ClSlaid commented on code in PR #10460:
URL: https://github.com/apache/datafusion/pull/10460#discussion_r1597540519
##########
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,
+ ))];
Review Comment:
> The difference is that the `select_expr.iter()...` makes a `vec` with the
same number of elements as `select_expr `
>
> To avoid a clone, perhaps you can use `into_iter` instead of `iter`, like
>
> ```rust
> let aggr_expr = select_expr
> .into_iter() // <---- Use into_iter() here
> .map(|e| {
> Expr::AggregateFunction(AggregateFunction::new(
> AggregateFunctionFunc::FirstValue,
> vec![e], // <--- to avoid a clone here
> false,
> None,
> sort_expr.clone(),
> None,
> ))
> })
> .collect::<Vec<Expr>>();
> ```
What bothers me most is that I cannot avoid cloning `sort_expr`. However, it
looks like the only way to make it correct.
--
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]
