From b41a39e24038876359aeb7ce2bbbb4de2234e5f3 Mon Sep 17 00:00:00 2001 From: Davies Liu Date: Thu, 6 Nov 2014 00:22:19 -0800 Subject: [SPARK-4186] add binaryFiles and binaryRecords in Python add binaryFiles() and binaryRecords() in Python ``` binaryFiles(self, path, minPartitions=None): :: Developer API :: Read a directory of binary files from HDFS, a local file system (available on all nodes), or any Hadoop-supported file system URI as a byte array. Each file is read as a single record and returned in a key-value pair, where the key is the path of each file, the value is the content of each file. Note: Small files are preferred, large file is also allowable, but may cause bad performance. binaryRecords(self, path, recordLength): Load data from a flat binary file, assuming each record is a set of numbers with the specified numerical format (see ByteBuffer), and the number of bytes per record is constant. :param path: Directory to the input data files :param recordLength: The length at which to split the records ``` Author: Davies Liu Closes #3078 from davies/binary and squashes the following commits: cd0bdbd [Davies Liu] Merge branch 'master' of github.com:apache/spark into binary 3aa349b [Davies Liu] add experimental notes 24e84b6 [Davies Liu] Merge branch 'master' of github.com:apache/spark into binary 5ceaa8a [Davies Liu] Merge branch 'master' of github.com:apache/spark into binary 1900085 [Davies Liu] bugfix bb22442 [Davies Liu] add binaryFiles and binaryRecords in Python --- .../main/scala/org/apache/spark/SparkContext.scala | 4 ++ .../apache/spark/api/java/JavaSparkContext.scala | 12 +++--- .../org/apache/spark/api/python/PythonRDD.scala | 45 ++++++++++++++-------- 3 files changed, 40 insertions(+), 21 deletions(-) (limited to 'core') diff --git a/core/src/main/scala/org/apache/spark/SparkContext.scala b/core/src/main/scala/org/apache/spark/SparkContext.scala index 3cdaa6a9cc..03ea672c81 100644 --- a/core/src/main/scala/org/apache/spark/SparkContext.scala +++ b/core/src/main/scala/org/apache/spark/SparkContext.scala @@ -560,6 +560,8 @@ class SparkContext(config: SparkConf) extends SparkStatusAPI with Logging { /** + * :: Experimental :: + * * Get an RDD for a Hadoop-readable dataset as PortableDataStream for each file * (useful for binary data) * @@ -602,6 +604,8 @@ class SparkContext(config: SparkConf) extends SparkStatusAPI with Logging { } /** + * :: Experimental :: + * * Load data from a flat binary file, assuming the length of each record is constant. * * @param path Directory to the input data files diff --git a/core/src/main/scala/org/apache/spark/api/java/JavaSparkContext.scala b/core/src/main/scala/org/apache/spark/api/java/JavaSparkContext.scala index e3aeba7e6c..5c6e8d32c5 100644 --- a/core/src/main/scala/org/apache/spark/api/java/JavaSparkContext.scala +++ b/core/src/main/scala/org/apache/spark/api/java/JavaSparkContext.scala @@ -21,11 +21,6 @@ import java.io.Closeable import java.util import java.util.{Map => JMap} -import java.io.DataInputStream - -import org.apache.hadoop.io.{BytesWritable, LongWritable} -import org.apache.spark.input.{PortableDataStream, FixedLengthBinaryInputFormat} - import scala.collection.JavaConversions import scala.collection.JavaConversions._ import scala.language.implicitConversions @@ -33,6 +28,7 @@ import scala.reflect.ClassTag import com.google.common.base.Optional import org.apache.hadoop.conf.Configuration +import org.apache.spark.input.PortableDataStream import org.apache.hadoop.mapred.{InputFormat, JobConf} import org.apache.hadoop.mapreduce.{InputFormat => NewInputFormat} @@ -286,6 +282,8 @@ class JavaSparkContext(val sc: SparkContext) new JavaPairRDD(sc.binaryFiles(path, minPartitions)) /** + * :: Experimental :: + * * Read a directory of binary files from HDFS, a local file system (available on all nodes), * or any Hadoop-supported file system URI as a byte array. Each file is read as a single * record and returned in a key-value pair, where the key is the path of each file, @@ -312,15 +310,19 @@ class JavaSparkContext(val sc: SparkContext) * * @note Small files are preferred; very large files but may cause bad performance. */ + @Experimental def binaryFiles(path: String): JavaPairRDD[String, PortableDataStream] = new JavaPairRDD(sc.binaryFiles(path, defaultMinPartitions)) /** + * :: Experimental :: + * * Load data from a flat binary file, assuming the length of each record is constant. * * @param path Directory to the input data files * @return An RDD of data with values, represented as byte arrays */ + @Experimental def binaryRecords(path: String, recordLength: Int): JavaRDD[Array[Byte]] = { new JavaRDD(sc.binaryRecords(path, recordLength)) } diff --git a/core/src/main/scala/org/apache/spark/api/python/PythonRDD.scala b/core/src/main/scala/org/apache/spark/api/python/PythonRDD.scala index e94ccdcd47..45beb8fc8c 100644 --- a/core/src/main/scala/org/apache/spark/api/python/PythonRDD.scala +++ b/core/src/main/scala/org/apache/spark/api/python/PythonRDD.scala @@ -21,6 +21,8 @@ import java.io._ import java.net._ import java.util.{List => JList, ArrayList => JArrayList, Map => JMap, Collections} +import org.apache.spark.input.PortableDataStream + import scala.collection.JavaConversions._ import scala.collection.mutable import scala.language.existentials @@ -395,22 +397,33 @@ private[spark] object PythonRDD extends Logging { newIter.asInstanceOf[Iterator[String]].foreach { str => writeUTF(str, dataOut) } - case pair: Tuple2[_, _] => - pair._1 match { - case bytePair: Array[Byte] => - newIter.asInstanceOf[Iterator[Tuple2[Array[Byte], Array[Byte]]]].foreach { pair => - dataOut.writeInt(pair._1.length) - dataOut.write(pair._1) - dataOut.writeInt(pair._2.length) - dataOut.write(pair._2) - } - case stringPair: String => - newIter.asInstanceOf[Iterator[Tuple2[String, String]]].foreach { pair => - writeUTF(pair._1, dataOut) - writeUTF(pair._2, dataOut) - } - case other => - throw new SparkException("Unexpected Tuple2 element type " + pair._1.getClass) + case stream: PortableDataStream => + newIter.asInstanceOf[Iterator[PortableDataStream]].foreach { stream => + val bytes = stream.toArray() + dataOut.writeInt(bytes.length) + dataOut.write(bytes) + } + case (key: String, stream: PortableDataStream) => + newIter.asInstanceOf[Iterator[(String, PortableDataStream)]].foreach { + case (key, stream) => + writeUTF(key, dataOut) + val bytes = stream.toArray() + dataOut.writeInt(bytes.length) + dataOut.write(bytes) + } + case (key: String, value: String) => + newIter.asInstanceOf[Iterator[(String, String)]].foreach { + case (key, value) => + writeUTF(key, dataOut) + writeUTF(value, dataOut) + } + case (key: Array[Byte], value: Array[Byte]) => + newIter.asInstanceOf[Iterator[(Array[Byte], Array[Byte])]].foreach { + case (key, value) => + dataOut.writeInt(key.length) + dataOut.write(key) + dataOut.writeInt(value.length) + dataOut.write(value) } case other => throw new SparkException("Unexpected element type " + first.getClass) -- cgit v1.2.3