dataframe-arrow/org.jetbrains.kotlinx.dataframe.io/ArrowWriterImpl

ArrowWriterImpl

internal class ArrowWriterImpl(val dataFrame: DataFrame<*>, val targetSchema: Schema, val mode: ArrowWriter.Mode, val mismatchSubscriber: (ConvertingMismatch) -> Unit = ignoreMismatchMessage) : ArrowWriter

Save dataFrame content in Apache Arrow format (can be written to File, ByteArray, OutputStream or raw Channel) with targetSchema. If dataFrame content does not match with targetSchema, behaviour is specified by mode, mismatches would be sent to mismatchSubscriber

Constructors

ArrowWriterImpl

constructor(dataFrame: DataFrame<*>, targetSchema: Schema, mode: ArrowWriter.Mode, mismatchSubscriber: (ConvertingMismatch) -> Unit = ignoreMismatchMessage)

Properties

dataFrame

open override val dataFrame: DataFrame<*>

mismatchSubscriber

open override val mismatchSubscriber: (ConvertingMismatch) -> Unit

mode

open override val mode: ArrowWriter.Mode

targetSchema

open override val targetSchema: Schema

Functions

allocateVectorSchemaRoot

open override fun allocateVectorSchemaRoot(): VectorSchemaRoot

Create Arrow VectorSchemaRoot with dataFrame content cast to targetSchema according to the mode.

open override fun close()

saveArrowFeatherToByteArray

open fun saveArrowFeatherToByteArray(): ByteArray

Save data to Arrow random access format, write to new ByteArray

saveArrowIPCToByteArray

open fun saveArrowIPCToByteArray(): ByteArray

Save data to Arrow interprocess streaming format, write to new ByteArray

writeArrowFeather

open fun writeArrowFeather(file: File)

Save data to Arrow random access format, write to new or existing file. If file exists, it would be recreated.

open fun writeArrowFeather(stream: OutputStream)

Save data to Arrow random access format, write to opened stream.

open fun writeArrowFeather(channel: WritableByteChannel)

Save data to Arrow random access format, write to opened channel.

open fun writeArrowFeather(path: Path)

Save data to Arrow random access format, write to a new or existing file on the given path. If file on the given path exists, it would be recreated.

writeArrowIPC

open fun writeArrowIPC(stream: OutputStream)

Save data to Arrow interprocess streaming format, write to opened stream.

open fun writeArrowIPC(channel: WritableByteChannel)

Save data to Arrow interprocess streaming format, write to opened channel.

open fun writeArrowIPC(file: File, append: Boolean = true)

Save data to Arrow interprocess streaming format, write to new or existing file. If file exists, it can be recreated or expanded.

open fun writeArrowIPC(path: Path, append: Boolean = true)

Save data to Arrow interprocess streaming format, write to a new or existing file on the given path. If file on the given path exists, it can be recreated or expanded.