Support arrays_overlap function

Author: erenavsarogullari

native/core/src/execution/planner.rs

@@ -65,6 +65,7 @@ use datafusion::{
     prelude::SessionContext,
 };
 use datafusion_comet_spark_expr::{create_comet_physical_fun, create_negate_expr};
+use datafusion_functions_nested::array_has::array_has_any_udf;
 use datafusion_functions_nested::concat::ArrayAppend;
 use datafusion_functions_nested::remove::array_remove_all_udf;
 use datafusion_physical_expr::aggregate::{AggregateExprBuilder, AggregateFunctionExpr};
@@ -765,6 +766,21 @@ impl PhysicalPlanner {
 
                 Ok(Arc::new(case_expr))
             }
+            ExprStruct::ArraysOverlap(expr) => {
+                let left_array_expr =
+                    self.create_expr(expr.left.as_ref().unwrap(), Arc::clone(&input_schema))?;
+                let right_array_expr =
+                    self.create_expr(expr.right.as_ref().unwrap(), Arc::clone(&input_schema))?;
+                let args = vec![Arc::clone(&left_array_expr), right_array_expr];
+                let datafusion_array_has_any = array_has_any_udf();
+                let array_has_any_expr = Arc::new(ScalarFunctionExpr::new(
+                    "array_has_any",
+                    datafusion_array_has_any,
+                    args,
+                    DataType::Boolean,
+                ));
+                Ok(array_has_any_expr)
+            }
             expr => Err(ExecutionError::GeneralError(format!(
                 "Not implemented: {:?}",
                 expr

native/proto/src/proto/expr.proto

@@ -86,6 +86,7 @@ message Expr {
     ArrayInsert array_insert = 59;
     BinaryExpr array_contains = 60;
     BinaryExpr array_remove = 61;
+    BinaryExpr arrays_overlap = 62;
   }
 }
 

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

@@ -2300,6 +2300,12 @@ object QueryPlanSerde extends Logging with ShimQueryPlanSerde with CometExprShim
             expr.children(1),
             inputs,
             (builder, binaryExpr) => builder.setArrayAppend(binaryExpr))
+        case ArraysOverlap(leftArrayExpr, rightArrayExpr) =>
+          createBinaryExpr(
+            leftArrayExpr,
+            rightArrayExpr,
+            inputs,
+            (builder, binaryExpr) => builder.setArraysOverlap(binaryExpr))
         case _ =>
           withInfo(expr, s"${expr.prettyName} is not supported", expr.children: _*)
           None

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

@@ -2568,4 +2568,21 @@ class CometExpressionSuite extends CometTestBase with AdaptiveSparkPlanHelper {
       }
     }
   }
+
+  test("arrays_overlap") {
+    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 arrays_overlap(array(_2, _3, _4), array(_3, _4)) from t1 where _2 is not null"))
+        checkSparkAnswerAndOperator(sql(
+          "SELECT arrays_overlap(array('a', null, cast(_1 as string)), array('b', cast(_1 as string), cast(_2 as string))) from t1 where _1 is not null"))
+        checkSparkAnswerAndOperator(sql(
+          "SELECT arrays_overlap(array('a', null), array('b', null)) from t1 where _1 is not null"))
+      }
+    }
+  }
+
 }