Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Fix](nereids)make agg output unchanged after normalized repeat #36207

Merged
merged 2 commits into from
Jun 13, 2024
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,7 @@
import org.apache.doris.nereids.util.PlanUtils.CollectNonWindowedAggFuncs;

import com.google.common.collect.ImmutableList;
import com.google.common.collect.ImmutableList.Builder;
import com.google.common.collect.ImmutableSet;
import com.google.common.collect.Lists;
import com.google.common.collect.Maps;
Expand Down Expand Up @@ -198,6 +199,8 @@ private static LogicalAggregate<Plan> normalizeRepeat(LogicalRepeat<Plan> repeat
.addAll(groupingSetsUsedSlot)
.addAll(allVirtualSlots)
.build();

normalizedAggOutput = getExprIdUnchangedNormalizedAggOutput(normalizedAggOutput, repeat.getOutputExpressions());
return new LogicalAggregate<>(normalizedAggGroupBy, (List) normalizedAggOutput,
Optional.of(normalizedRepeat), normalizedRepeat);
}
Expand Down Expand Up @@ -460,4 +463,25 @@ public Expression visitWindow(WindowExpression windowExpression, Map<Slot, Alias
return hasNewChildren ? windowExpression.withChildren(newChildren) : windowExpression;
}
}

private List<NamedExpression> getExprIdUnchangedNormalizedAggOutput(List<NamedExpression> normalizedAggOutput,
List<NamedExpression> originalAggOutput) {
Builder<NamedExpression> builder = new ImmutableList.Builder<>();
for (int i = 0; i < originalAggOutput.size(); i++) {
NamedExpression e = normalizedAggOutput.get(i);
// process Expression like Alias(SlotReference#0)#0
if (e instanceof Alias && e.child(0) instanceof SlotReference) {
SlotReference slotReference = (SlotReference) e.child(0);
if (slotReference.getExprId().equals(e.getExprId())) {
e = slotReference;
}
}
// Make the output ExprId unchanged
if (!e.getExprId().equals(originalAggOutput.get(i).getExprId())) {
e = new Alias(originalAggOutput.get(i).getExprId(), e, originalAggOutput.get(i).getName());
}
Comment on lines +479 to +482
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why changed in normalizeToUseSlotRef, could we ensure it not changed in normalizeToUseSlotRef?

builder.add(e);
}
return builder.build();
}
}
Original file line number Diff line number Diff line change
Expand Up @@ -7,3 +7,31 @@
29 -3 32
41 1 \N

-- !test_name_unchange --
\N \N
\N \N
-2169155 -2169155
-2169155 -2169155
-1760025 -1760025
-1760025 -1760025
-27328 -27328
-27328 -27328
-23380 -23380
-23380 -23380
-23025 -23025
-23025 -23025
-127 -127
-127 -127
-88 -88
-88 -88
-73 -73
-73 -73
25 25
25 25
5694 5694
5694 5694
29932 29932
29932 29932
5907087 5907087
5907087 5907087

Original file line number Diff line number Diff line change
Expand Up @@ -54,4 +54,39 @@ suite("grouping_normalize_test"){
sql("SELECT col_int_undef_signed, col_int_undef_signed2, SUM(pk) FROM grouping_normalize_test GROUP BY GROUPING SETS ((col_int_undef_signed, col_int_undef_signed2));")
notContains("VREPEAT_NODE")
}


sql "drop table if exists normalize_repeat_name_unchanged"
sql """create table normalize_repeat_name_unchanged (
col_int_undef_signed int/*agg_type_placeholder*/ ,
col_int_undef_signed2 int/*agg_type_placeholder*/ ,
col_float_undef_signed float/*agg_type_placeholder*/ ,
col_int_undef_signed3 int/*agg_type_placeholder*/ ,
col_int_undef_signed4 int/*agg_type_placeholder*/ ,
col_int_undef_signed5 int/*agg_type_placeholder*/ ,
pk int/*agg_type_placeholder*/
) engine=olap
distributed by hash(pk) buckets 10
properties("replication_num" = "1");"""
sql """
insert into normalize_repeat_name_unchanged(pk,col_int_undef_signed,col_int_undef_signed2,col_float_undef_signed,
col_int_undef_signed3,col_int_undef_signed4,col_int_undef_signed5) values (0,null,-27328,5595590,null,null,5767077),(1,3831,null,87,-14582,21,null),
(2,10131,5907087,28248,2473748,88,-18315),(3,2352090,5694,5173440,null,null,-31126),(4,-26805,29932,null,-55,3148,-6705245),(5,null,null,41,57,-3060427,null),
(6,118,25,3472000,-123,null,-2934940),(7,null,null,-109,112,-7344754,4326526),(8,null,-2169155,-19402,null,null,26943),(9,46,null,1736620,30084,13838,null),
(10,24708,null,null,-806832,-116,676),(11,2232,-23025,null,9665,-27413,13457),(12,-6,-127,-5007917,20521,-48,2709),(13,-72,-127,3258,null,-6394361,-5580),
(14,4494439,-1760025,-16580,66,6562396,-280256),(15,6099281,-73,-5376852,-303421,null,-1843),(16,122,-23380,null,7350221,111,null),
(17,null,null,11356,null,11799,108),(18,-91,-88,39,-29582,null,121),(19,4991662,null,-220,7593505,-54,4086882);"""

qt_test_name_unchange """
SELECT
col_int_undef_signed2 AS C1 ,
col_int_undef_signed2
FROM
normalize_repeat_name_unchanged
GROUP BY
GROUPING SETS (
(col_int_undef_signed2),
(col_int_undef_signed2))
order by 1,2
"""
}
Loading