test: pin the semconv family matrix on a live stack

A dedicated package runs SigNoz with resolve_semconv_families on and a
second instance with the flag at its default. The fleet has one
identity per state: OLD (old spelling only), NEW (current only), BOTH
(current staging and old production), NEITHER (keyless).

- 36 filter cells: nine operators, both spellings, both contexts. The
  result sets do not depend on the requested spelling; the current
  spelling wins on the conflict row; negative operators keep keyless
  rows, exactly like a single key.
- Group-by merges the fleet into one production group, a staging group,
  and a NULL group, and the group column carries the requested spelling.
- A bare name with the family under two contexts warns and keeps the
  resource side; the family survives the collision as one unit.
- Logs stay literal with the flag on, and everything stays literal with
  the flag off.

Verified: 42 passed against the live stack.

Assisted-by: Claude Fable 5

.githooks/install

@@ -1,30 +0,0 @@
-#!/bin/sh
-# Installs dispatcher stubs into the repository's hooks directory. Safe to run
-# repeatedly; invoked automatically from the Makefile and from the frontend
-# postinstall so no manual setup is required.
-
-set -e
-
-git rev-parse --git-dir >/dev/null 2>&1 || exit 0
-
-# core.hooksPath shadows the hooks directory git would otherwise use, and an
-# absolute value leaks one worktree's path into every other worktree.
-if [ -n "$(git config --get core.hooksPath 2>/dev/null || true)" ]; then
-	git config --unset-all core.hooksPath 2>/dev/null || true
-fi
-
-# Resolves to the common directory for linked worktrees, which is where git
-# looks for hooks regardless of which worktree the commit happens in.
-hooks_dir="$(git rev-parse --git-path hooks)"
-mkdir -p "$hooks_dir"
-
-for hook in pre-commit commit-msg; do
-	cat >"$hooks_dir/$hook" <<EOF
-#!/bin/sh
-# Generated by .githooks/install.
-root="\$(git rev-parse --show-toplevel)"
-[ -x "\$root/.githooks/$hook" ] || exit 0
-exec "\$root/.githooks/$hook" "\$@"
-EOF
-	chmod +x "$hooks_dir/$hook"
-done

.githooks/pre-commit

@@ -1,22 +0,0 @@
-#!/bin/sh
-
-set -e
-
-staged_go_files=$(git diff --cached --name-only --diff-filter=ACM -- '*.go')
-if [ -n "$staged_go_files" ]; then
-	if command -v gofmt >/dev/null 2>&1; then
-		echo "$staged_go_files" | xargs gofmt -l -w
-		echo "$staged_go_files" | xargs git add
-	else
-		echo "pre-commit: gofmt not found, skipping go formatting" >&2
-	fi
-fi
-
-staged_web_files=$(git diff --cached --name-only --diff-filter=ACM -- 'frontend/*')
-if [ -n "$staged_web_files" ]; then
-	if [ -d frontend/node_modules ] && command -v pnpm >/dev/null 2>&1; then
-		cd frontend && pnpm lint-staged
-	else
-		echo "pre-commit: frontend dependencies not installed, skipping lint-staged" >&2
-	fi
-fi

Makefile

@@ -35,13 +35,6 @@ DOCKERFILE_ENTERPRISE 			= $(SRC)/cmd/enterprise/Dockerfile
 DOCKER_REGISTRY_ENTERPRISE 		?= docker.io/signoz/signoz
 JS_BUILD_CONTEXT 				= $(SRC)/frontend
 
-##############################################################
-# git hooks
-##############################################################
-# Hooks live outside the working tree, so they cannot be checked in. Expanded
-# while this file is parsed, which wires them up on any make invocation.
-GITHOOKS_INSTALL := $(shell sh $(SRC)/.githooks/install >/dev/null 2>&1 || true)
-
 ##############################################################
 # directories
 ##############################################################

frontend/.husky/commit-msg

@@ -1,12 +1,7 @@
 #!/bin/sh
+. "$(dirname "$0")/_/husky.sh"
 
-set -e
-
-if [ -d frontend/node_modules ] && command -v pnpm >/dev/null 2>&1; then
-	(cd frontend && pnpm run commitlint --edit "$1")
-else
-	echo "commit-msg: frontend dependencies not installed, skipping commitlint" >&2
-fi
+cd frontend && pnpm run commitlint --edit $1
 
 branch="$(git rev-parse --abbrev-ref HEAD)"
 

frontend/.husky/pre-commit

@@ -0,0 +1,4 @@
+#!/bin/sh
+. "$(dirname "$0")/_/husky.sh"
+
+cd frontend && pnpm lint-staged

