apache · jatin510 · Feb 13, 2025 · Feb 18, 2025 · Feb 24, 2025 · Feb 24, 2025
diff --git a/datafusion-testing b/datafusion-testing
diff --git a/datafusion/common/src/config.rs b/datafusion/common/src/config.rs
@@ -231,7 +231,7 @@ config_namespace! {
     /// [`SessionConfig`]: https://docs.rs/datafusion/latest/datafusion/prelude/struct.SessionConfig.html
     pub struct SqlParserOptions {
         /// When set to true, SQL parser will parse float as decimal type
-        pub parse_float_as_decimal: bool, default = false
+        pub parse_float_as_decimal: bool, default = true
 
         /// When set to true, SQL parser will normalize ident (convert ident to lowercase when not quoted)
         pub enable_ident_normalization: bool, default = true

diff --git a/datafusion/core/tests/parquet/page_pruning.rs b/datafusion/core/tests/parquet/page_pruning.rs
@@ -689,7 +689,7 @@ async fn prune_f64_scalar_fun_and_gt() {
         Scenario::Float64,
         "SELECT * FROM t where abs(f - 1) <= 0.000001  and f >= 0.1",
         Some(0),
-        Some(10),
+        Some(0),
         1,
         5,
     )

diff --git a/datafusion/core/tests/parquet/row_group_pruning.rs b/datafusion/core/tests/parquet/row_group_pruning.rs
@@ -667,8 +667,8 @@ async fn prune_f64_scalar_fun_and_gt() {
         .with_scenario(Scenario::Float64)
         .with_query("SELECT * FROM t where abs(f - 1) <= 0.000001  and f >= 0.1")
         .with_expected_errors(Some(0))
-        .with_matched_by_stats(Some(2))
-        .with_pruned_by_stats(Some(2))
+        .with_matched_by_stats(Some(0))
+        .with_pruned_by_stats(Some(0))
         .with_matched_by_bloom_filter(Some(0))
         .with_pruned_by_bloom_filter(Some(0))
         .with_expected_rows(1)

diff --git a/datafusion/core/tests/user_defined/user_defined_scalar_functions.rs b/datafusion/core/tests/user_defined/user_defined_scalar_functions.rs
@@ -1071,11 +1071,11 @@ async fn create_scalar_function_from_sql_statement() -> Result<()> {
 
     assert_batches_eq!(
         &[
-            "+-----------------------------------+",
-            "| better_add(Float64(2),Float64(2)) |",
-            "+-----------------------------------+",
-            "| 4.0                               |",
-            "+-----------------------------------+",
+            "+---------------------------------------------------------------+",
+            "| better_add(Decimal128(Some(20),2,1),Decimal128(Some(20),2,1)) |",
+            "+---------------------------------------------------------------+",
+            "| 4.0                                                           |",
+            "+---------------------------------------------------------------+",
         ],
         &result
     );

diff --git a/datafusion/expr-common/src/type_coercion/binary.rs b/datafusion/expr-common/src/type_coercion/binary.rs
@@ -307,17 +307,25 @@ fn math_decimal_coercion(
         }
         // Unlike with comparison we don't coerce to a decimal in the case of floating point
         // numbers, instead falling back to floating point arithmetic instead
-        (Decimal128(_, _), Int8 | Int16 | Int32 | Int64) => {
-            Some((lhs_type.clone(), coerce_numeric_type_to_decimal(rhs_type)?))
-        }
-        (Int8 | Int16 | Int32 | Int64, Decimal128(_, _)) => {
-            Some((coerce_numeric_type_to_decimal(lhs_type)?, rhs_type.clone()))
-        }
-        (Decimal256(_, _), Int8 | Int16 | Int32 | Int64) => Some((
+        (
+            Decimal128(_, _),
+            Int8 | Int16 | Int32 | Int64 | UInt8 | UInt16 | UInt32 | UInt64,
+        ) => Some((lhs_type.clone(), coerce_numeric_type_to_decimal(rhs_type)?)),
+        (
+            Int8 | Int16 | Int32 | Int64 | UInt8 | UInt16 | UInt32 | UInt64,
+            Decimal128(_, _),
+        ) => Some((coerce_numeric_type_to_decimal(lhs_type)?, rhs_type.clone())),
+        (
+            Decimal256(_, _),
+            Int8 | Int16 | Int32 | Int64 | UInt8 | UInt16 | UInt32 | UInt64,
+        ) => Some((
             lhs_type.clone(),
             coerce_numeric_type_to_decimal256(rhs_type)?,
         )),
-        (Int8 | Int16 | Int32 | Int64, Decimal256(_, _)) => Some((
+        (
+            Int8 | Int16 | Int32 | Int64 | UInt8 | UInt16 | UInt32 | UInt64,
+            Decimal256(_, _),
+        ) => Some((
             coerce_numeric_type_to_decimal256(lhs_type)?,
             rhs_type.clone(),
         )),
@@ -2156,6 +2164,18 @@ mod tests {
             DataType::Decimal128(10, 0),
             DataType::Decimal128(10, 2),
         );
+        test_math_decimal_coercion_rule(
+            DataType::UInt32,
+            DataType::Decimal128(10, 2),
+            DataType::Decimal128(10, 0),
+            DataType::Decimal128(10, 2),
+        );
+        test_math_decimal_coercion_rule(
+            DataType::Decimal128(10, 2),
+            DataType::UInt32,
+            DataType::Decimal128(10, 2),
+            DataType::Decimal128(10, 0),
+        );
 
