peter-toth commented on code in PR #10356:
URL: https://github.com/apache/datafusion/pull/10356#discussion_r1588889389
##########
datafusion/optimizer/src/analyzer/type_coercion.rs:
##########
@@ -99,25 +102,75 @@ fn analyze_internal(
// select t2.c2 from t1 where t1.c1 in (select t2.c1 from t2 where
t2.c2=t1.c3)
schema.merge(external_schema);
- let mut expr_rewrite = TypeCoercionRewriter { schema: &schema };
-
- let new_expr = plan
- .expressions()
- .into_iter()
- .map(|expr| {
- // ensure aggregate names don't change:
- // https://github.com/apache/datafusion/issues/3555
- rewrite_preserving_name(expr, &mut expr_rewrite)
- })
- .collect::<Result<Vec<_>>>()?;
-
- plan.with_new_exprs(new_expr, new_inputs)
+ let mut expr_rewrite = TypeCoercionRewriter::new(&schema);
+
+ let name_preserver = NamePreserver::new(&plan);
+ // apply coercion rewrite all expressions in the plan indivdually
+ plan.map_expressions(|expr| {
+ let original_name = name_preserver.save(&expr)?;
+ expr.rewrite(&mut expr_rewrite)?
+ .map_data(|expr| original_name.restore(expr))
+ })?
+ // coerce join expressions specially
+ .map_data(|plan| expr_rewrite.coerce_joins(plan))?
Review Comment:
Since `expr_rewrite.coerce_joins(plan)` can change the plan, shouldn't its
result be `Result<Transformed<LogicalPlan>>`? And then here we should probably
use `map_transformed()` instead of the current `map_data()`.
--
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