frontend/package.json

@@ -20,7 +20,8 @@
     "jest": "jest",
     "jest:coverage": "jest --coverage",
     "jest:watch": "jest --watch",
-    "postinstall": "pnpm i18n:generate-hash && (is-ci || sh ../.githooks/install) && node scripts/update-registry.cjs",
+    "postinstall": "pnpm i18n:generate-hash && (is-ci || pnpm husky:configure)  && node scripts/update-registry.cjs",
+    "husky:configure": "cd .. && husky install frontend/.husky && cd frontend && chmod ug+x .husky/*",
     "commitlint": "commitlint --edit $1",
     "test": "jest",
     "test:changedsince": "jest --changedSince=main --coverage --silent",
@@ -185,6 +186,7 @@
     "@typescript/native-preview": "7.0.0-dev.20260430.1",
     "eslint-plugin-sonarjs": "4.0.2",
     "glob": "^13.0.6",
+    "husky": "^7.0.4",
     "is-ci": "^3.0.1",
     "jest-environment-jsdom": "29.7.0",
     "jest-styled-components": "^7.2.0",

frontend/pnpm-lock.yaml

@@ -440,6 +440,9 @@ importers:
       glob:
         specifier: ^13.0.6
         version: 13.0.6
+      husky:
+        specifier: ^7.0.4
+        version: 7.0.4
       is-ci:
         specifier: ^3.0.1
         version: 3.0.1
@@ -5526,6 +5529,11 @@ packages:
     resolution: {integrity: sha512-eKCa6bwnJhvxj14kZk5NCPc6Hb6BdsU9DZcOnmQKSnO1VKrfV0zCvtttPZUsBvjmNDn8rpcJfpwSYnHBjc95MQ==}
     engines: {node: '>=18.18.0'}
 
+  husky@7.0.4:
+    resolution: {integrity: sha512-vbaCKN2QLtP/vD4yvs6iz6hBEo6wkSzs8HpRah1Z6aGmF2KW5PdYuAd7uX5a+OyBZHBhd+TFLqgjUgytQr4RvQ==}
+    engines: {node: '>=12'}
+    hasBin: true
+
   hyphenate-style-name@1.0.4:
     resolution: {integrity: sha512-ygGZLjmXfPHj+ZWh6LwbC37l43MhfztxetbFCoYTM2VjkIUpeHgSNn7QIyVFj7YQ1Wl9Cbw5sholVJPzWvC2MQ==}
 
@@ -14628,6 +14636,8 @@ snapshots:
 
   human-signals@8.0.1: {}
 
+  husky@7.0.4: {}
+
   hyphenate-style-name@1.0.4: {}
 
   i18next-browser-languagedetector@6.1.8:

frontend/src/container/PipelinePage/PipelineListsView/AddNewProcessor/ProcessorForm.tsx

@@ -2,17 +2,14 @@ import { useTranslation } from 'react-i18next';
 import { Input } from '@signozhq/ui/input';
 import { Switch } from '@signozhq/ui/switch';
 import { Form, Select, Space } from 'antd';
-import { FeatureKeys } from 'constants/features';
 import { ModalFooterTitle } from 'container/PipelinePage/styles';
-import { useAppContext } from 'providers/App/App';
 import { ProcessorData } from 'types/api/pipeline/def';
 
 import { formValidationRules } from '../config';
-import { ProcessorFormField } from './config';
+import { processorFields, ProcessorFormField } from './config';
 import CSVInput from './FormFields/CSVInput';
 import JsonFlattening from './FormFields/JsonFlattening';
 import { FormWrapper, PipelineIndexIcon, StyledSelect } from './styles';
-import { resolveProcessorFields } from './utils';
 
 import './styles.scss';
 
@@ -136,23 +133,16 @@ function ProcessorForm({
 	selectedProcessorData,
 	isAdd,
 }: ProcessorFormProps): JSX.Element {
-	const { featureFlags } = useAppContext();
-	const isBodyJsonEnabled =
-		featureFlags?.find((flag) => flag.name === FeatureKeys.USE_JSON_BODY)
-			?.active || false;
-
 	return (
 		<div className="processor-form-container">
-			{resolveProcessorFields(processorType, isBodyJsonEnabled).map(
-				(fieldData: ProcessorFormField) => (
-					<ProcessorFieldInput
-						key={fieldData.name + String(fieldData.initialValue)}
-						fieldData={fieldData}
-						selectedProcessorData={selectedProcessorData}
-						isAdd={isAdd}
-					/>
-				),
-			)}
+			{processorFields[processorType]?.map((fieldData: ProcessorFormField) => (
+				<ProcessorFieldInput
+					key={fieldData.name + String(fieldData.initialValue)}
+					fieldData={fieldData}
+					selectedProcessorData={selectedProcessorData}
+					isAdd={isAdd}
+				/>
+			))}
 		</div>
 	);
 }