         Ok(())
     }

diff --git a/datafusion/sqllogictest/test_files/arrow_typeof.slt b/datafusion/sqllogictest/test_files/arrow_typeof.slt
@@ -43,7 +43,7 @@ Int32
 query T
 SELECT arrow_typeof(1.0)
 ----
-Float64
+Decimal128(2, 1)
 
 # arrow_typeof_f32
 query T

diff --git a/datafusion/sqllogictest/test_files/coalesce.slt b/datafusion/sqllogictest/test_files/coalesce.slt
@@ -41,15 +41,15 @@ select
   arrow_typeof(coalesce(1, 2.0))
 ;
 ----
-1 Float64
+1 Decimal128(21, 1)
 
 query RT
 select
   coalesce(2.0, 1),
   arrow_typeof(coalesce(2.0, 1))
 ;
 ----
-2 Float64
+2 Decimal128(21, 1)
 
 query RT
 select
@@ -322,7 +322,7 @@ select coalesce(null, 34, arrow_cast(123, 'Dictionary(Int32, Int8)'));
 query RT
 select coalesce(2.0, 1, '3'), arrow_typeof(coalesce(2.0, 1, '3'));
 ----
-2 Float64
+2 Decimal128(21,1)
 
 # explicitly cast to Int8, and it will implicitly cast to Int64
 query IT

diff --git a/datafusion/sqllogictest/test_files/decimal.slt b/datafusion/sqllogictest/test_files/decimal.slt
@@ -584,7 +584,7 @@ Decimal128(12, 2) 400420638.54
 query TR
 select arrow_typeof(cast(400420638.54 as decimal(12,2)) * 1.0), cast(400420638.54 as decimal(12,2)) * 1.0;
 ----
-Float64 400420638.54
+Decimal128(15, 3) 400420638.54
 
 
 query TB

diff --git a/datafusion/sqllogictest/test_files/explain.slt b/datafusion/sqllogictest/test_files/explain.slt
@@ -395,10 +395,10 @@ query TT
 explain select struct(1, 2.3, 'abc');
 ----
 logical_plan
-01)Projection: Struct({c0:1,c1:2.3,c2:abc}) AS struct(Int64(1),Float64(2.3),Utf8("abc"))
+01)Projection: Struct({c0:1,c1:2.3,c2:abc}) AS struct(Int64(1),Decimal128(Some(23),2,1),Utf8("abc"))
 02)--EmptyRelation
 physical_plan
-01)ProjectionExec: expr=[{c0:1,c1:2.3,c2:abc} as struct(Int64(1),Float64(2.3),Utf8("abc"))]
+01)ProjectionExec: expr=[{c0:1,c1:2.3,c2:abc} as struct(Int64(1),Decimal128(Some(23),2,1),Utf8("abc"))]
 02)--PlaceholderRowExec
 
 

diff --git a/datafusion/sqllogictest/test_files/expr.slt b/datafusion/sqllogictest/test_files/expr.slt
@@ -587,7 +587,7 @@ select repeat('-1.2', arrow_cast(3, 'Int32'));
 ----
 -1.2-1.2-1.2
 
-query error DataFusion error: Error during planning: Internal error: Expect TypeSignatureClass::Native\(LogicalType\(Native\(Int64\), Int64\)\) but received NativeType::Float64, DataType: Float64
+query error DataFusion error: Error during planning: Internal error: Expect TypeSignatureClass::Native(LogicalType(Native(Int64), Int64)) but received NativeType::Decimal(2, 1), DataType: Decimal128(2, 1).
 select repeat('-1.2', 3.2);
 
 query T

diff --git a/datafusion/sqllogictest/test_files/operator.slt b/datafusion/sqllogictest/test_files/operator.slt
@@ -88,7 +88,7 @@ select
     arrow_typeof(decimal + 2.0)
 from numeric_types;
 ----
-Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64
+Decimal128(5, 1) Decimal128(7, 1) Decimal128(12, 1) Decimal128(22, 1) Decimal128(5, 1) Decimal128(7, 1) Decimal128(12, 1) Decimal128(22, 1) Float32 Float64 Decimal128(6, 2)
 
