[SPARK-2330] Spark shell has weird scala semantics

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Duplicate
• Affects Version/s: 0.9.1, 1.0.0
• Component/s: Spark Core
• Labels: scala shell

Environment

Ubuntu 14.04 with spark-x.x.x-bin-hadoop2

• Assignee: Unassigned
• Reporter: Andrea Ferretti
• Created: 30/Jun/14 14:10
• Updated: 03/Jul/14 18:22
• Resolved: 03/Jul/14 18:22

Description

Normal scala expressions are interpreted in a strange way in the spark shell. For instance

case class Foo(x: Int)
def print(f: Foo) = f.x
val f = Foo(3)
print(f)
<console>:24: error: type mismatch;
 found   : Foo
 required: Foo

For another example

trait Currency
case object EUR extends Currency
case object USD extends Currency

def nextCurrency: Currency = nextInt(2) match {
  case 0 => EUR
  case _ => USD
}

<console>:22: error: type mismatch;
 found   : EUR.type
 required: Currency
         case 0 => EUR

<console>:24: error: type mismatch;
 found   : USD.type
 required: Currency
         case _ => USD

Issue Links

• duplicates: SPARK-1199 Type mismatch in Spark shell when using case class defined in shell