fix(explain): render aggregate expressions correctly when CSE is applied

.gitignore

@@ -75,3 +75,8 @@ rat.txt
 
 # data generated by examples
 datafusion-examples/examples/datafusion-examples/
+
+output.txt
+test_output.txt
+datafusion/sqllogictest/test_files/scratch/
+

datafusion/physical-plan/src/aggregates/mod.rs

@@ -16,7 +16,6 @@
 // under the License.
 
 //! Aggregates functionalities
-
 use std::any::Any;
 use std::sync::Arc;
 
@@ -1070,15 +1069,32 @@ impl DisplayAs for AggregateExec {
     ) -> std::fmt::Result {
         match t {
             DisplayFormatType::Default | DisplayFormatType::Verbose => {
-                let format_expr_with_alias =
-                    |(e, alias): &(Arc<dyn PhysicalExpr>, String)| -> String {
-                        let e = e.to_string();
-                        if &e != alias {
-                            format!("{e} as {alias}")
-                        } else {
-                            e
-                        }
-                    };
+	fn normalize_literal_display(s: &str) -> String {
+                    // already quoted → keep
+                    if s.starts_with('"') || s.starts_with('\'') {
+                        return s.to_string();
+                    }
+
+                    // numeric literal → keep unquoted
+                    if s.chars().all(|c| c.is_ascii_digit()) {
+                        return s.to_string();
+                    }
+
+                    // everything else → quote
+                    format!("\"{s}\"")
+                }
+let format_expr_with_alias =
+    |(expr, alias): &(Arc<dyn PhysicalExpr>, String)| -> String {
+        let display = expr.human_display().to_string();
+
+        if display.is_empty() {
+            alias.clone()
+        } else if display == *alias {
+            display
+        } else {
+            format!("{display} as {alias}")
+        }
+    };
 
                 write!(f, "AggregateExec: mode={:?}", self.mode)?;
                 let g: Vec<String> = if self.group_by.is_single() {
@@ -1116,8 +1132,14 @@ impl DisplayAs for AggregateExec {
                 let a: Vec<String> = self
                     .aggr_expr
                     .iter()
-                    .map(|agg| agg.name().to_string())
+                    .map(|agg| {
+                        let expr = agg.human_display().to_string();
+                        let alias = agg.name().to_string();
+
+                        if expr.is_empty() { alias.clone() } else { expr }
+                    })
                     .collect();
+
                 write!(f, ", aggr=[{}]", a.join(", "))?;
                 if let Some(limit) = self.limit {
                     write!(f, ", lim=[{limit}]")?;

datafusion/sqllogictest/test_files/aggregate.slt

@@ -7730,7 +7730,8 @@ logical_plan
 15)------------EmptyRelation: rows=1
 physical_plan
 01)ProjectionExec: expr=[last_value(a.foo) ORDER BY [a.foo ASC NULLS LAST]@1 as last_value(a.foo) ORDER BY [a.foo ASC NULLS LAST], sum(DISTINCT Int64(1))@2 as sum(DISTINCT Int64(1))]
-02)--AggregateExec: mode=FinalPartitioned, gby=[id@0 as id], aggr=[last_value(a.foo) ORDER BY [a.foo ASC NULLS LAST], sum(DISTINCT Int64(1))], ordering_mode=Sorted
+02)--AggregateExec: mode=FinalPartitioned, gby=[id@0 as id], aggr=[last_value(a.foo) ORDER BY [a.foo ASC NULLS LAST], sum(DISTINCT 1) as sum(DISTINCT Int64(1))]
+, ordering_mode=Sorted
 03)----RepartitionExec: partitioning=Hash([id@0], 4), input_partitions=5
 04)------AggregateExec: mode=Partial, gby=[id@0 as id], aggr=[last_value(a.foo) ORDER BY [a.foo ASC NULLS LAST], sum(DISTINCT Int64(1))], ordering_mode=Sorted
 05)--------UnionExec
@@ -7856,7 +7857,7 @@ logical_plan
 01)Aggregate: groupBy=[[]], aggr=[[count(Int64(1)), count(Int64(2))]]
 02)--TableScan: t projection=[]
 physical_plan
-01)AggregateExec: mode=Single, gby=[], aggr=[count(Int64(1)), count(Int64(2))]
+01)AggregateExec: mode=Single, gby=[], aggr=[count(1) as count(Int64(1)), count(2) as count(Int64(2))]
 02)--DataSourceExec: partitions=1, partition_sizes=[1]
 
