Compare commits

..

20 Commits

Author SHA1 Message Date
nityanandagohain
b818ff5fc4 fix: address comments 2026-04-29 17:19:19 +05:30
nityanandagohain
e7d729ab5d Merge remote-tracking branch 'origin/main' into issue_4203 2026-04-29 16:51:49 +05:30
Piyush Singariya
a672335a33 fix: Body Search warning with FTS in JSON Logs (#10807)
Some checks failed
build-staging / prepare (push) Has been cancelled
build-staging / js-build (push) Has been cancelled
build-staging / go-build (push) Has been cancelled
build-staging / staging (push) Has been cancelled
Release Drafter / update_release_draft (push) Has been cancelled
* fix: fts warning miss in direct text search

* fix: comments

* test: added one more test variation

* ci: go lint

* fix: fts warning update

* fix: integration tests

* fix: go test and fmtlint
2026-04-29 08:50:28 +00:00
Nityananda Gohain
ed812ad1c8 Merge branch 'main' into issue_4203 2026-04-24 11:25:38 +05:30
nityanandagohain
3b82c2ce43 fix: restrict merging to only span data 2026-04-24 11:25:11 +05:30
nityanandagohain
214980ddad Merge remote-tracking branch 'origin/main' into issue_4203 2026-04-24 10:22:33 +05:30
nityanandagohain
a7b69a2678 fix: py-fmt 2026-04-21 12:13:47 +05:30
nityanandagohain
73c82f50a9 Merge remote-tracking branch 'origin/main' into issue_4203 2026-04-21 11:49:52 +05:30
nityanandagohain
2593c5eb91 fix: linting issues 2026-04-13 15:44:43 +05:30
Nityananda Gohain
b6b2d36baa Merge branch 'main' into issue_4203 2026-04-10 17:15:08 +05:30
nityanandagohain
a444a039f9 Merge remote-tracking branch 'origin/issue_4203' into issue_4203 2026-04-10 17:13:22 +05:30
nityanandagohain
bfb050ec17 fix: add changes 2026-04-10 16:57:50 +05:30
nityanandagohain
ff3e87f70c Merge remote-tracking branch 'origin/main' into issue_4203 2026-04-09 21:29:11 +05:30
Nityananda Gohain
9ac02ebe00 Merge branch 'main' into issue_4203 2026-03-25 15:50:04 +05:30
nityanandagohain
fbdd0bebbc Merge remote-tracking branch 'origin/main' into issue_4203 2026-03-25 15:21:52 +05:30
nityanandagohain
b2245b48fe fix: retain existing behaviour 2026-03-23 11:03:34 +05:30
Nityananda Gohain
87e654fc73 chore: add comment
Co-authored-by: Tushar Vats <tushar@signoz.io>
2026-03-18 16:54:09 +05:30
nityanandagohain
0ee31ce440 chore: fix tests 2026-03-17 18:16:51 +05:30
nityanandagohain
63e681b87b chore: add integration tests 2026-03-17 15:38:00 +05:30
nityanandagohain
28375c8c1e chore: send all data for trace list api 2026-03-13 19:31:59 +05:30
16 changed files with 496 additions and 256 deletions

View File

@@ -265,6 +265,15 @@ func (q *builderQuery[T]) executeWithContext(ctx context.Context, query string,
return nil, err
}
// TODO: This should move to readAsRaw function in consume.go but for now we are keeping it here since it's only relevant for traces
if q.spec.Signal == telemetrytypes.SignalTraces {
if raw, ok := payload.(*qbtypes.RawData); ok {
for _, rr := range raw.Rows {
mergeSpanAttributeColumns(rr.Data)
}
}
}
return &qbtypes.Result{
Type: q.kind,
Value: payload,

View File

@@ -431,6 +431,45 @@ func readAsRaw(rows driver.Rows, queryName string) (*qbtypes.RawData, error) {
}, nil
}
// mergeSpanAttributeColumns merges the typed ClickHouse span attribute columns
// (attributes_string, attributes_number, attributes_bool, resources_string) into
// unified "attributes" and "resource" keys, removing the raw columns.
func mergeSpanAttributeColumns(data map[string]any) {
attrStr := data["attributes_string"]
attrNum := data["attributes_number"]
attrBool := data["attributes_bool"]
// todo(nitya): move to resource json
resStr := data["resources_string"]
attributes := make(map[string]any)
if m, ok := attrStr.(map[string]string); ok {
for k, v := range m {
attributes[k] = v
}
}
if m, ok := attrNum.(map[string]float64); ok {
for k, v := range m {
attributes[k] = v
}
}
if m, ok := attrBool.(map[string]bool); ok {
for k, v := range m {
attributes[k] = v
}
}
delete(data, "attributes_string")
delete(data, "attributes_number")
delete(data, "attributes_bool")
data["attributes"] = attributes
resource := map[string]string{}
if m, ok := resStr.(map[string]string); ok {
resource = m
}
data["resource"] = resource
delete(data, "resources_string")
}
// numericAsFloat converts numeric types to float64 efficiently.
func numericAsFloat(v any) float64 {
switch x := v.(type) {

View File

@@ -85,6 +85,13 @@ func (q *traceOperatorQuery) executeWithContext(ctx context.Context, query strin
return nil, err
}
// TODO: This should move to readAsRaw function in consume.go but for now we can keep it here since it's only relevant for traces
if raw, ok := payload.(*qbtypes.RawData); ok {
for _, rr := range raw.Rows {
mergeSpanAttributeColumns(rr.Data)
}
}
return &qbtypes.Result{
Type: q.kind,
Value: payload,

