maytasm commented on code in PR #18426:
URL: https://github.com/apache/druid/pull/18426#discussion_r2301500029
##########
extensions-core/datasketches/src/main/java/org/apache/druid/query/aggregation/datasketches/hll/sql/HllPostAggExprMacros.java:
##########
@@ -88,5 +105,44 @@ public ExprEval eval(ObjectBinding bindings)
return round ? ExprEval.of(Math.round(estimate)) : ExprEval.of(estimate);
}
}
+
+ public static class HllSketchEstimateWithErrorBoundsExpr extends
ExprMacroTable.BaseScalarMacroFunctionExpr
+ {
+ private Expr estimateExpr;
+ private Expr numStdDev;
+
+ public
HllSketchEstimateWithErrorBoundsExpr(HllSketchEstimateWithErrorBoundsExprMacro
macro, List<Expr> args)
+ {
+ super(macro, args);
+ this.estimateExpr = args.get(0);
+ if (args.size() == 2) {
+ numStdDev = args.get(1);
+ }
+ }
+
+ @Nullable
+ @Override
+ public ExpressionType getOutputType(InputBindingInspector inspector)
+ {
+ return ExpressionType.DOUBLE_ARRAY;
+ }
+
+ @Override
+ public ExprEval eval(ObjectBinding bindings)
+ {
+ int numStdDevs =
HllSketchToEstimateWithBoundsPostAggregator.DEFAULT_NUM_STD_DEVS;
+ ExprEval eval = estimateExpr.eval(bindings);
+ if (numStdDev != null) {
+ numStdDevs = numStdDev.eval(bindings).asInt();
+ }
+
+ final Object valObj = eval.value();
+ if (valObj == null) {
+ return ExprEval.ofDoubleArray(new Double[]{0.0D, 0.0D, 0.0D});
+ }
+ HllSketchHolder sketch = HllSketchHolder.fromObj(valObj);
+ return ExprEval.ofDoubleArray(new Double[]{sketch.getEstimate(),
sketch.getLowerBound(numStdDevs), sketch.getUpperBound(numStdDevs)});
Review Comment:
Yeah. HLL and Theta Sketch are different functions. The inconsistency
already exists with the PostAggs (group by) versions for these functions. I am
keeping the HLL consistence between using as an expression and as an PostAggs
and similarly the Theta Sketch consistence between using as an expression and
as an PostAggs.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
