Compare commits

...

11 Commits

Author SHA1 Message Date
Piyush Singariya
c0deb2a935 fix: tests 2026-02-02 13:48:18 +05:30
Piyush Singariya
e7923b757d Merge branch 'main' into json-qb-integration 2026-02-02 12:47:22 +05:30
Piyush Singariya
a7bc55bac9 revert: old json QB 2026-01-22 12:53:47 +05:30
Piyush Singariya
55ead8fb20 test: integration tests fixed 2026-01-21 22:34:22 +05:30
Piyush Singariya
0210f02397 Merge branch 'has-jsonqb' into json-qb-integration 2026-01-20 14:28:57 +05:30
Piyush Singariya
e8e4897cc8 fix: tests GroupBy 2026-01-20 12:10:44 +05:30
Piyush Singariya
96fb88aaee fix: ignored .vscode in gitignore 2026-01-20 11:47:34 +05:30
Piyush Singariya
5a00e6c2cd Merge branch 'main' into has-jsonqb 2026-01-20 11:44:32 +05:30
Piyush Singariya
e2500cff7d fix: tests expected queries and values 2026-01-20 11:40:56 +05:30
Piyush Singariya
4864c3bc37 feat: has JSON QB 2026-01-20 11:23:50 +05:30
Piyush Singariya
d5ac192dc4 feat: integration tests for new JSON QB 2026-01-19 12:22:46 +05:30
14 changed files with 1950 additions and 36 deletions

View File

@@ -18,13 +18,17 @@ type responseerroradditional struct {
func AsJSON(cause error) *JSON {
// See if this is an instance of the base error or not
_, c, m, _, u, a := Unwrapb(cause)
_, c, m, err, u, a := Unwrapb(cause)
rea := make([]responseerroradditional, len(a))
for k, v := range a {
rea[k] = responseerroradditional{v}
}
if err != nil {
rea = append(rea, responseerroradditional{err.Error()})
}
return &JSON{
Code: c.String(),
Message: m,

View File

@@ -61,13 +61,11 @@ var (
}
)
type fieldMapper struct {
}
type fieldMapper struct {}
func NewFieldMapper() qbtypes.FieldMapper {
return &fieldMapper{}
}
func (m *fieldMapper) getColumn(_ context.Context, key *telemetrytypes.TelemetryFieldKey) (*schema.Column, error) {
switch key.FieldContext {
case telemetrytypes.FieldContextResource:
@@ -254,12 +252,27 @@ func (m *fieldMapper) buildFieldForJSON(key *telemetrytypes.TelemetryFieldKey) (
"plan length is less than 2 for promoted path: %s", key.Name)
}
// promoted column first then body_json column
// TODO(Piyush): Change this in future for better performance
expr = fmt.Sprintf("coalesce(%s, %s)",
fmt.Sprintf("dynamicElement(%s, '%s')", plan[1].FieldPath(), plan[1].TerminalConfig.ElemType.StringValue()),
expr,
node := plan[1]
promotedExpr := fmt.Sprintf(
"dynamicElement(%s, '%s')",
node.FieldPath(),
node.TerminalConfig.ElemType.StringValue(),
)
// dynamicElement returns NULL for scalar types or an empty array for array types.
if node.TerminalConfig.ElemType.IsArray {
expr = fmt.Sprintf(
"if(length(%s) > 0, %s, %s)",
promotedExpr,
promotedExpr,
expr,
)
} else {
// promoted column first then body_json column
// TODO(Piyush): Change this in future for better performance
expr = fmt.Sprintf("coalesce(%s, %s)", promotedExpr, expr)
}
}
return expr, nil
@@ -282,7 +295,6 @@ func (m *fieldMapper) buildArrayConcat(plan telemetrytypes.JSONAccessPlan) (stri
// Build arrayMap expressions for ALL available branches at the root level.
// Iterate branches in deterministic order (JSON then Dynamic) so generated SQL
// is stable across environments; map iteration order is random in Go.
var arrayMapExpressions []string
for _, node := range plan {
for _, branchType := range node.BranchesInOrder() {

View File

@@ -73,10 +73,10 @@ func (c *jsonConditionBuilder) buildTerminalCondition(node *telemetrytypes.JSONA
// switch operator for array membership checks
switch operator {
case qbtypes.FilterOperatorContains, qbtypes.FilterOperatorIn:
operator = qbtypes.FilterOperatorEqual
case qbtypes.FilterOperatorNotContains, qbtypes.FilterOperatorNotIn:
operator = qbtypes.FilterOperatorNotEqual
case qbtypes.FilterOperatorContains:
operator = qbtypes.FilterOperatorIn
case qbtypes.FilterOperatorNotContains:
operator = qbtypes.FilterOperatorNotIn
}
}
@@ -191,13 +191,14 @@ func (c *jsonConditionBuilder) buildArrayMembershipCondition(node *telemetrytype
arrayExpr = typedArrayExpr()
}
fieldExpr, value := querybuilder.DataTypeCollisionHandledFieldName(&localKeyCopy, value, "x", operator)
key := "x"
fieldExpr, value := querybuilder.DataTypeCollisionHandledFieldName(&localKeyCopy, value, key, operator)
op, err := c.applyOperator(sb, fieldExpr, operator, value)
if err != nil {
return "", err
}
return fmt.Sprintf("arrayExists(%s -> %s, %s)", fieldExpr, op, arrayExpr), nil
return fmt.Sprintf("arrayExists(%s -> %s, %s)", key, op, arrayExpr), nil
}
// recurseArrayHops recursively builds array traversal conditions