View File

@@ -4,6 +4,10 @@ const (
TrueConditionLiteral = "true"
SkipConditionLiteral = "__skip__"
ErrorConditionLiteral = "__skip_because_of_error__"
// BodyFullTextSearchDefaultWarning is emitted when a full-text search or "body" searches are hit
// with New JSON Body enhancements.
BodyFullTextSearchDefaultWarning = "Full text searches default to `body.message:string`. Use `body.<key>` to search a different field inside body"
)
var (

View File

@@ -362,6 +362,10 @@ func (v *filterExpressionVisitor) VisitPrimary(ctx *grammar.PrimaryContext) any
v.errors = append(v.errors, fmt.Sprintf("failed to build full text search condition: %s", err.Error()))
return ErrorConditionLiteral
}
if v.bodyJSONEnabled && v.fullTextColumn.Name == "body" {
v.warnings = append(v.warnings, BodyFullTextSearchDefaultWarning)
}
return cond
}
@@ -717,6 +721,10 @@ func (v *filterExpressionVisitor) VisitFullText(ctx *grammar.FullTextContext) an
return ErrorConditionLiteral
}
if v.bodyJSONEnabled && v.fullTextColumn.Name == "body" {
v.warnings = append(v.warnings, BodyFullTextSearchDefaultWarning)
}
return cond
}

View File

@@ -894,12 +894,12 @@ func TestAdjustKey(t *testing.T) {
func TestStmtBuilderBodyField(t *testing.T) {
cases := []struct {
name string
requestType qbtypes.RequestType
query qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]
name string
requestType qbtypes.RequestType
query qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]
enableUseJSONBody bool
expected qbtypes.Statement
expectedErr error
expected qbtypes.Statement
expectedErr error
}{
{
name: "body_exists",
@@ -1039,15 +1039,15 @@ func TestStmtBuilderBodyField(t *testing.T) {
func TestStmtBuilderBodyFullTextSearch(t *testing.T) {
cases := []struct {
name string
requestType qbtypes.RequestType
query qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]
name string
requestType qbtypes.RequestType
query qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]
enableUseJSONBody bool
expected qbtypes.Statement
expectedErr error
expected qbtypes.Statement
expectedErr error
}{
{
name: "body_contains",
name: "fts",
requestType: qbtypes.RequestTypeRaw,
query: qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]{
Signal: telemetrytypes.SignalLogs,
@@ -1056,13 +1056,30 @@ func TestStmtBuilderBodyFullTextSearch(t *testing.T) {
},
enableUseJSONBody: true,
expected: qbtypes.Statement{
Query: "SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body_v2 as body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE match(LOWER(body_v2.message), LOWER(?)) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"error", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Query: "SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body_v2 as body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE match(LOWER(body_v2.message), LOWER(?)) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"error", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{querybuilder.BodyFullTextSearchDefaultWarning},
},
expectedErr: nil,
},
{
name: "body_contains_disabled",
name: "fts_2",
requestType: qbtypes.RequestTypeRaw,
query: qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]{
Signal: telemetrytypes.SignalLogs,
Filter: &qbtypes.Filter{Expression: "error"},
Limit: 10,
},
enableUseJSONBody: true,
expected: qbtypes.Statement{
Query: "SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body_v2 as body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE match(LOWER(body_v2.message), LOWER(?)) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"error", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{querybuilder.BodyFullTextSearchDefaultWarning},
},
expectedErr: nil,
},
{
name: "fts_disabled",
requestType: qbtypes.RequestTypeRaw,
query: qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]{
Signal: telemetrytypes.SignalLogs,

View File

@@ -1,6 +1,50 @@
package telemetrytraces
import "github.com/SigNoz/signoz/pkg/types/telemetrytypes"
import (
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
)
const (
// Internal Columns.
SpanTimestampBucketStartColumn = "ts_bucket_start"
SpanResourceFingerPrintColumn = "resource_fingerprint"
// Intrinsic Columns.
SpanTimestampColumn = "timestamp"
SpanTraceIDColumn = "trace_id"
SpanSpanIDColumn = "span_id"
SpanTraceStateColumn = "trace_state"
SpanParentSpanIDColumn = "parent_span_id"
SpanFlagsColumn = "flags"
SpanNameColumn = "name"
SpanKindColumn = "kind"
SpanKindStringColumn = "kind_string"
SpanDurationNanoColumn = "duration_nano"
SpanStatusCodeColumn = "status_code"
SpanStatusMessageColumn = "status_message"
SpanStatusCodeStringColumn = "status_code_string"
SpanEventsColumn = "events"
SpanLinksColumn = "links"
// Calculated Columns.
SpanResponseStatusCodeColumn = "response_status_code"
SpanExternalHTTPURLColumn = "external_http_url"
SpanHTTPURLColumn = "http_url"
SpanExternalHTTPMethodColumn = "external_http_method"
SpanHTTPMethodColumn = "http_method"
SpanHTTPHostColumn = "http_host"
SpanDBNameColumn = "db_name"
SpanDBOperationColumn = "db_operation"
SpanHasErrorColumn = "has_error"
SpanIsRemoteColumn = "is_remote"
// Contextual Columns.
SpanAttributesStringColumn = "attributes_string"
SpanAttributesNumberColumn = "attributes_number"
SpanAttributesBoolColumn = "attributes_bool"
SpanResourcesStringColumn = "resources_string"
)
var (
IntrinsicFields = map[string]telemetrytypes.TelemetryFieldKey{
@@ -334,6 +378,51 @@ var (
SpanSearchScopeRoot = "isroot"
SpanSearchScopeEntryPoint = "isentrypoint"
// IntrinsicSpanFields lists the intrinsic span columns, in the order they
// should appear when a raw query expands its SelectFields.
IntrinsicSpanFields = []telemetrytypes.TelemetryFieldKey{
{Name: SpanTimestampColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanTraceIDColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanSpanIDColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanTraceStateColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanParentSpanIDColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanFlagsColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanNameColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanKindColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanKindStringColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanDurationNanoColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanStatusCodeColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanStatusMessageColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanStatusCodeStringColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanEventsColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanLinksColumn, FieldContext: telemetrytypes.FieldContextSpan},
}
// CalculatedSpanFields lists the calculated/derived span columns, in the
// order they should appear when a raw query expands its SelectFields.
CalculatedSpanFields = []telemetrytypes.TelemetryFieldKey{
{Name: SpanResponseStatusCodeColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanExternalHTTPURLColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanHTTPURLColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanExternalHTTPMethodColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanHTTPMethodColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanHTTPHostColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanDBNameColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanDBOperationColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanHasErrorColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanIsRemoteColumn, FieldContext: telemetrytypes.FieldContextSpan},
}
// ContextualSpanColumns lists the typed attribute and resource columns
// selected raw (rather than via ColumnExpressionFor) so that consume.go
// can merge them into unified "attributes" and "resource" maps.
ContextualSpanColumns = []string{
SpanAttributesStringColumn,
SpanAttributesNumberColumn,
SpanAttributesBoolColumn,
SpanResourcesStringColumn,
}
DefaultFields = map[string]telemetrytypes.TelemetryFieldKey{
"timestamp": {
Name: "timestamp",

