adding benchmarking test and fixed some compilation issues

Author: Jolanrensen

core/src/test/kotlin/org/jetbrains/kotlinx/dataframe/io/DoubleParserTests.kt

@@ -120,6 +120,7 @@ class DoubleParserTests {
             "100123.35",
             "-204,235.23",
             "1.234e3",
+            "3e-04", // failed with old double parser
         )
 
         val expectedDoubles = listOf(
@@ -128,6 +129,7 @@ class DoubleParserTests {
             100_123.35,
             -204_235.23,
             1.234e3,
+            3e-04,
         )
 
         // CharSequence

dataframe-csv/build.gradle.kts

@@ -11,6 +11,7 @@ plugins {
         alias(ktlint)
         alias(jupyter.api)
         alias(docProcessor)
+        alias(kotlinx.benchmark)
     }
     idea
 }
@@ -39,12 +40,19 @@ dependencies {
     implementation(libs.kotlin.coroutinesCore)
 
     testApi(project(":core"))
+    testImplementation(libs.kotlinx.benchmark.runtime)
     testImplementation(libs.junit)
     testImplementation(libs.kotestAssertions) {
         exclude("org.jetbrains.kotlin", "kotlin-stdlib-jdk8")
     }
 }
 
+benchmark {
+    targets {
+        register("test")
+    }
+}
+
 val generatedSourcesFolderName = "generated-sources"
 
 // Backup the kotlin source files location
@@ -155,7 +163,6 @@ kotlin {
     explicitApi()
     sourceSets.all {
         languageSettings {
-            enableLanguageFeature("ExplicitBackingFields")
         }
     }
 }

dataframe-csv/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/io/ListSink.kt

@@ -66,9 +66,10 @@ internal class ListSink(val columnIndex: Int, val dataType: DataType) : SinkSour
         )
     }
 
-    @Suppress("MUST_BE_INITIALIZED_OR_BE_ABSTRACT", "EXPLICIT_BACKING_FIELDS_UNSUPPORTED")
+    private val _data: MutableList<Any?> = mutableListOf()
+
     val data: List<Any?>
-        field = mutableListOf()
+        get() = _data
 
     var hasNulls: Boolean = false
         private set
@@ -121,13 +122,12 @@ internal class ListSink(val columnIndex: Int, val dataType: DataType) : SinkSour
         destEnd: Int,
         isNull: BooleanArray,
     ) {
-        data as MutableList<Any?>
         while (data.size < destBegin) {
-            data += null
+            _data += null
             hasNulls = true
         }
         for ((srcIndex, _) in (destBegin..<destEnd).withIndex()) {
-            data += getValue(src, srcIndex, isNull)
+            _data += getValue(src, srcIndex, isNull)
         }
     }
 
@@ -137,9 +137,8 @@ internal class ListSink(val columnIndex: Int, val dataType: DataType) : SinkSour
         destEnd: Int,
         isNull: BooleanArray,
     ) {
-        data as MutableList<Any?>
         for ((srcIndex, destIndex) in (destBegin..<destEnd).withIndex()) {
-            data[destIndex] = getValue(src, srcIndex, isNull)
+            _data[destIndex] = getValue(src, srcIndex, isNull)
         }
     }
 

dataframe-csv/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/csv.kt

@@ -1,7 +1,8 @@
+@file:JvmName("CsvDeephavenKt")
+
 package org.jetbrains.kotlinx.dataframe.io
 
 import org.jetbrains.kotlinx.dataframe.DataFrame
-import org.jetbrains.kotlinx.dataframe.api.ParserOptions
 import org.jetbrains.kotlinx.dataframe.codeGen.AbstractDefaultReadMethod
 import org.jetbrains.kotlinx.dataframe.codeGen.DefaultReadDfMethod
 import org.jetbrains.kotlinx.dataframe.impl.io.DelimParams

dataframe-csv/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/readCsv.kt

@@ -1,3 +1,5 @@
+@file:JvmName("ReadCsvDeephavenKt")
+
 package org.jetbrains.kotlinx.dataframe.io
 
 import io.deephaven.csv.CsvSpecs

dataframe-csv/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/readDelim.kt

@@ -1,3 +1,5 @@
+@file:JvmName("ReadDelimDeephavenKt")
+
 package org.jetbrains.kotlinx.dataframe.io
 
 import io.deephaven.csv.CsvSpecs

dataframe-csv/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/readTsv.kt

@@ -1,3 +1,5 @@
+@file:JvmName("ReadTsvDeephavenKt")
+
 package org.jetbrains.kotlinx.dataframe.io
 
 import io.deephaven.csv.CsvSpecs

dataframe-csv/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/tsv.kt

@@ -1,3 +1,5 @@
+@file:JvmName("TsvDeephavenKt")
+
 package org.jetbrains.kotlinx.dataframe.io
 
 import org.jetbrains.kotlinx.dataframe.DataFrame

dataframe-csv/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/writeCsv.kt

@@ -1,3 +1,5 @@
+@file:JvmName("WriteCsvDeephavenKt")
+
 package org.jetbrains.kotlinx.dataframe.io
 
 import org.apache.commons.csv.CSVFormat

dataframe-csv/src/main/kotlin/org/jetbrains/kotlinx/dataframe/io/writeTsv.kt

@@ -1,3 +1,5 @@
+@file:JvmName("WriteTsvDeephavenKt")
+
 package org.jetbrains.kotlinx.dataframe.io
 
 import org.apache.commons.csv.CSVFormat

dataframe-csv/src/test/kotlin/org/jetbrains/kotlinx/dataframe/io/BenchmarkTest.kt

@@ -0,0 +1,55 @@
+package org.jetbrains.kotlinx.dataframe.io
+
+import org.jetbrains.kotlinx.dataframe.DataFrame
+import org.openjdk.jmh.annotations.Benchmark
+import org.openjdk.jmh.annotations.BenchmarkMode
+import org.openjdk.jmh.annotations.Measurement
+import org.openjdk.jmh.annotations.Mode
+import org.openjdk.jmh.annotations.Param
+import org.openjdk.jmh.annotations.Scope
+import org.openjdk.jmh.annotations.Setup
+import org.openjdk.jmh.annotations.State
+import org.openjdk.jmh.annotations.TearDown
+import org.openjdk.jmh.annotations.Warmup
+import java.io.File
+import java.util.concurrent.TimeUnit
+
+@BenchmarkMode(Mode.SingleShotTime)
+@Warmup(iterations = 10, time = 5, timeUnit = TimeUnit.SECONDS)
+@Measurement(iterations = 20, timeUnit = TimeUnit.SECONDS)
+@State(Scope.Benchmark)
+open class BenchmarkTest {
+
+    @Param("small", "medium", "large")
+    var type = ""
+    var file: File? = null
+
+    @Setup
+    fun setup() {
+        System.setProperty("org.slf4j.simpleLogger.defaultLogLevel", "info")
+        file = File(
+            "src/test/resources/" + when (type) {
+                "small" -> "testCSV.csv"
+                "medium" -> "gross-domestic-product-june-2024-quarter.csv"
+                "large" -> "largeCsv.csv.gz"
+                else -> throw IllegalArgumentException("Invalid type")
+            },
+        )
+    }
+
+    @TearDown
+    fun tearDown() {
+        file = null
+    }
+
+    @Benchmark
+    fun apache() {
+        DataFrame.readCSV(file!!)
+    }
+
+    @OptIn(ExperimentalCsv::class)
+    @Benchmark
+    fun deephaven() {
+        DataFrame.readCsv(file!!)
+    }
+}