wip extracting json module

Author: Jolanrensen

build.gradle.kts

@@ -54,6 +54,7 @@ dependencies {
     api(projects.dataframeExcel)
     api(projects.dataframeJdbc)
     api(projects.dataframeCsv)
+    api(projects.dataframeJson)
 
     // experimental, so not included by default:
     // api(projects.dataframeOpenapi)
@@ -64,6 +65,7 @@ dependencies {
     kover(projects.dataframeOpenapi)
     kover(projects.dataframeJdbc)
     kover(projects.dataframeCsv)
+    kover(projects.dataframeJson)
     kover(projects.plugins.kotlinDataframe)
     kover(projects.dataframeJupyter)
 }

core/build.gradle.kts

@@ -65,8 +65,6 @@ dependencies {
 
     api(libs.commonsCsv)
     implementation(libs.commonsIo)
-    implementation(libs.serialization.core)
-    implementation(libs.serialization.json)
     implementation(libs.fastDoubleParser)
 
     api(libs.kotlin.datetimeJvm)

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/annotations/ImportDataSchema.kt

@@ -5,7 +5,6 @@ import org.jetbrains.kotlinx.dataframe.api.KeyValueProperty
 import org.jetbrains.kotlinx.dataframe.columns.ColumnGroup
 import org.jetbrains.kotlinx.dataframe.columns.FrameColumn
 import org.jetbrains.kotlinx.dataframe.documentation.UnifyingNumbers
-import org.jetbrains.kotlinx.dataframe.io.JSON
 
 /**
  * Annotation preprocessing will generate a DataSchema interface from the data at `path`.
@@ -73,8 +72,11 @@ public annotation class JdbcOptions(
 )
 
 public annotation class JsonOptions(
-    /** Allows the choice of how to handle type clashes when reading a JSON file. */
-    public val typeClashTactic: JSON.TypeClashTactic = JSON.TypeClashTactic.ARRAY_AND_VALUE_COLUMNS,
+    /**
+     * Allows the choice of how to handle type clashes when reading a JSON file.
+     * Must be either [TypeClashTactics.ARRAY_AND_VALUE_COLUMNS] or [TypeClashTactics.ANY_COLUMNS]
+     * */
+    public val typeClashTactic: String = TypeClashTactics.ARRAY_AND_VALUE_COLUMNS,
     /**
      * List of [JsonPath]s where instead of a [ColumnGroup], a [FrameColumn]<[KeyValueProperty]>
      *     will be created.
@@ -85,4 +87,9 @@ public annotation class JsonOptions(
     public val keyValuePaths: Array<String> = [],
     /** Whether to [unify the numbers that are read][UnifyingNumbers]. `true` by default. */
     public val unifyNumbers: Boolean = true,
-)
+) {
+    public object TypeClashTactics {
+        public const val ARRAY_AND_VALUE_COLUMNS: String = "ARRAY_AND_VALUE_COLUMNS"
+        public const val ANY_COLUMNS: String = "ANY_COLUMNS"
+    }
+}

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/api/chunked.kt

@@ -15,10 +15,14 @@ import org.jetbrains.kotlinx.dataframe.type
  * Creates a [FrameColumn] from [this] by splitting the dataframe into
  * smaller ones, with their number of rows at most [size].
  */
-public fun <T> DataFrame<T>.chunked(size: Int, name: String = "groups"): FrameColumn<T> {
-    val startIndices = (0 until nrow step size)
-    return this.chunkedImpl(startIndices, name)
-}
+public fun <T> DataFrame<T>.chunked(size: Int, name: String = "groups"): FrameColumn<T> =
+    chunked(
+        startIndices = 0 until nrow step size,
+        name = name,
+    )
+
+public fun <T> DataFrame<T>.chunked(startIndices: Iterable<Int>, name: String = "groups"): FrameColumn<T> =
+    chunkedImpl(startIndices, name)
 
 public fun <T> DataColumn<T>.chunked(size: Int): ValueColumn<List<T>> {
     val values = toList().chunked(size)

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/codeGen/DefaultReadDfMethods.kt

@@ -23,7 +23,6 @@ private const val CAST = "cast"
 private const val VERIFY = "verify" // cast(true) is obscure, i think it's better to use named argument here
 private const val READ_CSV = "readCSV"
 private const val READ_TSV = "readTSV"
-private const val READ_JSON = "readJson"
 private const val READ_JDBC = "readJdbc"
 
 public abstract class AbstractDefaultReadMethod(
@@ -82,13 +81,6 @@ public abstract class AbstractDefaultReadMethod(
     override val additionalImports: List<String> = listOf("import org.jetbrains.kotlinx.dataframe.io.$methodName")
 }
 
