.SemconvResolution{{
Requested: "db.system.name",
Current: "db.system.name",
Members: []string{"db.system.name", "db.system"},
Kind: "attribute",
}}, semconvResolutionsForRequest(req), "current log attribute should identify its complete family")
}
func TestSemconvResolutionsIgnoresRawSQL(t *testing.T) {
req := &qbtypes.QueryRangeRequest{
CompositeQuery: qbtypes.CompositeQuery{Queries: []qbtypes.QueryEnvelope{{
Spec: qbtypes.ClickHouseQuery{Query: "SELECT attributes_string['deployment.environment']"},
}}},
}
assert.Empty(t, semconvResolutionsForRequest(req), "raw SQL is not rewritten and should not report a resolution")
}
func TestSemconvResolutionsRequiresNameBoundary(t *testing.T) {
req := &qbtypes.QueryRangeRequest{
CompositeQuery: qbtypes.CompositeQuery{Queries: []qbtypes.QueryEnvelope{{
Spec: qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]{
Signal: telemetrytypes.SignalTraces,
Filter: &qbtypes.Filter{Expression: "custom.deployment.environment = 'prod'"},
},
}}},
}
assert.Empty(t, semconvResolutionsForRequest(req), "a family name embedded in a larger custom key must not match")
}

View File

@@ -52,10 +52,10 @@ import (
"github.com/SigNoz/signoz/pkg/query-service/constants"
chErrors "github.com/SigNoz/signoz/pkg/query-service/errors"
"github.com/SigNoz/signoz/pkg/query-service/metrics"
"github.com/SigNoz/signoz/pkg/query-service/model"
v3 "github.com/SigNoz/signoz/pkg/query-service/model/v3"
"github.com/SigNoz/signoz/pkg/query-service/utils"
"github.com/SigNoz/signoz/pkg/semconv"
)
const (
@@ -3190,11 +3190,6 @@ func (r *ClickHouseReader) GetMetricAttributeValues(ctx context.Context, orgID v
var rows driver.Rows
var attributeValues v3.FilterAttributeValueResponse
normalized := true
if constants.IsDotMetricsEnabled {
normalized = false
}
reductionEnabled := r.fl.BooleanOrEmpty(ctx, flagger.FeatureEnableMetricsReduction, featuretypes.NewFlaggerEvaluationContext(orgID))
if reductionEnabled {
@@ -3205,8 +3200,7 @@ func (r *ClickHouseReader) GetMetricAttributeValues(ctx context.Context, orgID v
if req.Limit != 0 {
query = query + fmt.Sprintf(" LIMIT %d;", req.Limit)
}
names := []string{req.AggregateAttribute}
names = append(names, metrics.GetTransitionedMetric(req.AggregateAttribute, normalized))
names := semconv.MetricNames(req.AggregateAttribute)
rows, err = r.db.Query(ctx, query, req.FilterAttributeKey, names, req.FilterAttributeKey, fmt.Sprintf("%%%s%%", req.SearchText), common.PastDayRoundOff())

View File

@@ -1,27 +0,0 @@
package metrics
var MetricsUnderTransition = map[string]string{
"k8s_pod_cpu_utilization": "k8s_pod_cpu_usage",
"k8s_node_cpu_utilization": "k8s_node_cpu_usage",
"container_cpu_utilization": "container_cpu_usage",
}
var DotMetricsUnderTransition = map[string]string{
"k8s.pod.cpu.utilization": "k8s.pod.cpu.usage",
"k8s.node.cpu.utilization": "k8s.node.cpu.usage",
"container.cpu.utilization": "container.cpu.usage",
}
func GetTransitionedMetric(metric string, normalized bool) string {
if normalized {
if _, ok := MetricsUnderTransition[metric]; ok {
return MetricsUnderTransition[metric]
}
return metric
} else {
if _, ok := DotMetricsUnderTransition[metric]; ok {
return DotMetricsUnderTransition[metric]
}
return metric
}
}

View File

@@ -10,8 +10,8 @@ import (
"log/slog"
"github.com/SigNoz/signoz/pkg/query-service/constants"
"github.com/SigNoz/signoz/pkg/query-service/metrics"
v3 "github.com/SigNoz/signoz/pkg/query-service/model/v3"
"github.com/SigNoz/signoz/pkg/semconv"
)
// ValidateAndCastValue validates and casts the value of a key to the corresponding data type of the key
@@ -234,12 +234,12 @@ func ClickHouseFormattedValue(v interface{}) string {
func ClickHouseFormattedMetricNames(v interface{}) string {
if name, ok := v.(string); ok {
transitionedMetrics := metrics.GetTransitionedMetric(name, !constants.IsDotMetricsEnabled)
if transitionedMetrics != name {
return ClickHouseFormattedValue([]interface{}{transitionedMetrics})
} else {
return ClickHouseFormattedValue([]interface{}{name})
members := semconv.MetricNames(name)
values := make([]interface{}, 0, len(members))
for _, member := range members {
values = append(values, member)
}
return ClickHouseFormattedValue(values)
}
return ClickHouseFormattedValue(v)

View File

@@ -2,13 +2,29 @@ package querybuilder
import (
"fmt"
"strings"
schema "github.com/SigNoz/signoz-otel-collector/cmd/signozschemamigrator/schema_migrator"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/semconv"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
)
func physicalSemconvMembers(key *telemetrytypes.TelemetryFieldKey) []string {
if len(key.SemconvMembers) > 0 {
return key.SemconvMembers
}
if key.FieldContext != telemetrytypes.FieldContextResource && key.FieldContext != telemetrytypes.FieldContextAttribute {
return []string{key.Name}
}
return semconv.AttributeMembers(telemetrytypes.FieldKeySelector{
Name: key.Name,
Signal: key.Signal,
FieldContext: key.FieldContext,
})
}
// ExistsExpression renders the existence predicate for a key resolved to the given
// columns (negated when exists is false). Comparisons are against constants rendered
// as literals, so the expression carries no bind args and can guard column expressions
@@ -43,11 +59,26 @@ func ExistsExpression(columns []*schema.Column, key *telemetrytypes.TelemetryFie
if len(evolutionsEntries) > 0 && evolutionsEntries[0] != nil {
columnName = evolutionsEntries[0].ColumnName
}
rawPath := fmt.Sprintf("%s.`%s`", columnName, key.Name)
if exists {
return rawPath + " IS NOT NULL", nil
members := physicalSemconvMembers(key)
paths := make([]string, 0, len(members))
for _, member := range members {
paths = append(paths, fmt.Sprintf("%s.`%s`", columnName, member))
}
return rawPath + " IS NULL", nil
if len(paths) == 1 {
if exists {
return paths[0] + " IS NOT NULL", nil
}
return paths[0] + " IS NULL", nil
}
guards := make([]string, 0, len(paths))
for _, path := range paths {
guards = append(guards, path+" IS NOT NULL")
}
rawPath := "(" + strings.Join(guards, " OR ") + ")"
if exists {
return rawPath, nil
}
return "NOT " + rawPath, nil
case schema.ColumnTypeEnumString,
schema.ColumnTypeEnumFixedString:
if exists {
@@ -88,8 +119,16 @@ func ExistsExpression(columns []*schema.Column, key *telemetrytypes.TelemetryFie
switch valueType := column.Type.(schema.MapColumnType).ValueType; valueType.GetType() {
case schema.ColumnTypeEnumString, schema.ColumnTypeEnumBool, schema.ColumnTypeEnumFloat64:
leftOperand := fmt.Sprintf("mapContains(%s, '%s')", column.Name, key.Name)
if key.Materialized {
members := physicalSemconvMembers(key)
operands := make([]string, 0, len(members))
for _, member := range members {
operands = append(operands, fmt.Sprintf("mapContains(%s, '%s')", column.Name, member))
}
leftOperand := strings.Join(operands, " OR ")
if len(operands) > 1 {
leftOperand = "(" + leftOperand + ")"
}
if key.Materialized && (len(members) == 1 || key.MaterializedSemconv) {
leftOperand = telemetrytypes.FieldKeyToMaterializedColumnNameForExists(key)
}
if exists {

View File

@@ -39,6 +39,7 @@ type filterExpressionVisitor struct {
fullTextColumn *telemetrytypes.TelemetryFieldKey
skipResourceFilter bool
skipFullTextFilter bool
exactSemconv bool
variables map[string]qbtypes.VariableItem
keysWithWarnings map[string]bool
@@ -59,6 +60,7 @@ type FilterExprVisitorOpts struct {
FullTextColumn *telemetrytypes.TelemetryFieldKey
SkipResourceFilter bool
SkipFullTextFilter bool
ExactSemconv bool
Variables map[string]qbtypes.VariableItem
StartNs uint64
EndNs uint64
@@ -76,6 +78,7 @@ func newFilterExpressionVisitor(opts FilterExprVisitorOpts) *filterExpressionVis
fullTextColumn: opts.FullTextColumn,
skipResourceFilter: opts.SkipResourceFilter,
skipFullTextFilter: opts.SkipFullTextFilter,
exactSemconv: opts.ExactSemconv,
variables: opts.Variables,
keysWithWarnings: make(map[string]bool),
startNs: opts.StartNs,
@@ -380,7 +383,7 @@ func (v *filterExpressionVisitor) VisitPrimary(ctx *grammar.PrimaryContext) any
// VisitComparison handles all comparison operators.
func (v *filterExpressionVisitor) VisitComparison(ctx *grammar.ComparisonContext) any {
key := v.Visit(ctx.Key()).(*telemetrytypes.TelemetryFieldKey)
matching := MatchingFieldKeys(key, v.fieldKeys)
matching := v.matchingFieldKeys(key)
// Handle EXISTS specially
if ctx.EXISTS() != nil {
@@ -731,7 +734,7 @@ func (v *filterExpressionVisitor) VisitFunctionCall(ctx *grammar.FunctionCallCon
return ErrorConditionLiteral
}
conds, ok := v.buildConditions(key, MatchingFieldKeys(key, v.fieldKeys), operator, value)
conds, ok := v.buildConditions(key, v.matchingFieldKeys(key), operator, value)
if !ok {
return ErrorConditionLiteral
}
@@ -924,7 +927,7 @@ func (v *filterExpressionVisitor) VisitKey(ctx *grammar.KeyContext) any {
// buildConditions invokes the condition builder for a filter term, folding its
// warnings/errors into visitor state; returns false if an error was recorded.
func (v *filterExpressionVisitor) buildConditions(key *telemetrytypes.TelemetryFieldKey, matching []*telemetrytypes.TelemetryFieldKey, op qbtypes.FilterOperator, value any) ([]string, bool) {
conds, warns, err := v.conditionBuilder.ConditionFor(v.context, v.orgID, v.startNs, v.endNs, key, v.fieldKeys, qbtypes.ConditionBuilderOptions{SkipResourceFilter: v.skipResourceFilter}, op, value, v.builder)
conds, warns, err := v.conditionBuilder.ConditionFor(v.context, v.orgID, v.startNs, v.endNs, key, v.fieldKeys, qbtypes.ConditionBuilderOptions{SkipResourceFilter: v.skipResourceFilter, ExactSemconv: v.exactSemconv}, op, value, v.builder)
if err != nil {
_, _, _, _, errURL, _ := errors.Unwrapb(err)
assignIfEmpty(&v.mainErrorURL, errURL)
@@ -983,12 +986,24 @@ 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 {
return matchingFieldKeys(field, fieldKeys, true)
}
// MatchingFieldKeysExact matches only the requested physical spelling.
func MatchingFieldKeysExact(field *telemetrytypes.TelemetryFieldKey, fieldKeys map[string][]*telemetrytypes.TelemetryFieldKey) []*telemetrytypes.TelemetryFieldKey {
return matchingFieldKeys(field, fieldKeys, false)
}
func matchingFieldKeys(field *telemetrytypes.TelemetryFieldKey, fieldKeys map[string][]*telemetrytypes.TelemetryFieldKey, resolveSemconv bool) []*telemetrytypes.TelemetryFieldKey {
selector := telemetrytypes.FieldKeySelector{
Name: field.Name,
Signal: field.Signal,
FieldContext: field.FieldContext,
}
members := semconv.Members(semconv.KindAttribute, selector)
members := []string{field.Name}
if resolveSemconv {
members = semconv.AttributeMembers(selector)
}
isFamily := len(members) > 1
fieldKeysForName := make([]*telemetrytypes.TelemetryFieldKey, 0)
indexByIdentity := make(map[string]int)
@@ -1005,13 +1020,13 @@ func MatchingFieldKeys(field *telemetrytypes.TelemetryFieldKey, fieldKeys map[st
// 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 {
if resolveSemconv && memberName != field.Name {
itemSelector := telemetrytypes.FieldKeySelector{
Name: field.Name,
Signal: item.Signal,
FieldContext: item.FieldContext,
}
if !slices.Contains(semconv.Members(semconv.KindAttribute, itemSelector), memberName) {
if !slices.Contains(semconv.AttributeMembers(itemSelector), memberName) {
continue
}
}
@@ -1038,6 +1053,8 @@ func MatchingFieldKeys(field *telemetrytypes.TelemetryFieldKey, fieldKeys map[st
if isFamily {
resolved.Name = field.Name
resolved.SemconvMembers = slices.Clone(physicalMembers)
} else if !resolveSemconv {
resolved.SemconvMembers = []string{field.Name}
}
fieldKeysForName = append(fieldKeysForName, &resolved)
}
@@ -1060,3 +1077,25 @@ func MatchingFieldKeys(field *telemetrytypes.TelemetryFieldKey, fieldKeys map[st
return fieldKeysForName
}
func (v *filterExpressionVisitor) matchingFieldKeys(field *telemetrytypes.TelemetryFieldKey) []*telemetrytypes.TelemetryFieldKey {
if v.exactSemconv {
return MatchingFieldKeysExact(field, v.fieldKeys)
}
return MatchingFieldKeys(field, v.fieldKeys)
}
// ExactSemconvKeys returns copies pinned to their physical names, preventing a
// field mapper from expanding a synthesized or metadata-free key into a family.
func ExactSemconvKeys(keys []*telemetrytypes.TelemetryFieldKey) []*telemetrytypes.TelemetryFieldKey {
result := make([]*telemetrytypes.TelemetryFieldKey, 0, len(keys))
for _, key := range keys {
if key == nil {
continue
}
resolved := *key
resolved.SemconvMembers = []string{key.Name}
result = append(result, &resolved)
}
return result
}

