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

ARROW_PARQUET_DEFAULT_BATCH_SIZE

const val ARROW_PARQUET_DEFAULT_BATCH_SIZE: Long = 32768

Number of rows readParquet pulls from Arrow Dataset per scan batch, when no batchSize is given.

Public so that the value can be referenced instead of copied — the documentation samples used to hard-code their own copy of it, which silently drifted out of date.