org.apache.accumulo.core.file
Class FileUtil

java.lang.Object
  org.apache.accumulo.core.file.FileUtil

public class FileUtil
extends java.lang.Object

Nested Class Summary

static class

FileUtil.FileInfo

Constructor Summary

FileUtil()

Method Summary

static double	estimatePercentageLTE(org.apache.hadoop.io.Text prevEndRow, org.apache.hadoop.io.Text endRow, java.util.Collection<java.lang.String> mapFiles, org.apache.hadoop.io.Text splitRow)
static java.util.Map<KeyExtent,java.lang.Long>	estimateSizes(org.apache.hadoop.fs.Path mapFile, long fileSize, java.util.List<KeyExtent> extents, org.apache.hadoop.conf.Configuration conf, org.apache.hadoop.fs.FileSystem fs)
static org.apache.hadoop.io.WritableComparable<Key>	findLastKey(java.util.Collection<java.lang.String> mapFiles)
static java.util.SortedMap<java.lang.Double,Key>	findMidPoint(org.apache.hadoop.io.Text prevEndRow, org.apache.hadoop.io.Text endRow, java.util.Collection<java.lang.String> mapFiles, double minSplit)
static java.util.SortedMap<java.lang.Double,Key>	findMidPoint(org.apache.hadoop.io.Text prevEndRow, org.apache.hadoop.io.Text endRow, java.util.Collection<java.lang.String> mapFiles, double minSplit, boolean useIndex)
static java.util.Collection<java.lang.String>	reduceFiles(org.apache.hadoop.conf.Configuration conf, org.apache.hadoop.fs.FileSystem fs, org.apache.hadoop.io.Text prevEndRow, org.apache.hadoop.io.Text endRow, java.util.Collection<java.lang.String> mapFiles, int maxFiles, java.lang.String tmpDir, int pass)
static java.util.Map<java.lang.String,FileUtil.FileInfo>	tryToGetFirstAndLastRows(java.util.Set<java.lang.String> mapfiles)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

FileUtil

public FileUtil()

Method Detail

reduceFiles

public static java.util.Collection<java.lang.String> reduceFiles(org.apache.hadoop.conf.Configuration conf,
                                                                 org.apache.hadoop.fs.FileSystem fs,
                                                                 org.apache.hadoop.io.Text prevEndRow,
                                                                 org.apache.hadoop.io.Text endRow,
                                                                 java.util.Collection<java.lang.String> mapFiles,
                                                                 int maxFiles,
                                                                 java.lang.String tmpDir,
                                                                 int pass)
                                                          throws java.io.IOException

Throws:

java.io.IOException

findMidPoint

public static java.util.SortedMap<java.lang.Double,Key> findMidPoint(org.apache.hadoop.io.Text prevEndRow,
                                                                     org.apache.hadoop.io.Text endRow,
                                                                     java.util.Collection<java.lang.String> mapFiles,
                                                                     double minSplit)
                                                              throws java.io.IOException

Throws:

java.io.IOException

estimatePercentageLTE

public static double estimatePercentageLTE(org.apache.hadoop.io.Text prevEndRow,
                                           org.apache.hadoop.io.Text endRow,
                                           java.util.Collection<java.lang.String> mapFiles,
                                           org.apache.hadoop.io.Text splitRow)
                                    throws java.io.IOException

Throws:

java.io.IOException

findMidPoint

public static java.util.SortedMap<java.lang.Double,Key> findMidPoint(org.apache.hadoop.io.Text prevEndRow,
                                                                     org.apache.hadoop.io.Text endRow,
                                                                     java.util.Collection<java.lang.String> mapFiles,
                                                                     double minSplit,
                                                                     boolean useIndex)
                                                              throws java.io.IOException

Parameters:

mapFiles - - list MapFiles to find the mid point key ISSUES : This method used the index files to find the mid point. If the map files have different index intervals this method will not return an accurate mid point. Also, it would be tricky to use this method in conjunction with an in memory map because the indexing interval is unknown.

Throws:

java.io.IOException

tryToGetFirstAndLastRows

public static java.util.Map<java.lang.String,FileUtil.FileInfo> tryToGetFirstAndLastRows(java.util.Set<java.lang.String> mapfiles)

findLastKey

public static org.apache.hadoop.io.WritableComparable<Key> findLastKey(java.util.Collection<java.lang.String> mapFiles)
                                                                throws java.io.IOException

Throws:

java.io.IOException

estimateSizes

public static java.util.Map<KeyExtent,java.lang.Long> estimateSizes(org.apache.hadoop.fs.Path mapFile,
                                                                    long fileSize,
                                                                    java.util.List<KeyExtent> extents,
                                                                    org.apache.hadoop.conf.Configuration conf,
                                                                    org.apache.hadoop.fs.FileSystem fs)
                                                             throws java.io.IOException

Throws:

java.io.IOException