frontend/src/container/PipelinePage/PipelineListsView/AddNewProcessor/utils.ts

@@ -1,24 +0,0 @@
-import { processorFields, ProcessorFormField } from './config';
-
-const BODY_PARSE_FROM = 'body';
-const JSON_BODY_PARSE_FROM = 'body.message';
-
-// With use_json_body the collector normalizes every body into a map before user
-// operators run, so a parser pointed at `body` gets a map it cannot read and
-// silently extracts nothing. The log text lives at body.message.
-export function resolveProcessorFields(
-	processorType: string,
-	isBodyJsonEnabled: boolean,
-): Array<ProcessorFormField> {
-	const fields = processorFields[processorType] ?? [];
-
-	if (!isBodyJsonEnabled) {
-		return fields;
-	}
-
-	return fields.map((field) =>
-		field.name === 'parse_from' && field.initialValue === BODY_PARSE_FROM
-			? { ...field, initialValue: JSON_BODY_PARSE_FROM }
-			: field,
-	);
-}

frontend/src/container/PipelinePage/tests/processorFieldDefaults.test.ts

@@ -1,45 +0,0 @@
-import { processorFields } from '../PipelineListsView/AddNewProcessor/config';
-import { resolveProcessorFields } from '../PipelineListsView/AddNewProcessor/utils';
-
-const parseFromDefault = (
-	fields: ReturnType<typeof resolveProcessorFields>,
-): unknown => fields.find((field) => field.name === 'parse_from')?.initialValue;
-
-describe('resolveProcessorFields', () => {
-	it.each(['grok_parser', 'regex_parser', 'json_parser'])(
-		'defaults %s parse_from to body.message when use_json_body is on',
-		(processorType) => {
-			expect(parseFromDefault(resolveProcessorFields(processorType, true))).toBe(
-				'body.message',
-			);
-		},
-	);
-
-	it.each(['grok_parser', 'regex_parser', 'json_parser'])(
-		'keeps %s parse_from as body when use_json_body is off',
-		(processorType) => {
-			expect(parseFromDefault(resolveProcessorFields(processorType, false))).toBe(
-				'body',
-			);
-		},
-	);
-
-	it('leaves parse_from defaults that do not point at the body alone', () => {
-		expect(parseFromDefault(resolveProcessorFields('time_parser', true))).toBe(
-			'attributes.timestamp',
-		);
-		expect(
-			parseFromDefault(resolveProcessorFields('severity_parser', true)),
-		).toBe('attributes.logLevel');
-	});
-
-	it('does not mutate the shared config', () => {
-		resolveProcessorFields('grok_parser', true);
-
-		expect(parseFromDefault(processorFields.grok_parser)).toBe('body');
-	});
-
-	it('returns an empty list for an unknown processor type', () => {
-		expect(resolveProcessorFields('does_not_exist', true)).toStrictEqual([]);
-	});
-});

pkg/querier/builder_query.go

@@ -29,7 +29,6 @@ type builderQuery[T any] struct {
 	telemetryStore telemetrystore.TelemetryStore
 	orgID          valuer.UUID
 	stmtBuilder    qbtypes.StatementBuilder[T]
-	queryType      qbtypes.QueryType
 	spec           qbtypes.QueryBuilderQuery[T]
 	variables      map[string]qbtypes.VariableItem
 
@@ -52,7 +51,6 @@ func newBuilderQuery[T any](
 	telemetryStore telemetrystore.TelemetryStore,
 	orgID valuer.UUID,
 	stmtBuilder qbtypes.StatementBuilder[T],
-	queryType qbtypes.QueryType,
 	spec qbtypes.QueryBuilderQuery[T],
 	tr qbtypes.TimeRange,
 	kind qbtypes.RequestType,
@@ -64,7 +62,6 @@ func newBuilderQuery[T any](
 		telemetryStore: telemetryStore,
 		orgID:          orgID,
 		stmtBuilder:    stmtBuilder,
-		queryType:      queryType,
 		spec:           spec,
 		variables:      variables,
 		fromMS:         tr.From,
@@ -84,7 +81,7 @@ func (q *builderQuery[T]) Fingerprint() string {
 
 	// Create a deterministic fingerprint for builder queries
 	// This needs to include all fields that affect the query results
-	parts := []string{q.queryType.StringValue()}
+	parts := []string{"builder"}
 
 	// Add signal type
 	parts = append(parts, fmt.Sprintf("signal=%s", q.spec.Signal.StringValue()))

pkg/querier/builder_query_test.go

@@ -3,7 +3,6 @@ package querier
 import (
 	"strings"
 	"testing"
-	"time"
 