View File

@@ -316,7 +316,7 @@ func TestStatementBuilderListQueryBody(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(toFloat64(x) -> toFloat64(x) = ?, dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(toFloat64(x) -> toFloat64(x) = ?, arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> (toFloat64(x) = ?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> (toFloat64(x) = ?), arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{uint64(1747945619), uint64(1747983448), "%1.65%", 1.65, "%1.65%", 1.65, "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{"Key `education[].parameters` is ambiguous, found 2 different combinations of field context / data type: [name=education[].parameters,context=body,datatype=[]float64,jsondatatype=Array(Nullable(Float64)) name=education[].parameters,context=body,datatype=[]dynamic,jsondatatype=Array(Dynamic)]."},
},
@@ -345,7 +345,7 @@ func TestStatementBuilderListQueryBody(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> x = ?, dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> x = ?, arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> (x = ?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> (x = ?), arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{uint64(1747945619), uint64(1747983448), "%true%", true, "%true%", true, "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{"Key `education[].parameters` is ambiguous, found 2 different combinations of field context / data type: [name=education[].parameters,context=body,datatype=[]float64,jsondatatype=Array(Nullable(Float64)) name=education[].parameters,context=body,datatype=[]dynamic,jsondatatype=Array(Dynamic)]."},
},
@@ -360,7 +360,7 @@ func TestStatementBuilderListQueryBody(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(toString(x) -> toString(x) = ?, dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> x = ?, arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> (toString(x) = ?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> (x = ?), arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{uint64(1747945619), uint64(1747983448), "%passed%", "passed", "%passed%", "passed", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{"Key `education[].parameters` is ambiguous, found 2 different combinations of field context / data type: [name=education[].parameters,context=body,datatype=[]float64,jsondatatype=Array(Nullable(Float64)) name=education[].parameters,context=body,datatype=[]dynamic,jsondatatype=Array(Dynamic)]."},
},
@@ -389,7 +389,7 @@ func TestStatementBuilderListQueryBody(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.interests`-> arrayExists(`body_json.interests[].entities`-> arrayExists(`body_json.interests[].entities[].reviews`-> arrayExists(`body_json.interests[].entities[].reviews[].entries`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(Int64))')) OR arrayExists(toFloat64(x) -> toFloat64(x) = ?, dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(Int64))'))), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata`.`positions`, 'Array(JSON(max_dynamic_types=0, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews[].entries`.`metadata`, 'Array(JSON(max_dynamic_types=1, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews`.`entries`, 'Array(JSON(max_dynamic_types=2, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities`.`reviews`, 'Array(JSON(max_dynamic_types=4, max_dynamic_paths=0))')), dynamicElement(`body_json.interests`.`entities`, 'Array(JSON(max_dynamic_types=8, max_dynamic_paths=0))')), dynamicElement(body_json.`interests`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.interests`-> arrayExists(`body_json.interests[].entities`-> arrayExists(`body_json.interests[].entities[].reviews`-> arrayExists(`body_json.interests[].entities[].reviews[].entries`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(String))')) OR arrayExists(toFloat64OrNull(x) -> toFloat64OrNull(x) = ?, dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(String))'))), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata`.`positions`, 'Array(JSON(max_dynamic_types=0, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews[].entries`.`metadata`, 'Array(JSON(max_dynamic_types=1, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews`.`entries`, 'Array(JSON(max_dynamic_types=2, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities`.`reviews`, 'Array(JSON(max_dynamic_types=4, max_dynamic_paths=0))')), dynamicElement(`body_json.interests`.`entities`, 'Array(JSON(max_dynamic_types=8, max_dynamic_paths=0))')), dynamicElement(body_json.`interests`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.interests`-> arrayExists(`body_json.interests[].entities`-> arrayExists(`body_json.interests[].entities[].reviews`-> arrayExists(`body_json.interests[].entities[].reviews[].entries`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(Int64))')) OR arrayExists(x -> (toFloat64(x) = ?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(Int64))'))), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata`.`positions`, 'Array(JSON(max_dynamic_types=0, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews[].entries`.`metadata`, 'Array(JSON(max_dynamic_types=1, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews`.`entries`, 'Array(JSON(max_dynamic_types=2, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities`.`reviews`, 'Array(JSON(max_dynamic_types=4, max_dynamic_paths=0))')), dynamicElement(`body_json.interests`.`entities`, 'Array(JSON(max_dynamic_types=8, max_dynamic_paths=0))')), dynamicElement(body_json.`interests`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.interests`-> arrayExists(`body_json.interests[].entities`-> arrayExists(`body_json.interests[].entities[].reviews`-> arrayExists(`body_json.interests[].entities[].reviews[].entries`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(String))')) OR arrayExists(x -> (toFloat64OrNull(x) = ?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(String))'))), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata`.`positions`, 'Array(JSON(max_dynamic_types=0, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews[].entries`.