View File

@@ -731,6 +731,11 @@ func TestMatchingFieldKeysResolvesSemconvFamily(t *testing.T) {
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)
exactMatches := MatchingFieldKeysExact(requested, fieldKeys)
require.Len(t, exactMatches, 1, "exact lookup must return one field before its metadata is inspected")
assert.Equal(t, current.Name, exactMatches[0].Name)
assert.Equal(t, []string{current.Name}, exactMatches[0].SemconvMembers)
}
// ---------------------------------------------------------------------------

View File

@@ -2,21 +2,47 @@
package semconv
import "github.com/SigNoz/signoz/pkg/types/telemetrytypes"
var families = []Family{
{
Current: "container.cpu.usage",
Old: []string{"container.cpu.utilization"},
Kind: KindMetric,
Contexts: []telemetrytypes.FieldContext{telemetrytypes.FieldContextMetric},
Signals: []telemetrytypes.Signal{telemetrytypes.SignalMetrics},
ApplyToMetrics: nil,
},
{
Current: "db.system.name",
Old: []string{"db.system"},
Kind: KindAttribute,
Contexts: nil,
Signals: nil,
Contexts: []telemetrytypes.FieldContext{telemetrytypes.FieldContextAttribute, telemetrytypes.FieldContextResource},
Signals: []telemetrytypes.Signal{telemetrytypes.SignalLogs, telemetrytypes.SignalMetrics, telemetrytypes.SignalTraces},
ApplyToMetrics: nil,
},
{
Current: "deployment.environment.name",
Old: []string{"deployment.environment"},
Kind: KindAttribute,
Contexts: nil,
Signals: nil,
Contexts: []telemetrytypes.FieldContext{telemetrytypes.FieldContextAttribute, telemetrytypes.FieldContextResource},
Signals: []telemetrytypes.Signal{telemetrytypes.SignalLogs, telemetrytypes.SignalMetrics, telemetrytypes.SignalTraces},
ApplyToMetrics: nil,
},
{
Current: "k8s.node.cpu.usage",
Old: []string{"k8s.node.cpu.utilization"},
Kind: KindMetric,
Contexts: []telemetrytypes.FieldContext{telemetrytypes.FieldContextMetric},
Signals: []telemetrytypes.Signal{telemetrytypes.SignalMetrics},
ApplyToMetrics: nil,
},
{
Current: "k8s.pod.cpu.usage",
Old: []string{"k8s.pod.cpu.utilization"},
Kind: KindMetric,
Contexts: []telemetrytypes.FieldContext{telemetrytypes.FieldContextMetric},
Signals: []telemetrytypes.Signal{telemetrytypes.SignalMetrics},
ApplyToMetrics: nil,
},
}

View File

@@ -2,6 +2,7 @@ package semconv
import (
"slices"
"strings"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
@@ -26,6 +27,13 @@ type Family struct {
ValueMap map[string]string
}
type metricSpelling uint8
const (
metricSpellingDotted metricSpelling = iota
metricSpellingNormalized
)
var (
KindAttribute = Kind{String: valuer.NewString("attribute")}
KindMetric = Kind{String: valuer.NewString("metric")}
@@ -60,6 +68,88 @@ func Members(kind Kind, selector telemetrytypes.FieldKeySelector) []string {
return familyMembers[idx]
}
// AttributeMembers returns the physical attribute spellings that may represent
// selector.Name. Metrics have used both dotted and normalized label layouts;
// resource labels have additionally used a resource_ prefix. Keeping that
// storage detail here prevents metrics readers from maintaining local
// transition tables.
func AttributeMembers(selector telemetrytypes.FieldKeySelector) []string {
if selector.Signal != telemetrytypes.SignalMetrics {
return Members(KindAttribute, selector)
}
lookupSelector := selector
lookupSelector.Name = strings.TrimPrefix(selector.Name, "resource_")
family, style, ok := lookupMetricSpelling(KindAttribute, lookupSelector)
if !ok {
return []string{selector.Name}
}
logicalMembers := familyMembers[family]
result := make([]string, 0, len(logicalMembers)*4)
for _, member := range logicalMembers {
dotted := member
normalized := normalizeMetricSpelling(member)
variants := []string{dotted, normalized}
if style == metricSpellingNormalized {
variants[0], variants[1] = variants[1], variants[0]
}
if selector.FieldContext == telemetrytypes.FieldContextResource ||
selector.FieldContext == telemetrytypes.FieldContextUnspecified ||
strings.HasPrefix(selector.Name, "resource_") {
for _, variant := range variants {
result = appendUniqueString(result, "resource_"+variant)
}
}
for _, variant := range variants {
result = appendUniqueString(result, variant)
}
}
return result
}
// MetricNames returns the current and historical storage names for a metric.
// The input's dotted or normalized style is preserved because both layouts are
// valid metric identities and must not be mixed in one query.
func MetricNames(name string) []string {
selector := telemetrytypes.FieldKeySelector{
Name: name,
Signal: telemetrytypes.SignalMetrics,
FieldContext: telemetrytypes.FieldContextMetric,
}
family, style, ok := lookupMetricSpelling(KindMetric, selector)
if !ok {
return []string{name}
}
logicalMembers := familyMembers[family]
result := make([]string, 0, len(logicalMembers))
for _, member := range logicalMembers {
if style == metricSpellingNormalized {
member = normalizeMetricSpelling(member)
}
result = appendUniqueString(result, member)
}
return result
}
// CurrentAttribute returns the canonical dotted name for an attribute
// spelling, or selector.Name if no enabled family matches.
func CurrentAttribute(selector telemetrytypes.FieldKeySelector) string {
if selector.Signal != telemetrytypes.SignalMetrics {
return Current(KindAttribute, selector)
}
lookupSelector := selector
lookupSelector.Name = strings.TrimPrefix(selector.Name, "resource_")
family, _, ok := lookupMetricSpelling(KindAttribute, lookupSelector)
if !ok {
return selector.Name
}
return families[family].Current
}
// 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 {
@@ -103,6 +193,35 @@ func lookupIndex(kind Kind, selector telemetrytypes.FieldKeySelector) (int, bool
return 0, false
}
func lookupMetricSpelling(kind Kind, selector telemetrytypes.FieldKeySelector) (int, metricSpelling, bool) {
if idx, ok := lookupIndex(kind, selector); ok {
return idx, metricSpellingDotted, true
}
for idx, family := range families {
if !matchesSelector(family, kind, selector) {
continue
}
for _, member := range familyMembers[idx] {
if normalizeMetricSpelling(member) == selector.Name {
return idx, metricSpellingNormalized, true
}
}
}
return 0, metricSpellingDotted, false
}
func normalizeMetricSpelling(name string) string {
return strings.ReplaceAll(name, ".", "_")
}
func appendUniqueString(values []string, value string) []string {
if value == "" || slices.Contains(values, value) {
return values
}
return append(values, value)
}
func matchesSelector(family Family, kind Kind, selector telemetrytypes.FieldKeySelector) bool {
if family.Kind != kind {
return false

View File

@@ -57,3 +57,58 @@ func TestAllReturnsDefensiveCopies(t *testing.T) {
second := All()
assert.NotEqual(t, "mutated", second[0].Old[0], "All must not expose mutable generated data")
}
func TestAttributeMembersIncludesMetricResourceStorageSpellings(t *testing.T) {
selector := telemetrytypes.FieldKeySelector{
Name: "db.system.name",
Signal: telemetrytypes.SignalMetrics,
FieldContext: telemetrytypes.FieldContextResource,
}
assert.Equal(t, []string{
"resource_db.system.name", "resource_db_system_name", "db.system.name", "db_system_name",
"resource_db.system", "resource_db_system", "db.system", "db_system",
}, AttributeMembers(selector), "resource metric attributes should cover every historical storage layout")
}
func TestCurrentAttributeResolvesNormalizedMetricResourceSpelling(t *testing.T) {
selector := telemetrytypes.FieldKeySelector{
Name: "resource_db_system",
Signal: telemetrytypes.SignalMetrics,
FieldContext: telemetrytypes.FieldContextResource,
}
assert.Equal(t, "db.system.name", CurrentAttribute(selector), "normalized resource spelling should resolve to the dotted current name")
}
func TestAttributeMembersPreservesNormalizedMetricPointStyle(t *testing.T) {
selector := telemetrytypes.FieldKeySelector{
Name: "db_system",
Signal: telemetrytypes.SignalMetrics,
FieldContext: telemetrytypes.FieldContextAttribute,
}
assert.Equal(t, []string{
"db_system_name", "db.system.name", "db_system", "db.system",
}, AttributeMembers(selector), "normalized point attribute should remain the preferred storage spelling")
}
func TestMetricNamesPreservesDottedStyle(t *testing.T) {
assert.Equal(t,
[]string{"k8s.pod.cpu.usage", "k8s.pod.cpu.utilization"},
MetricNames("k8s.pod.cpu.usage"),
"dotted metric input should produce dotted family names",
)
}
func TestMetricNamesPreservesNormalizedStyle(t *testing.T) {
assert.Equal(t,
[]string{"k8s_pod_cpu_usage", "k8s_pod_cpu_utilization"},
MetricNames("k8s_pod_cpu_utilization"),
"normalized metric input should produce normalized family names",
)
}
func TestMetricNamesReturnsUnknownNameUnchanged(t *testing.T) {
assert.Equal(t, []string{"custom_metric"}, MetricNames("custom_metric"), "unknown metrics should not be expanded")
}