 	"github.com/SigNoz/signoz/pkg/querybuilder"
 	qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
@@ -21,8 +20,7 @@ func TestBuilderQueryFingerprint(t *testing.T) {
 		{
 			name: "fingerprint includes shiftby when ShiftBy field is set",
 			query: &builderQuery[qbtypes.MetricAggregation]{
-				queryType: qbtypes.QueryTypeBuilder,
-				kind:      qbtypes.RequestTypeTimeSeries,
+				kind: qbtypes.RequestTypeTimeSeries,
 				spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
 					Signal:  telemetrytypes.SignalMetrics,
 					ShiftBy: 3600,
@@ -42,8 +40,7 @@ func TestBuilderQueryFingerprint(t *testing.T) {
 		{
 			name: "fingerprint includes shiftby but not other functions",
 			query: &builderQuery[qbtypes.MetricAggregation]{
-				queryType: qbtypes.QueryTypeBuilder,
-				kind:      qbtypes.RequestTypeTimeSeries,
+				kind: qbtypes.RequestTypeTimeSeries,
 				spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
 					Signal:  telemetrytypes.SignalMetrics,
 					ShiftBy: 3600,
@@ -66,8 +63,7 @@ func TestBuilderQueryFingerprint(t *testing.T) {
 		{
 			name: "no shiftby in fingerprint when ShiftBy is zero",
 			query: &builderQuery[qbtypes.MetricAggregation]{
-				queryType: qbtypes.QueryTypeBuilder,
-				kind:      qbtypes.RequestTypeTimeSeries,
+				kind: qbtypes.RequestTypeTimeSeries,
 				spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
 					Signal:  telemetrytypes.SignalMetrics,
 					ShiftBy: 0,
@@ -98,29 +94,6 @@ func TestBuilderQueryFingerprint(t *testing.T) {
 	}
 }
 
-func TestBuilderQueryFingerprintQueryType(t *testing.T) {
-	spec := qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]{
-		Signal:       telemetrytypes.SignalTraces,
-		StepInterval: qbtypes.Step{Duration: 60 * time.Second},
-		Aggregations: []qbtypes.TraceAggregation{{Expression: "count()"}},
-		Filter:       &qbtypes.Filter{Expression: "gen_ai.request.model EXISTS"},
-	}
-	regular := &builderQuery[qbtypes.TraceAggregation]{
-		queryType: qbtypes.QueryTypeBuilder,
-		kind:      qbtypes.RequestTypeTimeSeries,
-		spec:      spec,
-	}
-	ai := &builderQuery[qbtypes.TraceAggregation]{
-		queryType: qbtypes.QueryTypeBuilderAI,
-		kind:      qbtypes.RequestTypeTimeSeries,
-		spec:      spec,
-	}
-
-	assert.True(t, strings.HasPrefix(regular.Fingerprint(), qbtypes.QueryTypeBuilder.StringValue()+"&"))
-	assert.True(t, strings.HasPrefix(ai.Fingerprint(), qbtypes.QueryTypeBuilderAI.StringValue()+"&"))
-	assert.NotEqual(t, regular.Fingerprint(), ai.Fingerprint())
-}
-
 func TestMakeBucketsOrder(t *testing.T) {
 	// Test that makeBuckets returns buckets in reverse chronological order by default
 	// Using milliseconds as input - need > 1 hour range to get multiple buckets

pkg/querier/querier.go

@@ -305,7 +305,7 @@ func (q *querier) buildQueries(
 			}
 			spec.ShiftBy = extractShiftFromBuilderQuery(spec)
 			timeRange := adjustTimeRangeForShift(spec, qbtypes.TimeRange{From: req.Start, To: req.End}, req.RequestType)
-			bq := newBuilderQuery(q.logger, q.telemetryStore, orgID, q.aiTraceStmtBuilder, query.Type, spec, timeRange, req.RequestType, tmplVars, builderConfig{})
+			bq := newBuilderQuery(q.logger, q.telemetryStore, orgID, q.aiTraceStmtBuilder, spec, timeRange, req.RequestType, tmplVars, builderConfig{})
 			queries[spec.Name] = bq
 			steps[spec.Name] = spec.StepInterval
 		case qbtypes.QueryTypeBuilder:
@@ -313,7 +313,7 @@ func (q *querier) buildQueries(
 			case qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]:
 				spec.ShiftBy = extractShiftFromBuilderQuery(spec)
 				timeRange := adjustTimeRangeForShift(spec, qbtypes.TimeRange{From: req.Start, To: req.End}, req.RequestType)
-				bq := newBuilderQuery(q.logger, q.telemetryStore, orgID, q.traceStmtBuilder, query.Type, spec, timeRange, req.RequestType, tmplVars, builderConfig{})
+				bq := newBuilderQuery(q.logger, q.telemetryStore, orgID, q.traceStmtBuilder, spec, timeRange, req.RequestType, tmplVars, builderConfig{})
 				queries[spec.Name] = bq
 				steps[spec.Name] = spec.StepInterval
 			case qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]:
@@ -323,7 +323,7 @@ func (q *querier) buildQueries(
 				if spec.Source == telemetrytypes.SourceAudit {
 					stmtBuilder = q.auditStmtBuilder
 				}
-				bq := newBuilderQuery(q.logger, q.telemetryStore, orgID, stmtBuilder, query.Type, spec, timeRange, req.RequestType, tmplVars, q.builderConfig)
+				bq := newBuilderQuery(q.logger, q.telemetryStore, orgID, stmtBuilder, spec, timeRange, req.RequestType, tmplVars, q.builderConfig)
 				queries[spec.Name] = bq
 				steps[spec.Name] = spec.StepInterval
 			case qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]:
@@ -340,9 +340,9 @@ func (q *querier) buildQueries(
 
