org.apache.hadoop.mapred
Class FileInputFormat<K,V>

java.lang.Object
  extended by org.apache.hadoop.mapred.FileInputFormat<K,V>
All Implemented Interfaces:
InputFormat<K,V>
Direct Known Subclasses:
AutoInputFormat, CombineFileInputFormat, KeyValueTextInputFormat, LineDocInputFormat, MultiFileInputFormat, NLineInputFormat, SequenceFileInputFormat, TeraInputFormat, TextInputFormat

public abstract class FileInputFormat<K,V>
extends Object
implements InputFormat<K,V>

A base class for file-based InputFormat.

FileInputFormat is the base class for all file-based InputFormats. This provides a generic implementation of getSplits(JobConf, int). Subclasses of FileInputFormat can also override the isSplitable(FileSystem, Path) method to ensure input-files are not split-up and are processed as a whole by Mappers.


Nested Class Summary
static class FileInputFormat.Counter
           
 
Field Summary
static org.apache.commons.logging.Log LOG
           
 
Constructor Summary
FileInputFormat()
           
 
Method Summary
static void addInputPath(JobConf conf, Path path)
          Add a Path to the list of inputs for the map-reduce job.
static void addInputPaths(JobConf conf, String commaSeparatedPaths)
          Add the given comma separated paths to the list of inputs for the map-reduce job.
protected  long computeSplitSize(long goalSize, long minSize, long blockSize)
           
protected  int getBlockIndex(BlockLocation[] blkLocations, long offset)
           
static PathFilter getInputPathFilter(JobConf conf)
          Get a PathFilter instance of the filter set for the input paths.
static Path[] getInputPaths(JobConf conf)
          Get the list of input Paths for the map-reduce job.
abstract  RecordReader<K,V> getRecordReader(InputSplit split, JobConf job, Reporter reporter)
          Get the RecordReader for the given InputSplit.
protected  String[] getSplitHosts(BlockLocation[] blkLocations, long offset, long splitSize, NetworkTopology clusterMap)
          This function identifies and returns the hosts that contribute most for a given split.
 InputSplit[] getSplits(JobConf job, int numSplits)
          Splits files returned by listStatus(JobConf) when they're too big.
protected  boolean isSplitable(FileSystem fs, Path filename)
          Is the given filename splitable? Usually, true, but if the file is stream compressed, it will not be.
protected  FileStatus[] listStatus(JobConf job)
          List input directories.
static void setInputPathFilter(JobConf conf, Class<? extends PathFilter> filter)
          Set a PathFilter to be applied to the input paths for the map-reduce job.
static void setInputPaths(JobConf conf, Path... inputPaths)
          Set the array of Paths as the list of inputs for the map-reduce job.
static void setInputPaths(JobConf conf, String commaSeparatedPaths)
          Sets the given comma separated paths as the list of inputs for the map-reduce job.
protected  void setMinSplitSize(long minSplitSize)
           
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

LOG

public static final org.apache.commons.logging.Log LOG
Constructor Detail

FileInputFormat

public FileInputFormat()
Method Detail

setMinSplitSize

protected void setMinSplitSize(long minSplitSize)

isSplitable

protected boolean isSplitable(FileSystem fs,
                              Path filename)
Is the given filename splitable? Usually, true, but if the file is stream compressed, it will not be. FileInputFormat implementations can override this and return false to ensure that individual input files are never split-up so that Mappers process entire files.

Parameters:
fs - the file system that the file is on
filename - the file name to check
Returns:
is this file splitable?

getRecordReader

public abstract RecordReader<K,V> getRecordReader(InputSplit split,
                                                  JobConf job,
                                                  Reporter reporter)
                                           throws IOException
Description copied from interface: InputFormat
Get the RecordReader for the given InputSplit.

It is the responsibility of the RecordReader to respect record boundaries while processing the logical split to present a record-oriented view to the individual task.

Specified by:
getRecordReader in interface InputFormat<K,V>
Parameters:
split - the InputSplit
job - the job that this split belongs to
Returns:
a RecordReader
Throws:
IOException

setInputPathFilter

public static void setInputPathFilter(JobConf conf,
                                      Class<? extends PathFilter> filter)
Set a PathFilter to be applied to the input paths for the map-reduce job.

Parameters:
filter - the PathFilter class use for filtering the input paths.

getInputPathFilter

public static PathFilter getInputPathFilter(JobConf conf)
Get a PathFilter instance of the filter set for the input paths.

Returns:
the PathFilter instance set for the job, NULL if none has been set.

listStatus

protected FileStatus[] listStatus(JobConf job)
                           throws IOException
List input directories. Subclasses may override to, e.g., select only files matching a regular expression.

Parameters:
job - the job to list input paths for
Returns:
array of FileStatus objects
Throws:
IOException - if zero items.

getSplits

public InputSplit[] getSplits(JobConf job,
                              int numSplits)
                       throws IOException
Splits files returned by listStatus(JobConf) when they're too big.

Specified by:
getSplits in interface InputFormat<K,V>
Parameters:
job - job configuration.
numSplits - the desired number of splits, a hint.
Returns:
an array of InputSplits for the job.
Throws:
IOException

computeSplitSize

protected long computeSplitSize(long goalSize,
                                long minSize,
                                long blockSize)

getBlockIndex

protected int getBlockIndex(BlockLocation[] blkLocations,
                            long offset)

setInputPaths

public static void setInputPaths(JobConf conf,
                                 String commaSeparatedPaths)
Sets the given comma separated paths as the list of inputs for the map-reduce job.

Parameters:
conf - Configuration of the job
commaSeparatedPaths - Comma separated paths to be set as the list of inputs for the map-reduce job.

addInputPaths

public static void addInputPaths(JobConf conf,
                                 String commaSeparatedPaths)
Add the given comma separated paths to the list of inputs for the map-reduce job.

Parameters:
conf - The configuration of the job
commaSeparatedPaths - Comma separated paths to be added to the list of inputs for the map-reduce job.

setInputPaths

public static void setInputPaths(JobConf conf,
                                 Path... inputPaths)
Set the array of Paths as the list of inputs for the map-reduce job.

Parameters:
conf - Configuration of the job.
inputPaths - the Paths of the input directories/files for the map-reduce job.

addInputPath

public static void addInputPath(JobConf conf,
                                Path path)
Add a Path to the list of inputs for the map-reduce job.

Parameters:
conf - The configuration of the job
path - Path to be added to the list of inputs for the map-reduce job.

getInputPaths

public static Path[] getInputPaths(JobConf conf)
Get the list of input Paths for the map-reduce job.

Parameters:
conf - The configuration of the job
Returns:
the list of input Paths for the map-reduce job.

getSplitHosts

protected String[] getSplitHosts(BlockLocation[] blkLocations,
                                 long offset,
                                 long splitSize,
                                 NetworkTopology clusterMap)
                          throws IOException
This function identifies and returns the hosts that contribute most for a given split. For calculating the contribution, rack locality is treated on par with host locality, so hosts from racks that contribute the most are preferred over hosts on racks that contribute less

Parameters:
blkLocations - The list of block locations
offset -
splitSize -
Returns:
array of hosts that contribute most to this split
Throws:
IOException


Copyright © 2009 The Apache Software Foundation