View File

@@ -78,6 +78,16 @@ func TestGetFieldKeyName(t *testing.T) {
expectedResult: "multiIf(resource.`deployment.environment` IS NOT NULL, resource.`deployment.environment`::String, `resource_string_deployment$$environment_exists`==true, `resource_string_deployment$$environment`, NULL)",
expectedError: nil,
},
{
name: "Contextual map column - attributes_string without span context does not short-circuit",
key: telemetrytypes.TelemetryFieldKey{
Name: SpanAttributesStringColumn,
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
expectedResult: "attributes_string['attributes_string']",
expectedError: nil,
},
{
name: "Non-existent column",
key: telemetrytypes.TelemetryFieldKey{

View File

@@ -4,7 +4,6 @@ import (
"context"
"fmt"
"log/slog"
"slices"
"strings"
"github.com/SigNoz/signoz/pkg/errors"
@@ -16,7 +15,6 @@ import (
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/huandu/go-sqlbuilder"
"golang.org/x/exp/maps"
)
var (
@@ -89,40 +87,13 @@ func (b *traceQueryStatementBuilder) Build(
return nil, err
}
/*
Adding a tech debt note here:
This piece of code is a hot fix and should be removed once we close issue: engineering-pod/issues/3622
*/
/*
-------------------------------- Start of tech debt ----------------------------
*/
isSelectFieldsEmpty := false
if requestType == qbtypes.RequestTypeRaw {
selectedFields := query.SelectFields
if len(selectedFields) == 0 {
sortedKeys := maps.Keys(DefaultFields)
slices.Sort(sortedKeys)
for _, key := range sortedKeys {
selectedFields = append(selectedFields, DefaultFields[key])
}
query.SelectFields = selectedFields
}
selectFieldKeys := []string{}
for _, field := range selectedFields {
selectFieldKeys = append(selectFieldKeys, field.Name)
}
for _, x := range []string{"timestamp", "span_id", "trace_id"} {
if !slices.Contains(selectFieldKeys, x) {
query.SelectFields = append(query.SelectFields, DefaultFields[x])
}
}
isSelectFieldsEmpty = len(query.SelectFields) == 0
// we are expanding here to ensure that all the conflicts are taken care in adjustKeys
// i.e if there is a conflict we strip away context of the key in adjustKeys
query = b.expandRawSelectFields(query)
}
/*
-------------------------------- End of tech debt ----------------------------
*/
query = b.adjustKeys(ctx, keys, query, requestType)
@@ -131,7 +102,7 @@ func (b *traceQueryStatementBuilder) Build(
switch requestType {
case qbtypes.RequestTypeRaw:
return b.buildListQuery(ctx, q, query, start, end, keys, variables)
return b.buildListQuery(ctx, q, query, start, end, keys, variables, isSelectFieldsEmpty)
case qbtypes.RequestTypeTimeSeries:
return b.buildTimeSeriesQuery(ctx, q, query, start, end, keys, variables)
case qbtypes.RequestTypeScalar:
@@ -295,6 +266,7 @@ func (b *traceQueryStatementBuilder) buildListQuery(
start, end uint64,
keys map[string][]*telemetrytypes.TelemetryFieldKey,
variables map[string]qbtypes.VariableItem,
isSelectFieldsEmpty bool,
) (*qbtypes.Statement, error) {
var (
@@ -309,7 +281,6 @@ func (b *traceQueryStatementBuilder) buildListQuery(
cteArgs = append(cteArgs, args)
}
// TODO: should we deprecate `SelectFields` and return everything from a span like we do for logs?
for _, field := range query.SelectFields {
colExpr, err := b.fm.ColumnExpressionFor(ctx, start, end, &field, keys)
if err != nil {
@@ -318,6 +289,12 @@ func (b *traceQueryStatementBuilder) buildListQuery(
sb.SelectMore(colExpr)
}
if isSelectFieldsEmpty {
for _, col := range ContextualSpanColumns {
sb.SelectMore(col)
}
}
// From table
sb.From(fmt.Sprintf("%s.%s", DBName, SpanIndexV3TableName))
@@ -844,3 +821,30 @@ func (b *traceQueryStatementBuilder) buildResourceFilterCTE(
variables,
)
}
// expandRawSelectFields populates SelectFields for raw (list view) queries.
// It must be called before adjustKeys so that normalization runs over the full set.
func (b *traceQueryStatementBuilder) expandRawSelectFields(query qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]) qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation] {
if len(query.SelectFields) == 0 {
selectFields := make([]telemetrytypes.TelemetryFieldKey, 0, len(IntrinsicSpanFields)+len(CalculatedSpanFields))
selectFields = append(selectFields, IntrinsicSpanFields...)
selectFields = append(selectFields, CalculatedSpanFields...)
query.SelectFields = selectFields
return query
}
selectFields := []telemetrytypes.TelemetryFieldKey{
{Name: SpanTimestampColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanTraceIDColumn, FieldContext: telemetrytypes.FieldContextSpan},
{Name: SpanSpanIDColumn, FieldContext: telemetrytypes.FieldContextSpan},
}
for _, field := range query.SelectFields {
// TODO(tvats): If a user specifies attribute.timestamp in the select fields, this loop will basically ignore it, as we already added a field by default. This can be fixed once we close https://github.com/SigNoz/engineering-pod/issues/3693
if field.Name == SpanTimestampColumn || field.Name == SpanTraceIDColumn || field.Name == SpanSpanIDColumn {
continue
}
selectFields = append(selectFields, field)
}
query.SelectFields = selectFields
return query
}