View File

@@ -10,6 +10,7 @@ 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/statementbuilder"
"github.com/SigNoz/signoz/pkg/telemetryschema/metricstelemetryschema"
"github.com/SigNoz/signoz/pkg/types/metrictypes"
@@ -31,6 +32,15 @@ const (
OthersMultiTemporality = `IF(LOWER(temporality) LIKE LOWER('delta'), %s, %s) AS per_series_value`
)
func metricNameValues(name string) []any {
members := semconv.MetricNames(name)
values := make([]any, 0, len(members))
for _, member := range members {
values = append(values, member)
}
return values
}
type StatementBuilder struct {
logger *slog.Logger
metadataStore telemetrytypes.MetadataStore
@@ -341,7 +351,7 @@ func (b *StatementBuilder) buildReducedTimeSeriesCTE(
sb.SelectMore(col)
}
sb.Where(
sb.In("metric_name", query.Aggregations[0].MetricName),
sb.In("metric_name", metricNameValues(query.Aggregations[0].MetricName)...),
sb.GTE("unix_milli", start),
sb.LTE("unix_milli", end),
)
@@ -385,7 +395,7 @@ func (b *StatementBuilder) buildReducedSpatialAggFastPath(
sb.From(fmt.Sprintf("%s.%s AS points FINAL", metricstelemetryschema.DBName, metricstelemetryschema.WhichReducedSamplesTableToUse(agg.Type)))
sb.JoinWithOption(sqlbuilder.InnerJoin, timeSeriesCTE, "points.reduced_fingerprint = filtered_time_series.fingerprint")
sb.Where(
sb.In("metric_name", agg.MetricName),
sb.In("metric_name", metricNameValues(agg.MetricName)...),
sb.GTE("unix_milli", start),
sb.LT("unix_milli", end),
)
@@ -427,7 +437,7 @@ func (b *StatementBuilder) buildReducedTemporalAggregationCTE(
sb.From(fmt.Sprintf("%s.%s AS points FINAL", metricstelemetryschema.DBName, metricstelemetryschema.WhichReducedSamplesTableToUse(agg.Type)))
sb.JoinWithOption(sqlbuilder.InnerJoin, timeSeriesCTE, "points.reduced_fingerprint = filtered_time_series.fingerprint")
sb.Where(
sb.In("metric_name", agg.MetricName),
sb.In("metric_name", metricNameValues(agg.MetricName)...),
sb.GTE("unix_milli", start),
sb.LT("unix_milli", end),
)
@@ -505,7 +515,7 @@ func (b *StatementBuilder) buildTemporalAggDeltaFastPath(
sb.From(fmt.Sprintf("%s.%s AS points", metricstelemetryschema.DBName, samplesTable))
sb.JoinWithOption(sqlbuilder.InnerJoin, timeSeriesCTE, "points.fingerprint = filtered_time_series.fingerprint")
sb.Where(
sb.In("metric_name", query.Aggregations[0].MetricName),
sb.In("metric_name", metricNameValues(query.Aggregations[0].MetricName)...),
sb.GTE("unix_milli", start),
sb.LT("unix_milli", end),
)
@@ -560,7 +570,7 @@ func (b *StatementBuilder) buildTimeSeriesCTE(
}
sb.Where(
sb.In("metric_name", query.Aggregations[0].MetricName),
sb.In("metric_name", metricNameValues(query.Aggregations[0].MetricName)...),
sb.GTE("unix_milli", start),
sb.LTE("unix_milli", end),
)
@@ -638,7 +648,7 @@ func (b *StatementBuilder) buildTemporalAggDelta(
sb.From(fmt.Sprintf("%s.%s AS points", metricstelemetryschema.DBName, samplesTable))
sb.JoinWithOption(sqlbuilder.InnerJoin, timeSeriesCTE, "points.fingerprint = filtered_time_series.fingerprint")
sb.Where(
sb.In("metric_name", query.Aggregations[0].MetricName),
sb.In("metric_name", metricNameValues(query.Aggregations[0].MetricName)...),
sb.GTE("unix_milli", start),
sb.LT("unix_milli", end),
)
@@ -679,7 +689,7 @@ func (b *StatementBuilder) buildTemporalAggCumulativeOrUnspecified(
baseSb.From(fmt.Sprintf("%s.%s AS points", metricstelemetryschema.DBName, samplesTable))
baseSb.JoinWithOption(sqlbuilder.InnerJoin, timeSeriesCTE, "points.fingerprint = filtered_time_series.fingerprint")
baseSb.Where(
baseSb.In("metric_name", query.Aggregations[0].MetricName),
baseSb.In("metric_name", metricNameValues(query.Aggregations[0].MetricName)...),
baseSb.GTE("unix_milli", start),
baseSb.LT("unix_milli", end),
)
@@ -770,7 +780,7 @@ func (b *StatementBuilder) buildTemporalAggForMultipleTemporalities(
sb.From(fmt.Sprintf("%s.%s AS points", metricstelemetryschema.DBName, samplesTable))
sb.JoinWithOption(sqlbuilder.InnerJoin, timeSeriesCTE, "points.fingerprint = filtered_time_series.fingerprint")
sb.Where(
sb.In("metric_name", query.Aggregations[0].MetricName),
sb.In("metric_name", metricNameValues(query.Aggregations[0].MetricName)...),
sb.GTE("unix_milli", start),
sb.LT("unix_milli", end),
)

View File

@@ -13,9 +13,21 @@ import (
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes/telemetrytypestest"
"github.com/SigNoz/signoz/pkg/valuer"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestMetricNameValuesResolveFamily(t *testing.T) {
assert.Equal(t,
[]any{"k8s.pod.cpu.usage", "k8s.pod.cpu.utilization"},
metricNameValues("k8s.pod.cpu.usage"),
)
assert.Equal(t,
[]any{"container_cpu_usage", "container_cpu_utilization"},
metricNameValues("container_cpu_utilization"),
)
}
func TestStatementBuilder(t *testing.T) {
cases := []struct {
name string

