Feat: Support array_join

Author: erenavsarogullari

native/core/src/execution/planner.rs

@@ -67,6 +67,7 @@ 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_functions_nested::string::array_to_string_udf;
 use datafusion_physical_expr::aggregate::{AggregateExprBuilder, AggregateFunctionExpr};
 
 use crate::execution::shuffle::CompressionCodec;
@@ -765,6 +766,21 @@ impl PhysicalPlanner {
 
                 Ok(Arc::new(case_expr))
             }
+            ExprStruct::ArrayJoin(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), key_expr];
+                let datafusion_array_to_string = array_to_string_udf();
+                let array_intersect_expr = Arc::new(ScalarFunctionExpr::new(
+                    "array_join",
+                    datafusion_array_to_string,
+                    args,
+                    DataType::Utf8,
+                ));
+                Ok(array_intersect_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 array_join = 63;
   }
 }
 

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

@@ -2284,6 +2284,12 @@ object QueryPlanSerde extends Logging with ShimQueryPlanSerde with CometExprShim
             expr.children(1),
             inputs,
             (builder, binaryExpr) => builder.setArrayAppend(binaryExpr))
+        case _ if expr.prettyName == "array_join" =>
+          createBinaryExpr(
+            expr.children(0),
+            expr.children(1),
+            inputs,
+            (builder, binaryExpr) => builder.setArrayJoin(binaryExpr))
         case _ =>
           withInfo(expr, s"${expr.prettyName} is not supported", expr.children: _*)
           None

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

@@ -2545,4 +2545,20 @@ class CometExpressionSuite extends CometTestBase with AdaptiveSparkPlanHelper {
       }
     }
   }
+
+  test("array_join") {
+    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_join(array(cast(_1 as string), cast(_2 as string), cast(_6 as string)), ' @ ') from t1"))
+        checkSparkAnswerAndOperator(
+          sql("SELECT array_join(array('hello', 'world', cast(_2 as string)), ' ') from t1 where _2 is not null"))
+        checkSparkAnswerAndOperator(
+          sql("SELECT array_join(array('hello', '-', 'world', cast(_2 as string)), ' ') from t1"))
+      }
+    }
+  }
 }