From fab48f54b1f82ff34e317b4ab586241039b9c756 Mon Sep 17 00:00:00 2001
From: starocean999 <40539150+starocean999@users.noreply.github.com>
Date: Mon, 18 Mar 2024 12:08:06 +0800
Subject: [PATCH] [enhancement](nereids)simplify OneRowRelation scalar subquery
 (#32276)

select count() from t where dt > (select '2024-02-02 00:00:00');
-->
select count() from t where dt > '2024-02-02 00:00:00';
---
 .../nereids/rules/analysis/SubExprAnalyzer.java | 17 +++++++++++++++++
 .../nereids_p0/subquery/test_subquery.groovy    | 16 ++++++++++++++++
 2 files changed, 33 insertions(+)

diff --git a/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/analysis/SubExprAnalyzer.java b/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/analysis/SubExprAnalyzer.java
index 047b0efbd2..c3c17c7834 100644
--- a/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/analysis/SubExprAnalyzer.java
+++ b/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/analysis/SubExprAnalyzer.java
@@ -20,6 +20,7 @@ package org.apache.doris.nereids.rules.analysis;
 import org.apache.doris.nereids.CascadesContext;
 import org.apache.doris.nereids.analyzer.Scope;
 import org.apache.doris.nereids.exceptions.AnalysisException;
+import org.apache.doris.nereids.trees.expressions.Alias;
 import org.apache.doris.nereids.trees.expressions.Exists;
 import org.apache.doris.nereids.trees.expressions.Expression;
 import org.apache.doris.nereids.trees.expressions.InSubquery;
@@ -33,7 +34,9 @@ import org.apache.doris.nereids.trees.expressions.visitor.DefaultExpressionRewri
 import org.apache.doris.nereids.trees.plans.Plan;
 import org.apache.doris.nereids.trees.plans.logical.LogicalAggregate;
 import org.apache.doris.nereids.trees.plans.logical.LogicalLimit;
+import org.apache.doris.nereids.trees.plans.logical.LogicalOneRowRelation;
 import org.apache.doris.nereids.trees.plans.logical.LogicalPlan;
+import org.apache.doris.nereids.trees.plans.logical.LogicalProject;
 
 import com.google.common.collect.ImmutableList;
 import com.google.common.collect.ImmutableSet;
@@ -106,6 +109,20 @@ class SubExprAnalyzer<T> extends DefaultExpressionRewriter<T> {
         checkHasAgg(analyzedResult);
         checkHasNoGroupBy(analyzedResult);
 
+        // if scalar subquery is like select '2024-02-02 00:00:00'
+        // we can just return the constant expr '2024-02-02 00:00:00'
+        if (analyzedResult.getLogicalPlan() instanceof LogicalProject) {
+            LogicalProject project = (LogicalProject) analyzedResult.getLogicalPlan();
+            if (project.child() instanceof LogicalOneRowRelation
+                    && project.getProjects().size() == 1
+                    && project.getProjects().get(0) instanceof Alias) {
+                Alias alias = (Alias) project.getProjects().get(0);
+                if (alias.isConstant()) {
+                    return alias.child();
+                }
+            }
+        }
+
         return new ScalarSubquery(analyzedResult.getLogicalPlan(), analyzedResult.getCorrelatedSlots());
     }
 
diff --git a/regression-test/suites/nereids_p0/subquery/test_subquery.groovy b/regression-test/suites/nereids_p0/subquery/test_subquery.groovy
index 6e29d2205c..45825a0acb 100644
--- a/regression-test/suites/nereids_p0/subquery/test_subquery.groovy
+++ b/regression-test/suites/nereids_p0/subquery/test_subquery.groovy
@@ -275,4 +275,20 @@ suite("test_subquery") {
     qt_select_assert_num_row """SELECT * FROM table_100_undef_partitions2_keys3_properties4_distributed_by5 AS t1 WHERE t1.`pk` IN (0, 6, 8, 9, 5)  OR  t1.`pk`  -  0  <  (SELECT `pk` FROM table_5_undef_partitions2_keys3_properties4_distributed_by5 AS t2 WHERE t2.pk = 9) order by t1.pk;"""
     sql """drop table if exists table_100_undef_partitions2_keys3_properties4_distributed_by5"""
     sql """drop table if exists table_5_undef_partitions2_keys3_properties4_distributed_by5"""
+    
+    sql """drop table if exists scalar_subquery_t"""
+    sql """create table scalar_subquery_t (id int , name varchar(32), dt datetime) 
+            partition by range(dt) (from ('2024-02-01 00:00:00') to ('2024-02-07 00:00:00') interval 1 day) 
+            distributed by hash(id) buckets 1 
+            properties ("replication_num"="1");"""
+    sql """insert into scalar_subquery_t values (1, 'Tom', '2024-02-01 23:12:42'), (2, 'Jelly', '2024-02-02 13:53:32'), (3, 'Kat', '2024-02-03 06:42:21');"""
+    explain {
+        sql("""select count(*) from scalar_subquery_t where dt >  (select '2024-02-02 00:00:00');""")
+        contains("partitions=2/")
+    }
+    explain {
+        sql("""select count(*) from scalar_subquery_t where dt >  (select '2024-02-02 00:00:00' from scalar_subquery_t);""")
+        contains("partitions=3/")
+    }
+    sql """drop table if exists scalar_subquery_t"""
 }