Github user gatorsmile commented on a diff in the pull request:
https://github.com/apache/spark/pull/10678#discussion_r51356217
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
---
@@ -521,38 +522,99 @@ class Analyzer(
*/
object ResolveSortReferences extends Rule[LogicalPlan] {
def apply(plan: LogicalPlan): LogicalPlan = plan resolveOperators {
- case s @ Sort(ordering, global, p @ Project(projectList, child))
- if !s.resolved && p.resolved =>
- val (newOrdering, missing) = resolveAndFindMissing(ordering, p,
child)
+ case s @ Sort(_, _, a: Aggregate) if a.resolved =>
--- End diff --
```
select sum(product) over () from windowData where product > 3
group by product having avg(month) > 0
order by avg(area), product
```
This is not resolvable unless we reimplement the rule
`ResolveAggregateFunctions`. So far, the rule assumes Sort and Aggregate are
adjacent. If you read the plan, you will understand this is not true. This is
not a simple change. Do you want me to do it in a separate JIRA?
```
'Sort ['avg('area) ASC,'product ASC], true
+- Project [_c0#61L]
+- Project [_w0#62L,_c0#61L,_c0#61L]
+- Window [_w0#62L], [(sum(_w0#62L),mode=Complete,isDistinct=false)
windowspecdefinition(ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING)
AS _c0#61L]
+- Project [_w0#62L]
+- Filter havingCondition#63: boolean
+- Aggregate [product#2], [cast(((avg(cast(month#0 as
bigint)),mode=Complete,isDistinct=false) > cast(0 as double)) as boolean) AS
havingCondition#63,cast(product#2 as bigint) AS _w0#62L]
+- Filter (product#2 > 3)
+- Subquery windowdata
+- LogicalRDD [month#0,area#1,product#2],
MapPartitionsRDD[1] at apply at Transformer.scala:22
```
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