-internal class DefaultReadJsonMethod(path: String?, arguments: MethodArguments) :
-    AbstractDefaultReadMethod(
-        path = path,
-        arguments = arguments,
-        methodName = READ_JSON,
-    )
-
 internal class DefaultReadCsvMethod(path: String?, arguments: MethodArguments) :
     AbstractDefaultReadMethod(path, arguments, READ_CSV)
 

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/columns/ColumnGroup.kt

@@ -9,6 +9,8 @@ import org.jetbrains.kotlinx.dataframe.annotations.HasSchema
 import org.jetbrains.kotlinx.dataframe.api.ColumnsSelectionDsl
 import org.jetbrains.kotlinx.dataframe.api.asColumnGroup
 import org.jetbrains.kotlinx.dataframe.api.columnGroup
+import org.jetbrains.kotlinx.dataframe.impl.schema.extractSchema
+import org.jetbrains.kotlinx.dataframe.schema.DataFrameSchema
 import kotlin.reflect.KProperty
 
 /**

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/documentation/UnifyingNumbers.kt

@@ -24,7 +24,7 @@ import org.jetbrains.kotlinx.dataframe.impl.UnifiedNumberTypeOptions
  *
  * At the bottom of the graph is [Nothing]. This can be interpreted as `null`.
  */
-internal interface UnifyingNumbers {
+public interface UnifyingNumbers {
 
     /**
      * ```
@@ -47,5 +47,6 @@ internal interface UnifyingNumbers {
      *       Nothing?
      * ```
      */
-    interface Graph
+    @ExcludeFromSources
+    private interface Graph
 }

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/ColumnDataCollector.kt

@@ -14,7 +14,7 @@ import kotlin.reflect.full.isSubclassOf
 import kotlin.reflect.full.withNullability
 import kotlin.reflect.jvm.jvmErasure
 
-internal interface DataCollector<T> {
+public interface DataCollector<T> {
 
     public val data: List<T?>
     public val hasNulls: Boolean

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/Utils.kt

@@ -150,6 +150,11 @@ internal fun <T> catchSilent(body: () -> T): T? =
 internal fun Iterable<KClass<*>>.commonType(nullable: Boolean, upperBound: KType? = null) =
     commonParents(this).createType(nullable, upperBound)
 
+// helper overload for friend modules
+@JvmName("commonTypeOverload")
+internal fun commonType(types: Iterable<KType?>, useStar: Boolean = true) =
+    types.commonType(useStar)
+
 /**
  * Returns the common supertype of the given types.
  *
@@ -276,6 +281,10 @@ internal fun <T> DataFrame<T>.splitByIndices(startIndices: Sequence<Int>): Seque
         }
     }
 
+// helper overload for friend modules
+@JvmName("splitByIndicesOverload")
+internal fun <T> splitByIndices(list: List<T>, startIndices: Sequence<Int>) = list.splitByIndices(startIndices)
+
 internal fun <T> List<T>.splitByIndices(startIndices: Sequence<Int>): Sequence<List<T>> =
     (startIndices + size).zipWithNext { start, endExclusive ->
         subList(start, endExclusive)

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/io/image.kt

@@ -58,3 +58,24 @@ internal fun BufferedImage.toByteArray(format: String = DEFAULT_IMG_FORMAT): Byt
         ImageIO.write(this, format, bos)
         bos.toByteArray()
     }
+
+// helper overload for friend modules
+internal fun resizeKeepingAspectRatio(
+    image: BufferedImage,
+    maxSize: Int,
+    resultImageType: Int = BufferedImage.TYPE_INT_ARGB,
+    interpolation: Any = RenderingHints.VALUE_INTERPOLATION_NEAREST_NEIGHBOR,
+    renderingQuality: Any = RenderingHints.VALUE_RENDER_QUALITY,
+    antialiasing: Any = RenderingHints.VALUE_ANTIALIAS_ON,
+    observer: ImageObserver? = null,
+) = image.resizeKeepingAspectRatio(
+    maxSize = maxSize,
+    resultImageType = resultImageType,
+    interpolation = interpolation,
+    renderingQuality = renderingQuality,
+    antialiasing = antialiasing,
+    observer = observer,
+)
+
+// helper overload for friend modules
+internal fun toByteArray(image: BufferedImage, format: String = DEFAULT_IMG_FORMAT) = image.toByteArray(format)