[SPARK-1995][SQL] system function upper and lower can be supported

I don't know whether it's time to implement system function about string operation in spark sql now.

Author: egraldlo <[email protected]>

Closes #936 from egraldlo/stringoperator and squashes the following commits:

3c6c60a [egraldlo] Add UPPER, LOWER, MAX and MIN into hive parser
ea76d0a [egraldlo] modify the formatting issues
b49f25e [egraldlo] modify the formatting issues
1f0bbb5 [egraldlo] system function upper and lower supported
13d3267 [egraldlo] system function upper and lower supported

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/SqlParser.scala

@@ -124,6 +124,8 @@ class SqlParser extends StandardTokenParsers with PackratParsers {
   protected val OVERWRITE = Keyword("OVERWRITE")
   protected val LIKE = Keyword("LIKE")
   protected val RLIKE = Keyword("RLIKE")
+  protected val UPPER = Keyword("UPPER")
+  protected val LOWER = Keyword("LOWER")
   protected val REGEXP = Keyword("REGEXP")
   protected val ORDER = Keyword("ORDER")
   protected val OUTER = Keyword("OUTER")
@@ -329,6 +331,8 @@ class SqlParser extends StandardTokenParsers with PackratParsers {
     AVG ~> "(" ~> expression <~ ")" ^^ { case exp => Average(exp) } |
     MIN ~> "(" ~> expression <~ ")" ^^ { case exp => Min(exp) } |
     MAX ~> "(" ~> expression <~ ")" ^^ { case exp => Max(exp) } |
+    UPPER ~> "(" ~> expression <~ ")" ^^ { case exp => Upper(exp) } |
+    LOWER ~> "(" ~> expression <~ ")" ^^ { case exp => Lower(exp) } |
     IF ~> "(" ~> expression ~ "," ~ expression ~ "," ~ expression <~ ")" ^^ {
       case c ~ "," ~ t ~ "," ~ f => If(c,t,f)
     } |

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

@@ -70,6 +70,22 @@ trait StringRegexExpression {
   }
 }
 
+trait CaseConversionExpression {
+  self: UnaryExpression =>
+
+  type EvaluatedType = Any
+
+  def convert(v: String): String
+
+  def nullable: Boolean = child.nullable
+  def dataType: DataType = StringType
+
+  override def eval(input: Row): Any = {
+    val converted = child.eval(input)
+    convert(converted.toString)
+  }
+}
+
 /**
  * Simple RegEx pattern matching function
  */
@@ -115,3 +131,19 @@ case class RLike(left: Expression, right: Expression)
   override def escape(v: String): String = v
   override def matches(regex: Pattern, str: String): Boolean = regex.matcher(str).find(0)
 }
+
+/**
+ * A function that converts the characters of a string to uppercase.
+ */
+case class Upper(child: Expression) extends UnaryExpression with CaseConversionExpression {
+
+  override def convert(v: String): String = v.toUpperCase()
+}
+
+/**
+ * A function that converts the characters of a string to lowercase.
+ */
+case class Lower(child: Expression) extends UnaryExpression with CaseConversionExpression {
+
+  override def convert(v: String): String = v.toLowerCase()
+}

sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala

@@ -313,4 +313,27 @@ class SQLQuerySuite extends QueryTest {
         (3, "C"),
         (4, "D")))
   }
+
+  test("system function upper()") {
+    checkAnswer(
+      sql("SELECT n,UPPER(l) FROM lowerCaseData"),
+      Seq(
+        (1, "A"),
+        (2, "B"),
+        (3, "C"),
+        (4, "D")))
+  }
+
+  test("system function lower()") {
+    checkAnswer(
+      sql("SELECT N,LOWER(L) FROM upperCaseData"),
+      Seq(
+        (1, "a"),
+        (2, "b"),
+        (3, "c"),
+        (4, "d"),
+        (5, "e"),
+        (6, "f")))
+  }  
+
 }

sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveQl.scala

@@ -781,6 +781,10 @@ private[hive] object HiveQl {
   val COUNT = "(?i)COUNT".r
   val AVG = "(?i)AVG".r
   val SUM = "(?i)SUM".r
+  val MAX = "(?i)MAX".r
+  val MIN = "(?i)MIN".r
+  val UPPER = "(?i)UPPER".r
+  val LOWER = "(?i)LOWER".r
   val RAND = "(?i)RAND".r
   val AND = "(?i)AND".r
   val OR = "(?i)OR".r
@@ -817,7 +821,13 @@ private[hive] object HiveQl {
     case Token("TOK_FUNCTIONDI", Token(COUNT(), Nil) :: args) => CountDistinct(args.map(nodeToExpr))
     case Token("TOK_FUNCTION", Token(SUM(), Nil) :: arg :: Nil) => Sum(nodeToExpr(arg))
     case Token("TOK_FUNCTIONDI", Token(SUM(), Nil) :: arg :: Nil) => SumDistinct(nodeToExpr(arg))
-
+    case Token("TOK_FUNCTION", Token(MAX(), Nil) :: arg :: Nil) => Max(nodeToExpr(arg))
+    case Token("TOK_FUNCTION", Token(MIN(), Nil) :: arg :: Nil) => Min(nodeToExpr(arg))
+
+    /* System functions about string operations */
+    case Token("TOK_FUNCTION", Token(UPPER(), Nil) :: arg :: Nil) => Upper(nodeToExpr(arg))
+    case Token("TOK_FUNCTION", Token(LOWER(), Nil) :: arg :: Nil) => Lower(nodeToExpr(arg))
+
     /* Casts */
     case Token("TOK_FUNCTION", Token("TOK_STRING", Nil) :: arg :: Nil) =>
       Cast(nodeToExpr(arg), StringType)