View File

@@ -439,7 +439,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT name AS `name`, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, duration_nano AS `duration_nano`, `attribute_number_cart$$items_count` AS `cart.items_count`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp AS `timestamp`, trace_id AS `trace_id`, span_id AS `span_id`, name AS `name`, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, duration_nano AS `duration_nano`, `attribute_number_cart$$items_count` AS `cart.items_count` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -468,7 +468,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT duration_nano AS `duration_nano`, name AS `name`, response_status_code AS `response_status_code`, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, span_id AS `span_id`, timestamp AS `timestamp`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? ORDER BY attributes_string['user.id'] AS `user.id` desc LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp AS `timestamp`, trace_id AS `trace_id`, span_id AS `span_id`, trace_state AS `trace_state`, parent_span_id AS `parent_span_id`, flags AS `flags`, name AS `name`, kind AS `kind`, kind_string AS `kind_string`, duration_nano AS `duration_nano`, status_code AS `status_code`, status_message AS `status_message`, status_code_string AS `status_code_string`, events AS `events`, links AS `links`, response_status_code AS `response_status_code`, external_http_url AS `external_http_url`, http_url AS `http_url`, external_http_method AS `external_http_method`, http_method AS `http_method`, http_host AS `http_host`, db_name AS `db_name`, db_operation AS `db_operation`, has_error AS `has_error`, is_remote AS `is_remote`, attributes_string, attributes_number, attributes_bool, resources_string FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? ORDER BY attributes_string['user.id'] AS `user.id` desc LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -512,7 +512,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, response_status_code AS `responseStatusCode`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp AS `timestamp`, trace_id AS `trace_id`, span_id AS `span_id`, name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, response_status_code AS `responseStatusCode` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -556,7 +556,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, multiIf(toString(`attribute_string_mixed$$materialization$$key`) != '', toString(`attribute_string_mixed$$materialization$$key`), toString(multiIf(resource.`mixed.materialization.key` IS NOT NULL, resource.`mixed.materialization.key`::String, mapContains(resources_string, 'mixed.materialization.key'), resources_string['mixed.materialization.key'], NULL)) != '', toString(multiIf(resource.`mixed.materialization.key` IS NOT NULL, resource.`mixed.materialization.key`::String, mapContains(resources_string, 'mixed.materialization.key'), resources_string['mixed.materialization.key'], NULL)), NULL) AS `mixed.materialization.key`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp AS `timestamp`, trace_id AS `trace_id`, span_id AS `span_id`, name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, multiIf(toString(`attribute_string_mixed$$materialization$$key`) != '', toString(`attribute_string_mixed$$materialization$$key`), toString(multiIf(resource.`mixed.materialization.key` IS NOT NULL, resource.`mixed.materialization.key`::String, mapContains(resources_string, 'mixed.materialization.key'), resources_string['mixed.materialization.key'], NULL)) != '', toString(multiIf(resource.`mixed.materialization.key` IS NOT NULL, resource.`mixed.materialization.key`::String, mapContains(resources_string, 'mixed.materialization.key'), resources_string['mixed.materialization.key'], NULL)), NULL) AS `mixed.materialization.key` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -601,7 +601,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, `attribute_string_mixed$$materialization$$key` AS `mixed.materialization.key`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp AS `timestamp`, trace_id AS `trace_id`, span_id AS `span_id`, name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, `attribute_string_mixed$$materialization$$key` AS `mixed.materialization.key` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -711,7 +711,7 @@ func TestStatementBuilderListQueryWithCorruptData(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "SELECT duration_nano AS `duration_nano`, name AS `name`, response_status_code AS `response_status_code`, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, span_id AS `span_id`, timestamp AS `timestamp`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Query: "SELECT timestamp AS `timestamp`, trace_id AS `trace_id`, span_id AS `span_id`, trace_state AS `trace_state`, parent_span_id AS `parent_span_id`, flags AS `flags`, name AS `name`, kind AS `kind`, kind_string AS `kind_string`, duration_nano AS `duration_nano`, status_code AS `status_code`, status_message AS `status_message`, status_code_string AS `status_code_string`, events AS `events`, links AS `links`, response_status_code AS `response_status_code`, external_http_url AS `external_http_url`, http_url AS `http_url`, external_http_method AS `external_http_method`, http_method AS `http_method`, http_host AS `http_host`, db_name AS `db_name`, db_operation AS `db_operation`, has_error AS `has_error`, is_remote AS `is_remote`, attributes_string, attributes_number, attributes_bool, resources_string FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -744,7 +744,7 @@ func TestStatementBuilderListQueryWithCorruptData(t *testing.T) {
}},
},
expected: qbtypes.Statement{
Query: "SELECT duration_nano AS `duration_nano`, name AS `name`, response_status_code AS `response_status_code`, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, span_id AS `span_id`, timestamp AS `timestamp`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? ORDER BY timestamp AS `timestamp` asc LIMIT ?",
Query: "SELECT timestamp AS `timestamp`, trace_id AS `trace_id`, span_id AS `span_id`, trace_state AS `trace_state`, parent_span_id AS `parent_span_id`, flags AS `flags`, name AS `name`, kind AS `kind`, kind_string AS `kind_string`, duration_nano AS `duration_nano`, status_code AS `status_code`, status_message AS `status_message`, status_code_string AS `status_code_string`, events AS `events`, links AS `links`, response_status_code AS `response_status_code`, external_http_url AS `external_http_url`, http_url AS `http_url`, external_http_method AS `external_http_method`, http_method AS `http_method`, http_host AS `http_host`, db_name AS `db_name`, db_operation AS `db_operation`, has_error AS `has_error`, is_remote AS `is_remote`, attributes_string, attributes_number, attributes_bool, resources_string FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? ORDER BY timestamp AS `timestamp` asc LIMIT ?",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,