 ############### Subtraction ###############
 
@@ -145,7 +145,7 @@ select
     arrow_typeof(decimal - 2.0)
 from numeric_types;
 ----
-Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64
+Decimal128(5, 1) Decimal128(7, 1) Decimal128(12, 1) Decimal128(22, 1) Decimal128(5, 1) Decimal128(7, 1) Decimal128(12, 1) Decimal128(22, 1) Float32 Float64 Decimal128(6, 2)
 
 ############### Multiplication ###############
 
@@ -202,7 +202,7 @@ select
     arrow_typeof(decimal * 2.0)
 from numeric_types;
 ----
-Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64 Float64
+Decimal128(6, 1) Decimal128(8, 1) Decimal128(13, 1) Decimal128(23, 1) Decimal128(6, 1) Decimal128(8, 1) Decimal128(13, 1) Decimal128(23, 1) Float32 Float64 Decimal128(8, 3)
 
 ############### Division ###############
 

diff --git a/datafusion/sqllogictest/test_files/options.slt b/datafusion/sqllogictest/test_files/options.slt
@@ -89,12 +89,12 @@ drop table a
 query RR
 select 10000000000000000000.01, -10000000000000000000.01
 ----
-10000000000000000000 -10000000000000000000 
+10000000000000000000.01 -10000000000000000000.01
 
 query TT
 select arrow_typeof(10000000000000000000.01), arrow_typeof(-10000000000000000000.01)
 ----
-Float64 Float64
+Decimal128(22, 2) Decimal128(22, 2)
 
 # select 0, i64::MIN, i64::MIN-1, i64::MAX, i64::MAX + 1, u64::MAX, u64::MAX + 1
 query IIRIIIR

diff --git a/datafusion/sqllogictest/test_files/select.slt b/datafusion/sqllogictest/test_files/select.slt
@@ -423,7 +423,7 @@ SELECT * FROM (VALUES (1,'a'),(2,NULL)) AS t(c1, c2)
 query TT
 EXPLAIN VALUES (1, 'a', -1, 1.1),(NULL, 'b', -3, 0.5)
 ----
-logical_plan Values: (Int64(1), Utf8("a"), Int64(-1), Float64(1.1)), (Int64(NULL), Utf8("b"), Int64(-3), Float64(0.5))
+logical_plan Values: (Int64(1), Utf8("a"), Int64(-1), Decimal128(Some(11),2,1)), (Int64(NULL), Utf8("b"), Int64(-3), Decimal128(Some(5),2,1) AS Decimal128(Some(5),1,1))
 physical_plan DataSourceExec: partitions=1, partition_sizes=[1]
 
 query TT

diff --git a/datafusion/sqllogictest/test_files/struct.slt b/datafusion/sqllogictest/test_files/struct.slt
@@ -465,8 +465,8 @@ select * from t;
 query T
 select arrow_typeof(column1) from t;
 ----
-Struct([Field { name: "r", data_type: Utf8, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }, Field { name: "c", data_type: Float64, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }])
-Struct([Field { name: "r", data_type: Utf8, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }, Field { name: "c", data_type: Float64, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }])
+Struct([Field { name: "r", data_type: Utf8, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }, Field { name: "c", data_type: Decimal128(21, 1), nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }])
+Struct([Field { name: "r", data_type: Utf8, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }, Field { name: "c", data_type: Decimal128(21, 1), nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }])
 
 statement ok
 drop table t;