`metadata`, 'Array(JSON(max_dynamic_types=1, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews`.`entries`, 'Array(JSON(max_dynamic_types=2, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities`.`reviews`, 'Array(JSON(max_dynamic_types=4, max_dynamic_paths=0))')), dynamicElement(`body_json.interests`.`entities`, 'Array(JSON(max_dynamic_types=8, max_dynamic_paths=0))')), dynamicElement(body_json.`interests`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{uint64(1747945619), uint64(1747983448), "%4%", float64(4), "%4%", float64(4), "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{"Key `interests[].entities[].reviews[].entries[].metadata[].positions[].ratings` is ambiguous, found 2 different combinations of field context / data type: [name=interests[].entities[].reviews[].entries[].metadata[].positions[].ratings,context=body,datatype=[]int64,jsondatatype=Array(Nullable(Int64)) name=interests[].entities[].reviews[].entries[].metadata[].positions[].ratings,context=body,datatype=[]string,jsondatatype=Array(Nullable(String))]."},
},
@@ -404,7 +404,7 @@ func TestStatementBuilderListQueryBody(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.interests`-> arrayExists(`body_json.interests[].entities`-> arrayExists(`body_json.interests[].entities[].reviews`-> arrayExists(`body_json.interests[].entities[].reviews[].entries`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(Int64))')) OR arrayExists(toString(x) -> toString(x) = ?, dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(Int64))'))), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata`.`positions`, 'Array(JSON(max_dynamic_types=0, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews[].entries`.`metadata`, 'Array(JSON(max_dynamic_types=1, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews`.`entries`, 'Array(JSON(max_dynamic_types=2, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities`.`reviews`, 'Array(JSON(max_dynamic_types=4, max_dynamic_paths=0))')), dynamicElement(`body_json.interests`.`entities`, 'Array(JSON(max_dynamic_types=8, max_dynamic_paths=0))')), dynamicElement(body_json.`interests`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.interests`-> arrayExists(`body_json.interests[].entities`-> arrayExists(`body_json.interests[].entities[].reviews`-> arrayExists(`body_json.interests[].entities[].reviews[].entries`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(String))')) OR arrayExists(x -> x = ?, dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(String))'))), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata`.`positions`, 'Array(JSON(max_dynamic_types=0, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews[].entries`.`metadata`, 'Array(JSON(max_dynamic_types=1, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews`.`entries`, 'Array(JSON(max_dynamic_types=2, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities`.`reviews`, 'Array(JSON(max_dynamic_types=4, max_dynamic_paths=0))')), dynamicElement(`body_json.interests`.`entities`, 'Array(JSON(max_dynamic_types=8, max_dynamic_paths=0))')), dynamicElement(body_json.`interests`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.interests`-> arrayExists(`body_json.interests[].entities`-> arrayExists(`body_json.interests[].entities[].reviews`-> arrayExists(`body_json.interests[].entities[].reviews[].entries`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(Int64))')) OR arrayExists(x -> (toString(x) = ?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(Int64))'))), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata`.`positions`, 'Array(JSON(max_dynamic_types=0, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews[].entries`.`metadata`, 'Array(JSON(max_dynamic_types=1, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews`.`entries`, 'Array(JSON(max_dynamic_types=2, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities`.`reviews`, 'Array(JSON(max_dynamic_types=4, max_dynamic_paths=0))')), dynamicElement(`body_json.interests`.`entities`, 'Array(JSON(max_dynamic_types=8, max_dynamic_paths=0))')), dynamicElement(body_json.`interests`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))'))) OR (arrayExists(`body_json.interests`-> arrayExists(`body_json.interests[].entities`-> arrayExists(`body_json.interests[].entities[].reviews`-> arrayExists(`body_json.interests[].entities[].reviews[].entries`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata`-> arrayExists(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(String))')) OR arrayExists(x -> (x = ?), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata[].positions`.`ratings`, 'Array(Nullable(String))'))), dynamicElement(`body_json.interests[].entities[].reviews[].entries[].metadata`.`positions`, 'Array(JSON(max_dynamic_types=0, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews[].entries`.`metadata`, 'Array(JSON(max_dynamic_types=1, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities[].reviews`.`entries`, 'Array(JSON(max_dynamic_types=2, max_dynamic_paths=0))')), dynamicElement(`body_json.interests[].entities`.`reviews`, 'Array(JSON(max_dynamic_types=4, max_dynamic_paths=0))')), dynamicElement(`body_json.interests`.`entities`, 'Array(JSON(max_dynamic_types=8, max_dynamic_paths=0))')), dynamicElement(body_json.`interests`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{uint64(1747945619), uint64(1747983448), "%Good%", "Good", "%Good%", "Good", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{"Key `interests[].entities[].reviews[].entries[].metadata[].positions[].ratings` is ambiguous, found 2 different combinations of field context / data type: [name=interests[].entities[].reviews[].entries[].metadata[].positions[].ratings,context=body,datatype=[]int64,jsondatatype=Array(Nullable(Int64)) name=interests[].entities[].reviews[].entries[].metadata[].positions[].ratings,context=body,datatype=[]string,jsondatatype=Array(Nullable(String))]."},
},
@@ -492,7 +492,7 @@ func TestStatementBuilderListQueryBodyPromoted(t *testing.T) {
disableBodyJSONQuery(t)
}()
statementBuilder := buildJSONTestStatementBuilder(t, "education")
statementBuilder := buildJSONTestStatementBuilder(t, "education", "tags")