 				if spec.Source == telemetrytypes.SourceMeter {
 					event.Source = telemetrytypes.SourceMeter.StringValue()
-					bq = newBuilderQuery(q.logger, q.telemetryStore, orgID, q.meterStmtBuilder, query.Type, spec, timeRange, req.RequestType, tmplVars, builderConfig{})
+					bq = newBuilderQuery(q.logger, q.telemetryStore, orgID, q.meterStmtBuilder, spec, timeRange, req.RequestType, tmplVars, builderConfig{})
 				} else {
-					bq = newBuilderQuery(q.logger, q.telemetryStore, orgID, q.metricStmtBuilder, query.Type, spec, timeRange, req.RequestType, tmplVars, builderConfig{})
+					bq = newBuilderQuery(q.logger, q.telemetryStore, orgID, q.metricStmtBuilder, spec, timeRange, req.RequestType, tmplVars, builderConfig{})
 				}
 
 				queries[spec.Name] = bq
@@ -618,7 +618,7 @@ func (q *querier) QueryRawStream(ctx context.Context, orgID valuer.UUID, req *qb
 			if spec.Source == telemetrytypes.SourceAudit {
 				liveTailStmtBuilder = q.auditStmtBuilder
 			}
-			bq := newBuilderQuery(q.logger, q.telemetryStore, orgID, liveTailStmtBuilder, query.Type, spec, timeRange, req.RequestType, map[string]qbtypes.VariableItem{
+			bq := newBuilderQuery(q.logger, q.telemetryStore, orgID, liveTailStmtBuilder, spec, timeRange, req.RequestType, map[string]qbtypes.VariableItem{
 				"id": {
 					Value: updatedLogID,
 				},
@@ -941,9 +941,8 @@ func (q *querier) createRangedQuery(_ valuer.UUID, originalQuery qbtypes.Query,
 		specCopy := qt.spec.Copy()
 		specCopy.ShiftBy = extractShiftFromBuilderQuery(specCopy)
 		adjustedTimeRange := adjustTimeRangeForShift(specCopy, timeRange, qt.kind)
-		// reuse the original query's statement builder and type so an AI query
-		// keeps its AI builder and cache key
-		return newBuilderQuery(q.logger, q.telemetryStore, qt.orgID, qt.stmtBuilder, qt.queryType, specCopy, adjustedTimeRange, qt.kind, qt.variables, qt.builderConfig)
+		// reuse the original query's statement builder so an AI query keeps its AI builder
+		return newBuilderQuery(q.logger, q.telemetryStore, qt.orgID, qt.stmtBuilder, specCopy, adjustedTimeRange, qt.kind, qt.variables, builderConfig{})
 
 	case *builderQuery[qbtypes.LogAggregation]:
 		specCopy := qt.spec.Copy()
@@ -953,16 +952,16 @@ func (q *querier) createRangedQuery(_ valuer.UUID, originalQuery qbtypes.Query,
 		if qt.spec.Source == telemetrytypes.SourceAudit {
 			shiftStmtBuilder = q.auditStmtBuilder
 		}
-		return newBuilderQuery(q.logger, q.telemetryStore, qt.orgID, shiftStmtBuilder, qt.queryType, specCopy, adjustedTimeRange, qt.kind, qt.variables, q.builderConfig)
+		return newBuilderQuery(q.logger, q.telemetryStore, qt.orgID, shiftStmtBuilder, specCopy, adjustedTimeRange, qt.kind, qt.variables, q.builderConfig)
 
 	case *builderQuery[qbtypes.MetricAggregation]:
 		specCopy := qt.spec.Copy()
 		specCopy.ShiftBy = extractShiftFromBuilderQuery(specCopy)
 		adjustedTimeRange := adjustTimeRangeForShift(specCopy, timeRange, qt.kind)
 		if qt.spec.Source == telemetrytypes.SourceMeter {
-			return newBuilderQuery(q.logger, q.telemetryStore, qt.orgID, q.meterStmtBuilder, qt.queryType, specCopy, adjustedTimeRange, qt.kind, qt.variables, builderConfig{})
+			return newBuilderQuery(q.logger, q.telemetryStore, qt.orgID, q.meterStmtBuilder, specCopy, adjustedTimeRange, qt.kind, qt.variables, builderConfig{})
 		}
-		return newBuilderQuery(q.logger, q.telemetryStore, qt.orgID, q.metricStmtBuilder, qt.queryType, specCopy, adjustedTimeRange, qt.kind, qt.variables, builderConfig{})
+		return newBuilderQuery(q.logger, q.telemetryStore, qt.orgID, q.metricStmtBuilder, specCopy, adjustedTimeRange, qt.kind, qt.variables, builderConfig{})
 	case *traceOperatorQuery:
 		specCopy := qt.spec.Copy()
 		return &traceOperatorQuery{