View File

@@ -413,21 +413,18 @@ func (b *traceOperatorCTEBuilder) buildFinalQuery(ctx context.Context, selectFro
}
func (b *traceOperatorCTEBuilder) buildListQuery(ctx context.Context, selectFromCTE string) (*qbtypes.Statement, error) {
keySelectors := b.getKeySelectors()
for _, field := range b.operator.SelectFields {
keySelectors = append(keySelectors, &telemetrytypes.FieldKeySelector{
Name: field.Name,
Signal: telemetrytypes.SignalTraces,
FieldContext: field.FieldContext,
FieldDataType: field.FieldDataType,
})
}
keys, _, err := b.stmtBuilder.metadataStore.GetKeysMulti(ctx, keySelectors)
if err != nil {
return nil, err
}
sb := sqlbuilder.NewSelectBuilder()
// Select core fields
sb.Select(
"timestamp",
"trace_id",
"span_id",
"name",
"duration_nano",
"parent_span_id",
)
coreFields := []string{"trace_id", "span_id", "name", "duration_nano", "parent_span_id"}
selectedFields := map[string]bool{
"timestamp": true,
"trace_id": true,
@@ -437,15 +434,23 @@ func (b *traceOperatorCTEBuilder) buildListQuery(ctx context.Context, selectFrom
"parent_span_id": true,
}
// Inner SELECT reads from the CTE and renames timestamp→ts.
// This breaks the `ORDER BY col AS `col`` pattern that triggers a
// CH 25.12.5 distributed-analyzer regression (NOT_FOUND_COLUMN_IN_BLOCK /
// timestamp renamed to timestamp_0). See ClickHouse/ClickHouse#103508.
innerSB := sqlbuilder.NewSelectBuilder()
innerSB.Select("timestamp AS ts")
innerSB.SelectMore(coreFields...)
// Get keys for selectFields
keySelectors := b.getKeySelectors()
for _, field := range b.operator.SelectFields {
keySelectors = append(keySelectors, &telemetrytypes.FieldKeySelector{
Name: field.Name,
Signal: telemetrytypes.SignalTraces,
FieldContext: field.FieldContext,
FieldDataType: field.FieldDataType,
})
}
var additionalSelectedFields []string
keys, _, err := b.stmtBuilder.metadataStore.GetKeysMulti(ctx, keySelectors)
if err != nil {
return nil, err
}
// Add selectFields using ColumnExpressionFor since we now have all base table columns
for _, field := range b.operator.SelectFields {
if selectedFields[field.Name] {
continue
@@ -456,60 +461,41 @@ func (b *traceOperatorCTEBuilder) buildListQuery(ctx context.Context, selectFrom
slog.String("field", field.Name), errors.Attr(err))
continue
}
innerSB.SelectMore(colExpr)
sb.SelectMore(colExpr)
selectedFields[field.Name] = true
additionalSelectedFields = append(additionalSelectedFields, field.Name)
}
// Also expose any explicit ORDER BY fields that aren't already selected,
// so the outer query can reference them by alias name.
sb.From(selectFromCTE)
// Add order by support using ColumnExpressionFor
orderApplied := false
for _, orderBy := range b.operator.Order {
if selectedFields[orderBy.Key.Name] {
continue
}
colExpr, err := b.stmtBuilder.fm.ColumnExpressionFor(ctx, b.start, b.end, &orderBy.Key.TelemetryFieldKey, keys)
if err != nil {
return nil, err
}
innerSB.SelectMore(colExpr)
selectedFields[orderBy.Key.Name] = true
sb.OrderBy(fmt.Sprintf("%s %s", colExpr, orderBy.Direction.StringValue()))
orderApplied = true
}
innerSB.From(selectFromCTE)
innerSQL, innerArgs := innerSB.BuildWithFlavor(sqlbuilder.ClickHouse)
// Outer SELECT reads from the inner subquery and re-exposes timestamp via
// the ts alias. ORDER BY uses the alias name directly — no AS-alias in the
// ORDER BY position — which is the pattern that avoids the CH regression.
outerSB := sqlbuilder.NewSelectBuilder()
outerSB.Select("ts AS timestamp")
outerSB.SelectMore(coreFields...)
for _, name := range additionalSelectedFields {
outerSB.SelectMore(fmt.Sprintf("`%s`", name))
}
outerSB.From(fmt.Sprintf("(%s) AS t", innerSQL))
if len(b.operator.Order) > 0 {
for _, orderBy := range b.operator.Order {
outerSB.OrderBy(fmt.Sprintf("`%s` %s", orderBy.Key.Name, orderBy.Direction.StringValue()))
}
} else {
outerSB.OrderBy("timestamp DESC")
if !orderApplied {
sb.OrderBy("timestamp DESC")
}
if b.operator.Limit > 0 {
outerSB.Limit(b.operator.Limit)
sb.Limit(b.operator.Limit)
} else {
outerSB.Limit(100)
}
if b.operator.Offset > 0 {
outerSB.Offset(b.operator.Offset)
sb.Limit(100)
}
outerSQL, outerArgs := outerSB.BuildWithFlavor(sqlbuilder.ClickHouse)
if b.operator.Offset > 0 {
sb.Offset(b.operator.Offset)
}
sql, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
return &qbtypes.Statement{
Query: outerSQL,
Args: append(innerArgs, outerArgs...),
Query: sql,
Args: args,
}, nil
}

