Java Code Examples for org.apache.flink.core.fs.FileInputSplit#getLength()

private DataFileReader<E> initReader(FileInputSplit split) throws IOException {
	DatumReader<E> datumReader;

	if (org.apache.avro.generic.GenericRecord.class == avroValueType) {
		datumReader = new GenericDatumReader<E>();
	} else {
		datumReader = org.apache.avro.specific.SpecificRecordBase.class.isAssignableFrom(avroValueType)
			? new SpecificDatumReader<E>(avroValueType) : new ReflectDatumReader<E>(avroValueType);
	}
	if (LOG.isInfoEnabled()) {
		LOG.info("Opening split {}", split);
	}

	SeekableInput in = new FSDataInputStreamWrapper(stream, split.getPath().getFileSystem().getFileStatus(split.getPath()).getLen());
	DataFileReader<E> dataFileReader = (DataFileReader) DataFileReader.openReader(in, datumReader);

	if (LOG.isDebugEnabled()) {
		LOG.debug("Loaded SCHEMA: {}", dataFileReader.getSchema());
	}

	end = split.getStart() + split.getLength();
	recordsReadSinceLastSync = 0;
	return dataFileReader;
}
 

private Tuple2<Long, Long> getOffsetAndLengthForSplit(FileInputSplit split, List<StripeInformation> stripes) {
	long splitStart = split.getStart();
	long splitEnd = splitStart + split.getLength();

	long readStart = Long.MAX_VALUE;
	long readEnd = Long.MIN_VALUE;

	for (StripeInformation s : stripes) {
		if (splitStart <= s.getOffset() && s.getOffset() < splitEnd) {
			// stripe starts in split, so it is included
			readStart = Math.min(readStart, s.getOffset());
			readEnd = Math.max(readEnd, s.getOffset() + s.getLength());
		}
	}

	if (readStart < Long.MAX_VALUE) {
		// at least one split is included
		return Tuple2.of(readStart, readEnd - readStart);
	} else {
		return Tuple2.of(0L, 0L);
	}
}
 

private DataFileReader<E> initReader(FileInputSplit split) throws IOException {
	DatumReader<E> datumReader;

	if (org.apache.avro.generic.GenericRecord.class == avroValueType) {
		datumReader = new GenericDatumReader<E>();
	} else {
		datumReader = org.apache.avro.specific.SpecificRecordBase.class.isAssignableFrom(avroValueType)
			? new SpecificDatumReader<E>(avroValueType) : new ReflectDatumReader<E>(avroValueType);
	}
	if (LOG.isInfoEnabled()) {
		LOG.info("Opening split {}", split);
	}

	SeekableInput in = new FSDataInputStreamWrapper(stream, split.getPath().getFileSystem().getFileStatus(split.getPath()).getLen());
	DataFileReader<E> dataFileReader = (DataFileReader) DataFileReader.openReader(in, datumReader);

	if (LOG.isDebugEnabled()) {
		LOG.debug("Loaded SCHEMA: {}", dataFileReader.getSchema());
	}

	end = split.getStart() + split.getLength();
	recordsReadSinceLastSync = 0;
	return dataFileReader;
}
 

private Tuple2<Long, Long> getOffsetAndLengthForSplit(FileInputSplit split, List<StripeInformation> stripes) {
	long splitStart = split.getStart();
	long splitEnd = splitStart + split.getLength();

	long readStart = Long.MAX_VALUE;
	long readEnd = Long.MIN_VALUE;

	for (StripeInformation s : stripes) {
		if (splitStart <= s.getOffset() && s.getOffset() < splitEnd) {
			// stripe starts in split, so it is included
			readStart = Math.min(readStart, s.getOffset());
			readEnd = Math.max(readEnd, s.getOffset() + s.getLength());
		}
	}

	if (readStart < Long.MAX_VALUE) {
		// at least one split is included
		return Tuple2.of(readStart, readEnd - readStart);
	} else {
		return Tuple2.of(0L, 0L);
	}
}
 

/**
 * Opens an input stream to the file defined in the input format.
 * The stream is positioned at the beginning of the given split.
 * <p>
 * The stream is actually opened in an asynchronous thread to make sure any interruptions to the thread 
 * working on the input format do not reach the file system.
 */
@Override
public void open(FileInputSplit fileSplit) throws IOException {

	this.currentSplit = fileSplit;
	this.splitStart = fileSplit.getStart();
	this.splitLength = fileSplit.getLength();

	if (LOG.isDebugEnabled()) {
		LOG.debug("Opening input split " + fileSplit.getPath() + " [" + this.splitStart + "," + this.splitLength + "]");
	}

	
	// open the split in an asynchronous thread
	final InputSplitOpenThread isot = new InputSplitOpenThread(fileSplit, this.openTimeout);
	isot.start();
	
	try {
		this.stream = isot.waitForCompletion();
		this.stream = decorateInputStream(this.stream, fileSplit);
	}
	catch (Throwable t) {
		throw new IOException("Error opening the Input Split " + fileSplit.getPath() + 
				" [" + splitStart + "," + splitLength + "]: " + t.getMessage(), t);
	}
	
	// get FSDataInputStream
	if (this.splitStart != 0) {
		this.stream.seek(this.splitStart);
	}
}
 

@Override
public void open(FileInputSplit fileSplit) throws IOException {
	LOG.debug("Opening ORC file {}", fileSplit.getPath());
	this.reader = new OrcRowSplitReader(
			conf,
			schema,
			selectedFields,
			conjunctPredicates,
			batchSize,
			fileSplit.getPath(),
			fileSplit.getStart(),
			fileSplit.getLength());
}
 

