add codegen soundex

prabeesh · Jul 27, 2015 · 2538908 · 2538908
1 parent d15d329
commit 2538908
Show file tree

Hide file tree

Showing 3 changed files with 10 additions and 3 deletions.
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringOperations.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringOperations.scala
@@ -794,6 +794,9 @@ case class SoundEx(child: Expression) extends UnaryExpression with ExpectsInputT
   override def nullSafeEval(input: Any): Any = {
     input.asInstanceOf[UTF8String].soundex()
   }
+  override def genCode(ctx: CodeGenContext, ev: GeneratedExpressionCode): String = {
+    nullSafeCodeGen(ctx, ev, c => s"${ev.primitive} = $c.soundex();")
+  }
 }
 
 /**

diff --git a/...yst/src/test/scala/org/apache/spark/sql/catalyst/expressions/StringExpressionsSuite.scala b/...yst/src/test/scala/org/apache/spark/sql/catalyst/expressions/StringExpressionsSuite.scala
@@ -335,11 +335,12 @@ class StringExpressionsSuite extends SparkFunSuite with ExpressionEvalHelper {
     checkEvaluation(SoundEx(Literal("ZIN")), "Z500", create_row("s1"))
     checkEvaluation(SoundEx(Literal("SU")), "S000", create_row("s2"))
     checkEvaluation(SoundEx(Literal("")), "", create_row("s3"))
-    checkEvaluation(SoundEx(Literal(null)), null, create_row("s4"))
+    checkEvaluation(SoundEx(Literal.create(null, StringType)), null, create_row("s4"))
+
     // scalastyle:off
     // non ascii characters are not allowed in the code, so we disable the scalastyle here.
-    checkEvaluation(SoundEx(Literal("再见")), "再见", create_row("s5"))
-    checkEvaluation(SoundEx(Literal("z再見")), "z再見", create_row("s6"))
+    checkEvaluation(SoundEx(Literal("测试")), "测试", create_row("s5"))
+    checkEvaluation(SoundEx(Literal("z測試")), "z測試", create_row("s6"))
     checkEvaluation(SoundEx(Literal("Tschüss")), "Tschüss", create_row("s7"))
     // scalastyle:on
     checkEvaluation(SoundEx(Literal("zZ")), "z000", create_row("s8"))

diff --git a/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java b/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java
@@ -641,6 +641,9 @@ public int hashCode() {
    * https://en.wikipedia.org/wiki/Soundex
    */
   public UTF8String soundex() {
+    if(this == null) {
+      return null;
+    }
     if (numBytes == 0) {
       return UTF8String.fromBytes(new byte[0]);
     }