amaliujia commented on a change in pull request #2107:
URL: https://github.com/apache/calcite/pull/2107#discussion_r474304854
##########
File path: core/src/test/resources/sql/blank.iq
##########
@@ -90,16 +90,20 @@ insert into table2 values (NULL, 1), (2, 1);
!set lateDecorrelate true
select i, j from table1 where table1.j NOT IN (select i from table2 where
table1.i=table2.j);
EnumerableCalc(expr#0..7=[{inputs}], expr#8=[0], expr#9=[=($t3, $t8)],
expr#10=[IS NULL($t1)], expr#11=[IS NOT NULL($t7)], expr#12=[<($t4, $t3)],
expr#13=[OR($t10, $t11, $t12)], expr#14=[IS NOT TRUE($t13)], expr#15=[OR($t9,
$t14)], proj#0..1=[{exprs}], $condition=[$t15])
- EnumerableHashJoin(condition=[AND(=($0, $6), =($1, $5))], joinType=[left])
- EnumerableHashJoin(condition=[=($0, $2)], joinType=[left])
- EnumerableTableScan(table=[[BLANK, TABLE1]])
- EnumerableAggregate(group=[{1}], c=[COUNT()], ck=[COUNT($0)])
- EnumerableCalc(expr#0..1=[{inputs}], expr#2=[IS NOT NULL($t1)],
proj#0..1=[{exprs}], $condition=[$t2])
- EnumerableTableScan(table=[[BLANK, TABLE2]])
- EnumerableCalc(expr#0..1=[{inputs}], expr#2=[true], proj#0..2=[{exprs}])
- EnumerableAggregate(group=[{0, 1}])
- EnumerableCalc(expr#0..1=[{inputs}], expr#2=[IS NOT NULL($t1)],
expr#3=[IS NOT NULL($t0)], expr#4=[AND($t2, $t3)], proj#0..1=[{exprs}],
$condition=[$t4])
- EnumerableTableScan(table=[[BLANK, TABLE2]])
+ EnumerableMergeJoin(condition=[AND(=($0, $6), =($1, $5))], joinType=[left])
Review comment:
Just an observation, looks like in many cases, Calcite believes merge
join are better than hash 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.
For queries about this service, please contact Infrastructure at:
[email protected]