View File

@@ -153,15 +153,19 @@ 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{
func attributeSemconvMembers(name string, signal telemetrytypes.Signal, fieldContext telemetrytypes.FieldContext) []string {
return semconv.AttributeMembers(telemetrytypes.FieldKeySelector{
Name: name,
Signal: telemetrytypes.SignalTraces,
Signal: signal,
FieldContext: fieldContext,
})
}
func traceSemconvDuplicateFactor() int {
func traceSemconvMembers(name string, fieldContext telemetrytypes.FieldContext) []string {
return attributeSemconvMembers(name, telemetrytypes.SignalTraces, fieldContext)
}
func semconvDuplicateFactor(signal telemetrytypes.Signal) int {
factor := 1
for _, family := range semconv.All() {
if family.Kind != semconv.KindAttribute {
@@ -169,43 +173,66 @@ func traceSemconvDuplicateFactor() int {
}
if _, ok := semconv.Lookup(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
Name: family.Current,
Signal: telemetrytypes.SignalTraces,
Signal: signal,
}); ok {
factor = max(factor, len(family.Old)+1)
factor = max(factor, len(attributeSemconvMembers(family.Current, signal, telemetrytypes.FieldContextUnspecified)))
}
}
return factor
}
func traceSemconvDuplicateFactor() int {
return semconvDuplicateFactor(telemetrytypes.SignalTraces)
}
func inStrings(sb *sqlbuilder.SelectBuilder, column string, values []string) string {
args := make([]any, 0, len(values))
for _, value := range values {
args = append(args, value)
}
return sb.In(column, args...)
}
// 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 {
return canonicalizeSemconvKeys(keys, telemetrytypes.SignalTraces)
}
// canonicalizeSemconvKeys presents one current-name key for each enabled
// attribute family and records only the physical spellings actually present in
// metadata. That lets field mappers retain their optimized single-key SQL for
// homogeneous ranges while mixed ranges coalesce current-first.
func canonicalizeSemconvKeys(keys []*telemetrytypes.TelemetryFieldKey, signal telemetrytypes.Signal) []*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 {
if key.Signal != signal {
result = append(result, key)
continue
}
family, ok := semconv.Lookup(semconv.KindAttribute, telemetrytypes.FieldKeySelector{
selector := telemetrytypes.FieldKeySelector{
Name: key.Name,
Signal: telemetrytypes.SignalTraces,
Signal: signal,
FieldContext: key.FieldContext,
})
if !ok {
}
members := semconv.AttributeMembers(selector)
current := semconv.CurrentAttribute(selector)
if len(members) == 1 && current == key.Name {
result = append(result, key)
continue
}
resolved := *key
resolved.Name = family.Current
resolved.Name = current
resolved.SemconvMembers = []string{key.Name}
identity := resolved.Name + ";" + resolved.Signal.StringValue() + ";" + resolved.FieldContext.StringValue() + ";" + resolved.FieldDataType.StringValue()
fromCurrent := key.Name == family.Current
physicalName := strings.TrimPrefix(key.Name, "resource_")
fromCurrent := physicalName == current || physicalName == strings.ReplaceAll(current, ".", "_")
if index, found := indexByIdentity[identity]; found {
physicalMembers := result[index].SemconvMembers
@@ -238,7 +265,7 @@ func canonicalizeTraceSemconvKeys(keys []*telemetrytypes.TelemetryFieldKey) []*t
present[member] = true
}
ordered := make([]string, 0, len(key.SemconvMembers))
for _, member := range traceSemconvMembers(key.Name, key.FieldContext) {
for _, member := range attributeSemconvMembers(key.Name, signal, key.FieldContext) {
if present[member] {
ordered = append(ordered, member)
}
@@ -569,10 +596,19 @@ func (t *telemetryMetaStore) getLogsKeys(ctx context.Context, orgID valuer.UUID,
continue
}
fieldKeyConds := []string{}
members := attributeSemconvMembers(sel.Name, telemetrytypes.SignalLogs, fieldContext)
if sel.SelectorMatchType == telemetrytypes.FieldSelectorMatchTypeExact {
fieldKeyConds = append(fieldKeyConds, sb.E("name", sel.Name))
memberValues := make([]any, 0, len(members))
for _, member := range members {
memberValues = append(memberValues, member)
}
fieldKeyConds = append(fieldKeyConds, sb.In("name", memberValues...))
} else {
fieldKeyConds = append(fieldKeyConds, sb.ILike("name", "%"+escapeForLike(sel.Name)+"%"))
memberConds := make([]string, 0, len(members))
for _, member := range members {
memberConds = append(memberConds, sb.ILike("name", "%"+escapeForLike(member)+"%"))
}
fieldKeyConds = append(fieldKeyConds, sb.Or(memberConds...))
}
if sel.FieldDataType != telemetrytypes.FieldDataTypeUnspecified {
fieldKeyConds = append(fieldKeyConds, sb.E("datatype", sel.FieldDataType.TagDataType()))
@@ -665,6 +701,7 @@ func (t *telemetryMetaStore) getLogsKeys(ctx context.Context, orgID valuer.UUID,
limit = 1000
}
dbLimit := limit * semconvDuplicateFactor(telemetrytypes.SignalLogs)
mainQuery := fmt.Sprintf(`
SELECT tag_key, tag_type, tag_data_type, max(priority) as priority
FROM (
@@ -673,7 +710,7 @@ func (t *telemetryMetaStore) getLogsKeys(ctx context.Context, orgID valuer.UUID,
GROUP BY tag_key, tag_type, tag_data_type
ORDER BY priority
LIMIT %d
`, strings.Join(queries, " UNION ALL "), limit+1)
`, strings.Join(queries, " UNION ALL "), dbLimit+1)
rows, err := t.telemetrystore.ClickhouseDB().Query(ctx, mainQuery, allArgs...)
if err != nil {
@@ -694,7 +731,7 @@ func (t *telemetryMetaStore) getLogsKeys(ctx context.Context, orgID valuer.UUID,
for rows.Next() {
rowCount++
// reached the limit, we know there are more results
if rowCount > limit {
if rowCount > dbLimit {
break
}
@@ -742,7 +779,16 @@ func (t *telemetryMetaStore) getLogsKeys(ctx context.Context, orgID valuer.UUID,
}
// hit the limit? (only counting DB results)
complete := rowCount <= limit
complete := rowCount <= dbLimit
keys = canonicalizeSemconvKeys(keys, telemetrytypes.SignalLogs)
if len(keys) > limit {
keys = keys[:limit]
complete = false
}
mapOfKeys = make(map[string]*telemetrytypes.TelemetryFieldKey, len(keys))
for _, key := range keys {
mapOfKeys[key.Name+";"+key.FieldContext.StringValue()+";"+key.FieldDataType.StringValue()] = key
}
staticKeys := []string{}
staticKeys = append(staticKeys, maps.Keys(logstelemetryschema.IntrinsicFields)...)
@@ -1050,10 +1096,19 @@ func (t *telemetryMetaStore) getMetricsKeys(ctx context.Context, fieldKeySelecto
conds := []string{}
for _, fieldKeySelector := range fieldKeySelectors {
fieldConds := []string{}
members := attributeSemconvMembers(fieldKeySelector.Name, telemetrytypes.SignalMetrics, fieldKeySelector.FieldContext)
if fieldKeySelector.SelectorMatchType == telemetrytypes.FieldSelectorMatchTypeExact {
fieldConds = append(fieldConds, sb.E("attr_name", fieldKeySelector.Name))
memberValues := make([]any, 0, len(members))
for _, member := range members {
memberValues = append(memberValues, member)
}
fieldConds = append(fieldConds, sb.In("attr_name", memberValues...))
} else {
fieldConds = append(fieldConds, sb.ILike("attr_name", "%"+escapeForLike(fieldKeySelector.Name)+"%"))
memberConds := make([]string, 0, len(members))
for _, member := range members {
memberConds = append(memberConds, sb.ILike("attr_name", "%"+escapeForLike(member)+"%"))
}
fieldConds = append(fieldConds, sb.Or(memberConds...))
}
fieldConds = append(fieldConds, sb.NotLike("attr_name", "\\_\\_%"))
@@ -1069,7 +1124,12 @@ func (t *telemetryMetaStore) getMetricsKeys(ctx context.Context, fieldKeySelecto
if fieldKeySelector.MetricContext != nil {
if fieldKeySelector.MetricContext.MetricName != "" {
fieldConds = append(fieldConds, sb.E("metric_name", fieldKeySelector.MetricContext.MetricName))
metricNames := semconv.MetricNames(fieldKeySelector.MetricContext.MetricName)
metricValues := make([]any, 0, len(metricNames))
for _, metricName := range metricNames {
metricValues = append(metricValues, metricName)
}
fieldConds = append(fieldConds, sb.In("metric_name", metricValues...))
}
if fieldKeySelector.MetricContext.MetricNamespace != "" {
fieldConds = append(fieldConds, sb.Like("metric_name", escapeForLike(fieldKeySelector.MetricContext.MetricNamespace)+"%"))
@@ -1091,7 +1151,8 @@ func (t *telemetryMetaStore) getMetricsKeys(ctx context.Context, fieldKeySelecto
mainSb.GroupBy("name", "field_context", "field_data_type")
mainSb.OrderBy("priority")
// query one extra to check if we hit the limit
mainSb.Limit(limit + 1)
dbLimit := limit * semconvDuplicateFactor(telemetrytypes.SignalMetrics)
mainSb.Limit(dbLimit + 1)
query, args := mainSb.BuildWithFlavor(sqlbuilder.ClickHouse)
@@ -1106,7 +1167,7 @@ func (t *telemetryMetaStore) getMetricsKeys(ctx context.Context, fieldKeySelecto
for rows.Next() {
rowCount++
// reached the limit, we know there are more results
if rowCount > limit {
if rowCount > dbLimit {
break
}
@@ -1131,7 +1192,12 @@ func (t *telemetryMetaStore) getMetricsKeys(ctx context.Context, fieldKeySelecto
}
// hit the limit?
complete := rowCount <= limit
complete := rowCount <= dbLimit
keys = canonicalizeSemconvKeys(keys, telemetrytypes.SignalMetrics)
if len(keys) > limit {
keys = keys[:limit]
complete = false
}
return keys, complete, nil
}
@@ -1153,16 +1219,30 @@ func (t *telemetryMetaStore) getMeterSourceMetricKeys(ctx context.Context, field
var limit int
for _, fieldKeySelector := range fieldKeySelectors {
fieldConds := []string{}
members := attributeSemconvMembers(fieldKeySelector.Name, telemetrytypes.SignalMetrics, fieldKeySelector.FieldContext)
if fieldKeySelector.SelectorMatchType == telemetrytypes.FieldSelectorMatchTypeExact {
fieldConds = append(fieldConds, sb.E("attr_name", fieldKeySelector.Name))
memberValues := make([]any, 0, len(members))
for _, member := range members {
memberValues = append(memberValues, member)
}
fieldConds = append(fieldConds, sb.In("attr_name", memberValues...))
} else {
fieldConds = append(fieldConds, sb.Like("attr_name", "%"+fieldKeySelector.Name+"%"))
memberConds := make([]string, 0, len(members))
for _, member := range members {
memberConds = append(memberConds, sb.Like("attr_name", "%"+member+"%"))
}
fieldConds = append(fieldConds, sb.Or(memberConds...))
}
fieldConds = append(fieldConds, sb.NotLike("attr_name", "\\_\\_%"))
if fieldKeySelector.MetricContext != nil {
if fieldKeySelector.MetricContext.MetricName != "" {
fieldConds = append(fieldConds, sb.E("metric_name", fieldKeySelector.MetricContext.MetricName))
metricNames := semconv.MetricNames(fieldKeySelector.MetricContext.MetricName)
metricValues := make([]any, 0, len(metricNames))
for _, metricName := range metricNames {
metricValues = append(metricValues, metricName)
}
fieldConds = append(fieldConds, sb.In("metric_name", metricValues...))
}
if fieldKeySelector.MetricContext.MetricNamespace != "" {
fieldConds = append(fieldConds, sb.Like("metric_name", escapeForLike(fieldKeySelector.MetricContext.MetricNamespace)+"%"))
@@ -1177,7 +1257,8 @@ func (t *telemetryMetaStore) getMeterSourceMetricKeys(ctx context.Context, field
limit = 1000
}
sb.Limit(limit)
dbLimit := limit * semconvDuplicateFactor(telemetrytypes.SignalMetrics)
sb.Limit(dbLimit + 1)
query, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
rows, err := t.telemetrystore.ClickhouseDB().Query(ctx, query, args...)
@@ -1191,7 +1272,7 @@ func (t *telemetryMetaStore) getMeterSourceMetricKeys(ctx context.Context, field
for rows.Next() {
rowCount++
// reached the limit, we know there are more results
if rowCount > limit {
if rowCount > dbLimit {
break
}
@@ -1200,9 +1281,14 @@ func (t *telemetryMetaStore) getMeterSourceMetricKeys(ctx context.Context, field
if err != nil {
return nil, false, errors.Wrap(err, errors.TypeInternal, errors.CodeInternal, ErrFailedToGetMeterKeys.Error())
}
fieldContext := telemetrytypes.FieldContextAttribute
if strings.HasPrefix(name, "resource_") {
fieldContext = telemetrytypes.FieldContextResource
}
keys = append(keys, &telemetrytypes.TelemetryFieldKey{
Name: name,
Signal: telemetrytypes.SignalMetrics,
Name: name,
Signal: telemetrytypes.SignalMetrics,
FieldContext: fieldContext,
})
}
@@ -1211,7 +1297,12 @@ func (t *telemetryMetaStore) getMeterSourceMetricKeys(ctx context.Context, field
}
// hit the limit?
complete := rowCount <= limit
complete := rowCount <= dbLimit
keys = canonicalizeSemconvKeys(keys, telemetrytypes.SignalMetrics)
if len(keys) > limit {
keys = keys[:limit]
complete = false
}
return keys, complete, nil
@@ -1570,7 +1661,6 @@ func (t *telemetryMetaStore) getRelatedValues(ctx context.Context, orgID valuer.
if limit == 0 {
limit = 50
}
// query one extra to check if we hit the limit
sb.Limit(limit + 1)
query, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
@@ -1725,7 +1815,12 @@ func (t *telemetryMetaStore) getLogFieldValues(ctx context.Context, fieldValueSe
sb := sqlbuilder.Select("DISTINCT string_value, number_value").From(t.logsDBName + "." + t.logsFieldsTblName)
if fieldValueSelector.Name != "" {
sb.Where(sb.E("tag_key", fieldValueSelector.Name))
members := attributeSemconvMembers(fieldValueSelector.Name, telemetrytypes.SignalLogs, fieldValueSelector.FieldContext)
memberValues := make([]any, 0, len(members))
for _, member := range members {
memberValues = append(memberValues, member)
}
sb.Where(sb.In("tag_key", memberValues...))
}
if fieldValueSelector.FieldContext != telemetrytypes.FieldContextUnspecified {
@@ -1923,7 +2018,9 @@ func (t *telemetryMetaStore) getMetricFieldValues(ctx context.Context, orgID val
From(t.metricsDBName + "." + t.metricsFieldsTblName)
if fieldValueSelector.Name != "" {
sb.Where(sb.E("attr_name", fieldValueSelector.Name))
sb.Where(inStrings(sb, "attr_name", attributeSemconvMembers(
fieldValueSelector.Name, telemetrytypes.SignalMetrics, fieldValueSelector.FieldContext,
)))
}
if fieldValueSelector.FieldContext != telemetrytypes.FieldContextUnspecified {
@@ -1935,7 +2032,7 @@ func (t *telemetryMetaStore) getMetricFieldValues(ctx context.Context, orgID val
}
if fieldValueSelector.MetricContext != nil && fieldValueSelector.MetricContext.MetricName != "" {
sb.Where(sb.E("metric_name", fieldValueSelector.MetricContext.MetricName))
sb.Where(inStrings(sb, "metric_name", semconv.MetricNames(fieldValueSelector.MetricContext.MetricName)))
}
if fieldValueSelector.MetricContext != nil && fieldValueSelector.MetricContext.MetricNamespace != "" {
sb.Where(sb.Like("metric_name", escapeForLike(fieldValueSelector.MetricContext.MetricNamespace)+"%"))
@@ -2069,7 +2166,7 @@ func (t *telemetryMetaStore) getIntrinsicMetricFieldValuesForTable(ctx context.C
From(t.metricsDBName + "." + tableName)
if fieldValueSelector.MetricContext != nil && fieldValueSelector.MetricContext.MetricName != "" {
sb.Where(sb.E("metric_name", fieldValueSelector.MetricContext.MetricName))
sb.Where(inStrings(sb, "metric_name", semconv.MetricNames(fieldValueSelector.MetricContext.MetricName)))
}
if fieldValueSelector.MetricContext != nil && fieldValueSelector.MetricContext.MetricNamespace != "" {
sb.Where(sb.Like("metric_name", escapeForLike(fieldValueSelector.MetricContext.MetricNamespace)+"%"))
@@ -2132,12 +2229,14 @@ func (t *telemetryMetaStore) getMeterSourceMetricFieldValues(ctx context.Context
From(t.meterDBName + "." + t.meterFieldsTblName)
if fieldValueSelector.Name != "" {
sb.Where(sb.E("attr.1", fieldValueSelector.Name))
sb.Where(inStrings(sb, "attr.1", attributeSemconvMembers(
fieldValueSelector.Name, telemetrytypes.SignalMetrics, fieldValueSelector.FieldContext,
)))
}
sb.Where(sb.NotLike("attr.1", "\\_\\_%"))
if fieldValueSelector.MetricContext != nil && fieldValueSelector.MetricContext.MetricName != "" {
sb.Where(sb.E("metric_name", fieldValueSelector.MetricContext.MetricName))
sb.Where(inStrings(sb, "metric_name", semconv.MetricNames(fieldValueSelector.MetricContext.MetricName)))
}
if fieldValueSelector.MetricContext != nil && fieldValueSelector.MetricContext.MetricNamespace != "" {
sb.Where(sb.Like("metric_name", escapeForLike(fieldValueSelector.MetricContext.MetricNamespace)+"%"))
@@ -2156,8 +2255,10 @@ func (t *telemetryMetaStore) getMeterSourceMetricFieldValues(ctx context.Context
if limit == 0 {
limit = 50
}
// query one extra to check if we hit the limit
sb.Limit(limit + 1)
// A value can be present under several physical spellings; over-fetch and
// de-duplicate after scanning so one family cannot consume the result limit.
dbLimit := limit * semconvDuplicateFactor(telemetrytypes.SignalMetrics)
sb.Limit(dbLimit + 1)
query, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
rows, err := t.telemetrystore.ClickhouseDB().Query(ctx, query, args...)
@@ -2167,11 +2268,13 @@ func (t *telemetryMetaStore) getMeterSourceMetricFieldValues(ctx context.Context
defer rows.Close()
values := &telemetrytypes.TelemetryFieldValues{}
seen := make(map[string]bool)
rowCount := 0
uniqueCount := 0
for rows.Next() {
rowCount++
// reached the limit, we know there are more results
if rowCount > limit {
if rowCount > dbLimit {
break
}
@@ -2180,12 +2283,16 @@ func (t *telemetryMetaStore) getMeterSourceMetricFieldValues(ctx context.Context
return nil, false, errors.Wrap(err, errors.TypeInternal, errors.CodeInternal, ErrFailedToGetMeterValues.Error())
}
if len(attribute) > 1 {
values.StringValues = append(values.StringValues, attribute[1])
if !seen[attribute[1]] && uniqueCount < limit {
values.StringValues = append(values.StringValues, attribute[1])
seen[attribute[1]] = true
uniqueCount++
}
}
}
// hit the limit?
complete := rowCount <= limit
complete := rowCount <= dbLimit && uniqueCount < limit
return values, complete, nil
}

