[SPARK-3412][SQL]add missing row api

chenghao-intel assigned this to me, check PR #2284 for previous discussion

Author: Daoyuan Wang <[email protected]>

Closes #2529 from adrian-wang/rowapi and squashes the following commits:

c6594b2 [Daoyuan Wang] using boxed
7b7e6e3 [Daoyuan Wang] update pattern match
7a39456 [Daoyuan Wang] rename file and refresh getAs[T]
4c18c29 [Daoyuan Wang] remove setAs[T] and null judge
1614493 [Daoyuan Wang] add missing row api

Author: adrian-wang

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/Projection.scala

@@ -137,6 +137,9 @@ class JoinedRow extends Row {
   def getString(i: Int): String =
     if (i < row1.size) row1.getString(i) else row2.getString(i - row1.size)
 
+  override def getAs[T](i: Int): T =
+    if (i < row1.size) row1.getAs[T](i) else row2.getAs[T](i - row1.size)
+
   def copy() = {
     val totalSize = row1.size + row2.size
     val copiedValues = new Array[Any](totalSize)
@@ -226,6 +229,9 @@ class JoinedRow2 extends Row {
   def getString(i: Int): String =
     if (i < row1.size) row1.getString(i) else row2.getString(i - row1.size)
 
+  override def getAs[T](i: Int): T =
+    if (i < row1.size) row1.getAs[T](i) else row2.getAs[T](i - row1.size)
+
   def copy() = {
     val totalSize = row1.size + row2.size
     val copiedValues = new Array[Any](totalSize)
@@ -309,6 +315,9 @@ class JoinedRow3 extends Row {
   def getString(i: Int): String =
     if (i < row1.size) row1.getString(i) else row2.getString(i - row1.size)
 
+  override def getAs[T](i: Int): T =
+    if (i < row1.size) row1.getAs[T](i) else row2.getAs[T](i - row1.size)
+
   def copy() = {
     val totalSize = row1.size + row2.size
     val copiedValues = new Array[Any](totalSize)
@@ -392,6 +401,9 @@ class JoinedRow4 extends Row {
   def getString(i: Int): String =
     if (i < row1.size) row1.getString(i) else row2.getString(i - row1.size)
 
+  override def getAs[T](i: Int): T =
+    if (i < row1.size) row1.getAs[T](i) else row2.getAs[T](i - row1.size)
+
   def copy() = {
     val totalSize = row1.size + row2.size
     val copiedValues = new Array[Any](totalSize)
@@ -475,6 +487,9 @@ class JoinedRow5 extends Row {
   def getString(i: Int): String =
     if (i < row1.size) row1.getString(i) else row2.getString(i - row1.size)
 
+  override def getAs[T](i: Int): T =
+    if (i < row1.size) row1.getAs[T](i) else row2.getAs[T](i - row1.size)
+
   def copy() = {
     val totalSize = row1.size + row2.size
     val copiedValues = new Array[Any](totalSize)

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/Row.scala

@@ -64,6 +64,7 @@ trait Row extends Seq[Any] with Serializable {
   def getShort(i: Int): Short
   def getByte(i: Int): Byte
   def getString(i: Int): String
+  def getAs[T](i: Int): T = apply(i).asInstanceOf[T]
 
   override def toString() =
     s"[${this.mkString(",")}]"
@@ -118,6 +119,7 @@ object EmptyRow extends Row {
   def getShort(i: Int): Short = throw new UnsupportedOperationException
   def getByte(i: Int): Byte = throw new UnsupportedOperationException
   def getString(i: Int): String = throw new UnsupportedOperationException
+  override def getAs[T](i: Int): T = throw new UnsupportedOperationException
 
   def copy() = this
 }
@@ -217,19 +219,19 @@ class GenericMutableRow(size: Int) extends GenericRow(size) with MutableRow {
   /** No-arg constructor for serialization. */
   def this() = this(0)
 
-  override def setBoolean(ordinal: Int,value: Boolean): Unit = { values(ordinal) = value }
-  override def setByte(ordinal: Int,value: Byte): Unit = { values(ordinal) = value }
-  override def setDouble(ordinal: Int,value: Double): Unit = { values(ordinal) = value }
-  override def setFloat(ordinal: Int,value: Float): Unit = { values(ordinal) = value }
-  override def setInt(ordinal: Int,value: Int): Unit = { values(ordinal) = value }
-  override def setLong(ordinal: Int,value: Long): Unit = { values(ordinal) = value }
-  override def setString(ordinal: Int,value: String): Unit = { values(ordinal) = value }
+  override def setBoolean(ordinal: Int, value: Boolean): Unit = { values(ordinal) = value }
+  override def setByte(ordinal: Int, value: Byte): Unit = { values(ordinal) = value }
+  override def setDouble(ordinal: Int, value: Double): Unit = { values(ordinal) = value }
+  override def setFloat(ordinal: Int, value: Float): Unit = { values(ordinal) = value }
+  override def setInt(ordinal: Int, value: Int): Unit = { values(ordinal) = value }
+  override def setLong(ordinal: Int, value: Long): Unit = { values(ordinal) = value }
+  override def setString(ordinal: Int, value: String): Unit = { values(ordinal) = value }
 
   override def setNullAt(i: Int): Unit = { values(i) = null }
 
-  override def setShort(ordinal: Int,value: Short): Unit = { values(ordinal) = value }
+  override def setShort(ordinal: Int, value: Short): Unit = { values(ordinal) = value }
 
-  override def update(ordinal: Int,value: Any): Unit = { values(ordinal) = value }
+  override def update(ordinal: Int, value: Any): Unit = { values(ordinal) = value }
 
   override def copy() = new GenericRow(values.clone())
 }

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/SpecificRow.scala renamed to sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/SpecificMutableRow.scala

@@ -233,9 +233,9 @@ final class SpecificMutableRow(val values: Array[MutableValue]) extends MutableR
 
   override def iterator: Iterator[Any] = values.map(_.boxed).iterator
 
-  def setString(ordinal: Int, value: String) = update(ordinal, value)
+  override def setString(ordinal: Int, value: String) = update(ordinal, value)
 
-  def getString(ordinal: Int) = apply(ordinal).asInstanceOf[String]
+  override def getString(ordinal: Int) = apply(ordinal).asInstanceOf[String]
 
   override def setInt(ordinal: Int, value: Int): Unit = {
     val currentValue = values(ordinal).asInstanceOf[MutableInt]
@@ -306,4 +306,8 @@ final class SpecificMutableRow(val values: Array[MutableValue]) extends MutableR
   override def getByte(i: Int): Byte = {
     values(i).asInstanceOf[MutableByte].value
   }
+
+  override def getAs[T](i: Int): T = {
+    values(i).boxed.asInstanceOf[T]
+  }
 }