= Benchmark against random names

benchmark/src/main/scala/org/globalnames/ScientificNameParserBenchmark.scala

@@ -3,24 +3,50 @@ package org.globalnames
 import java.util.concurrent.TimeUnit
 
 import org.openjdk.jmh.annotations._
-import org.globalnames.parser.ScientificNameParser.{instance ⇒ snp, Result}
+import org.globalnames.parser.ScientificNameParser.{Result, instance => snp}
 
-@Warmup(iterations = 5, time = 1, timeUnit = TimeUnit.SECONDS)
-@Measurement(iterations = 5, time = 1, timeUnit = TimeUnit.SECONDS)
+import scala.io.Source
+import scala.util.Random
+
+@Warmup(iterations = 3, time = 1, timeUnit = TimeUnit.SECONDS)
+@Measurement(iterations = 3, time = 1, timeUnit = TimeUnit.SECONDS)
 @Fork(1)
 @BenchmarkMode(Array(Mode.AverageTime))
 @State(Scope.Benchmark)
 @OutputTimeUnit(TimeUnit.NANOSECONDS)
 class ScientificNameParserBenchmark {
-  @Param(Array("Helvella macropus",
-               "Helvella leucophaea Pers., 1801",
-               "Helvella lycoperdoides Scop. 1772",
-               "Helvella lilacina Wulfen, 1786",
-               "Helvella leucopus var. populina I. Arroyo & Calonge 2000"))
+  import ScientificNameParserBenchmark.allNames
+
+  @Param(Array("0", "1", "2", "3", "4", "5", "6", "7", "8", "9", "10"))
+  var seed: Int = _
+
   var name: String = _
 
+  @Setup
+  def setup(): Unit = {
+    val rand = new Random(seed)
+    name = allNames(rand.nextInt(allNames.size)).verbatim
+    println(name)
+  }
+
   @Benchmark
-  def fromString: Result = {
+  def fromString: Result= {
     snp.fromString(name)
   }
 }
+
+object ScientificNameParserBenchmark {
+  case class ExpectedName(verbatim: String, json: String, simple: String)
+
+  def expectedNames(filePath: String): Vector[ExpectedName] = {
+    Source.fromFile(filePath, "UTF-8")
+      .getLines
+      .takeWhile { _.trim != "__END__" }
+      .withFilter { line => !(line.isEmpty || ("#\r\n\f\t" contains line.charAt(0))) }
+      .sliding(3, 3)
+      .map { ls => ExpectedName(ls(0), ls(1), ls(2)) }
+      .toVector
+  }
+
+  val allNames = expectedNames("../parser/src/test/resources/test_data.txt").drop(200)
+}