Files
signoz/pkg/querybuilder/exists_expr.go
Nityananda Gohain 8371a70801 perf(querybuilder): compare materialized exists columns explicitly (#12978)
<!--A few plain bullets saying what changed and why, for a reviewer
skimming it - not a wall of text, not a restatement of the diff, not
generated boilerplate.-->
#### Description

Materialized existence checks now render as an explicit comparison
instead of a bare bool column. Results are unchanged; only skip-index
usage improves.

  ```sql
  -- before
  WHERE `attribute_string_gen_ai$$request$$model_exists`
     OR `attribute_string_gen_ai$$provider$$name` = 'anthropic'

  -- after
  WHERE `attribute_string_gen_ai$$request$$model_exists` = true
     OR `attribute_string_gen_ai$$provider$$name` = 'anthropic'
  ```

  <details>
<summary>EXPLAIN indexes = 1 (trace-matching phase, 123M
spans)</summary>

  Before: bare `col_exists`
  ```
  Name: idx_gen_ai_span_exists
  Granules: 15193/15193
  Name: <Combined skip indexes>
  Granules: 15193/15193
  ```

  After: `col_exists = true`
  ```
  Name: idx_gen_ai_span_exists
  Granules: 15193/15193
  Name: <Combined skip indexes>
  Granules: 488/15193
  ```
  </details>

----
- ClickHouse can use a different skip index for each side of an OR and
union the results, but it can't when one side is a bare bool column.
Comparing with `= true` fixes that.
- This shape comes from the AI explorer trace list with a span filter: a
trace qualifies when it has a gen_ai span *and* a span matching the
filter (possibly different spans), so the WHERE is `(gen_ai gate) OR
<filter>` followed by a HAVING.
- Needs the gen_ai materialized columns and `idx_gen_ai_span_exists`
from SigNoz/signoz-otel-collector#929; without them there's no index to
combine.

<!--Reference issues using `Closes #issue-number` to enable automatic
closure on merge. -->
#### Issues closed by this PR
Part of https://github.com/SigNoz/nerve-pod/issues/282

<!--Anything reviewers should keep in mind while reviewing -->
#### Additional Information
- Benchmarked the AI trace list filtered on `gen_ai.provider.name`
against a 123M-span table (direct I/O, caches off): from ~30M spans in
the window, latency drops 16–17% and CPU 35–38%, with ~25x fewer rows
read (123M spans: 510 → 427 ms, 1.5 → 0.9 sCPU). The saved time and CPU
keep growing with span count, so larger windows save more.
- Single-condition filters (`gen_ai.request.model EXISTS` in dashboard
panels, the AND-ed gate in AI aggregations) already pruned with the bare
form; no change there.
2026-09-24 13:15:32 +00:00

112 lines
4.3 KiB
Go

package querybuilder
import (
"fmt"
schema "github.com/SigNoz/signoz-otel-collector/cmd/signozschemamigrator/schema_migrator"
"github.com/SigNoz/signoz/pkg/clickhousesql"
"github.com/SigNoz/signoz/pkg/errors"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
)
// ExistsExpression renders the existence predicate for a key resolved to the given
// columns (negated when exists is false). Comparisons are against constants rendered
// as literals, so the expression carries no bind args and can guard column expressions
// directly. Signal-specific presence checks (body JSON paths, label maps) are handled
// by the field mappers before falling through to this.
func ExistsExpression(columns []*schema.Column, key *telemetrytypes.TelemetryFieldKey, tsStart, tsEnd uint64, fieldExpression string, exists bool) (string, error) {
newColumns, evolutionsEntries, err := qbtypes.SelectEvolutionsForColumns(columns, key.Evolutions, tsStart, tsEnd)
if err != nil {
return "", err
}
if len(newColumns) == 0 {
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "no valid evolution found for field %s in the given time range", key.Name)
}
comparison := func(operator string, value string) string {
return fmt.Sprintf("%s %s %s", fieldExpression, operator, value)
}
if len(newColumns) > 1 {
if exists {
return fieldExpression + " IS NOT NULL", nil
}
return fieldExpression + " IS NULL", nil
}
column := newColumns[0]
switch column.Type.GetType() {
case schema.ColumnTypeEnumJSON:
// the ::String cast in the value expression folds NULL to '', so the
// presence check must address the raw JSON path
columnName := column.Name
if len(evolutionsEntries) > 0 && evolutionsEntries[0] != nil {
columnName = evolutionsEntries[0].ColumnName
}
rawPath := fmt.Sprintf("%s.%s", columnName, clickhousesql.Identifier(key.Name))
if exists {
return rawPath + " IS NOT NULL", nil
}
return rawPath + " IS NULL", nil
case schema.ColumnTypeEnumString,
schema.ColumnTypeEnumFixedString:
if exists {
return comparison("<>", "''"), nil
}
return comparison("=", "''"), nil
case schema.ColumnTypeEnumDateTime64:
zero := fmt.Sprintf("toDateTime64(0, %d)", column.Type.(schema.DateTime64ColumnType).Precision)
if exists {
return comparison("<>", zero), nil
}
return comparison("=", zero), nil
case schema.ColumnTypeEnumLowCardinality:
switch elementType := column.Type.(schema.LowCardinalityColumnType).ElementType; elementType.GetType() {
case schema.ColumnTypeEnumString:
if exists {
return comparison("<>", "''"), nil
}
return comparison("=", "''"), nil
default:
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "exists operator is not supported for low cardinality column type %s", elementType)
}
case schema.ColumnTypeEnumUInt64,
schema.ColumnTypeEnumUInt32,
schema.ColumnTypeEnumUInt8,
schema.ColumnTypeEnumInt8,
schema.ColumnTypeEnumInt16,
schema.ColumnTypeEnumBool:
if exists {
return comparison("<>", "0"), nil
}
return comparison("=", "0"), nil
case schema.ColumnTypeEnumArray:
if exists {
return fmt.Sprintf("notEmpty(%s)", fieldExpression), nil
}
return fmt.Sprintf("empty(%s)", fieldExpression), nil
case schema.ColumnTypeEnumMap:
keyType := column.Type.(schema.MapColumnType).KeyType
if _, ok := keyType.(schema.LowCardinalityColumnType); !ok {
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "key type %s is not supported for map column type %s", keyType, column.Type)
}
switch valueType := column.Type.(schema.MapColumnType).ValueType; valueType.GetType() {
case schema.ColumnTypeEnumString, schema.ColumnTypeEnumBool, schema.ColumnTypeEnumFloat64:
if key.Materialized {
return telemetrytypes.FieldKeyToMaterializedExistsCondition(key, exists), nil
}
leftOperand := fmt.Sprintf("mapContains(%s, %s)", column.Name, clickhousesql.StringLiteral(key.Name))
if exists {
return leftOperand, nil
}
return "NOT " + leftOperand, nil
default:
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "exists operator is not supported for map column type %s", valueType)
}
default:
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "exists operator is not supported for column type %s", column.Type)
}
}