 query II

datafusion/sqllogictest/test_files/joins.slt

@@ -4367,7 +4367,7 @@ logical_plan
 08)----------TableScan: my_catalog.my_schema.table_with_many_types projection=[binary_col]
 physical_plan
 01)ProjectionExec: expr=[count(Int64(1))@0 as count(*)]
-02)--AggregateExec: mode=Single, gby=[], aggr=[count(Int64(1))]
+02)--AggregateExec: mode=Single, gby=[], aggr=[count(1) as count(Int64(1))]
 03)----ProjectionExec: expr=[]
 04)------HashJoinExec: mode=CollectLeft, join_type=Inner, on=[(binary_col@0, binary_col@0)]
 05)--------DataSourceExec: partitions=1, partition_sizes=[1]

datafusion/sqllogictest/test_files/limit.slt

@@ -371,7 +371,8 @@ logical_plan
 06)----------TableScan: t1 projection=[a]
 physical_plan
 01)ProjectionExec: expr=[count(Int64(1))@0 as count(*)]
-02)--AggregateExec: mode=Final, gby=[], aggr=[count(Int64(1))]
+02)--AggregateExec: mode=Final, gby=[], aggr=[count(Int64(1)) as count(*)]
+
 03)----CoalescePartitionsExec
 04)------AggregateExec: mode=Partial, gby=[], aggr=[count(Int64(1))]
 05)--------RepartitionExec: partitioning=RoundRobinBatch(4), input_partitions=1

datafusion/sqllogictest/test_files/scratch/aggregate_repartition/dim.csv

@@ -0,0 +1,6 @@
+env,value,category
+prod,100,A
+dev,200,B
+test,150,A
+prod,300,C
+dev,250,B

datafusion/sqllogictest/test_files/scratch/copy/table.csv

@@ -0,0 +1,2 @@
+1,Foo
+2,Bar

datafusion/sqllogictest/test_files/scratch/copy/table.json

@@ -0,0 +1,2 @@
+{"col1":1,"col2":"Foo"}
+{"col1":2,"col2":"Bar"}

datafusion/sqllogictest/test_files/scratch/copy/table_csv_with_options/KouMFRaB6OvzjCNJ_0.csv

@@ -0,0 +1,3 @@
+col1;col2
+1;Foo
+2;Bar

datafusion/sqllogictest/test_files/scratch/copy/table_json/YkEunS6mAK2DzS9o_0.json

@@ -0,0 +1,2 @@
+{"col1":1,"col2":"Foo"}
+{"col1":2,"col2":"Bar"}

datafusion/sqllogictest/test_files/scratch/csv_files/csv_partitions/1.csv

@@ -0,0 +1,4 @@
+1,aaa,100,1
+2,bbb,200,1
+3,ccc,300,1
+4,ddd,400,1

datafusion/sqllogictest/test_files/scratch/csv_files/csv_partitions/2.csv

@@ -0,0 +1,4 @@
+5,eee,500,2
+6,fff,600,2
+7,ggg,700,2
+8,hhh,800,2

datafusion/sqllogictest/test_files/scratch/csv_files/file_with_comments.csv

@@ -0,0 +1,3 @@
+column1,column2
+#second line is a comment
+2,3

datafusion/sqllogictest/test_files/scratch/csv_files/table_with_double_quotes.csv

@@ -0,0 +1,5 @@
+c1,c2
+id0,"""value0"""
+id1,"""value1"""
+id2,"""value2"""
+id3,"""value3"""

datafusion/sqllogictest/test_files/scratch/csv_files/table_with_escaped_quotes.csv

@@ -0,0 +1,5 @@
+c1,c2
+id0,"#"value0#""
+id1,"#"value1#""
+id2,"#"value2#""
+id3,"#"value3#""

datafusion/sqllogictest/test_files/scratch/csv_files/table_with_necessary_quoting.csv