cases := []struct {
name string
requestType qbtypes.RequestType
@@ -500,6 +500,20 @@ func TestStatementBuilderListQueryBodyPromoted(t *testing.T) {
expected qbtypes.Statement
expectedErr error
}{
{
name: "Has Array promoted uses body fallback",
requestType: qbtypes.RequestTypeRaw,
query: qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]{
Signal: telemetrytypes.SignalLogs,
Filter: &qbtypes.Filter{Expression: "has(body.tags, 'production')"},
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND has(if(length(dynamicElement(body_json_promoted.`tags`, 'Array(Nullable(String))')) > 0, dynamicElement(body_json_promoted.`tags`, 'Array(Nullable(String))'), dynamicElement(body_json.`tags`, 'Array(Nullable(String))')), ?) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{uint64(1747945619), uint64(1747983448), "production", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
},
expectedErr: nil,
},
{
name: "Key inside Array(JSON) exists",
requestType: qbtypes.RequestTypeRaw,
@@ -551,7 +565,7 @@ func TestStatementBuilderListQueryBodyPromoted(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(toFloat64(x) -> toFloat64(x) = ?, dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(toFloat64(x) -> toFloat64(x) = ?, dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'))) OR (arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(toFloat64(x) -> toFloat64(x) = ?, arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(toFloat64(x) -> toFloat64(x) = ?, arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> (toFloat64(x) = ?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> (toFloat64(x) = ?), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'))) OR (arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> (toFloat64(x) = ?), arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> (toFloat64(x) = ?), arrayMap(x->dynamicElement(x, 'Float64'), arrayFilter(x->(dynamicType(x) = 'Float64'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{uint64(1747945619), uint64(1747983448), "%1.65%", 1.65, "%1.65%", 1.65, "%1.65%", 1.65, "%1.65%", 1.65, "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{"Key `education[].parameters` is ambiguous, found 2 different combinations of field context / data type: [name=education[].parameters,context=body,datatype=[]float64,materialized=true,jsondatatype=Array(Nullable(Float64)) name=education[].parameters,context=body,datatype=[]dynamic,materialized=true,jsondatatype=Array(Dynamic)]."},
},
@@ -580,7 +594,7 @@ func TestStatementBuilderListQueryBodyPromoted(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> x = ?, dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> x = ?, dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'))) OR (arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> x = ?, arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> x = ?, arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> (x = ?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> (x = ?), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'))) OR (arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> (x = ?), arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> (x = ?), arrayMap(x->dynamicElement(x, 'Bool'), arrayFilter(x->(dynamicType(x) = 'Bool'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{uint64(1747945619), uint64(1747983448), "%true%", true, "%true%", true, "%true%", true, "%true%", true, "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{"Key `education[].parameters` is ambiguous, found 2 different combinations of field context / data type: [name=education[].parameters,context=body,datatype=[]float64,materialized=true,jsondatatype=Array(Nullable(Float64)) name=education[].parameters,context=body,datatype=[]dynamic,materialized=true,jsondatatype=Array(Dynamic)]."},
},
@@ -595,7 +609,7 @@ func TestStatementBuilderListQueryBodyPromoted(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(toString(x) -> toString(x) = ?, dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(toString(x) -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(toString(x) -> toString(x) = ?, dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'))) OR (arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> x = ?, arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> x = ?, arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE true AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, body_json, body_json_promoted, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND ((arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> (toString(x) = ?), dynamicElement(`body_json.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(x -> LOWER(toString(x)) LIKE LOWER(?), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))')) OR arrayExists(x -> (toString(x) = ?), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Nullable(Float64))'))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'))) OR (arrayExists(`body_json.education`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> (x = ?), arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=0))')) OR arrayExists(`body_json_promoted.education`-> (arrayExists(x -> LOWER(x) LIKE LOWER(?), arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)')))) OR arrayExists(x -> (x = ?), arrayMap(x->dynamicElement(x, 'String'), arrayFilter(x->(dynamicType(x) = 'String'), dynamicElement(`body_json_promoted.education`.`parameters`, 'Array(Dynamic)'))))), dynamicElement(body_json_promoted.`education`, 'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))')))) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{uint64(1747945619), uint64(1747983448), "%passed%", "passed", "%passed%", "passed", "%passed%", "passed", "%passed%", "passed", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
Warnings: []string{"Key `education[].parameters` is ambiguous, found 2 different combinations of field context / data type: [name=education[].parameters,context=body,datatype=[]float64,materialized=true,jsondatatype=Array(Nullable(Float64)) name=education[].parameters,context=body,datatype=[]dynamic,materialized=true,jsondatatype=Array(Dynamic)]."},
},

