why does takeOnePointedGraph return Try?

[antonkulaga]

I see that takeOnePointedGraph

  def takeOnePointedGraph: Try[PointedGraph[Rdf]] = {
    val it = nodes.iterator
    if (!it.hasNext) {
      Failure(WrongExpectation("expected exactly one node but got 0"))
    } else {
      val first = it.next
      Success(PointedGraph(first, graph))
    }
  }

I wonder, why does it return try? In most of scala collection such thing is called heapOption and returns Option. I suggest that we change is in a same way.

[ghost]

@antonkulaga Some collection methods return an Option (eg headOption, as you say), others throw exceptions (eg head). So takeOnePointedGraph returning Try is just a nicer way of having the exception. The reason why both head methods are needed so the client can decide if the value is optional or not.

But I guess you already know this. What you need is PointedGraphs.asOption

[antonkulaga]

@inthenow I simply do not like how method are named in PointedGraphs. There are conventions in scala collections, so head and headOption are short names that are obvious for the users. And actually verbose

def takeOnePointedGraph

means that we take head (not an arbitary onePointedGraph)

[bblfish]

@antonkulaga In RDF triples are not ordered. So this method will not necessarily return the same PG for each call. It partly depends on the underlying lib.