View File

@@ -130,6 +130,44 @@ func TestCanonicalizeTraceSemconvKeys(t *testing.T) {
assert.Equal(t, "deployment.environment", result[2].Name, "phase 1 must not rewrite raw log metadata")
}
func TestCanonicalizeLogAndMetricSemconvKeys(t *testing.T) {
logKeys := canonicalizeSemconvKeys([]*telemetrytypes.TelemetryFieldKey{
{
Name: "db.system",
Signal: telemetrytypes.SignalLogs,
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
{
Name: "db.system.name",
Signal: telemetrytypes.SignalLogs,
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
}, telemetrytypes.SignalLogs)
require.Len(t, logKeys, 1)
assert.Equal(t, "db.system.name", logKeys[0].Name)
assert.Equal(t, []string{"db.system.name", "db.system"}, logKeys[0].SemconvMembers)
metricKeys := canonicalizeSemconvKeys([]*telemetrytypes.TelemetryFieldKey{
{
Name: "resource_db_system",
Signal: telemetrytypes.SignalMetrics,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
{
Name: "db.system.name",
Signal: telemetrytypes.SignalMetrics,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
}, telemetrytypes.SignalMetrics)
require.Len(t, metricKeys, 1)
assert.Equal(t, "db.system.name", metricKeys[0].Name)
assert.Equal(t, []string{"db.system.name", "resource_db_system"}, metricKeys[0].SemconvMembers)
}
func TestGetSpanFieldValuesMergesSemconvFamily(t *testing.T) {
mockTelemetryStore := telemetrystoretest.New(telemetrystore.Config{}, &regexMatcher{})
mock := mockTelemetryStore.Mock()

View File

@@ -0,0 +1,156 @@
package telemetrymetadata
import (
"context"
"fmt"
"slices"
"strings"
"github.com/huandu/go-sqlbuilder"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/semconv"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
)
type semconvMigrationRow struct {
current string
old string
signal string
service string
resourceSets uint64
lastSeenUnixMilli int64
}
// GetSemconvMigrationReport derives an old-only service report from the
// generated family registry and attributes_metadata. The latter is already a
// deduplicated set of resource/attribute fingerprints, so this audit avoids a
// scan of the raw telemetry tables.
func (t *telemetryMetaStore) GetSemconvMigrationReport(
ctx context.Context,
_ valuer.UUID,
startUnixMilli, endUnixMilli int64,
) (*telemetrytypes.GettableSemconvMigrationReport, error) {
query, args := t.semconvMigrationReportQuery(startUnixMilli, endUnixMilli)
report := &telemetrytypes.GettableSemconvMigrationReport{
StartUnixMilli: startUnixMilli,
EndUnixMilli: endUnixMilli,
Entries: []*telemetrytypes.SemconvMigrationReportEntry{},
}
if query == "" {
return report, nil
}
rows, err := t.telemetrystore.ClickhouseDB().Query(ctx, query, args...)
if err != nil {
return nil, errors.Wrap(err, errors.TypeInternal, errors.CodeInternal, "failed to build semantic-convention migration report")
}
defer rows.Close()
grouped := make(map[string]*telemetrytypes.SemconvMigrationReportEntry)
serviceSets := make(map[string]map[string]struct{})
for rows.Next() {
var row semconvMigrationRow
if err := rows.Scan(
&row.current,
&row.old,
&row.signal,
&row.service,
&row.resourceSets,
&row.lastSeenUnixMilli,
); err != nil {
return nil, errors.Wrap(err, errors.TypeInternal, errors.CodeInternal, "failed to scan semantic-convention migration report")
}
identity := row.current + "\x00" + row.old + "\x00" + row.signal
entry, ok := grouped[identity]
if !ok {
entry = &telemetrytypes.SemconvMigrationReportEntry{
Current: row.current,
Old: row.old,
Signal: row.signal,
Services: []string{},
}
grouped[identity] = entry
serviceSets[identity] = make(map[string]struct{})
report.Entries = append(report.Entries, entry)
}
serviceSets[identity][row.service] = struct{}{}
entry.ResourceSets += row.resourceSets
entry.LastSeenUnixMilli = max(entry.LastSeenUnixMilli, row.lastSeenUnixMilli)
}
if err := rows.Err(); err != nil {
return nil, errors.Wrap(err, errors.TypeInternal, errors.CodeInternal, "failed to read semantic-convention migration report")
}
for identity, entry := range grouped {
for service := range serviceSets[identity] {
entry.Services = append(entry.Services, service)
}
slices.Sort(entry.Services)
}
slices.SortFunc(report.Entries, func(a, b *telemetrytypes.SemconvMigrationReportEntry) int {
return strings.Compare(a.Current+"\x00"+a.Old+"\x00"+a.Signal, b.Current+"\x00"+b.Old+"\x00"+b.Signal)
})
return report, nil
}
func (t *telemetryMetaStore) semconvMigrationReportQuery(startUnixMilli, endUnixMilli int64) (string, []any) {
builders := make([]sqlbuilder.Builder, 0)
for _, family := range semconv.All() {
if family.Kind != semconv.KindAttribute {
continue
}
for _, old := range family.Old {
sb := sqlbuilder.NewSelectBuilder()
sb.Select(
fmt.Sprintf("%s AS current_name", sb.Var(family.Current)),
fmt.Sprintf("%s AS old_name", sb.Var(old)),
"data_source",
"if(empty(resource_attributes['service.name']), '<unknown>', resource_attributes['service.name']) AS service_name",
"uniqExact(tuple(resource_fingerprint, attrs_fingerprint)) AS resource_sets",
"max(unix_milli) AS last_seen_unix_milli",
)
sb.From(t.relatedMetadataDBName + "." + t.relatedMetadataTblName)
sb.Where(sb.GE("unix_milli", startUnixMilli))
sb.Where(sb.LE("unix_milli", endUnixMilli))
if len(family.Signals) > 0 {
signals := make([]any, 0, len(family.Signals))
for _, signal := range family.Signals {
signals = append(signals, signal.StringValue())
}
sb.Where(sb.In("data_source", signals...))
}
oldPresence := semconvMetadataPresenceConditions(sb, old)
currentPresence := semconvMetadataPresenceConditions(sb, family.Current)
sb.Where(sb.Or(oldPresence...))
sb.Where(fmt.Sprintf("NOT (%s)", sb.Or(currentPresence...)))
sb.GroupBy("data_source", "service_name")
builders = append(builders, sb)
}
}
if len(builders) == 0 {
return "", nil
}
union := sqlbuilder.UnionAll(builders...)
return union.BuildWithFlavor(sqlbuilder.ClickHouse)
}
func semconvMetadataPresenceConditions(sb *sqlbuilder.SelectBuilder, name string) []string {
spellings := []string{name, strings.ReplaceAll(name, ".", "_")}
spellings = append(spellings, "resource_"+name, "resource_"+strings.ReplaceAll(name, ".", "_"))
spellings = slices.Compact(spellings)
conditions := make([]string, 0, len(spellings)*2)
for _, spelling := range spellings {
conditions = append(conditions,
fmt.Sprintf("mapContains(resource_attributes, %s)", sb.Var(spelling)),
fmt.Sprintf("mapContains(attributes, %s)", sb.Var(spelling)),
)
}
return conditions
}

View File

@@ -89,10 +89,12 @@ func (v *TelemetryFieldVisitor) VisitColumnDef(expr *parser.ColumnDef) error {
// Create and store the TelemetryFieldKey
field := &telemetrytypes.TelemetryFieldKey{
Name: fieldName,
FieldContext: fieldContext,
FieldDataType: fieldDataType,
Materialized: true,
Name: fieldName,
FieldContext: fieldContext,
FieldDataType: fieldDataType,
Materialized: true,
MaterializedColumnName: columnName,
MaterializedSemconv: strings.Count(defaultExprStr, "['") > 1,
}
v.Fields = append(v.Fields, field)