View File

@@ -8,7 +8,6 @@ import (
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
)
func parseStrValue(valueStr string, operator qbtypes.FilterOperator) (telemetrytypes.FieldDataType, any) {
@@ -49,7 +48,15 @@ func InferDataType(value any, operator qbtypes.FilterOperator, key *telemetrytyp
if len(v) > 0 {
valueType, _ = InferDataType(v[0], operator, key)
}
return valueType, v
// Between/In/NotIn compare a scalar field to multiple values - use scalar type for the field expression.
// For array paths (key has [*] or []), In means array membership so keep array type.
if operator == qbtypes.FilterOperatorBetween || operator == qbtypes.FilterOperatorIn || operator == qbtypes.FilterOperatorNotIn {
if !strings.HasSuffix(key.Name, telemetrytypes.ArrayAnyIndex) && !strings.HasSuffix(key.Name, telemetrytypes.ArraySep) {
return valueType, v
}
}
// convert the scaler type to the array type
return telemetrytypes.ScalerFieldTypeToArrayFieldType[valueType], v
case uint8, uint16, uint32, uint64, int, int8, int16, int32, int64:
valueType = telemetrytypes.FieldDataTypeInt64
case float32, float64:
@@ -60,11 +67,6 @@ func InferDataType(value any, operator qbtypes.FilterOperator, key *telemetrytyp
valueType = telemetrytypes.FieldDataTypeBool
}
// check if it is array
if strings.HasSuffix(key.Name, "[*]") || strings.HasSuffix(key.Name, "[]") {
valueType = telemetrytypes.FieldDataType{String: valuer.NewString(fmt.Sprintf("[]%s", valueType.StringValue()))}
}
return valueType, value
}

View File

@@ -112,7 +112,8 @@ func (t *telemetryMetaStore) buildBodyJSONPaths(ctx context.Context,
}
for _, fieldKey := range fieldKeys {
fieldKey.Materialized = promoted.Contains(fieldKey.Name)
promotedKey := strings.Split(fieldKey.Name, telemetrytypes.ArraySep)[0]
fieldKey.Materialized = promoted.Contains(promotedKey)
fieldKey.Indexes = indexes[fieldKey.Name]
}
@@ -501,7 +502,8 @@ func (t *telemetryMetaStore) GetPromotedPaths(ctx context.Context, paths ...stri
sb := sqlbuilder.Select("path").From(fmt.Sprintf("%s.%s", DBName, PromotedPathsTableName))
pathConditions := []string{}
for _, path := range paths {
pathConditions = append(pathConditions, sb.Equal("path", path))
split := strings.Split(path, telemetrytypes.ArraySep)
pathConditions = append(pathConditions, sb.Equal("path", split[0]))
}
sb.Where(sb.Or(pathConditions...))

View File

@@ -93,6 +93,14 @@ var (
FieldDataTypeArrayFloat64: "Array(Float64)",
FieldDataTypeArrayBool: "Array(Bool)",
}
ScalerFieldTypeToArrayFieldType = map[FieldDataType]FieldDataType{
FieldDataTypeString: FieldDataTypeArrayString,
FieldDataTypeBool: FieldDataTypeArrayBool,
FieldDataTypeNumber: FieldDataTypeArrayNumber,
FieldDataTypeInt64: FieldDataTypeArrayInt64,
FieldDataTypeFloat64: FieldDataTypeArrayFloat64,
}
)
func (f FieldDataType) CHDataType() string {

View File

@@ -65,6 +65,7 @@ func TestJSONTypeSet() (map[string][]JSONDataType, MetadataStore) {
"interests[].entities[].reviews[].entries[].metadata[].positions[].unit": {String},
"interests[].entities[].reviews[].entries[].metadata[].positions[].ratings": {ArrayInt64, ArrayString},
"message": {String},
"tags": {ArrayString},
}
return types, nil

View File

@@ -1,5 +1,19 @@
import json
import os
import tempfile
import pytest
if not os.environ.get("DOCKER_CONFIG"):
os.environ["DOCKER_CONFIG"] = tempfile.mkdtemp(prefix="docker-config-")
os.environ.setdefault("DOCKER_CREDENTIAL_STORE", "")
os.environ.setdefault("DOCKER_CREDENTIAL_HELPER", "")
docker_config_path = os.path.join(os.environ["DOCKER_CONFIG"], "config.json")
if not os.path.exists(docker_config_path):
with open(docker_config_path, "w", encoding="utf-8") as config_file:
json.dump({"auths": {}, "credsStore": ""}, config_file)
pytest_plugins = [
"fixtures.auth",
"fixtures.clickhouse",
@@ -19,6 +33,7 @@ pytest_plugins = [
"fixtures.idp",
"fixtures.idputils",
"fixtures.notification_channel",
"fixtures.jsontypeexporter",
]
@@ -56,7 +71,7 @@ def pytest_addoption(parser: pytest.Parser):
parser.addoption(
"--clickhouse-version",
action="store",
default="25.5.6",
default="25.8.6",
help="clickhouse version",
)
parser.addoption(
@@ -71,3 +86,5 @@ def pytest_addoption(parser: pytest.Parser):
default="v0.129.7",
help="schema migrator version",
)

