[fix](Nereids)Fix the bug that count(*) does not push down for tables…

… with only one column. (#25222) after pr #22115 . Fixed the bug that when selecting count(*) from table, if the table has only one column, the aggregate count is not pushed down.
apache · Oct 11, 2023 · d9357cc · d9357cc
1 parent d9fccf5
commit d9357cc
Show file tree

Hide file tree

Showing 3 changed files with 110 additions and 0 deletions.
diff --git a/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/RuleType.java b/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/RuleType.java
@@ -324,6 +324,7 @@ public enum RuleType {
     LOGICAL_ASSERT_NUM_ROWS_TO_PHYSICAL_ASSERT_NUM_ROWS(RuleTypeClass.IMPLEMENTATION),
     STORAGE_LAYER_AGGREGATE_WITHOUT_PROJECT(RuleTypeClass.IMPLEMENTATION),
     STORAGE_LAYER_AGGREGATE_WITH_PROJECT(RuleTypeClass.IMPLEMENTATION),
+    STORAGE_LAYER_AGGREGATE_WITHOUT_PROJECT_FOR_FILE_SCAN(RuleTypeClass.IMPLEMENTATION),
     STORAGE_LAYER_AGGREGATE_WITH_PROJECT_FOR_FILE_SCAN(RuleTypeClass.IMPLEMENTATION),
     COUNT_ON_INDEX(RuleTypeClass.IMPLEMENTATION),
     COUNT_ON_INDEX_WITHOUT_PROJECT(RuleTypeClass.IMPLEMENTATION),

diff --git a/...core/src/main/java/org/apache/doris/nereids/rules/implementation/AggregateStrategies.java b/...core/src/main/java/org/apache/doris/nereids/rules/implementation/AggregateStrategies.java
@@ -161,6 +161,13 @@ public List<Rule> buildRules() {
                     return storageLayerAggregate(agg, project, olapScan, ctx.cascadesContext);
                 })
             ),
+            RuleType.STORAGE_LAYER_AGGREGATE_WITHOUT_PROJECT_FOR_FILE_SCAN.build(
+                logicalAggregate(
+                    logicalFileScan()
+                )
+                    .when(agg -> agg.isNormalized() && enablePushDownNoGroupAgg())
+                    .thenApply(ctx -> storageLayerAggregate(ctx.root, null, ctx.root.child(), ctx.cascadesContext))
+            ),
             RuleType.STORAGE_LAYER_AGGREGATE_WITH_PROJECT_FOR_FILE_SCAN.build(
                 logicalAggregate(
                     logicalProject(

diff --git a/regression-test/suites/external_table_p2/hive/test_select_count_optimize.groovy b/regression-test/suites/external_table_p2/hive/test_select_count_optimize.groovy
@@ -88,6 +88,108 @@ suite("test_select_count_optimize", "p2") {
 
         qt_sql """ select count(*) as a from tpch_1000.nation group by n_regionkey order by a;""" 
 
+        explain {
+
+            sql "select count(*) from tpch_1000_parquet.nation;"
+
+            contains "pushdown agg=COUNT"
+        } 
+
+        explain {
+
+            sql "select count(1) from tpch_1000_parquet.nation;"
+
+            contains "pushdown agg=COUNT"
+        } 
+
+
+        explain {
+
+            sql "select count(2) from tpch_1000_parquet.nation;"
+
+            contains "pushdown agg=COUNT"
+        } 
+
+
+        explain {
+
+            sql "select count(n_name) from tpch_1000_parquet.nation;"
+
+            notContains "pushdown agg=COUNT"
+        }
+
+        explain {
+
+            sql "select count(n_name) from tpch_1000_parquet.nation  where n_nationkey = 1;"
+
+            notContains "pushdown agg=COUNT"
+        }
+
+        explain {
+
+            sql "select count(*) from tpch_1000_parquet.nation group by n_regionkey ;"
+
+            notContains "pushdown agg=COUNT"
+        }
+
+
+        explain {
+
+            sql " select count(*) from  multi_catalog.test_csv_format_error; "
+
+            contains "pushdown agg=COUNT"
+        } 
+
+        explain {
+            sql "select count(*) from  multi_catalog.hits_orc ; "
+
+            contains "pushdown agg=COUNT"
+
+        }
+
+
+        explain {
+            sql "select count(*) from  multi_catalog.hits_orc ; "
+
+            contains "pushdown agg=COUNT"
+
+        }
+
+
+        explain {
+
+            sql "select count(*) from multi_catalog.parquet_one_column;"
+
+            contains "pushdown agg=COUNT"
+        }
+
+        explain {
+
+            sql "select count(col1) from multi_catalog.parquet_one_column;"
+
+            notContains "pushdown agg=COUNT"
+        }
+
+        explain {
+
+            sql "select count(*) from multi_catalog.parquet_two_column;"
+
+            contains "pushdown agg=COUNT"
+        }
+        explain {
+
+            sql "select count(*) from multi_catalog.parquet_two_column where col1 = 1;"
+
+            notContains "pushdown agg=COUNT"
+        }
+
+
+        explain {
+
+            sql "select count(*) from  multi_catalog.logs2_orc;"
+
+            contains "pushdown agg=COUNT"
+        }
     }
 }