Alexey Novakov novakov-alexey-zz

From `scala.xml.pull` to `javax.xml.stream.events`

Want to process XML that's too large to fit in memory?

The Scala standard library used to offer scala.xml.pull for this. It became part of the scala-xml library when scala-xml became separate. But then scala.xml.pull got deprecated (in scala.xml 1.1.1) and finally removed entirely (in scala-xml 2.0.0-M1). The recommended replacement is javax.xml.stream.events.

I had some old code that used scala.xml.pull to digest my iTunes Music Library.xml and print various statistics. Recently I converted it to Scala 3, so I decided to get off the deprecated API at the same time.

So, here is the before-and-after. Perhaps this will help other users of scala.xml.pull who want to convert.

	observations.foldLeft(
	List[Observation]() -> WindowStats(0, Int.MaxValue, Int.MinValue, 0)
	) {
	case ((series, stats), ob) =>

	@tailrec
	def calcStats(
	st: WindowStats,
	obs: List[Observation]): (WindowStats, List[Observation]) = {

	/**
	*
	* @param edgeTo a sequence which represents the last edge
	* on the shortest path from 'sourceV' to vertex i.
	* None means there is no path to vertex i
	* @param distTo a sequence of distances from source vertex to a specific i vertex
	*/
	class ShortestPathCalc(edgeTo: Seq[Option[DirectedEdge]], distTo: Seq[Double]) {
	/**
	*

	object ShortestPath {
	/**
	* Function tries to find a shortest path from source vertex
	* to all other vertices in the graph
	*
	* @param g EdgeWeightedDigraph to find a shortest path
	* @param sourceV source vertex to find a shortest path from
	* @return return either error as string when input parameters
	* are invalid or return shortest path result
	*/

	/**
	* Algorithm is based on Java version implemented
	* by Princeton University https://algs4.cs.princeton.edu/44sp/
	*/

	final case class DirectedEdge(from: Int, to: Int, weight: Double)

	final case class EdgeWeightedDigraph(adj: Map[Int, List[DirectedEdge]] = Map.empty)

	object EdgeWeightedDigraphOps {

	lazy val root = (project in file(".")).settings(
	inThisBuild(List(organization := "org.alexeyn", scalaVersion := "2.12.7")),
	name := "akka-crud-service",
	libraryDependencies ++= Seq(
	...
	),
	dockerBaseImage := "openjdk:8-jre-alpine"
	).enablePlugins(JavaAppPackaging, AshScriptPlugin)

	case class Observation(time: Long, value: Double)

	// contains 4 additional columns we need to calculate for each event
	case class WindowStats(sum: Double, min: Double, max: Double, count: Int)

	import java.time.LocalDate
	import java.time.format.DateTimeFormatter

	import akka.http.scaladsl.marshallers.sprayjson.SprayJsonSupport
	import spray.json.{DefaultJsonProtocol, DeserializationException, JsString, JsValue, JsonFormat, RootJsonFormat}

	trait JsonCodes extends SprayJsonSupport with DefaultJsonProtocol {
	implicit val trip: RootJsonFormat[Trip] = jsonFormat7(Trip)
	implicit val trips: RootJsonFormat[Trips] = jsonFormat1(Trips)
	implicit val commandResult: RootJsonFormat[CommandResult] = jsonFormat1(CommandResult)

	object Main extends App with StrictLogging {
	implicit val system: ActorSystem = ActorSystem("crud-service")
	implicit val materializer: ActorMaterializer = ActorMaterializer()
	implicit val executionContext: ExecutionContext = system.dispatcher

	val (server, cfg) = AppConfig.load.fold(e => sys.error(e.toString), identity)
	val mod = new Module(true, cfg)
	val serverBinding = Http().bindAndHandle(mod.routes, server.host, server.port)

	serverBinding.onComplete {

	class Module(createSchema: Boolean = true, cfg: Config = ConfigFactory.load())(
	implicit system: ActorSystem,
	executionContext: ExecutionContext
	) extends StrictLogging {

	val db = Database.forConfig("storage", cfg)
	val dao = new TripDao(db)
	val service = new TripService(dao)
	val routes: Route = concat(QueryRoutes.routes(service), CommandRoutes.routes(service))

Alexey Novakov novakov-alexey-zz

From scala.xml.pull to javax.xml.stream.events

From `scala.xml.pull` to `javax.xml.stream.events`