View File

@@ -0,0 +1,408 @@
"""
Simpler version of jsontypeexporter for test fixtures.
This exports JSON type metadata to the path_types table by parsing JSON bodies
and extracting all paths with their types, similar to how the real jsontypeexporter works.
"""
import datetime
import json
from abc import ABC
from typing import TYPE_CHECKING, Any, Callable, Dict, Generator, List, Optional, Set, Union
import numpy as np
import pytest
from fixtures import types
if TYPE_CHECKING:
from fixtures.logs import Logs
class JSONPathType(ABC):
"""Represents a JSON path with its type information"""
path: str
type: str
last_seen: np.uint64
def __init__(
self,
path: str,
type: str, # pylint: disable=redefined-builtin
last_seen: Optional[datetime.datetime] = None,
) -> None:
self.path = path
self.type = type
if last_seen is None:
last_seen = datetime.datetime.now()
self.last_seen = np.uint64(int(last_seen.timestamp() * 1e9))
def np_arr(self) -> np.array:
"""Return path type data as numpy array for database insertion"""
return np.array([self.path, self.type, self.last_seen])
# Constants matching jsontypeexporter
ARRAY_SEPARATOR = "[]." # Used in paths like "education[].name"
ARRAY_SUFFIX = "[]" # Used when traversing into array element objects
def _infer_array_type(elements: List[Any]) -> Optional[str]:
"""
Infer array type from array elements, matching jsontypeexporter's inferArrayMask logic.
Returns None if no valid type can be inferred.
"""
if len(elements) == 0:
return None
# Collect element types (matching Go: types := make([]pcommon.ValueType, 0, s.Len()))
types = []
for elem in elements:
if elem is None:
continue
if isinstance(elem, dict):
types.append("JSON")
elif isinstance(elem, str):
types.append("String")
elif isinstance(elem, bool):
types.append("Bool")
elif isinstance(elem, float):
types.append("Float64")
elif isinstance(elem, int):
types.append("Int64")
if len(types) == 0:
return None
# Get unique types (matching Go: unique := make(map[pcommon.ValueType]struct{}, len(types)))
unique = set(types)
# Classify types (matching Go logic)
has_json = "JSON" in unique
has_primitive = any(t in unique for t in ["String", "Bool", "Float64", "Int64"])
if has_json:
# If only JSON → Array(JSON) (no primitive types)
if not has_primitive:
return "Array(JSON)"
# If there's JSON + any primitive → Dynamic
return "Array(Dynamic)"
# ---- Primitive Type Resolution ----
if "String" in unique:
# Strings cannot coerce with any other primitive
if len(unique) > 1:
return "Array(Dynamic)"
return "Array(Nullable(String))"
if "Float64" in unique:
return "Array(Nullable(Float64))"
if "Int64" in unique:
return "Array(Nullable(Int64))"
if "Bool" in unique:
return "Array(Nullable(Bool))"
return "Array(Dynamic)"
def _python_type_to_clickhouse_type(value: Any) -> str:
"""
Convert Python type to ClickHouse JSON type string.
Maps Python types to ClickHouse JSON data types.
"""
if value is None:
return "String" # Default for null values
if isinstance(value, bool):
return "Bool"
elif isinstance(value, int):
return "Int64"
elif isinstance(value, float):
return "Float64"
elif isinstance(value, str):
return "String"
elif isinstance(value, list):
# Use the sophisticated array type inference
array_type = _infer_array_type(value)
return array_type if array_type else "Array(Dynamic)"
elif isinstance(value, dict):
return "JSON"
else:
return "String" # Default fallback
def _extract_json_paths(
obj: Any,
current_path: str = "",
path_types: Optional[Dict[str, Set[str]]] = None,
level: int = 0,
) -> Dict[str, Set[str]]:
"""
Recursively extract all paths and their types from a JSON object.
Matches jsontypeexporter's analyzePValue logic.
Args:
obj: The JSON object to traverse
current_path: Current path being built (e.g., "user.name")
path_types: Dictionary mapping paths to sets of types found
level: Current nesting level (for depth limiting)
Returns:
Dictionary mapping paths to sets of type strings
"""
if path_types is None:
path_types = {}
if obj is None:
if current_path:
if current_path not in path_types:
path_types[current_path] = set()
path_types[current_path].add("String") # Null defaults to String
return path_types
if isinstance(obj, dict):
# For objects, add the object itself and recurse into keys
if current_path:
if current_path not in path_types:
path_types[current_path] = set()
path_types[current_path].add("JSON")
for key, value in obj.items():
# Build the path for this key
if current_path:
new_path = f"{current_path}.{key}"
else:
new_path = key
# Recurse into the value
_extract_json_paths(value, new_path, path_types, level + 1)
elif isinstance(obj, list):
# Skip empty arrays
if len(obj) == 0:
return path_types
# Collect types from array elements (matching Go: types := make([]pcommon.ValueType, 0, s.Len()))
types = []
for item in obj:
if isinstance(item, dict):
# When traversing into array element objects, use ArraySuffix ([])
# This matches: prefix+ArraySuffix in the Go code
# Example: if current_path is "education", we use "education[]" to traverse into objects
array_prefix = current_path + ARRAY_SUFFIX if current_path else ""
for key, value in item.items():
if array_prefix:
# Use array separator: education[].name
array_path = f"{array_prefix}.{key}"
else:
array_path = key
# Recurse without increasing level (matching Go behavior)
_extract_json_paths(value, array_path, path_types, level)
types.append("JSON")
elif isinstance(item, list):
# Arrays inside arrays are not supported - skip the whole path
# Matching Go: e.logger.Error("arrays inside arrays are not supported!", ...); return nil
return path_types
elif isinstance(item, str):
types.append("String")
elif isinstance(item, bool):
types.append("Bool")