View File

@@ -67,7 +67,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT ts AS timestamp, trace_id, span_id, name, duration_nano, parent_span_id, `service.name` FROM (SELECT timestamp AS ts, trace_id, span_id, name, duration_nano, parent_span_id, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name` FROM A_DIR_DESC_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name` FROM A_DIR_DESC_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -104,7 +104,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id) SELECT ts AS timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM (SELECT timestamp AS ts, trace_id, span_id, name, duration_nano, parent_span_id FROM A_INDIR_DESC_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_INDIR_DESC_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "gateway", "%service.name%", "%service.name\":\"gateway%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "database", "%service.name%", "%service.name\":\"database%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 5},
},
expectedErr: nil,
@@ -141,7 +141,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_AND_B AS (SELECT l.* FROM A AS l INNER JOIN B AS r ON l.trace_id = r.trace_id) SELECT ts AS timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM (SELECT timestamp AS ts, trace_id, span_id, name, duration_nano, parent_span_id FROM A_AND_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_AND_B AS (SELECT l.* FROM A AS l INNER JOIN B AS r ON l.trace_id = r.trace_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_AND_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 15},
},
expectedErr: nil,
@@ -178,7 +178,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_OR_B AS (SELECT * FROM A UNION DISTINCT SELECT * FROM B) SELECT ts AS timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM (SELECT timestamp AS ts, trace_id, span_id, name, duration_nano, parent_span_id FROM A_OR_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_OR_B AS (SELECT * FROM A UNION DISTINCT SELECT * FROM B) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_OR_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 20},
},
expectedErr: nil,
@@ -215,7 +215,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_not_B AS (SELECT l.* FROM A AS l WHERE l.trace_id GLOBAL NOT IN (SELECT DISTINCT trace_id FROM B)) SELECT ts AS timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM (SELECT timestamp AS ts, trace_id, span_id, name, duration_nano, parent_span_id FROM A_not_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_not_B AS (SELECT l.* FROM A AS l WHERE l.trace_id GLOBAL NOT IN (SELECT DISTINCT trace_id FROM B)) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_not_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -380,72 +380,11 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), __resource_filter_C AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), C AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_C) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_D AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), D AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_D) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), C_DIR_DESC_D AS (SELECT p.* FROM C AS p INNER JOIN D AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), A_DIR_DESC_B_AND_C_DIR_DESC_D AS (SELECT l.* FROM A_DIR_DESC_B AS l INNER JOIN C_DIR_DESC_D AS r ON l.trace_id = r.trace_id) SELECT ts AS timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM (SELECT timestamp AS ts, trace_id, span_id, name, duration_nano, parent_span_id FROM A_DIR_DESC_B_AND_C_DIR_DESC_D) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), __resource_filter_C AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), C AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_C) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_D AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), D AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_D) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), C_DIR_DESC_D AS (SELECT p.* FROM C AS p INNER JOIN D AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), A_DIR_DESC_B_AND_C_DIR_DESC_D AS (SELECT l.* FROM A_DIR_DESC_B AS l INNER JOIN C_DIR_DESC_D AS r ON l.trace_id = r.trace_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_DIR_DESC_B_AND_C_DIR_DESC_D ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "auth", "%service.name%", "%service.name\":\"auth%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "database", "%service.name%", "%service.name\":\"database%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 5},
},
expectedErr: nil,
},
{
// order-by field (http.request.method) is not present in SelectFields;
// it must be included in the inner SELECT so the outer ORDER BY can
// reference it by alias, but must NOT appear in the outer SELECT list.
name: "order by field not in select fields",
requestType: qbtypes.RequestTypeRaw,
operator: qbtypes.QueryBuilderTraceOperator{
Expression: "A => B",
SelectFields: []telemetrytypes.TelemetryFieldKey{
{
Name: "service.name",
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
},
Order: []qbtypes.OrderBy{
{
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: "http.request.method",
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
},
Direction: qbtypes.OrderDirectionDesc,
},
},
Limit: 10,
},
compositeQuery: &qbtypes.CompositeQuery{
Queries: []qbtypes.QueryEnvelope{
{
Type: qbtypes.QueryTypeBuilder,
Spec: qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]{
Name: "A",
Signal: telemetrytypes.SignalTraces,
Filter: &qbtypes.Filter{
Expression: "service.name = 'frontend'",
},
},
},
{
Type: qbtypes.QueryTypeBuilder,
Spec: qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]{
Name: "B",
Signal: telemetrytypes.SignalTraces,
Filter: &qbtypes.Filter{
Expression: "service.name = 'backend'",
},
},
},
},
},
expected: qbtypes.Statement{
// http.request.method is in the inner SELECT (so ORDER BY can reach it)
// but is absent from the outer SELECT column list — only the ORDER BY clause references it.
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT ts AS timestamp, trace_id, span_id, name, duration_nano, parent_span_id, `service.name` FROM (SELECT timestamp AS ts, trace_id, span_id, name, duration_nano, parent_span_id, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, attributes_string['http.request.method'] AS `http.request.method` FROM A_DIR_DESC_B) AS t ORDER BY `http.request.method` desc LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
},
}
fm := NewFieldMapper()