View File

@@ -5,8 +5,25 @@ import (
"testing"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestExtractFieldKeyPreservesHistoricalMaterializedColumnName(t *testing.T) {
statement := `CREATE TABLE signoz_traces.signoz_index_v3
(
attributes_string Map(LowCardinality(String), String),
` + "`attribute_string_db$$system`" + ` LowCardinality(String)
DEFAULT if(mapContains(attributes_string, 'db.system.name'), attributes_string['db.system.name'], attributes_string['db.system'])
) ENGINE = MergeTree ORDER BY tuple()`
keys, err := ExtractFieldKeysFromTblStatement(statement)
require.NoError(t, err, "table statement should parse")
require.Len(t, keys, 1, "table statement should contain one materialized key")
assert.Equal(t, "db.system.name", keys[0].Name)
assert.Equal(t, "attribute_string_db$$system", keys[0].MaterializedColumnName)
assert.True(t, keys[0].MaterializedSemconv)
}
func TestExtractFieldKeysFromTblStatement(t *testing.T) {
var statement = `CREATE TABLE signoz_logs.logs_v2

View File

@@ -453,6 +453,9 @@ func (c *conditionBuilder) ConditionFor(
sb *sqlbuilder.SelectBuilder,
) ([]string, []string, error) {
matches := querybuilder.MatchingFieldKeys(key, fieldKeys)
if options.ExactSemconv {
matches = querybuilder.MatchingFieldKeysExact(key, fieldKeys)
}
skipResourceFilter := options.SkipResourceFilter
// search() resolves its own (optional) scope; handle it before key resolution.
@@ -499,6 +502,9 @@ func (c *conditionBuilder) ConditionFor(
warnings = append(warnings, querybuilder.NewKeyNotFoundWarning(key.Name))
}
}
if options.ExactSemconv {
keys = querybuilder.ExactSemconvKeys(keys)
}
if skipResourceFilter && !synthesized {
filtered := make([]*telemetrytypes.TelemetryFieldKey, 0, len(keys))

View File

@@ -10,6 +10,7 @@ import (
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/flagger"
"github.com/SigNoz/signoz/pkg/querybuilder"
"github.com/SigNoz/signoz/pkg/semconv"
"github.com/SigNoz/signoz/pkg/types/featuretypes"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
@@ -67,6 +68,20 @@ type fieldMapper struct {
fl flagger.Flagger
}
func logSemconvMembers(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.AttributeMembers(telemetrytypes.FieldKeySelector{
Name: key.Name,
Signal: telemetrytypes.SignalLogs,
FieldContext: key.FieldContext,
})
}
func NewFieldMapper(fl flagger.Flagger) qbtypes.FieldMapper {
return &fieldMapper{fl: fl}
}
@@ -141,8 +156,20 @@ func (m *fieldMapper) FieldFor(ctx context.Context, orgID valuer.UUID, tsStart,
case schema.ColumnTypeEnumJSON:
switch key.FieldContext {
case telemetrytypes.FieldContextResource:
exprs = append(exprs, fmt.Sprintf("%s.`%s`::String", columnName, key.Name))
existExpr = append(existExpr, fmt.Sprintf("%s.`%s` IS NOT NULL", columnName, key.Name))
members := logSemconvMembers(key)
if len(members) > 1 {
values := make([]string, 0, len(members))
guards := make([]string, 0, len(members))
for _, member := range members {
values = append(values, fmt.Sprintf("NULLIF(%s.`%s`::String, '')", columnName, member))
guards = append(guards, fmt.Sprintf("%s.`%s` IS NOT NULL", columnName, member))
}
exprs = append(exprs, "COALESCE("+strings.Join(values, ", ")+")")
existExpr = append(existExpr, "("+strings.Join(guards, " OR ")+")")
} else {
exprs = append(exprs, fmt.Sprintf("%s.`%s`::String", columnName, members[0]))
existExpr = append(existExpr, fmt.Sprintf("%s.`%s` IS NOT NULL", columnName, members[0]))
}
case telemetrytypes.FieldContextBody:
if key.Name == messageSubField {
exprs = append(exprs, messageSubColumn)
@@ -181,13 +208,32 @@ func (m *fieldMapper) FieldFor(ctx context.Context, orgID valuer.UUID, tsStart,
switch valueType := column.Type.(schema.MapColumnType).ValueType; valueType.GetType() {
case schema.ColumnTypeEnumString, schema.ColumnTypeEnumBool, schema.ColumnTypeEnumFloat64:
// a key could have been materialized, if so return the materialized column name
if key.Materialized {
members := logSemconvMembers(key)
if key.Materialized && (len(members) == 1 || key.MaterializedSemconv) {
exprs = append(exprs, telemetrytypes.FieldKeyToMaterializedColumnName(key))
existExpr = append(existExpr, telemetrytypes.FieldKeyToMaterializedColumnNameForExists(key))
} else 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+1)
for i, member := range members {
branches = append(branches, guards[i], fmt.Sprintf("%s['%s']", columnName, member))
}
exprs = append(exprs, "multiIf("+strings.Join(branches, ", ")+", NULL)")
}
existExpr = append(existExpr, "("+strings.Join(guards, " OR ")+")")
} else {
exprs = append(exprs, fmt.Sprintf("%s['%s']", columnName, key.Name))
existExpr = append(existExpr, fmt.Sprintf("mapContains(%s, '%s')", columnName, key.Name))
exprs = append(exprs, fmt.Sprintf("%s['%s']", columnName, members[0]))
existExpr = append(existExpr, fmt.Sprintf("mapContains(%s, '%s')", columnName, members[0]))
}
default:
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "exists operator is not supported for map column type %s", valueType)

View File

@@ -2,6 +2,7 @@ package logstelemetryschema
import (
"context"
"strings"
"testing"
"time"
@@ -14,6 +15,32 @@ import (
"github.com/stretchr/testify/require"
)
func TestFieldForSemconvFamily(t *testing.T) {
ctx := context.Background()
fm := NewFieldMapper(flaggertest.New(t)).(*fieldMapper)
key := &telemetrytypes.TelemetryFieldKey{
Name: "db.system.name",
Signal: telemetrytypes.SignalLogs,
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
expression, err := fm.FieldFor(ctx, valuer.UUID{}, 0, 0, key)
require.NoError(t, err)
assert.Equal(t, "COALESCE(NULLIF(attributes_string['db.system.name'], ''), NULLIF(attributes_string['db.system'], ''))", expression)
assert.Less(t, strings.Index(expression, "db.system.name"), strings.Index(expression, "db.system']"), "current spelling must win")
exists, err := fm.existsExpressionFor(ctx, valuer.UUID{}, 0, 0, key, true)
require.NoError(t, err)
assert.Equal(t, "(mapContains(attributes_string, 'db.system.name') OR mapContains(attributes_string, 'db.system'))", exists)
exact := *key
exact.SemconvMembers = []string{"db.system"}
expression, err = fm.FieldFor(ctx, valuer.UUID{}, 0, 0, &exact)
require.NoError(t, err)
assert.Equal(t, "attributes_string['db.system']", expression)
}
func TestGetColumn(t *testing.T) {
ctx := context.Background()

View File

@@ -4,6 +4,7 @@ import (
"context"
"fmt"
"slices"
"strings"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/querybuilder"
@@ -135,10 +136,22 @@ func (c *conditionBuilder) conditionFor(
return "true", nil
}
if operator == qbtypes.FilterOperatorExists {
return fmt.Sprintf("has(JSONExtractKeys(labels), '%s')", key.Name), nil
members := metricAttributeMembers(key)
guards := make([]string, 0, len(members))
for _, member := range members {
guards = append(guards, fmt.Sprintf("has(JSONExtractKeys(labels), '%s')", member))
}
return fmt.Sprintf("not has(JSONExtractKeys(labels), '%s')", key.Name), nil
guard := strings.Join(guards, " OR ")
if len(guards) > 1 {
guard = "(" + guard + ")"
}
if operator == qbtypes.FilterOperatorExists {
return guard, nil
}
if len(guards) == 1 {
return "not " + guard, nil
}
return "NOT " + guard, nil
}
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "unsupported operator: %v", operator)
}
@@ -151,7 +164,7 @@ func (c *conditionBuilder) ConditionFor(
endNs uint64,
key *telemetrytypes.TelemetryFieldKey,
fieldKeys map[string][]*telemetrytypes.TelemetryFieldKey,
_ qbtypes.ConditionBuilderOptions,
options qbtypes.ConditionBuilderOptions,
operator qbtypes.FilterOperator,
value any,
sb *sqlbuilder.SelectBuilder,
@@ -162,7 +175,15 @@ func (c *conditionBuilder) ConditionFor(
return nil, nil, err
}
requestedKey := *key
if requestedKey.Signal == telemetrytypes.SignalUnspecified {
requestedKey.Signal = telemetrytypes.SignalMetrics
}
key = &requestedKey
keys := querybuilder.MatchingFieldKeys(key, fieldKeys)
if options.ExactSemconv {
keys = querybuilder.MatchingFieldKeysExact(key, fieldKeys)
}
var warnings []string
if len(keys) == 0 {
if _, isColumn := timeSeriesV4Columns[key.Name]; isColumn {
@@ -180,6 +201,9 @@ func (c *conditionBuilder) ConditionFor(
}
}
}
if options.ExactSemconv {
keys = querybuilder.ExactSemconvKeys(keys)
}
conds := make([]string, 0, len(keys))
for _, k := range keys {

View File

@@ -390,3 +390,45 @@ func TestConditionForKeyNotInMetadata(t *testing.T) {
})
}
}
func TestConditionForSemconvMetricLabels(t *testing.T) {
ctx := context.Background()
fm := NewFieldMapper()
conditionBuilder := NewConditionBuilder(fm)
requested := telemetrytypes.TelemetryFieldKey{
Name: "db.system.name",
Signal: telemetrytypes.SignalMetrics,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
current := requested
legacyNormalized := requested
legacyNormalized.Name = "resource_db_system"
fieldKeys := map[string][]*telemetrytypes.TelemetryFieldKey{
current.Name: {&current},
legacyNormalized.Name: {&legacyNormalized},
}
sb := sqlbuilder.NewSelectBuilder()
conditions, warnings, err := conditionBuilder.ConditionFor(
ctx, valuer.UUID{}, 0, 0, &requested, fieldKeys, qbtypes.ConditionBuilderOptions{},
qbtypes.FilterOperatorEqual, "postgresql", sb,
)
require.NoError(t, err)
assert.Empty(t, warnings)
sb.Where(conditions...)
query, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
assert.Contains(t, query, "COALESCE(NULLIF(JSONExtractString(labels, 'db.system.name'), ''), NULLIF(JSONExtractString(labels, 'resource_db_system'), '')) = ?")
assert.Equal(t, []any{"postgresql"}, args)
sb = sqlbuilder.NewSelectBuilder()
conditions, _, err = conditionBuilder.ConditionFor(
ctx, valuer.UUID{}, 0, 0, &requested, fieldKeys, qbtypes.ConditionBuilderOptions{ExactSemconv: true},
qbtypes.FilterOperatorExists, nil, sb,
)
require.NoError(t, err)
sb.Where(conditions...)
query, _ = sb.BuildWithFlavor(sqlbuilder.ClickHouse)
assert.Contains(t, query, "has(JSONExtractKeys(labels), 'db.system.name')")
assert.NotContains(t, query, "resource_db_system")
}