elif isinstance(item, float):
types.append("Float64")
elif isinstance(item, int):
types.append("Int64")
# Infer array type from collected types (matching Go: if mask := inferArrayMask(types); mask != 0)
if len(types) > 0:
array_type = _infer_array_type(obj)
if array_type and current_path:
if current_path not in path_types:
path_types[current_path] = set()
path_types[current_path].add(array_type)
else:
# Primitive value (string, number, bool)
if current_path:
if current_path not in path_types:
path_types[current_path] = set()
obj_type = _python_type_to_clickhouse_type(obj)
path_types[current_path].add(obj_type)
return path_types
def _parse_json_bodies_and_extract_paths(
json_bodies: List[str],
timestamp: Optional[datetime.datetime] = None,
) -> List[JSONPathType]:
"""
Parse JSON bodies and extract all paths with their types.
This mimics the behavior of jsontypeexporter.
Args:
json_bodies: List of JSON body strings to parse
timestamp: Timestamp to use for last_seen (defaults to now)
Returns:
List of JSONPathType objects with all discovered paths and types
"""
if timestamp is None:
timestamp = datetime.datetime.now()
# Aggregate all paths and their types across all JSON bodies
all_path_types: Dict[str, Set[str]] = {}
for json_body in json_bodies:
try:
parsed = json.loads(json_body)
_extract_json_paths(parsed, "", all_path_types, level=0)
except (json.JSONDecodeError, TypeError):
# Skip invalid JSON
continue
# Convert to list of JSONPathType objects
# Each path can have multiple types, so we create one JSONPathType per type
path_type_objects: List[JSONPathType] = []
for path, types_set in all_path_types.items():
for type_str in types_set:
path_type_objects.append(
JSONPathType(path=path, type=type_str, last_seen=timestamp)
)
return path_type_objects
@pytest.fixture(name="export_json_types", scope="function")
def export_json_types(
clickhouse: types.TestContainerClickhouse,
request: pytest.FixtureRequest, # To access migrator fixture
) -> Generator[Callable[[Union[List[JSONPathType], List[str], List[Any]]], None], Any, None]:
"""
Fixture for exporting JSON type metadata to the path_types table.
This is a simpler version of jsontypeexporter for test fixtures.
The function can accept:
1. List of JSONPathType objects (manual specification)
2. List of JSON body strings (auto-extract paths)
3. List of Logs objects (extract from body_json field)
Usage examples:
# Manual specification
export_json_types([
JSONPathType(path="user.name", type="String"),
JSONPathType(path="user.age", type="Int64"),
])
# Auto-extract from JSON strings
export_json_types([
'{"user": {"name": "alice", "age": 25}}',
'{"user": {"name": "bob", "age": 30}}',
])
# Auto-extract from Logs objects
export_json_types(logs_list)
"""
# Ensure migrator has run to create the table
try:
request.getfixturevalue("migrator")
except Exception:
# If migrator fixture is not available, that's okay - table might already exist
pass
def _export_json_types(
data: Union[List[JSONPathType], List[str], List[Any]], # List[Logs] but avoiding circular import
) -> None:
"""
Export JSON type metadata to signoz_metadata.distributed_json_path_types table.
This table stores path and type information for body JSON fields.
"""
path_types: List[JSONPathType] = []
if len(data) == 0:
return
# Determine input type and convert to JSONPathType list
first_item = data[0]
if isinstance(first_item, JSONPathType):
# Already JSONPathType objects
path_types = data # type: ignore
elif isinstance(first_item, str):
# List of JSON strings - parse and extract paths
path_types = _parse_json_bodies_and_extract_paths(data) # type: ignore
else:
# Assume it's a list of Logs objects - extract body_json
json_bodies: List[str] = []
for log in data: # type: ignore
# Try to get body_json attribute
if hasattr(log, "body_json") and log.body_json:
json_bodies.append(log.body_json)
elif hasattr(log, "body") and log.body:
# Fallback to body if body_json not available
try:
# Try to parse as JSON
json.loads(log.body)
json_bodies.append(log.body)
except (json.JSONDecodeError, TypeError):
pass
if json_bodies:
path_types = _parse_json_bodies_and_extract_paths(json_bodies)
if len(path_types) == 0:
return
clickhouse.conn.insert(
database="signoz_metadata",
table="distributed_json_path_types",
data=[path_type.np_arr() for path_type in path_types],
column_names=[
"path",
"type",
"last_seen",
],
)
yield _export_json_types
# Cleanup - truncate the local table after tests (following pattern from logs fixture)
clickhouse.conn.query(
f"TRUNCATE TABLE signoz_metadata.json_path_types ON CLUSTER '{clickhouse.env['SIGNOZ_TELEMETRYSTORE_CLICKHOUSE_CLUSTER']}' SYNC"
)
@pytest.fixture(name="export_promoted_paths", scope="function")
def export_promoted_paths(
clickhouse: types.TestContainerClickhouse,
request: pytest.FixtureRequest, # To access migrator fixture
) -> Generator[Callable[[List[str]], None], Any, None]:
"""
Fixture for exporting promoted JSON paths to the promoted paths table.
"""
# Ensure migrator has run to create the table
try:
request.getfixturevalue("migrator")
except Exception:
# If migrator fixture is not available, that's okay - table might already exist
pass
def _export_promoted_paths(paths: List[str]) -> None:
if len(paths) == 0:
return
now_ms = int(datetime.datetime.now().timestamp() * 1000)
rows = [(path, now_ms) for path in paths]
clickhouse.conn.insert(
database="signoz_metadata",
table="distributed_json_promoted_paths",
data=rows,
column_names=[
"path",
"created_at",
],
)
yield _export_promoted_paths
clickhouse.conn.query(
f"TRUNCATE TABLE signoz_metadata.json_promoted_paths ON CLUSTER '{clickhouse.env['SIGNOZ_TELEMETRYSTORE_CLICKHOUSE_CLUSTER']}' SYNC"
)