View File

@@ -72,7 +72,6 @@ class TraceOperatorQuery:
return_spans_from: str
limit: int | None = None
order: list[OrderBy] | None = None
select_fields: list[TelemetryFieldKey] | None = None
def to_dict(self) -> dict:
spec: dict[str, Any] = {
@@ -84,8 +83,6 @@ class TraceOperatorQuery:
spec["limit"] = self.limit
if self.order:
spec["order"] = [o.to_dict() if hasattr(o, "to_dict") else o for o in self.order]
if self.select_fields:
spec["selectFields"] = [f.to_dict() for f in self.select_fields]
return {"type": "builder_trace_operator", "spec": spec}

View File

@@ -481,25 +481,24 @@ def test_traces_list(
"name": "A",
"signal": "traces",
"disabled": False,
"selectFields": [
{"name": "span_id"},
{"name": "span.timestamp"},
{"name": "trace_id"},
],
"order": [{"key": {"name": "timestamp"}, "direction": "desc"}],
"limit": 1,
},
},
HTTPStatus.OK,
lambda x: [
x[3].duration_nano,
x[3].name,
x[3].response_status_code,
x[3].service_name,
x[3].span_id,
format_timestamp(x[3].timestamp),
x[3].trace_id,
], # type: Callable[[List[Traces]], List[Any]]
),
# Case 2: order by attribute timestamp field which is there in attributes as well
# This should break but it doesn't because attribute.timestamp gets adjusted to timestamp
# because of default trace.timestamp gets added by default and bug in field mapper picks
# instrinsic field
# attribute.timestamp gets adjusted to span.timestamp
pytest.param(
{
"type": "builder_query",
@@ -507,16 +506,19 @@ def test_traces_list(
"name": "A",
"signal": "traces",
"disabled": False,
"order": [{"key": {"name": "attribute.timestamp"}, "direction": "desc"}],
"selectFields": [
{"name": "span_id"},
{"name": "span.timestamp"},
{"name": "trace_id"},
],
"order": [
{"key": {"name": "attribute.timestamp"}, "direction": "desc"}
],
"limit": 1,
},
},
HTTPStatus.OK,
lambda x: [
x[3].duration_nano,
x[3].name,
x[3].response_status_code,
x[3].service_name,
x[3].span_id,
format_timestamp(x[3].timestamp),
x[3].trace_id,
@@ -542,7 +544,7 @@ def test_traces_list(
], # type: Callable[[List[Traces]], List[Any]]
),
# Case 4: select attribute.timestamp with empty order by
# This doesn't return any data because of where_clause using aliased timestamp
# This returns the one span which has attribute.timestamp
pytest.param(
{
"type": "builder_query",
@@ -556,7 +558,11 @@ def test_traces_list(
},
},
HTTPStatus.OK,
lambda x: [], # type: Callable[[List[Traces]], List[Any]]
lambda x: [
x[0].span_id,
format_timestamp(x[0].timestamp),
x[0].trace_id,
], # type: Callable[[List[Traces]], List[Any]]
),
# Case 5: select timestamp with timestamp order by
pytest.param(
@@ -693,6 +699,112 @@ def test_traces_list_with_corrupt_data(
assert data[key] == value
@pytest.mark.parametrize(
"select_fields,status_code,expected_keys",
[
pytest.param(
[],
HTTPStatus.OK,
[
# all intrinsic column
"timestamp",
"trace_id",
"span_id",
"trace_state",
"parent_span_id",
"flags",
"name",
"kind",
"kind_string",
"duration_nano",
"status_code",
"status_message",
"status_code_string",
"events",
"links",
# all calculated columns
"response_status_code",
"external_http_url",
"http_url",
"external_http_method",
"http_method",
"http_host",
"db_name",
"db_operation",
"has_error",
"is_remote",
# all contextual columns (merged in response layer)
"attributes",
"resource",
],
),
pytest.param(
[
{"name": "service.name"},
],
HTTPStatus.OK,
["timestamp", "trace_id", "span_id", "service.name"],
),
],
)
def test_traces_list_with_select_fields(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[List[Traces]], None],
select_fields: List[dict],
status_code: HTTPStatus,
expected_keys: List[str],
) -> None:
"""
Setup:
Insert 4 traces with different attributes.
Tests:
1. Empty select fields should return all the fields.
2. Non empty select field should return the select field along with timestamp, trace_id and span_id.
"""
traces = (
generate_traces_with_corrupt_metadata()
) # using this as the data doesn't matter
insert_traces(traces)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
payload = {
"type": "builder_query",
"spec": {
"name": "A",
"signal": "traces",
"selectFields": select_fields,
"order": [{"key": {"name": "timestamp"}, "direction": "desc"}],
"limit": 1,
},
}
response = make_query_request(
signoz,
token,
start_ms=int(
(datetime.now(tz=UTC) - timedelta(minutes=5)).timestamp() * 1000
),
end_ms=int(datetime.now(tz=UTC).timestamp() * 1000),
request_type="raw",
queries=[payload],
)
assert response.status_code == status_code
if response.status_code == HTTPStatus.OK:
data = response.json()
assert len(data["data"]["data"]["results"][0]["rows"][0]["data"].keys()) == len(
expected_keys
)
assert set(data["data"]["data"]["results"][0]["rows"][0]["data"].keys()) == set(
expected_keys
)
@pytest.mark.parametrize(
"order_by,aggregation_alias,expected_status",
[

