feat: add support for array_remove expression

native/core/src/execution/planner.rs

@@ -66,6 +66,7 @@ use datafusion::{
 };
 use datafusion_comet_spark_expr::{create_comet_physical_fun, create_negate_expr};
 use datafusion_functions_nested::concat::ArrayAppend;
+use datafusion_functions_nested::remove::array_remove_all_udf;
 use datafusion_physical_expr::aggregate::{AggregateExprBuilder, AggregateFunctionExpr};
 
 use crate::execution::shuffle::CompressionCodec;
@@ -735,6 +736,36 @@ impl PhysicalPlanner {
                 ));
                 Ok(array_has_expr)
             }
+            ExprStruct::ArrayRemove(expr) => {
+                let src_array_expr =
+                    self.create_expr(expr.left.as_ref().unwrap(), Arc::clone(&input_schema))?;
+                let key_expr =
+                    self.create_expr(expr.right.as_ref().unwrap(), Arc::clone(&input_schema))?;
+                let args = vec![Arc::clone(&src_array_expr), Arc::clone(&key_expr)];
+                let return_type = src_array_expr.data_type(&input_schema)?;
+
+                let datafusion_array_remove = array_remove_all_udf();
+
+                let array_remove_expr: Arc<dyn PhysicalExpr> = Arc::new(ScalarFunctionExpr::new(
+                    "array_remove",
+                    datafusion_array_remove,
+                    args,
+                    return_type,
+                ));
+                let is_null_expr: Arc<dyn PhysicalExpr> = Arc::new(IsNullExpr::new(key_expr));
+
+                let null_literal_expr: Arc<dyn PhysicalExpr> =
+                    Arc::new(Literal::new(ScalarValue::Null));
+
+                let case_expr = CaseExpr::try_new(
+                    None,
+                    vec![(is_null_expr, null_literal_expr)],
+                    Some(array_remove_expr),
+                )?;
+
+                Ok(Arc::new(case_expr))
+                // Ok(array_remove_expr)
+            }
             expr => Err(ExecutionError::GeneralError(format!(
                 "Not implemented: {:?}",
                 expr

native/proto/src/proto/expr.proto

@@ -85,6 +85,7 @@ message Expr {
     BinaryExpr array_append = 58;
     ArrayInsert array_insert = 59;
     BinaryExpr array_contains = 60;
+    BinaryExpr array_remove = 61;
   }
 }
 

spark/src/main/scala/org/apache/comet/serde/QueryPlanSerde.scala

@@ -2266,6 +2266,12 @@ object QueryPlanSerde extends Logging with ShimQueryPlanSerde with CometExprShim
             withInfo(expr, "unsupported arguments for GetArrayStructFields", child)
             None
           }
+        case expr if expr.prettyName == "array_remove" =>
+          createBinaryExpr(
+            expr.children(0),
+            expr.children(1),
+            inputs,
+            (builder, binaryExpr) => builder.setArrayRemove(binaryExpr))
         case expr if expr.prettyName == "array_contains" =>
           createBinaryExpr(
             expr.children(0),
@@ -2859,7 +2865,11 @@ object QueryPlanSerde extends Logging with ShimQueryPlanSerde with CometExprShim
           case RightOuter => JoinType.RightOuter
           case FullOuter => JoinType.FullOuter
           case LeftSemi => JoinType.LeftSemi
-          case LeftAnti => JoinType.LeftAnti
+          // TODO: DF SMJ with join condition fails TPCH q21
+          case LeftAnti if condition.isEmpty => JoinType.LeftAnti
+          case LeftAnti =>
+            withInfo(join, "LeftAnti SMJ join with condition is not supported")
+            return None
           case _ =>
             // Spark doesn't support other join types
             withInfo(op, s"Unsupported join type ${join.joinType}")

spark/src/test/scala/org/apache/comet/CometExpressionSuite.scala

@@ -2529,4 +2529,16 @@ class CometExpressionSuite extends CometTestBase with AdaptiveSparkPlanHelper {
         spark.sql("SELECT array_contains((CASE WHEN _2 =_3 THEN array(_4) END), _4) FROM t1"));
     }
   }
+  test("array_remove") {
+    Seq(true, false).foreach { dictionaryEnabled =>
+      withTempDir { dir =>
+        val path = new Path(dir.toURI.toString, "test.parquet")
+        makeParquetFileAllTypes(path, dictionaryEnabled, 10000)
+        spark.read.parquet(path.toString).createOrReplaceTempView("t1");
+        checkSparkAnswerAndOperator(sql("SELECT array_remove(array(_2, _3,_4), _2) from t1"))
+        checkSparkAnswerAndOperator(sql("SELECT array_remove(array(_2, _3,_4), 3) from t1"))
+        checkSparkAnswerAndOperator(sql("SELECT array_remove(array(_2, _3,_4), 10) from t1"))
+      }
+    }
+  }
 }