Update deps, stabilize flat format's order

Issue: RiverBench/RiverBench#120

build.sbt

@@ -5,12 +5,12 @@ resolvers +=
   "Sonatype OSS Snapshots" at "https://s01.oss.sonatype.org/content/repositories/snapshots"
 
 lazy val circeV = "0.14.7"
-lazy val jellyV = "0.14.1+12-0d137338-SNAPSHOT"
-lazy val jenaV = "4.10.0"
-lazy val pekkoV = "1.0.2"
+lazy val jellyV = "2.0.0"
+lazy val jenaV = "5.1.0"
+lazy val pekkoV = "1.1.0"
 lazy val pekkoHttpV = "1.0.1"
 lazy val pekkoConnV = "1.0.2"
-lazy val rdf4jV = "4.3.11"
+lazy val rdf4jV = "5.0.2"
 lazy val icu4jV = "74.2"
 
 lazy val root = (project in file("."))
@@ -20,7 +20,7 @@ lazy val root = (project in file("."))
 
     // Scala 3 or not Scala at all
     libraryDependencies ++= Seq(
-      "com.google.guava" % "guava" % "33.2.0-jre",
+      "com.google.guava" % "guava" % "33.2.1-jre",
       "com.ibm.icu" % "icu4j" % icu4jV,
       "eu.ostrzyciel.jelly" %% "jelly-stream" % jellyV,
       "eu.ostrzyciel.jelly" %% "jelly-jena" % jellyV,
@@ -62,5 +62,6 @@ lazy val root = (project in file("."))
     // emit deprecated warnings
     scalacOptions ++= Seq(
       "-deprecation",
+      "-Werror",
     ),
   )

src/main/scala/commands/PackageCommand.scala

@@ -7,10 +7,11 @@ import util.io.*
 import eu.ostrzyciel.jelly.core.proto.v1.PhysicalStreamType
 import eu.ostrzyciel.jelly.core.{JellyOptions, LogicalStreamTypeFactory}
 import eu.ostrzyciel.jelly.stream.{EncoderFlow, JellyIo}
+import org.apache.jena.graph.GraphMemFactory
 import org.apache.jena.rdf.model.{ModelFactory, Resource}
 import org.apache.jena.riot.{Lang, RDFParser, RDFWriter}
 import org.apache.jena.riot.lang.LabelToNode
-import org.apache.jena.riot.system.{ErrorHandlerFactory, FactoryRDFStd}
+import org.apache.jena.riot.system.{ErrorHandlerFactory, FactoryRDFStd, StreamRDFWriter}
 import org.apache.jena.sparql.core.{DatasetGraph, DatasetGraphFactory}
 import org.apache.pekko.stream.*
 import org.apache.pekko.stream.connectors.file.TarArchiveMetadata
@@ -20,14 +21,15 @@ import org.apache.pekko.{Done, NotUsed}
 import org.eclipse.rdf4j.model.vocabulary.XSD
 import org.eclipse.rdf4j.rio
 
-import java.io.{ByteArrayInputStream, FileOutputStream}
+import java.io.{ByteArrayInputStream, ByteArrayOutputStream, FileOutputStream}
 import java.nio.file.{FileSystems, Path}
 import java.util.UUID
 import scala.concurrent.Future
 import scala.jdk.CollectionConverters.*
 
 object PackageCommand extends Command:
   import eu.ostrzyciel.jelly.convert.jena.{*, given}
+  import RdfOrdering.given
 
   sealed trait DistType(val weight: Int)
   object DistType:
@@ -48,6 +50,8 @@ object PackageCommand extends Command:
     val repoDir = FileSystems.getDefault.getPath(args(1))
     val sourceArchiveFile = FileSystems.getDefault.getPath(args(2))
     val outDir = FileSystems.getDefault.getPath(args(3))
+
+    GraphMemFactory.setDftGraphSameTerm(false)
 
     val metadata = MetadataReader.read(repoDir)
     val stats = new StatCounterSuite(metadata.elementCount)
@@ -206,11 +210,12 @@ object PackageCommand extends Command:
   Flow[(DatasetGraph, Long), (Long, StatCounterSuite.Result), NotUsed] =
     Flow[(DatasetGraph, Long)]
       .wireTap((_, num) => if (num + 1) % 100_000 == 0 then println(s"Stats stream at: ${num + 1}"))
-      .splitAfter(SubstreamCancelStrategy.propagate)((_, num) =>
+      .splitAfter((_, num) =>
         val shouldSplit = Constants.packageSizes.contains(num + 1)
         if shouldSplit then println(s"Splitting stats stream at ${num + 1}")
         shouldSplit
       )
+      .withAttributes(ActorAttributes.supervisionStrategy(Supervision.stoppingDecider))
       .map((ds, num) => {
         stats.add(ds)
         num + 1
@@ -334,19 +339,20 @@ object PackageCommand extends Command:
     }
 