@@ -0,0 +1,5 @@
+int_col|string_col
+1|~e|e|e~
+2|~f|f|f~
+3|~g|g|g~
+4|~h|h|h~

datafusion/sqllogictest/test_files/scratch/dynamic_file/csv_partitions/partition_col=1/CEbHQ5jJPBvyZRFQ.csv

@@ -0,0 +1,5 @@
+int_col,string_col,bigint_col
+1,aaa,100
+2,bbb,200
+3,ccc,300
+4,ddd,400

datafusion/sqllogictest/test_files/scratch/dynamic_file/csv_partitions/partition_col=2/YhkE7pQeNE9nkkeV.csv

@@ -0,0 +1,5 @@
+int_col,string_col,bigint_col
+5,eee,500
+6,fff,600
+7,ggg,700
+8,hhh,800

datafusion/sqllogictest/test_files/scratch/dynamic_file/json_partitions/partition_col=1/Up012XvC6WJPVh28.json

@@ -0,0 +1,4 @@
+{"int_col":1,"string_col":"aaa","bigint_col":100}
+{"int_col":2,"string_col":"bbb","bigint_col":200}
+{"int_col":3,"string_col":"ccc","bigint_col":300}
+{"int_col":4,"string_col":"ddd","bigint_col":400}

datafusion/sqllogictest/test_files/scratch/dynamic_file/json_partitions/partition_col=2/SawIykjk0NuenfMR.json

@@ -0,0 +1,4 @@
+{"int_col":5,"string_col":"eee","bigint_col":500}
+{"int_col":6,"string_col":"fff","bigint_col":600}
+{"int_col":7,"string_col":"ggg","bigint_col":700}
+{"int_col":8,"string_col":"hhh","bigint_col":800}

datafusion/sqllogictest/test_files/scratch/explain_tree/table1.csv

@@ -0,0 +1,4 @@
+column1,column2,column3,column4
+1,foo,1,2023-01-01
+2,bar,2,2023-01-02
+3,baz,3,2023-01-03

datafusion/sqllogictest/test_files/scratch/explain_tree/table4.json

@@ -0,0 +1,3 @@
+{"int_col":1,"string_col":"foo","bigint_col":1,"date_col":"2023-01-01"}
+{"int_col":2,"string_col":"bar","bigint_col":2,"date_col":"2023-01-02"}
+{"int_col":3,"string_col":"baz","bigint_col":3,"date_col":"2023-01-03"}

datafusion/sqllogictest/test_files/scratch/group_by/timestamp_table/0.csv

@@ -0,0 +1,11 @@
+t1,c2
+2020-12-10T00:00:00,0
+2020-12-11T00:00:00,1
+2020-12-12T00:00:00,2
+2020-12-13T00:00:00,3
+2020-12-14T00:00:00,4
+2020-12-15T00:00:00,5
+2020-12-16T00:00:00,6
+2020-12-17T00:00:00,7
+2020-12-18T00:00:00,8
+2020-12-19T00:00:00,9

datafusion/sqllogictest/test_files/scratch/group_by/timestamp_table/1.csv

@@ -0,0 +1,11 @@
+t1,c2
+2020-12-10T00:00:00,0
+2020-12-11T00:00:00,1
+2020-12-12T00:00:00,2
+2020-12-13T00:00:00,3
+2020-12-14T00:00:00,4
+2020-12-15T00:00:00,5
+2020-12-16T00:00:00,6
+2020-12-17T00:00:00,7
+2020-12-18T00:00:00,8
+2020-12-19T00:00:00,9

datafusion/sqllogictest/test_files/scratch/group_by/timestamp_table/2.csv

@@ -0,0 +1,11 @@
+t1,c2
+2020-12-10T00:00:00,0
+2020-12-11T00:00:00,1
+2020-12-12T00:00:00,2
+2020-12-13T00:00:00,3
+2020-12-14T00:00:00,4
+2020-12-15T00:00:00,5
+2020-12-16T00:00:00,6
+2020-12-17T00:00:00,7
+2020-12-18T00:00:00,8
+2020-12-19T00:00:00,9