View File

@@ -4,8 +4,10 @@ import (
"context"
"fmt"
"slices"
"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"
@@ -38,6 +40,23 @@ var (
type fieldMapper struct{}
func metricAttributeMembers(key *telemetrytypes.TelemetryFieldKey) []string {
if key.FieldContext != telemetrytypes.FieldContextResource &&
key.FieldContext != telemetrytypes.FieldContextScope &&
key.FieldContext != telemetrytypes.FieldContextAttribute &&
key.FieldContext != telemetrytypes.FieldContextUnspecified {
return []string{key.Name}
}
if len(key.SemconvMembers) > 0 {
return key.SemconvMembers
}
return semconv.AttributeMembers(telemetrytypes.FieldKeySelector{
Name: key.Name,
Signal: telemetrytypes.SignalMetrics,
FieldContext: key.FieldContext,
})
}
// CandidateKeys returns nil: metrics has no attribute-map fallback, so a context-missing
// key stays unresolved and the caller errors.
func (m *fieldMapper) CandidateKeys(_ context.Context, _ valuer.UUID, _ *telemetrytypes.TelemetryFieldKey, _ any, _ map[string][]*telemetrytypes.TelemetryFieldKey) []*telemetrytypes.TelemetryFieldKey {
@@ -80,19 +99,30 @@ func (m *fieldMapper) FieldFor(ctx context.Context, _ valuer.UUID, startNs, endN
switch key.FieldContext {
case telemetrytypes.FieldContextResource, telemetrytypes.FieldContextScope, telemetrytypes.FieldContextAttribute:
return fmt.Sprintf("JSONExtractString(%s, '%s')", columns[0].Name, key.Name), nil
return metricLabelExpression(columns[0].Name, metricAttributeMembers(key)), nil
case telemetrytypes.FieldContextMetric:
return columns[0].Name, nil
case telemetrytypes.FieldContextUnspecified:
if slices.Contains(IntrinsicFields, key.Name) {
return columns[0].Name, nil
}
return fmt.Sprintf("JSONExtractString(%s, '%s')", columns[0].Name, key.Name), nil
return metricLabelExpression(columns[0].Name, metricAttributeMembers(key)), nil
}
return columns[0].Name, nil
}
func metricLabelExpression(columnName string, members []string) string {
if len(members) == 1 {
return fmt.Sprintf("JSONExtractString(%s, '%s')", columnName, members[0])
}
values := make([]string, 0, len(members))
for _, member := range members {
values = append(values, fmt.Sprintf("NULLIF(JSONExtractString(%s, '%s'), '')", columnName, member))
}
return "COALESCE(" + strings.Join(values, ", ") + ")"
}
func (m *fieldMapper) ColumnFor(ctx context.Context, _ valuer.UUID, tsStart, tsEnd uint64, key *telemetrytypes.TelemetryFieldKey) ([]*schema.Column, error) {
return m.getColumn(ctx, tsStart, tsEnd, key)
}

View File

@@ -2,6 +2,7 @@ package metricstelemetryschema
import (
"context"
"strings"
"testing"
schema "github.com/SigNoz/signoz-otel-collector/cmd/signozschemamigrator/schema_migrator"
@@ -12,6 +13,32 @@ import (
"github.com/stretchr/testify/require"
)
func TestMetricLabelSemconvSpellings(t *testing.T) {
fm := NewFieldMapper()
key := &telemetrytypes.TelemetryFieldKey{
Name: "db.system.name",
Signal: telemetrytypes.SignalMetrics,
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
}
expression, err := fm.FieldFor(context.Background(), valuer.UUID{}, 0, 0, key)
require.NoError(t, err)
for _, member := range []string{
"resource_db.system.name", "resource_db_system_name", "db.system.name", "db_system_name",
"resource_db.system", "resource_db_system", "db.system", "db_system",
} {
assert.Contains(t, expression, "'"+member+"'")
}
assert.Less(t, strings.Index(expression, "resource_db.system.name"), strings.Index(expression, "resource_db.system'"))
exact := *key
exact.SemconvMembers = []string{"resource_db_system"}
expression, err = fm.FieldFor(context.Background(), valuer.UUID{}, 0, 0, &exact)
require.NoError(t, err)
assert.Equal(t, "JSONExtractString(labels, 'resource_db_system')", expression)
}
func TestGetColumn(t *testing.T) {
ctx := context.Background()

View File

@@ -220,6 +220,9 @@ func (c *conditionBuilder) ConditionFor(
}
matches := querybuilder.MatchingFieldKeys(key, fieldKeys)
if options.ExactSemconv {
matches = querybuilder.MatchingFieldKeysExact(key, fieldKeys)
}
skipResourceFilter := options.SkipResourceFilter
keys, warning := querybuilder.ResolveKeys(key, matches)
@@ -265,6 +268,9 @@ func (c *conditionBuilder) ConditionFor(
synthesized = true
warnings = append(warnings, querybuilder.NewKeyNotFoundWarning(key.Name))
}
if options.ExactSemconv {
keys = querybuilder.ExactSemconvKeys(keys)
}
// When a resource sub-query already covers the term, drop resource keys from the main
// query. Synthesized keys are exempt: the sub-query skips keys absent from metadata.

View File

@@ -362,7 +362,10 @@ func (m *fieldMapper) resolveColumnExprs(
switch valueType := column.Type.(schema.MapColumnType).ValueType; valueType.GetType() {
case schema.ColumnTypeEnumString, schema.ColumnTypeEnumFloat64, schema.ColumnTypeEnumBool:
members := traceSemconvMembers(key)
if len(members) > 1 {
if key.Materialized && (len(members) == 1 || key.MaterializedSemconv) {
exprs = append(exprs, telemetrytypes.FieldKeyToMaterializedColumnName(key))
existExprs = append(existExprs, telemetrytypes.FieldKeyToMaterializedColumnNameForExists(key))
} else if len(members) > 1 {
guards := make([]string, 0, len(members))
for _, member := range members {
guards = append(guards, fmt.Sprintf("mapContains(%s, '%s')", columnName, member))
@@ -381,12 +384,6 @@ func (m *fieldMapper) resolveColumnExprs(
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, members[0]))
existExprs = append(existExprs, fmt.Sprintf("mapContains(%s, '%s')", columnName, members[0]))

View File

@@ -5,6 +5,8 @@ import (
"testing"
"time"
schema "github.com/SigNoz/signoz-otel-collector/cmd/signozschemamigrator/schema_migrator"
"github.com/SigNoz/signoz/pkg/querybuilder"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
@@ -348,3 +350,27 @@ func TestColumnExpressionForTimestampAttributeCollision(t *testing.T) {
assert.Contains(t, result, "attributes_number['timestamp']")
})
}
func TestDBSystemFamilyUsesSemconvAwareMaterializedColumn(t *testing.T) {
fm := NewFieldMapper()
key := &telemetrytypes.TelemetryFieldKey{
Name: "db.system.name",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
Materialized: true,
MaterializedColumnName: "attribute_string_db$$system",
MaterializedSemconv: true,
SemconvMembers: []string{"db.system.name", "db.system"},
}
expression, err := fm.FieldFor(context.Background(), valuer.UUID{}, 0, 0, key)
require.NoError(t, err)
assert.Equal(t, "`attribute_string_db$$system`", expression)
exists, err := querybuilder.ExistsExpression(
[]*schema.Column{indexV3Columns["attributes_string"]}, key, 0, 0, expression, true,
)
require.NoError(t, err)
assert.Equal(t, "`attribute_string_db$$system_exists`", exists)
}

View File

@@ -45,6 +45,10 @@ type ConditionBuilder interface {
type ConditionBuilderOptions struct {
// SkipResourceFilter drops the resource context from the candidate set.
SkipResourceFilter bool
// ExactSemconv disables semantic-convention family expansion. It is an
// internal escape hatch for diagnostics and migrations that must address one
// physical spelling only; public query APIs continue to resolve families.
ExactSemconv bool
}
type AggExprRewriter interface {
// Rewrite rewrites the aggregation expression to be used in the query.

View File

@@ -25,10 +25,22 @@ type Result struct {
}
type ExecStats struct {
RowsScanned uint64 `json:"rowsScanned"`
BytesScanned uint64 `json:"bytesScanned"`
DurationMS uint64 `json:"durationMs"`
StepIntervals map[string]uint64 `json:"stepIntervals,omitempty"`
RowsScanned uint64 `json:"rowsScanned"`
BytesScanned uint64 `json:"bytesScanned"`
DurationMS uint64 `json:"durationMs"`
StepIntervals map[string]uint64 `json:"stepIntervals,omitempty"`
SemconvResolutions []SemconvResolution `json:"semconvResolutions,omitempty"`
}
// SemconvResolution records a semantic-convention family that the query
// builder resolved. Requested preserves the spelling supplied by the caller so
// agents and editors can update their next query without changing response
// labels in the current response.
type SemconvResolution struct {
Requested string `json:"requested"`
Current string `json:"current"`
Members []string `json:"members"`
Kind string `json:"kind"`
}
var _ jsonschema.Preparer = &ExecStats{}

View File

@@ -46,6 +46,13 @@ type TelemetryFieldKey struct {
JSONPlan JSONAccessPlan `json:"-"`
Indexes []TelemetryFieldKeySkipIndex `json:"-"`
Materialized bool `json:"-"` // refers to promoted in case of body.... fields
// MaterializedColumnName preserves the physical column when its DEFAULT
// expression resolves a newer semantic-convention key than the historical
// column identifier (for example db.system.name in db$$system).
MaterializedColumnName string `json:"-"`
// MaterializedSemconv is true when the column's DEFAULT expression already
// coalesces every enabled family member and is therefore safe for a family query.
MaterializedSemconv bool `json:"-"`
Evolutions []*EvolutionEntry `json:"-"`
SemconvMembers []string `json:"-"`
@@ -127,6 +134,8 @@ func (f *TelemetryFieldKey) OverrideMetadataFrom(src *TelemetryFieldKey) {
f.FieldDataType = src.FieldDataType
f.Indexes = src.Indexes
f.Materialized = src.Materialized
f.MaterializedColumnName = src.MaterializedColumnName
f.MaterializedSemconv = src.MaterializedSemconv
f.JSONPlan = src.JSONPlan
f.Evolutions = src.Evolutions
f.SemconvMembers = src.SemconvMembers
@@ -204,6 +213,9 @@ func TelemetryFieldKeyToText(key *TelemetryFieldKey) string {
}
func FieldKeyToMaterializedColumnName(key *TelemetryFieldKey) string {
if key.MaterializedColumnName != "" {
return fmt.Sprintf("`%s`", key.MaterializedColumnName)
}
return fmt.Sprintf("`%s_%s_%s`",
key.FieldContext.String,
fieldDataTypes[key.FieldDataType.StringValue()].StringValue(),
@@ -212,6 +224,9 @@ func FieldKeyToMaterializedColumnName(key *TelemetryFieldKey) string {
}
func FieldKeyToMaterializedColumnNameForExists(key *TelemetryFieldKey) string {
if key.MaterializedColumnName != "" {
return fmt.Sprintf("`%s_exists`", key.MaterializedColumnName)
}
return fmt.Sprintf("`%s_%s_%s_exists`",
key.FieldContext.String,
fieldDataTypes[key.FieldDataType.StringValue()].StringValue(),
@@ -278,6 +293,28 @@ type GettableFieldValues struct {
Complete bool `json:"complete" required:"true"`
}
// PostableSemconvMigrationReportParams selects the metadata window used to
// find services that still emit only historical semantic-convention names.
type PostableSemconvMigrationReportParams struct {
StartUnixMilli int64 `query:"startUnixMilli"`
EndUnixMilli int64 `query:"endUnixMilli"`
}
type SemconvMigrationReportEntry struct {
Current string `json:"current"`
Old string `json:"old"`
Signal string `json:"signal"`
Services []string `json:"services"`
ResourceSets uint64 `json:"resourceSets"`
LastSeenUnixMilli int64 `json:"lastSeenUnixMilli"`
}
type GettableSemconvMigrationReport struct {
StartUnixMilli int64 `json:"startUnixMilli"`
EndUnixMilli int64 `json:"endUnixMilli"`
Entries []*SemconvMigrationReportEntry `json:"entries" required:"true"`
}
type PostableFieldValueParams struct {
PostableFieldKeysParams
Name string `query:"name"`