View File

@@ -1212,13 +1212,21 @@ def test_message_searches(
"aggregation": "count()",
"validate": lambda r: len(get_rows(r)) == 2 and set(_body_messages(r)) == payment_messages,
},
# FTS — bare keyword
# FTS — String bare keyword
{
"name": "msg.fts_quoted",
"requestType": "raw",
"expression": '"Payment"',
"aggregation": "count()",
"validate": lambda r: len(get_rows(r)) == 2 and all("Payment" in b.get("message", "") for b in _get_bodies(r)),
"validate": lambda r: len(get_rows(r)) == 2 and all("Payment" in b.get("message", "") for b in _get_bodies(r)) and r.json().get("data", {}).get("warning") is not None,
},
# FTS — bare keyword
{
"name": "msg.fts_quoted_without_quotes",
"requestType": "raw",
"expression": "Payment",
"aggregation": "count()",
"validate": lambda r: len(get_rows(r)) == 2 and all("Payment" in b.get("message", "") for b in _get_bodies(r)) and r.json().get("data", {}).get("warning") is not None,
},
# = operator via body.message — tests exact match path
{

View File

@@ -530,13 +530,11 @@ def test_export_traces_with_composite_query_trace_operator(
) -> None:
"""
Setup:
Insert a parent span and two child spans, all with an http.method attribute.
Insert multiple traces with parent-child relationships.
Tests:
1. Basic trace operator (A => B) returning parent spans, ordered by timestamp.
2. Same operator with selectFields=[service.name] and order by http.method, which is
NOT in selectFields — verifies the inner/outer subquery fix for the CH 25.12.5
NOT_FOUND_COLUMN_IN_BLOCK regression (ORDER BY col AS `col` in a CTE shape).
1. Export traces using trace operator in composite query (POST)
2. Verify trace operator query works correctly
"""
parent_trace_id = TraceIdGenerator.trace_id()
parent_span_id = TraceIdGenerator.span_id()
@@ -557,8 +555,12 @@ def test_export_traces_with_composite_query_trace_operator(
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "parent-service"},
attributes={"operation.type": "parent", "http.method": "GET"},
resources={
"service.name": "parent-service",
},
attributes={
"operation.type": "parent",
},
),
Traces(
timestamp=now - timedelta(seconds=9),
@@ -570,8 +572,12 @@ def test_export_traces_with_composite_query_trace_operator(
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "parent-service"},
attributes={"operation.type": "child", "http.method": "POST"},
resources={
"service.name": "parent-service",
},
attributes={
"operation.type": "child",
},
),
Traces(
timestamp=now - timedelta(seconds=7),
@@ -583,23 +589,31 @@ def test_export_traces_with_composite_query_trace_operator(
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "parent-service"},
attributes={"operation.type": "child", "http.method": "POST"},
resources={
"service.name": "parent-service",
},
attributes={
"operation.type": "child",
},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
url = signoz.self.host_configs["8080"].get("/api/v1/export_raw_data?format=jsonl")
# A: spans with operation.type = 'parent'
query_a = BuilderQuery(
signal="traces",
name="A",
limit=1000,
filter_expression="operation.type = 'parent'",
)
# B: spans with operation.type = 'child'
query_b = BuilderQuery(
signal="traces",
name="B",
@@ -607,50 +621,47 @@ def test_export_traces_with_composite_query_trace_operator(
filter_expression="operation.type = 'child'",
)
def export(operator: TraceOperatorQuery) -> list[dict]:
body = QueryRangeRequest(
start=start_ns,
end=end_ns,
queries=[query_a, query_b, operator],
).to_dict()
resp = requests.post(
url,
json=body,
timeout=10,
headers={"authorization": f"Bearer {token}", "Content-Type": "application/json"},
)
assert resp.status_code == HTTPStatus.OK, resp.text
assert resp.headers["Content-Type"] == "application/x-ndjson"
return [json.loads(line) for line in resp.text.strip().split("\n") if line]
# Test 1: basic trace operator ordered by timestamp
spans = export(
TraceOperatorQuery(
name="C",
expression="A => B",
return_spans_from="A",
limit=1000,
order=[OrderBy(TelemetryFieldKey("timestamp", "string", "span"), "desc")],
)
# Trace operator: find traces where A has a direct descendant B
query_c = TraceOperatorQuery(
name="C",
expression="A => B",
return_spans_from="A",
limit=1000,
order=[OrderBy(TelemetryFieldKey("timestamp", "string", "span"), "desc")],
)
assert len(spans) == 1
assert all(s.get("trace_id") == parent_trace_id for s in spans)
assert any(s.get("name") == "parent-operation" for s in spans)
# Test 2: order-by field (http.method) absent from selectFields
spans = export(
TraceOperatorQuery(
name="C",
expression="A => B",
return_spans_from="A",
limit=1000,
select_fields=[TelemetryFieldKey("service.name", "string", "resource")],
order=[OrderBy(TelemetryFieldKey("http.method", "string", "tag"), "desc")],
)
body = QueryRangeRequest(
start=start_ns,
end=end_ns,
queries=[query_a, query_b, query_c],
).to_dict()
url = signoz.self.host_configs["8080"].get("/api/v1/export_raw_data?format=jsonl")
response = requests.post(
url,
json=body,
timeout=10,
headers={
"authorization": f"Bearer {token}",
"Content-Type": "application/json",
},
)
assert len(spans) >= 1
assert all(s.get("trace_id") == parent_trace_id for s in spans)
assert any(s.get("name") == "parent-operation" for s in spans)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "application/x-ndjson"
# Parse JSONL content
jsonl_lines = response.text.strip().split("\n")
assert len(jsonl_lines) == 1, f"Expected at least 1 line, got {len(jsonl_lines)}"
# Verify all returned spans belong to the matched trace
json_objects = [json.loads(line) for line in jsonl_lines]
trace_ids = [obj.get("trace_id") for obj in json_objects]
assert all(tid == parent_trace_id for tid in trace_ids)
# Verify the parent span (returnSpansFrom = "A") is present
span_names = [obj.get("name") for obj in json_objects]
assert "parent-operation" in span_names
def test_export_traces_with_select_fields(