From a0f8a072e33842f19e53fd28d7578444d1d26cb3 Mon Sep 17 00:00:00 2001
From: Yuming Wang <wgyumg@gmail.com>
Date: Fri, 26 May 2017 20:59:14 -0700
Subject: [PATCH] [SPARK-20748][SQL] Add built-in SQL function CH[A]R.
## What changes were proposed in this pull request?
Add built-in SQL function `CH[A]R`:
For `CHR(bigint|double n)`, returns the ASCII character having the binary equivalent to `n`. If n is larger than 256 the result is equivalent to CHR(n % 256)
## How was this patch tested?
unit tests
Author: Yuming Wang <wgyumg@gmail.com>
Closes #18019 from wangyum/SPARK-20748.
---
.../catalyst/analysis/FunctionRegistry.scala | 2 +
.../expressions/stringExpressions.scala | 45 +++++++++++++++++++
.../expressions/StringExpressionsSuite.scala | 13 ++++++
3 files changed, 60 insertions(+)
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala
index 7521a7e124..a4c7f7a8de 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala
@@ -276,6 +276,8 @@ object FunctionRegistry {
// string functions
expression[Ascii]("ascii"),
+ expression[Chr]("char"),
+ expression[Chr]("chr"),
expression[Base64]("base64"),
expression[Concat]("concat"),
expression[ConcatWs]("concat_ws"),
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala
index 5598a14699..aba2f5f81f 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala
@@ -1267,6 +1267,51 @@ case class Ascii(child: Expression) extends UnaryExpression with ImplicitCastInp
}
}
+/**
+ * Returns the ASCII character having the binary equivalent to n.
+ * If n is larger than 256 the result is equivalent to chr(n % 256)
+ */
+// scalastyle:off line.size.limit
+@ExpressionDescription(
+ usage = "_FUNC_(expr) - Returns the ASCII character having the binary equivalent to `expr`. If n is larger than 256 the result is equivalent to chr(n % 256)",
+ extended = """
+ Examples:
+ > SELECT _FUNC_(65);
+ A
+ """)
+// scalastyle:on line.size.limit
+case class Chr(child: Expression) extends UnaryExpression with ImplicitCastInputTypes {
+
+ override def dataType: DataType = StringType
+ override def inputTypes: Seq[DataType] = Seq(LongType)
+
+ protected override def nullSafeEval(lon: Any): Any = {
+ val longVal = lon.asInstanceOf[Long]
+ if (longVal < 0) {
+ UTF8String.EMPTY_UTF8
+ } else if ((longVal & 0xFF) == 0) {
+ UTF8String.fromString(Character.MIN_VALUE.toString)
+ } else {
+ UTF8String.fromString((longVal & 0xFF).toChar.toString)
+ }
+ }
+
+ override def doGenCode(ctx: CodegenContext, ev: ExprCode): ExprCode = {
+ nullSafeCodeGen(ctx, ev, lon => {
+ s"""
+ if ($lon < 0) {
+ ${ev.value} = UTF8String.EMPTY_UTF8;
+ } else if (($lon & 0xFF) == 0) {
+ ${ev.value} = UTF8String.fromString(String.valueOf(Character.MIN_VALUE));
+ } else {
+ char c = (char)($lon & 0xFF);
+ ${ev.value} = UTF8String.fromString(String.valueOf(c));
+ }
+ """
+ })
+ }
+}
+
/**
* Converts the argument from binary to a base 64 string.
*/
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/StringExpressionsSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/StringExpressionsSuite.scala
index 26978a0482..9ae438d568 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/StringExpressionsSuite.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/StringExpressionsSuite.scala
@@ -263,6 +263,19 @@ class StringExpressionsSuite extends SparkFunSuite with ExpressionEvalHelper {
checkEvaluation(Ascii(Literal.create(null, StringType)), null, create_row("abdef"))
}
+ test("string for ascii") {
+ val a = 'a.long.at(0)
+ checkEvaluation(Chr(Literal(48L)), "0", create_row("abdef"))
+ checkEvaluation(Chr(a), "a", create_row(97L))
+ checkEvaluation(Chr(a), "a", create_row(97L + 256L))
+ checkEvaluation(Chr(a), "", create_row(-9L))
+ checkEvaluation(Chr(a), Character.MIN_VALUE.toString, create_row(0L))
+ checkEvaluation(Chr(a), Character.MIN_VALUE.toString, create_row(256L))
+ checkEvaluation(Chr(a), null, create_row(null))
+ checkEvaluation(Chr(a), 149.toChar.toString, create_row(149L))
+ checkEvaluation(Chr(Literal.create(null, LongType)), null, create_row("abdef"))
+ }
+
test("base64/unbase64 for string") {
val a = 'a.string.at(0)
val b = 'b.binary.at(0)
--
GitLab