tests/conftest.py

@@ -19,6 +19,7 @@ pytest_plugins = [
     "fixtures.traces",
     "fixtures.metrics",
     "fixtures.queriercommon",
+    "fixtures.semconvfamilies",
     "fixtures.metadata",
     "fixtures.meter",
     "fixtures.browser",

tests/fixtures/semconvfamilies.py

@@ -0,0 +1,75 @@
+"""Seed data for the semconv family matrix tests.
+
+Four identities cover every fleet state of the deployment.environment(.name)
+family. The tests assert which identities a filter returns, so BOTH (a row
+that carries the two spellings with different values) and NEITHER (a keyless
+row) are the point of most cases.
+
+Each row carries its family pairs in the resource attributes and in the span
+attributes, so one fleet serves the resource-context and attribute-context
+matrices. The same rows exist as logs for the logs literalness guard.
+"""
+
+from collections.abc import Callable, Generator
+from datetime import UTC, datetime, timedelta
+
+import pytest
+
+from fixtures.logs import Logs
+from fixtures.traces import TraceIdGenerator, Traces, TracesKind, TracesStatusCode
+
+PREFIX = "semconv-fam"
+CURRENT_KEY = "deployment.environment.name"
+OLD_KEY = "deployment.environment"
+
+# Row identities. The span name, the log body, and service.name are the identity.
+OLD = f"{PREFIX}-old"  # only the old spelling, value "production"
+NEW = f"{PREFIX}-new"  # only the current spelling, value "production"
+BOTH = f"{PREFIX}-both"  # current "staging" and old "production" - the conflict row
+NEITHER = f"{PREFIX}-neither"  # no member at all
+
+_ROWS = [
+    (OLD, {OLD_KEY: "production"}, timedelta(seconds=4)),
+    (NEW, {CURRENT_KEY: "production"}, timedelta(seconds=3)),
+    (BOTH, {CURRENT_KEY: "staging", OLD_KEY: "production"}, timedelta(seconds=2)),
+    (NEITHER, {}, timedelta(seconds=1)),
+]
+
+
+@pytest.fixture(name="family_fleet", scope="function")
+def family_fleet(
+    insert_logs: Callable[[list[Logs]], None],
+    insert_traces: Callable[[list[Traces]], None],
+) -> Generator[datetime]:
+    """Inserts one span and one log per identity and yields the base
+    timestamp."""
+    now = datetime.now(tz=UTC).replace(microsecond=0) - timedelta(minutes=1)
+
+    insert_traces(
+        [
+            Traces(
+                timestamp=now - offset,
+                duration=timedelta(milliseconds=10),
+                trace_id=TraceIdGenerator.trace_id(),
+                span_id=TraceIdGenerator.span_id(),
+                name=identity,
+                kind=TracesKind.SPAN_KIND_SERVER,
+                status_code=TracesStatusCode.STATUS_CODE_OK,
+                resources={"service.name": identity, **family},
+                attributes=dict(family),
+            )
+            for identity, family, offset in _ROWS
+        ]
+    )
+    insert_logs(
+        [
+            Logs(
+                timestamp=now - offset,
+                body=identity,
+                resources={"service.name": identity, **family},
+                attributes=dict(family),
+            )
+            for identity, family, offset in _ROWS
+        ]
+    )
+    yield now

