[VL] Support function ToPrettyString

backends-clickhouse/src/main/scala/org/apache/gluten/utils/CHExpressionUtil.scala

@@ -174,6 +174,7 @@ object CHExpressionUtil {
 
   final val CH_BLACKLIST_SCALAR_FUNCTION: Map[String, FunctionValidator] = Map(
     SPLIT_PART -> DefaultValidator(),
+    TO_PRETTY_STRING -> DefaultValidator(),
     TO_UNIX_TIMESTAMP -> UnixTimeStampValidator(),
     UNIX_TIMESTAMP -> UnixTimeStampValidator(),
     SEQUENCE -> SequenceValidator(),

gluten-ut/spark40/src/test/scala/org/apache/gluten/utils/velox/VeloxTestSettings.scala

@@ -943,6 +943,8 @@ class VeloxTestSettings extends BackendTestSettings {
     .exclude("SPARK-41048: Improve output partitioning and ordering with AQE cache")
     // Rewrite this test since it checks the physical operator which is changed in Gluten
     .exclude("SPARK-27439: Explain result should match collected result after view change")
+    // https://github.com/apache/incubator-gluten/issues/11570
+    .exclude("getRows: binary")
   enableSuite[GlutenDataFrameTimeWindowingSuite]
   enableSuite[GlutenDataFrameTungstenSuite]
   enableSuite[GlutenDataFrameWindowFunctionsSuite]

gluten-ut/spark40/src/test/scala/org/apache/spark/sql/GlutenDataFrameSuite.scala

@@ -34,6 +34,8 @@ import scala.util.Random
 
 class GlutenDataFrameSuite extends DataFrameSuite with GlutenSQLTestsTrait {
 
+  import testImplicits._
+
   testGluten("repartitionByRange") {
     val partitionNum = 10
     withSQLConf(
@@ -235,6 +237,16 @@ class GlutenDataFrameSuite extends DataFrameSuite with GlutenSQLTestsTrait {
     }
   }
 
+  testGluten("getRows: binary") {
+    val df = Seq(
+      ("12".getBytes(StandardCharsets.UTF_8), "ABC.".getBytes(StandardCharsets.UTF_8)),
+      ("34".getBytes(StandardCharsets.UTF_8), "12346".getBytes(StandardCharsets.UTF_8))
+    ).toDF()
+    val expectedAnswer =
+      Seq(Seq("_1", "_2"), Seq("[31 32]", "[41 42 43 2E]"), Seq("[33 34]", "[31 32 33 34 36]"))
+    assert(df.getRows(10, 20) === expectedAnswer)
+  }
+
   // TODO: fix in spark-4.0
   /** Failed to check WholeStageCodegenExec, so we rewrite the UT. */
   // testGluten("SPARK-22520: support code generation for large CaseWhen") {
@@ -252,7 +264,6 @@ class GlutenDataFrameSuite extends DataFrameSuite with GlutenSQLTestsTrait {
   //   // We check WholeStageTransformer instead of WholeStageCodegenExec
   //   assert(df.queryExecution.executedPlan.find(_.isInstanceOf[WholeStageTransformer]).isDefined)
   // }
-  import testImplicits._
 
   private lazy val person2: DataFrame = Seq(
     ("Bob", 16, 176),

gluten-ut/spark41/src/test/scala/org/apache/gluten/utils/velox/VeloxTestSettings.scala

@@ -913,6 +913,8 @@ class VeloxTestSettings extends BackendTestSettings {
     .exclude("SPARK-41048: Improve output partitioning and ordering with AQE cache")
     // Rewrite this test since it checks the physical operator which is changed in Gluten
     .exclude("SPARK-27439: Explain result should match collected result after view change")
+    // https://github.com/apache/incubator-gluten/issues/11570
+    .exclude("getRows: binary")
   enableSuite[GlutenDataFrameTimeWindowingSuite]
   enableSuite[GlutenDataFrameTungstenSuite]
   enableSuite[GlutenDataFrameWindowFunctionsSuite]

gluten-ut/spark41/src/test/scala/org/apache/spark/sql/GlutenDataFrameSuite.scala

@@ -394,6 +394,16 @@ class GlutenDataFrameSuite extends DataFrameSuite with GlutenSQLTestsTrait {
     }
   }
 
+  testGluten("getRows: binary") {
+    val df = Seq(
+      ("12".getBytes(StandardCharsets.UTF_8), "ABC.".getBytes(StandardCharsets.UTF_8)),
+      ("34".getBytes(StandardCharsets.UTF_8), "12346".getBytes(StandardCharsets.UTF_8))
+    ).toDF()
+    val expectedAnswer =
+      Seq(Seq("_1", "_2"), Seq("[31 32]", "[41 42 43 2E]"), Seq("[33 34]", "[31 32 33 34 36]"))
+    assert(df.getRows(10, 20) === expectedAnswer)
+  }
+
   // TODO: fix in spark-4.0
   // private def withExpr(newExpr: Expression): Column = new Column(newExpr)
 

shims/common/src/main/scala/org/apache/gluten/expression/ExpressionNames.scala

@@ -141,6 +141,7 @@ object ExpressionNames {
   final val MASK = "mask"
   final val FORMAT_STRING = "format_string"
   final val LUHN_CHECK = "luhn_check"
+  final val TO_PRETTY_STRING = "to_pretty_string"
 
   // URL functions
   final val PARSE_URL = "parse_url"

shims/spark35/src/main/scala/org/apache/gluten/sql/shims/spark35/Spark35Shims.scala

@@ -93,7 +93,8 @@ class Spark35Shims extends SparkShims {
       Sig[CheckOverflowInTableInsert](ExpressionNames.CHECK_OVERFLOW_IN_TABLE_INSERT),
       Sig[ArrayAppend](ExpressionNames.ARRAY_APPEND),
       Sig[UrlEncode](ExpressionNames.URL_ENCODE),
-      Sig[UrlDecode](ExpressionNames.URL_DECODE)
+      Sig[UrlDecode](ExpressionNames.URL_DECODE),
+      Sig[ToPrettyString](ExpressionNames.TO_PRETTY_STRING)
     )
   }
 

shims/spark40/src/main/scala/org/apache/gluten/sql/shims/spark40/Spark40Shims.scala

@@ -97,7 +97,8 @@ class Spark40Shims extends SparkShims {
       Sig[ArrayAppend](ExpressionNames.ARRAY_APPEND),
       Sig[UrlEncode](ExpressionNames.URL_ENCODE),
       Sig[KnownNotContainsNull](ExpressionNames.KNOWN_NOT_CONTAINS_NULL),
-      Sig[UrlDecode](ExpressionNames.URL_DECODE)
+      Sig[UrlDecode](ExpressionNames.URL_DECODE),
+      Sig[ToPrettyString](ExpressionNames.TO_PRETTY_STRING)
     )
   }
 

shims/spark41/src/main/scala/org/apache/gluten/sql/shims/spark41/Spark41Shims.scala

@@ -96,7 +96,8 @@ class Spark41Shims extends SparkShims {
       Sig[ArrayAppend](ExpressionNames.ARRAY_APPEND),
       Sig[UrlEncode](ExpressionNames.URL_ENCODE),
       Sig[KnownNotContainsNull](ExpressionNames.KNOWN_NOT_CONTAINS_NULL),
-      Sig[UrlDecode](ExpressionNames.URL_DECODE)
+      Sig[UrlDecode](ExpressionNames.URL_DECODE),
+      Sig[ToPrettyString](ExpressionNames.TO_PRETTY_STRING)
     )
   }