本文整理汇总了Scala中org.apache.parquet.hadoop.ParquetFileReader类的典型用法代码示例。如果您正苦于以下问题：Scala ParquetFileReader类的具体用法？Scala ParquetFileReader怎么用？Scala ParquetFileReader使用的例子？那么恭喜您, 这里精选的类代码示例或许可以为您提供帮助。
在下文中一共展示了ParquetFileReader类的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Scala代码示例。
示例1: ParquetUtils
//设置package包名称以及导入依赖的类
package com.scalagen.util
import scala.collection.JavaConverters._
import com.scalagen.data._
import com.scalagen.data.api.Source
import org.apache.hadoop.conf.Configuration
import org.apache.hadoop.fs._
import org.apache.parquet.hadoop.ParquetFileReader
import org.apache.parquet.schema.OriginalType._
import org.apache.parquet.schema.PrimitiveType.PrimitiveTypeName._
import org.apache.parquet.schema._
import org.slf4j.{Logger, LoggerFactory}
object ParquetUtils {
private val logger: Logger = LoggerFactory.getLogger(getClass)
private[scalagen] def makeSchema(s: String, sources: Seq[Source[_, _]], headers: Seq[String]): MessageType = {
logger.debug(s"Making schema for ${sources.mkString(", ")}")
val sourceTypes: Seq[Type] = sources.zip(headers).map {
case (s: Source[_, _], n: String) => sourceToParquetType(s, n)
case _ => throw new IllegalArgumentException("Bad input for parquet source types.")
}
new MessageType(s, sourceTypes: _*)
}
private[scalagen] def sourceToParquetType(s: Source[_, _], columnName: String): Type = {
s match {
case _: GaussianSource | _: RandomDouble => Types.required(DOUBLE).named(columnName)
case _: IncrementingSource | _: DeincrementingSource | _: RandomInt => Types.required(INT32).named(columnName)
case _: DateSource => Types.required(BINARY).as(UTF8).named(columnName)
case _: BernoulliSource => Types.required(BOOLEAN).named(columnName)
case _ => Types.required(BINARY).as(UTF8).named(columnName)
}
}
def parquetRowCount(s: String): Long = {
parquetRowCount(new Path(s))
}
def parquetRowCount(p: Path, conf: Configuration = new Configuration()): Long = {
val fs: FileSystem = p.getFileSystem(conf)
val status: FileStatus = fs.getFileStatus(p)
ParquetFileReader.readFooters(conf, status, false).asScala.head.getParquetMetadata.getBlocks.asScala.map(_.getRowCount).sum
}
}