tests/integration/tests/semconvfamilies/01_family_matrix.py

@@ -0,0 +1,220 @@
+"""The phase-1 matrix for semantic-convention family resolution.
+
+The package runs SigNoz with resolve_semconv_families on. The fleet in
+fixtures/semconvfamilies.py has one identity per state: OLD (old spelling
+only), NEW (current only), BOTH (current "staging" and old "production"),
+NEITHER (keyless). Each case asserts which identities a filter returns, with
+either spelling as the requested name and for both contexts.
+
+The pinned facts:
+  - Both spellings resolve to the same merged field; the result sets do not
+    depend on the requested spelling.
+  - The current spelling wins on a row that carries both (BOTH reads
+    "staging", never "production").
+  - Negative operators keep keyless rows (NEITHER), exactly like a single
+    key; presence stays an explicit EXISTS opt-in.
+  - Logs stay literal: only traces have family support today.
+  - With the flag off, everything stays literal.
+"""
+
+from collections.abc import Callable
+from datetime import datetime, timedelta
+from http import HTTPStatus
+
+import pytest
+
+from fixtures import types
+from fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD
+from fixtures.querier import (
+    RequestType,
+    build_aggregation,
+    build_group_by_field,
+    build_order_by,
+    build_raw_query,
+    build_traces_scalar_query,
+    get_column_data_from_response,
+    make_query_request,
+)
+from fixtures.semconvfamilies import (
+    BOTH,
+    CURRENT_KEY,
+    NEITHER,
+    NEW,
+    OLD,
+    OLD_KEY,
+    PREFIX,
+)
+
+FILTER_MATRIX = [
+    pytest.param("{key} = 'production'", {OLD, NEW}, id="eq_matches_either_spelling"),
+    pytest.param("{key} = 'staging'", {BOTH}, id="eq_current_wins_on_conflict"),
+    pytest.param("{key} != 'production'", {BOTH, NEITHER}, id="neq_keeps_keyless_and_conflict"),
+    pytest.param("{key} IN ['production', 'staging']", {OLD, NEW, BOTH}, id="in_matches_merged_value"),
+    pytest.param("{key} NOT IN ['production']", {BOTH, NEITHER}, id="not_in_keeps_keyless"),
+    pytest.param("{key} LIKE '%prod%'", {OLD, NEW}, id="like_matches_merged_value"),
+    pytest.param("{key} EXISTS", {OLD, NEW, BOTH}, id="exists_is_any_member"),
+    pytest.param("{key} NOT EXISTS", {NEITHER}, id="not_exists_is_no_member"),
+    pytest.param("{key} != 'production' AND {key} EXISTS", {BOTH}, id="neq_composed_with_exists"),
+]
+
+LITERAL_MATRIX = [
+    pytest.param("{key} = 'production'", {NEW}, id="literal_eq_reads_one_spelling"),
+    pytest.param("{key} != 'production'", {OLD, BOTH, NEITHER}, id="literal_neq_reads_one_spelling"),
+]
+
+
+def _trace_identities(
+    signoz: types.SigNoz,
+    token: str,
+    base: datetime,
+    expression: str,
+    signal: str = "traces",
+) -> set[str]:
+    identity_field = "span.name" if signal == "traces" else "body"
+    identity_column = "name" if signal == "traces" else "body"
+    response = make_query_request(
+        signoz,
+        token,
+        start_ms=int((base - timedelta(minutes=2)).timestamp() * 1000),
+        end_ms=int((base + timedelta(minutes=1)).timestamp() * 1000),
+        request_type=RequestType.RAW,
+        queries=[
+            build_raw_query(
+                "A",
+                signal,
+                limit=100,
+                filter_expression=expression,
+                order=[build_order_by("timestamp", "asc")],
+                select_fields=[{"name": identity_field}],
+            )
+        ],
+    )
+    assert response.status_code == HTTPStatus.OK, response.text
+    # Sets keep the assertion stable when the shared stack is reused and older
+    # rows with the same identities remain.
+    return {name for name in get_column_data_from_response(response.json(), identity_column) if name.startswith(PREFIX)}
+
+
+@pytest.mark.parametrize("expression_template,expected", FILTER_MATRIX)
+@pytest.mark.parametrize("requested_key", [CURRENT_KEY, OLD_KEY], ids=["current", "old"])
+@pytest.mark.parametrize("context", ["resource", "attribute"])
+def test_family_filters(
+    signoz: types.SigNoz,
+    create_user_admin: None,  # pylint: disable=unused-argument
+    get_token: Callable[[str, str], str],
+    family_fleet: datetime,
+    context: str,
+    requested_key: str,
+    expression_template: str,
+    expected: set[str],
+) -> None:
+    """One matrix cell: a filter on one spelling, in one context. The result
+    set is a property of the family, not of the requested spelling."""
+    token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
+    expression = expression_template.format(key=f"{context}.{requested_key}")
+    assert _trace_identities(signoz, token, family_fleet, expression) == expected, expression
+
+
+@pytest.mark.parametrize("expression_template,expected", LITERAL_MATRIX)
+def test_flag_off_stays_literal(
+    signoz_families_off: types.SigNoz,
+    create_user_admin: None,  # pylint: disable=unused-argument
+    get_token: Callable[[str, str], str],
+    family_fleet: datetime,
+    expression_template: str,
+    expected: set[str],
+) -> None:
+    """The same fleet through an instance with the flag at its default: the
+    current spelling reads only rows that carry the current spelling."""
+    token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
+    expression = expression_template.format(key=f"resource.{CURRENT_KEY}")
+    assert _trace_identities(signoz_families_off, token, family_fleet, expression) == expected, expression
+
+
+@pytest.mark.parametrize("expression_template,expected", LITERAL_MATRIX)
+def test_logs_stay_literal_with_flag_on(
+    signoz: types.SigNoz,
+    create_user_admin: None,  # pylint: disable=unused-argument
+    get_token: Callable[[str, str], str],
+    family_fleet: datetime,
+    expression_template: str,
+    expected: set[str],
+) -> None:
+    """Only traces have family support. The same filters on the logs copy of
+    the fleet behave literally even with the flag on."""
+    token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
+    expression = expression_template.format(key=f"resource.{CURRENT_KEY}")
+    assert _trace_identities(signoz, token, family_fleet, expression, signal="logs") == expected, expression
+
+
+def test_group_by_merges_and_echoes_requested_spelling(
+    signoz: types.SigNoz,
+    create_user_admin: None,  # pylint: disable=unused-argument
+    get_token: Callable[[str, str], str],
+    family_fleet: datetime,
+) -> None:
+    """Group by the current spelling over the fleet: OLD and NEW land in one
+    "production" group, BOTH lands in "staging", and the group column carries
+    the requested spelling."""
+    token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
+    response = make_query_request(
+        signoz,
+        token,
+        start_ms=int((family_fleet - timedelta(minutes=2)).timestamp() * 1000),
+        end_ms=int((family_fleet + timedelta(minutes=1)).timestamp() * 1000),
+        request_type=RequestType.SCALAR,
+        queries=[
+            build_traces_scalar_query(
+                [build_aggregation("count()")],
+                filter_expression=f"service.name LIKE '{PREFIX}%'",
+                group_by=[build_group_by_field(CURRENT_KEY, "string", "resource")],
+            )
+        ],
+    )
+    assert response.status_code == HTTPStatus.OK, response.text
+
+    result = response.json()["data"]["data"]["results"][0]
+    group_column = result["columns"][0]
+    assert group_column["name"] == CURRENT_KEY, group_column
+    assert group_column["columnType"] == "group", group_column
+
+    groups = {row[0] for row in result["data"]}
+    assert {"production", "staging"}.issubset(groups), groups
+    assert None in groups, groups
+
+
+def test_bare_name_prefers_resource_and_warns(
+    signoz: types.SigNoz,
+    create_user_admin: None,  # pylint: disable=unused-argument
+    get_token: Callable[[str, str], str],
+    family_fleet: datetime,
+) -> None:
+    """The fleet stores the family under the resource and the attribute
+    contexts, so a bare name is ambiguous. Resolution warns and keeps the
+    resource side; the family survives the collision as one unit."""
+    token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
+    response = make_query_request(
+        signoz,
+        token,
+        start_ms=int((family_fleet - timedelta(minutes=2)).timestamp() * 1000),
+        end_ms=int((family_fleet + timedelta(minutes=1)).timestamp() * 1000),
+        request_type=RequestType.RAW,
+        queries=[
+            build_raw_query(
+                "A",
+                "traces",
+                limit=100,
+                filter_expression=f"{CURRENT_KEY} = 'production'",
+                order=[build_order_by("timestamp", "asc")],
+                select_fields=[{"name": "span.name"}],
+            )
+        ],
+    )
+    assert response.status_code == HTTPStatus.OK, response.text
+
+    matched = {name for name in get_column_data_from_response(response.json(), "name") if name.startswith(PREFIX)}
+    assert matched == {OLD, NEW}
+
+    warning = response.json()["data"].get("warning") or {}
+    messages = " ".join(entry.get("message", "") for entry in warning.get("warnings", []))
+    assert "ambiguous" in messages.lower(), messages

