iwanttobepowerful commented on PR #4614:
URL: https://github.com/apache/calcite/pull/4614#issuecomment-3489838221
for sql:
```sql
SELECT deptno, (SELECT CASE WHEN SUM(sal) > 10 then 'VIP' else 'Regular' END
expr
FROM emp e
WHERE d.deptno = e.deptno) a
FROM dept d;
```
spark catalyst plan:
```
== Optimized Logical Plan ==
Project [deptno#575, if (isnull(alwaysTrue#621)) Regular else expr#613 AS
a#615] // <-------here is diff
+- Join LeftOuter, (deptno#575 = deptno#586)
:- Project [deptno#575]
: +- Relation spark_catalog.default.dept[deptno#575,dname#576,loc#577]
parquet
+- Aggregate [deptno#586], [CASE WHEN (sum(sal#584) > 10.0) THEN VIP ELSE
Regular END AS expr#613, deptno#586, true AS alwaysTrue#621] //
<-------here is diff
+- Project [sal#584, deptno#586]
+- Filter isnotnull(deptno#586)
+- Relation
spark_catalog.default.emp[empno#578,ename#579,job#580,age#581,mgr#582,hiredate#583,sal#584,comm#585,deptno#586,email#587,create_datetime#588,upsert_time#589]
parquet
```
but this pr plan:
```
LogicalProject(DEPTNO=[$0], A=[$3]), id = 56 // <-------here is diff
LogicalJoin(condition=[=($0, $4)], joinType=[left]), id = 55
LogicalTableScan(table=[[scott, DEPT]]), id = 11
LogicalProject(EXPR=[CASE(>($1, 10.00), 'VIP':VARCHAR(7),
'Regular':VARCHAR(7))], DEPTNO=[$0]), id = 54 // <-------here is diff
LogicalAggregate(group=[{0}], agg#0=[SUM($1)]), id = 53
LogicalProject(DEPTNO=[$7], SAL=[$5]), id = 52
LogicalFilter(condition=[IS NOT NULL($7)]), id = 50
LogicalTableScan(table=[[scott, EMP]]), id = 12
```
so maybe we should write some extra code to fix 'fast' path (i.e. no use
extra left join)
--
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]