datafusion/sqllogictest/test_files/scratch/group_by/timestamp_table/3.csv

@@ -0,0 +1,11 @@
+t1,c2
+2020-12-10T00:00:00,0
+2020-12-11T00:00:00,1
+2020-12-12T00:00:00,2
+2020-12-13T00:00:00,3
+2020-12-14T00:00:00,4
+2020-12-15T00:00:00,5
+2020-12-16T00:00:00,6
+2020-12-17T00:00:00,7
+2020-12-18T00:00:00,8
+2020-12-19T00:00:00,9

datafusion/sqllogictest/test_files/scratch/insert_to_external/insert_to_ordered/MAUz6B2UQasjNmCw_0.csv

@@ -0,0 +1,10 @@
+a,b
+1,5
+2,4
+3,3
+4,2
+5,1
+7,10
+7,9
+7,8
+7,7

datafusion/sqllogictest/test_files/scratch/insert_to_external/insert_to_partitioned/a=10/b=100/xhx0hF9GQxyt5r49.csv

@@ -0,0 +1,2 @@
+c
+1

datafusion/sqllogictest/test_files/scratch/insert_to_external/insert_to_partitioned/a=10/b=200/xhx0hF9GQxyt5r49.csv

@@ -0,0 +1,2 @@
+c
+1

datafusion/sqllogictest/test_files/scratch/insert_to_external/insert_to_partitioned/a=20/b=100/xhx0hF9GQxyt5r49.csv

@@ -0,0 +1,3 @@
+c
+1
+2

datafusion/sqllogictest/test_files/scratch/insert_to_external/insert_to_partitioned/a=20/b=200/xhx0hF9GQxyt5r49.csv

@@ -0,0 +1,3 @@
+c
+1
+2

datafusion/sqllogictest/test_files/scratch/insert_to_external/insert_to_partitioned/a=30/b=300/u3l2P52anQzcEZyC.csv

@@ -0,0 +1,2 @@
+c
+3

datafusion/sqllogictest/test_files/scratch/insert_to_external/insert_to_partitioned_json/a=2/fl6RsXA0i8jREdCZ.json

@@ -0,0 +1,2 @@
+{"b":"1"}
+{"b":"1"}

datafusion/sqllogictest/test_files/scratch/insert_to_external/insert_to_partitioned_json/a=4/fl6RsXA0i8jREdCZ.json

@@ -0,0 +1,2 @@
+{"b":"3"}
+{"b":"3"}

datafusion/sqllogictest/test_files/scratch/insert_to_external/insert_to_partitioned_json/a=6/fl6RsXA0i8jREdCZ.json

@@ -0,0 +1,2 @@
+{"b":"5"}
+{"b":"5"}

datafusion/sqllogictest/test_files/scratch/insert_to_external/single_csv_table.csv

@@ -0,0 +1,4 @@
+1,2
+3,4
+4,5
+6,7

datafusion/sqllogictest/test_files/scratch/joins/t1.csv

@@ -0,0 +1,6 @@
+column1
+1
+2
+3
+4
+5

datafusion/sqllogictest/test_files/scratch/joins/t2.csv

@@ -0,0 +1,6 @@
+column1
+5
+4
+3
+2
+1

datafusion/sqllogictest/test_files/scratch/repartition_scan/csv_table/1.csv

@@ -0,0 +1,6 @@
+column1
+1
+2
+3
+4
+5

datafusion/sqllogictest/test_files/scratch/repartition_scan/json_table/1.json

@@ -0,0 +1,5 @@
+{"column1":1}
+{"column1":2}
+{"column1":3}
+{"column1":4}
+{"column1":5}

datafusion/sqllogictest/test_files/window.slt

@@ -504,7 +504,7 @@ LIMIT 5
 #//     let actual = execute_to_batches(&ctx, sql).await;
 #//     let expected = vec![
 #//         "+----------------------------+-----------------+",
-#//         "| SUM(aggregate_test_100.c4) | COUNT(Int64(1)) |",
+#//         "| SUM(aggregate_test_100.c4) | count(1) as count(Int64(1))|",
 #//         "+----------------------------+-----------------+",
 #//         "| -33822      | 3|",
 #//         "| 20808       | 3|",