diff --git a/datafusion/sqllogictest/test_files/subquery.slt b/datafusion/sqllogictest/test_files/subquery.slt
@@ -227,15 +227,15 @@ query TT
 explain SELECT t1_id, (SELECT sum(t2_int * 1.0) + 1 FROM t2 WHERE t2.t2_id = t1.t1_id) as t2_sum from t1
 ----
 logical_plan
-01)Projection: t1.t1_id, __scalar_sq_1.sum(t2.t2_int * Float64(1)) + Int64(1) AS t2_sum
+01)Projection: t1.t1_id, __scalar_sq_1.sum(t2.t2_int * Decimal128(Some(10),2,1)) + Int64(1) AS t2_sum
 02)--Left Join: t1.t1_id = __scalar_sq_1.t2_id
 03)----TableScan: t1 projection=[t1_id]
 04)----SubqueryAlias: __scalar_sq_1
-05)------Projection: sum(t2.t2_int * Float64(1)) + Float64(1) AS sum(t2.t2_int * Float64(1)) + Int64(1), t2.t2_id
-06)--------Aggregate: groupBy=[[t2.t2_id]], aggr=[[sum(CAST(t2.t2_int AS Float64)) AS sum(t2.t2_int * Float64(1))]]
+05)------Projection: sum(t2.t2_int * Decimal128(Some(10),2,1)) + Decimal128(Some(1),20,0) AS sum(t2.t2_int * Decimal128(Some(10),2,1)) + Int64(1), t2.t2_id
+06)--------Aggregate: groupBy=[[t2.t2_id]], aggr=[[sum(CAST(t2.t2_int AS Decimal128(10, 0))) AS sum(t2.t2_int * Decimal128(Some(10),2,1))]]
 07)----------TableScan: t2 projection=[t2_id, t2_int]
 physical_plan
-01)ProjectionExec: expr=[t1_id@1 as t1_id, sum(t2.t2_int * Float64(1)) + Int64(1)@0 as t2_sum]
+01)ProjectionExec: expr=[t1_id@0 as t1_id, sum(t2.t2_int * Decimal128(Some(10),2,1)) + Int64(1)@1 as t2_sum]
 02)--CoalesceBatchesExec: target_batch_size=2
 03)----HashJoinExec: mode=CollectLeft, join_type=Right, on=[(t2_id@1, t1_id@0)], projection=[sum(t2.t2_int * Float64(1)) + Int64(1)@0, t1_id@2]
 04)------CoalescePartitionsExec
@@ -1450,10 +1450,10 @@ query TT
 explain SELECT a FROM t1 WHERE EXISTS (SELECT count(*) FROM t2)
 ----
 logical_plan
-01)LeftSemi Join: 
+01)LeftSemi Join:
 02)--TableScan: t1 projection=[a]
 03)--SubqueryAlias: __correlated_sq_1
-04)----Projection: 
+04)----Projection:
 05)------Aggregate: groupBy=[[]], aggr=[[count(Int64(1))]]
 06)--------TableScan: t2 projection=[]
 
@@ -1469,7 +1469,7 @@ statement count 0
 create table person(id int, last_name int, state int);
 
 query TT
-explain SELECT id FROM person p WHERE EXISTS 
+explain SELECT id FROM person p WHERE EXISTS
     (SELECT * FROM person WHERE last_name = p.last_name AND state = p.state)
 ----
 logical_plan

diff --git a/datafusion/sqllogictest/test_files/type_coercion.slt b/datafusion/sqllogictest/test_files/type_coercion.slt
@@ -142,9 +142,9 @@ EXPLAIN SELECT 1 a UNION ALL SELECT 1.1 a
 ----
 logical_plan
 01)Union
-02)--Projection: CAST(Int64(1) AS Float64) AS a
+02)--Projection: CAST(Int64(1) AS Decimal128(21, 1)) AS a
 03)----EmptyRelation
-04)--Projection: Float64(1.1) AS a
+04)--Projection: CAST(Decimal128(Some(11),2,1) AS Decimal128(21, 1)) AS a
 05)----EmptyRelation
 
 # union_with_null()

diff --git a/datafusion/sqllogictest/test_files/update.slt b/datafusion/sqllogictest/test_files/update.slt
@@ -31,7 +31,7 @@ explain update t1 set a=1, b=2, c=3.0, d=NULL;
 ----
 logical_plan
 01)Dml: op=[Update] table=[t1]
-02)--Projection: CAST(Int64(1) AS Int32) AS a, CAST(Int64(2) AS Utf8) AS b, Float64(3) AS c, CAST(NULL AS Int32) AS d
+02)--Projection: CAST(Int64(1) AS Int32) AS a, CAST(Int64(2) AS Utf8) AS b, CAST(Decimal128(Some(30),2,1) AS Float64) AS c, CAST(NULL AS Int32) AS d
 03)----TableScan: t1
 physical_plan_error This feature is not implemented: Unsupported logical plan: Dml(Update)
 
@@ -40,7 +40,7 @@ explain update t1 set a=c+1, b=a, c=c+1.0, d=b;
 ----
 logical_plan
 01)Dml: op=[Update] table=[t1]
-02)--Projection: CAST(t1.c + CAST(Int64(1) AS Float64) AS Int32) AS a, CAST(t1.a AS Utf8) AS b, t1.c + Float64(1) AS c, CAST(t1.b AS Int32) AS d
+02)--Projection: CAST(t1.c + CAST(Int64(1) AS Float64) AS Int32) AS a, CAST(t1.a AS Utf8) AS b, t1.c + CAST(Decimal128(Some(10),2,1) AS Float64) AS c, CAST(t1.b AS Int32) AS d
 03)----TableScan: t1
 physical_plan_error This feature is not implemented: Unsupported logical plan: Dml(Update)
-Original file line number
+Diff line change
@@ Expand Up / @@ -43,7 +43,7 @@ Int32 @@
     query T
     SELECT arrow_typeof(1.0)
     ----
-    Float64
+    Decimal128(2, 1)
     # arrow_typeof_f32
     query T
@@ Expand Down @@