tests/integration/tests/semconvfamilies/conftest.py

@@ -0,0 +1,56 @@
+import pytest
+from testcontainers.core.container import Network
+
+from fixtures import types
+from fixtures.signoz import create_signoz
+
+
+@pytest.fixture(name="signoz", scope="package")
+def signoz_semconv_families(
+    network: Network,
+    zeus: types.TestContainerDocker,
+    gateway: types.TestContainerDocker,
+    sqlstore: types.TestContainerSQL,
+    clickhouse: types.TestContainerClickhouse,
+    request: pytest.FixtureRequest,
+    pytestconfig: pytest.Config,
+) -> types.SigNoz:
+    """Package-scoped SigNoz with resolve_semconv_families on."""
+    return create_signoz(
+        network=network,
+        zeus=zeus,
+        gateway=gateway,
+        sqlstore=sqlstore,
+        clickhouse=clickhouse,
+        request=request,
+        pytestconfig=pytestconfig,
+        cache_key="signoz-semconv-families",
+        env_overrides={
+            "SIGNOZ_FLAGGER_CONFIG_BOOLEAN_RESOLVE__SEMCONV__FAMILIES": True,
+        },
+    )
+
+
+@pytest.fixture(name="signoz_families_off", scope="package")
+def signoz_families_off(
+    network: Network,
+    zeus: types.TestContainerDocker,
+    gateway: types.TestContainerDocker,
+    sqlstore: types.TestContainerSQL,
+    clickhouse: types.TestContainerClickhouse,
+    request: pytest.FixtureRequest,
+    pytestconfig: pytest.Config,
+) -> types.SigNoz:
+    """A second instance with the flag at its default (off). It shares the
+    sqlstore and clickhouse, so the same admin token and seeded rows work."""
+    return create_signoz(
+        network=network,
+        zeus=zeus,
+        gateway=gateway,
+        sqlstore=sqlstore,
+        clickhouse=clickhouse,
+        request=request,
+        pytestconfig=pytestconfig,
+        cache_key="signoz-semconv-families-off",
+        env_overrides={},
+    )