View File

@@ -26,6 +26,10 @@ type MetadataStore interface {
// GetAllValues returns a list of all values.
GetAllValues(ctx context.Context, orgID valuer.UUID, fieldValueSelector *FieldValueSelector) (*TelemetryFieldValues, bool, error)
// GetSemconvMigrationReport returns services whose metadata contains an old
// semantic-convention name but no current member of that family.
GetSemconvMigrationReport(ctx context.Context, orgID valuer.UUID, startUnixMilli, endUnixMilli int64) (*GettableSemconvMigrationReport, error)
// FetchTemporality fetches the temporality for metric
FetchTemporality(ctx context.Context, orgID valuer.UUID, queryTimeRangeStartTs, queryTimeRangeEndTs uint64, metricName string) (metrictypes.Temporality, error)

View File

@@ -23,7 +23,19 @@ type MockMetadataStore struct {
ColumnEvolutionMetadataMap map[string][]*telemetrytypes.EvolutionEntry
LookupKeysMap map[telemetrytypes.MetricMetadataLookupKey]int64
// StaticFields holds signal-specific intrinsic field definitions (e.g. logstelemetryschema.IntrinsicFields).
StaticFields map[string]telemetrytypes.TelemetryFieldKey
StaticFields map[string]telemetrytypes.TelemetryFieldKey
SemconvMigrationReport *telemetrytypes.GettableSemconvMigrationReport
}
func (m *MockMetadataStore) GetSemconvMigrationReport(_ context.Context, _ valuer.UUID, startUnixMilli, endUnixMilli int64) (*telemetrytypes.GettableSemconvMigrationReport, error) {
if m.SemconvMigrationReport != nil {
return m.SemconvMigrationReport, nil
}
return &telemetrytypes.GettableSemconvMigrationReport{
StartUnixMilli: startUnixMilli,
EndUnixMilli: endUnixMilli,
Entries: []*telemetrytypes.SemconvMigrationReportEntry{},
}, nil
}
// NewMockMetadataStore creates a new instance of MockMetadataStore with initialized maps.

View File

@@ -7,5 +7,31 @@ default_enabled: false
families:
deployment.environment.name:
enabled: true
contexts: [resource, attribute]
signals: [traces, logs, metrics]
db.system.name:
enabled: true
contexts: [resource, attribute]
signals: [traces, logs, metrics]
# These metric renames predate the schema history vendored above. Keep them
# in the same generated registry so every v5 metric query uses one source of
# truth instead of the legacy hand-written transition table.
k8s.pod.cpu.usage:
enabled: true
kind: metric
old: [k8s.pod.cpu.utilization]
contexts: [metric]
signals: [metrics]
k8s.node.cpu.usage:
enabled: true
kind: metric
old: [k8s.node.cpu.utilization]
contexts: [metric]
signals: [metrics]
container.cpu.usage:
enabled: true
kind: metric
old: [container.cpu.utilization]
contexts: [metric]
signals: [metrics]

View File

@@ -0,0 +1,157 @@
"""Phase 2 semantic-convention checks across logs and metrics."""
from collections.abc import Callable
from datetime import UTC, datetime, timedelta
from http import HTTPStatus
import requests
from fixtures import querier, types
from fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD
from fixtures.logs import Logs
from fixtures.metrics import Metrics
DB_CURRENT = "db.system.name"
DB_OLD = "db.system"
METRIC_CURRENT = "container.cpu.usage"
METRIC_OLD = "container.cpu.utilization"
PREFIX = "semconv-phase2"
def _raw_log_bodies(
signoz: types.SigNoz,
token: str,
now: datetime,
expression: str,
) -> set[str]:
response = querier.make_query_request(
signoz,
token,
start_ms=int((now - timedelta(minutes=2)).timestamp() * 1000),
end_ms=int((now + timedelta(minutes=1)).timestamp() * 1000),
request_type=querier.RequestType.RAW,
queries=[querier.build_raw_query("A", "logs", limit=100, filter_expression=expression)],
)
assert response.status_code == HTTPStatus.OK, response.text
return {row["data"]["body"] for row in querier.get_rows(response)}
def _field_values(signoz: types.SigNoz, token: str, signal: str, name: str, context: str) -> set[str]:
response = requests.get(
signoz.self.host_configs["8080"].get("/api/v1/fields/values"),
timeout=5,
headers={"authorization": f"Bearer {token}"},
params={
"signal": signal,
"name": name,
"fieldContext": context,
"fieldDataType": "string",
},
)
assert response.status_code == HTTPStatus.OK, response.text
return set(response.json()["data"]["values"].get("stringValues") or [])
def test_logs_resolve_db_system_family(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[list[Logs]], None],
) -> None:
now = datetime.now(tz=UTC).replace(microsecond=0) - timedelta(minutes=1)
rows = [
("old", {DB_OLD: "postgresql"}),
("current", {DB_CURRENT: "postgresql"}),
("conflict", {DB_OLD: "mysql", DB_CURRENT: "postgresql"}),
("missing", {}),
]
insert_logs(
[
Logs(
timestamp=now + timedelta(seconds=index),
resources={"service.name": PREFIX, **attributes},
attributes=attributes,
body=f"{PREFIX}-{suffix}",
)
for index, (suffix, attributes) in enumerate(rows)
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
present = {f"{PREFIX}-old", f"{PREFIX}-current", f"{PREFIX}-conflict"}
for context in ("attribute", "resource"):
for requested in (DB_CURRENT, DB_OLD):
field = f"{context}.{requested}"
assert _raw_log_bodies(signoz, token, now, f'{field} = "postgresql"') == present
assert _raw_log_bodies(signoz, token, now, f"{field} EXISTS") == present
assert _raw_log_bodies(signoz, token, now, f"{field} NOT EXISTS") == {f"{PREFIX}-missing"}
assert _field_values(signoz, token, "logs", requested, context) == {"postgresql", "mysql"}
def test_metrics_resolve_label_and_metric_name_families(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_metrics: Callable[[list[Metrics]], None],
) -> None:
now = datetime.now(tz=UTC).replace(second=0, microsecond=0)
insert_metrics(
[
Metrics(
metric_name=METRIC_CURRENT,
labels={DB_CURRENT: "postgresql"},
timestamp=now - timedelta(seconds=3),
temporality="Unspecified",
type_="Gauge",
is_monotonic=False,
value=10,
),
Metrics(
metric_name=METRIC_OLD,
labels={"db_system": "mysql"},
timestamp=now - timedelta(seconds=2),
temporality="Unspecified",
type_="Gauge",
is_monotonic=False,
value=20,
),
Metrics(
metric_name=METRIC_OLD,
labels={DB_OLD: "mysql", DB_CURRENT: "postgresql", "series": "conflict"},
timestamp=now - timedelta(seconds=1),
temporality="Unspecified",
type_="Gauge",
is_monotonic=False,
value=30,
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
for metric_name in (METRIC_CURRENT, METRIC_OLD):
for requested_label in (DB_CURRENT, DB_OLD):
response = querier.make_scalar_query_request(
signoz,
token,
now,
[
querier.build_scalar_query(
name="A",
signal="metrics",
aggregations=[
querier.build_metrics_aggregation(
metric_name,
"latest",
"sum",
"unspecified",
reduce_to="last",
)
],
group_by=[querier.build_group_by_field(requested_label, "string", "attribute")],
filter_expression=f"attribute.{requested_label} EXISTS",
)
],
)
assert response.status_code == HTTPStatus.OK, response.text
data = {row[0]: row[-1] for row in querier.get_scalar_table_data(response.json())}
assert data == {"postgresql": 40.0, "mysql": 20.0}, (metric_name, requested_label, data)

View File

@@ -84,10 +84,7 @@ def semconv_phase1_data(
yield now
cluster = clickhouse.env["SIGNOZ_TELEMETRYSTORE_CLICKHOUSE_CLUSTER"]
clickhouse.conn.command(
f"ALTER TABLE signoz_traces.dependency_graph_minutes_v2 ON CLUSTER '{cluster}' "
f"DELETE WHERE startsWith(src, '{PREFIX}-map-') SETTINGS mutations_sync = 1"
)
clickhouse.conn.command(f"ALTER TABLE signoz_traces.dependency_graph_minutes_v2 ON CLUSTER '{cluster}' DELETE WHERE startsWith(src, '{PREFIX}-map-') SETTINGS mutations_sync = 1")
def _result(response: requests.Response) -> dict[str, Any]:
@@ -148,6 +145,31 @@ def test_semconv_phase1_mixed_sdk_generations( # pylint: disable=too-many-state
now = semconv_phase1_data
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# A builder response records the exact spelling that was resolved. Raw SQL
# and PromQL deliberately do not use this resolver.
resolution_response = make_query_request(
signoz,
token,
start_ms=int((now - timedelta(minutes=2)).timestamp() * 1000),
end_ms=int((now + timedelta(minutes=1)).timestamp() * 1000),
request_type=RequestType.RAW,
queries=[
BuilderQuery(
signal="traces",
name="A",
limit=1,
filter_expression=f"resource.{OLD} EXISTS",
).to_dict()
],
)
assert resolution_response.status_code == HTTPStatus.OK, resolution_response.text
assert {
"requested": OLD,
"current": CURRENT,
"members": [CURRENT, OLD],
"kind": "attribute",
} in resolution_response.json()["data"]["meta"]["semconvResolutions"]
# Resource and span-attribute paths share the same matrix. Run every
# operator with both the saved-query (old) and current request spellings.
for context in ("resource", "attribute"):
@@ -237,3 +259,16 @@ def test_semconv_phase1_mixed_sdk_generations( # pylint: disable=too-many-state
)
assert map_response.status_code == HTTPStatus.OK, map_response.text
assert {edge["parent"] for edge in map_response.json()} == {f"{PREFIX}-map-production"}
report_response = requests.get(
signoz.self.host_configs["8080"].get("/api/v1/fields/semconv-migration"),
timeout=30,
headers={"authorization": f"Bearer {token}"},
params={
"startUnixMilli": int((now - timedelta(minutes=2)).timestamp() * 1000),
"endUnixMilli": int((now + timedelta(minutes=1)).timestamp() * 1000),
},
)
assert report_response.status_code == HTTPStatus.OK, report_response.text
entry = next(item for item in report_response.json()["data"]["entries"] if item["current"] == CURRENT and item["old"] == OLD and item["signal"] == "traces")
assert set(entry["services"]) == {f"{PREFIX}-old", f"{PREFIX}-staging"}