alamb commented on code in PR #10460:
URL: https://github.com/apache/datafusion/pull/10460#discussion_r1598480059
##########
datafusion/optimizer/src/replace_distinct_aggregate.rs:
##########
@@ -88,60 +94,72 @@ impl OptimizerRule for ReplaceDistinctWithAggregate {
input,
schema,
})) => {
+ let expr_cnt = on_expr.len();
+
// Construct the aggregation expression to be used to fetch
the selected expressions.
- let aggr_expr = select_expr
- .iter()
- .map(|e| {
- Expr::AggregateFunction(AggregateFunction::new(
- AggregateFunctionFunc::FirstValue,
- vec![e.clone()],
- false,
- None,
- sort_expr.clone(),
- None,
- ))
- })
- .collect::<Vec<Expr>>();
+ let aggr_expr =
vec![Expr::AggregateFunction(AggregateFunction::new(
+ AggregateFunctionFunc::FirstValue,
+ select_expr,
+ false,
+ None,
+ sort_expr.clone(),
+ None,
+ ))];
+
+ let aggr_expr = normalize_cols(aggr_expr, input.as_ref())?;
+ let group_expr = normalize_cols(on_expr, input.as_ref())?;
// Build the aggregation plan
- let plan = LogicalPlanBuilder::from(input.as_ref().clone())
- .aggregate(on_expr.clone(), aggr_expr.to_vec())?
- .build()?;
+ let plan = LogicalPlan::Aggregate(Aggregate::try_new(
+ input, group_expr, aggr_expr,
+ )?);
+ let lpb = LogicalPlanBuilder::from(plan);
Review Comment:
```suggestion
// TODO use LogicalPlanBuilder directly rather than
recreating the Aggregate
// when https://github.com/apache/datafusion/issues/10485 is
available
let lpb = LogicalPlanBuilder::from(plan);
```
I filed https://github.com/apache/datafusion/issues/10485 to track this. I
think we could merge this PR as is (I added a comment) and improve the code as
a follow on PR
--
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: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