     Flow[(DatasetGraph, Long)]
-      .map((ds, _) => {
+      .map((ds, _) => ds.find().asScala.toSeq.sorted)
+      .map(quads => {
+        val os = ByteArrayOutputStream()
         if metadata.streamTypes.exists(_.elementType == ElementType.Triple) then
-          RDFWriter.create()
-            .lang(Lang.NTRIPLES)
-            .source(ds.getDefaultGraph)
-            .asString()
+          val writer = StreamRDFWriter.getWriterStream(os, Lang.NTRIPLES)
+          quads.foreach(q => writer.triple(q.asTriple))
+          writer.finish()
         else
-          RDFWriter.create()
-            .lang(Lang.NQUADS)
-            .source(ds)
-            .asString()
+          val writer = StreamRDFWriter.getWriterStream(os, Lang.NQUADS)
+          quads.foreach(q => writer.quad(q))
+          writer.finish()
+        os.toByteArray
       })
-      .map(ByteString.fromString)
+      .map(ByteString.fromArrayUnsafe)
       .toMat(StreamUtil.broadcastSink(sinks))(Keep.right)
 
   /**

src/main/scala/commands/PackageSchemaCommand.scala

@@ -41,14 +41,14 @@ object PackageSchemaCommand extends Command:
     for name <- toProcessNames do
       val inPath = repoDir.resolve(s"src/$name.ttl")
       val model = RDFDataMgr.loadModel(inPath.toString)
-      model.listSubjectsWithProperty(RDF.`type`, OWL.Ontology).asScala.toList match
+      model.listSubjectsWithProperty(RDF.`type`, OWL2.Ontology).asScala.toList match
         case List(ontology) =>
           // Update version IRI
           model.removeAll(ontology, OWL2.versionIRI, null)
           model.add(ontology, OWL2.versionIRI, model.createResource(s"${ontology.getURI}/$version"))
 
           // Update imports
-          ontology.listProperties(OWL.imports).asScala
+          ontology.listProperties(OWL2.imports).asScala
             .filter(_.getObject.isURIResource)
             .map(_.getObject.asResource)
             .filter(_.getURI.startsWith(schemaBase))
@@ -57,8 +57,8 @@ object PackageSchemaCommand extends Command:
             .toSeq
             .foreach { (oldRes, nameParts) =>
               val newRes = model.createResource(s"$schemaBase${nameParts(0)}/$version")
-              model.remove(ontology, OWL.imports, oldRes)
-              model.add(ontology, OWL.imports, newRes)
+              model.remove(ontology, OWL2.imports, oldRes)
+              model.add(ontology, OWL2.imports, newRes)
             }
         case _ => ()
 

src/main/scala/util/RdfOrdering.scala

@@ -0,0 +1,19 @@
+package io.github.riverbench.ci_worker
+package util
+
+import org.apache.jena.sparql.core.Quad
+
+object RdfOrdering:
+
+  given Ordering[Quad] with
+    def compare(x: Quad, y: Quad): Int =
+      val cmpGraph = x.getGraph.toString.compareTo(y.getGraph.toString)
+      if cmpGraph != 0 then cmpGraph
+      else
+        val cmpSubject = x.getSubject.toString.compareTo(y.getSubject.toString)
+        if cmpSubject != 0 then cmpSubject
+        else
+          val cmpPredicate = x.getPredicate.toString.compareTo(y.getPredicate.toString)
+          if cmpPredicate != 0 then cmpPredicate
+          else
+            x.getObject.toString.compareTo(y.getObject.toString)

src/main/scala/util/StatCounterSuite.scala

@@ -89,9 +89,9 @@ class StatCounterSuite(val size: Long):
 
   def add(ds: DatasetGraph): Unit =
     if ds.getDefaultGraph.isEmpty then
-      cGraphs.add(ds.listGraphNodes().asScala.map(_.toString(true)).toSeq)
+      cGraphs.add(ds.listGraphNodes().asScala.map(_.toString()).toSeq)
     else
-      cGraphs.add((ds.listGraphNodes().asScala.toSeq :+ DEFAULT_GRAPH).map(_.toString(true)))
+      cGraphs.add((ds.listGraphNodes().asScala.toSeq :+ DEFAULT_GRAPH).map(_.toString()))
 
     val subjects = mutable.Set[Node]()
     val predicates = mutable.Set[Node]()
@@ -131,7 +131,7 @@ class StatCounterSuite(val size: Long):
       else if n.isBlank then
         blankNodes += n.getBlankNodeLabel
       else if n.isLiteral then
-        val lit = n.toString(false)
+        val lit = n.toString()
         controlCharCount += countAsciiControlChars(lit)
         literals += lit
         if n.getLiteralLanguage != "" then
@@ -159,9 +159,9 @@ class StatCounterSuite(val size: Long):
 
     cQuotedTriples.lightAdd(quotedTripleCount)
 
-    cSubjects.addUnique(subjects.map(_.toString(true)))
-    cPredicates.addUnique(predicates.map(_.toString(true)))
-    cObjects.addUnique(objects.map(_.toString(true)))
+    cSubjects.addUnique(subjects.map(_.toString()))
+    cPredicates.addUnique(predicates.map(_.toString()))
+    cObjects.addUnique(objects.map(_.toString()))
     cStatements.lightAdd(stCount)
 
   private def countAsciiControlChars(s: String): Int =