Scala parsing, how to go through what has been collected - scala

Question

Using scala parser combinators I have parsed some text input and created some of my own types along the way. The result prints fine. Now I need to go through the output, which I presume is a nested structure that includes the types I created. How do I go about this?
This is how I call the parser:
GMParser1.parseItem(i_inputHard_2) match {
case GMParser1.Success(res, _) =>
println(">" + res + "< of type: " + res.getClass.getSimpleName)
case x =>
println("Could not parse the input string:" + x)
}
EDIT
What I get back from MsgResponse(O2 Flow|Off) is:
>(MsgResponse~(O2 Flow~Off))< of type: $tilde
And what I get back from WthrResponse(Id(Tube 25,Carbon Monoxide)|0.20) is:
>(WthrResponse~IdWithValue(Id(Tube 25,Carbon Monoxide),0.20))< of type: $tilde
Just to give context to the question here is some of the input parsing. I will want to get at Id:
trait Keeper
case class Id(leftContents:String,rightContents:String) extends Keeper
And here is Id being created:
def id = "Id(" ~> idContents <~ ")" ^^ { contents => Id(contents._1,contents._2) }
And here is the whole of the parser:
object GMParser1 extends RegexParsers {
override def skipWhitespace = false
def number = regex(new Regex("[-+]?(\\d*[.])?\\d+"))
def idContents = text ~ ("," ~> text)
def id = "Id(" ~> idContents <~ ")" ^^ { contents => Id(contents._1,contents._2) }
def text = """[A-Za-z0-9* ]+""".r
def wholeWord = """[A-Za-z]+""".r
def idBracketContents = id ~ ( "|" ~> number ) ^^ { contents => IdWithValue(contents._1,contents._2) }
def nonIdBracketContents = text ~ ( "|" ~> text )
def bracketContents = idBracketContents | nonIdBracketContents
def outerBrackets = "(" ~> bracketContents <~ ")"
def target = wholeWord ~ outerBrackets
def parseItem(str: String): ParseResult[Any] = parse(target, str)
trait Keeper
case class Id(leftContents:String,rightContents:String) extends Keeper
case class IdWithValue(leftContents:Id,numberContents:String) extends Keeper
}

Answer 1

The parser created by the ~ operator produces a value of the ~ case class. To get at its contents, you can pattern match on it like on any other case class (keeping in mind that its name is symbolic, so it's used infix).
So you can replace case GMParser1.Success(res, _) => ... with case GMParser1.Success(functionName ~ argument) => ... to get at the function name and argument (or whatever the semantics of wholeWord and bracketContents in wholeWord "(" bracketContents ")" are). You can then similarly use a nested pattern to get at the individual parts of the argument.
You could (and probably should) also use ^^ together with pattern matching in your rules to create a more meaningful AST structure that doesn't contain ~. This would be useful to distinguish a nonIdBracketContents result from a bracketContents result for example.