View File

@@ -99,6 +99,8 @@ class Logs(ABC):
severity_text: str
severity_number: np.uint8
body: str
body_json: str
body_json_promoted: str
attributes_string: dict[str, str]
attributes_number: dict[str, np.float64]
attributes_bool: dict[str, bool]
@@ -118,6 +120,8 @@ class Logs(ABC):
resources: dict[str, Any] = {},
attributes: dict[str, Any] = {},
body: str = "default body",
body_json: Optional[str] = None,
body_json_promoted: Optional[str] = None,
severity_text: str = "INFO",
trace_id: str = "",
span_id: str = "",
@@ -161,6 +165,32 @@ class Logs(ABC):
# Set body
self.body = body
# Set body_json - if body is JSON, parse and stringify it, otherwise use empty string
# ClickHouse accepts String input for JSON column
if body_json is not None:
self.body_json = body_json
else:
# Try to parse body as JSON, if successful use it, otherwise use empty string
try:
json.loads(body)
self.body_json = body
except (json.JSONDecodeError, TypeError):
self.body_json = ""
# Set body_json_promoted - must be valid JSON
# Tests will explicitly pass promoted column's content, but we validate it
if body_json_promoted is not None:
# Validate that it's valid JSON
try:
json.loads(body_json_promoted)
self.body_json_promoted = body_json_promoted
except (json.JSONDecodeError, TypeError):
# If invalid, default to empty JSON object
self.body_json_promoted = "{}"
else:
# Default to empty JSON object (valid JSON)
self.body_json_promoted = "{}"
# Process resources and attributes
self.resources_string = {k: str(v) for k, v in resources.items()}
@@ -318,6 +348,8 @@ class Logs(ABC):
self.severity_text,
self.severity_number,
self.body,
self.body_json,
self.body_json_promoted,
self.attributes_string,
self.attributes_number,
self.attributes_bool,
@@ -409,6 +441,8 @@ def insert_logs(
"severity_text",
"severity_number",
"body",
"body_json",
"body_json_promoted",
"attributes_string",
"attributes_number",
"attributes_bool",

View File

@@ -20,12 +20,14 @@ def migrator(
"""
def create() -> None:
version = request.config.getoption("--schema-migrator-version")
# Hardcode version for new QB tests
version = "v0.129.13-rc.1"
client = docker.from_env()
container = client.containers.run(
image=f"signoz/signoz-schema-migrator:{version}",
command=f"sync --replication=true --cluster-name=cluster --up= --dsn={clickhouse.env["SIGNOZ_TELEMETRYSTORE_CLICKHOUSE_DSN"]}",
environment={"ENABLE_LOGS_MIGRATIONS_V2": "1"},
detach=True,
auto_remove=False,
network=network.id,
@@ -44,6 +46,7 @@ def migrator(
container = client.containers.run(
image=f"signoz/signoz-schema-migrator:{version}",
command=f"async --replication=true --cluster-name=cluster --up= --dsn={clickhouse.env["SIGNOZ_TELEMETRYSTORE_CLICKHOUSE_DSN"]}",
environment={"ENABLE_LOGS_MIGRATIONS_V2": "1"},
detach=True,
auto_remove=False,
network=network.id,

View File

@@ -69,6 +69,7 @@ def signoz( # pylint: disable=too-many-arguments,too-many-positional-arguments
"SIGNOZ_GLOBAL_INGESTION__URL": "https://ingest.test.signoz.cloud",
"SIGNOZ_USER_PASSWORD_RESET_ALLOW__SELF": True,
"SIGNOZ_USER_PASSWORD_RESET_MAX__TOKEN__LIFETIME": "6h",
"BODY_JSON_QUERY_ENABLED": "true",
}
| sqlstore.env
| clickhouse.env

File diff suppressed because it is too large Load Diff