@Test
public void testReadRecordsLargerThanBuffer() throws IOException {
	final String myString = "aaaaaaaaaaaaaaaaaaaaa\n" +
							"bbbbbbbbbbbbbbbbbbbbbbbbb\n" +
							"ccccccccccccccccccc\n" +
							"ddddddddddddddddddddddddddddddddddd\n";

	final FileInputSplit split = createTempFile(myString);
	FileInputSplit split1 = new FileInputSplit(0, split.getPath(), 0, split.getLength() / 2, split.getHostnames());
	FileInputSplit split2 = new FileInputSplit(1, split.getPath(), split1.getLength(), split.getLength(), split.getHostnames());

	final Configuration parameters = new Configuration();

	format.setBufferSize(8);
	format.configure(parameters);

	String next;
	List<String> result = new ArrayList<String>();


	format.open(split1);
	while ((next = format.nextRecord(null)) != null) {
		result.add(next);
	}
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());
	format.close();

	format.open(split2);
	while ((next = format.nextRecord(null)) != null) {
		result.add(next);
	}
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());
	format.close();

	assertEquals(4, result.size());
	assertEquals(Arrays.asList(myString.split("\n")), result);
}
 

@Override
      public void open ( FileInputSplit split ) throws IOException {
          Path path = new Path(split.getPath().toString());
   if (Plan.conf == null)
Plan.conf = new Configuration();
          FileSystem fs = path.getFileSystem(Plan.conf);
          in = new SequenceFile.Reader(fs,path,Plan.conf);
          end = split.getStart()+split.getLength();
          if (split.getStart() > in.getPosition())
              in.sync(split.getStart());       // sync to start
          start = in.getPosition();
          more = start < end;
      }
 

private TimestampedFileInputSplit getTimestampedSplit(long modTime, FileInputSplit split) {
	Preconditions.checkNotNull(split);
	return new TimestampedFileInputSplit(
		modTime,
		split.getSplitNumber(),
		split.getPath(),
		split.getStart(),
		split.getLength(),
		split.getHostnames());
}
 

@Test
public void testReadRecordsLargerThanBuffer() throws IOException {
	final String myString = "aaaaaaaaaaaaaaaaaaaaa\n" +
							"bbbbbbbbbbbbbbbbbbbbbbbbb\n" +
							"ccccccccccccccccccc\n" +
							"ddddddddddddddddddddddddddddddddddd\n";

	final FileInputSplit split = createTempFile(myString);
	FileInputSplit split1 = new FileInputSplit(0, split.getPath(), 0, split.getLength() / 2, split.getHostnames());
	FileInputSplit split2 = new FileInputSplit(1, split.getPath(), split1.getLength(), split.getLength(), split.getHostnames());

	final Configuration parameters = new Configuration();

	format.setBufferSize(8);
	format.configure(parameters);

	String next;
	List<String> result = new ArrayList<String>();


	format.open(split1);
	while ((next = format.nextRecord(null)) != null) {
		result.add(next);
	}
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());
	format.close();

	format.open(split2);
	while ((next = format.nextRecord(null)) != null) {
		result.add(next);
	}
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());
	format.close();

	assertEquals(4, result.size());
	assertEquals(Arrays.asList(myString.split("\n")), result);
}
 

/**
 * Tests that the records are read correctly when the split boundary is in the middle of a record.
 */
@Test
public void testReadOverSplitBoundariesUnaligned() throws IOException {
	final String myString = "value1\nvalue2\nvalue3";
	final FileInputSplit split = createTempFile(myString);

	FileInputSplit split1 = new FileInputSplit(0, split.getPath(), 0, split.getLength() / 2, split.getHostnames());
	FileInputSplit split2 = new FileInputSplit(1, split.getPath(), split1.getLength(), split.getLength(), split.getHostnames());

	final Configuration parameters = new Configuration();

	format.configure(parameters);
	format.open(split1);

	assertEquals("value1", format.nextRecord(null));
	assertEquals("value2", format.nextRecord(null));
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());

	format.close();
	format.open(split2);

	assertEquals("value3", format.nextRecord(null));
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());

	format.close();
}
 

@PublicEvolving
@Override
public void reopen(FileInputSplit split, Long state) throws IOException {
	Preconditions.checkNotNull(split, "reopen() cannot be called on a null split.");
	Preconditions.checkNotNull(state, "reopen() cannot be called with a null initial state.");
	Preconditions.checkArgument(state == -1 || state >= split.getStart(),
		" Illegal offset "+ state +", smaller than the splits start=" + split.getStart());

	try {
		this.open(split);
	} finally {
		this.offset = state;
	}

	if (state > this.splitStart + split.getLength()) {
		this.end = true;
	} else if (state > split.getStart()) {
		initBuffers();

		this.stream.seek(this.offset);
		if (split.getLength() == -1) {
			// this is the case for unsplittable files
			fillBuffer(0);
		} else {
			this.splitLength = this.splitStart + split.getLength() - this.offset;
			if (splitLength <= 0) {
				this.end = true;
			}
		}
	}
}
 

/**
 * Opens an input stream to the file defined in the input format.
 * The stream is positioned at the beginning of the given split.
 * <p>
 * The stream is actually opened in an asynchronous thread to make sure any interruptions to the thread 
 * working on the input format do not reach the file system.
 */
@Override
public void open(FileInputSplit fileSplit) throws IOException {

	this.currentSplit = fileSplit;
	this.splitStart = fileSplit.getStart();
	this.splitLength = fileSplit.getLength();

	if (LOG.isDebugEnabled()) {
		LOG.debug("Opening input split " + fileSplit.getPath() + " [" + this.splitStart + "," + this.splitLength + "]");
	}

	
	// open the split in an asynchronous thread
	final InputSplitOpenThread isot = new InputSplitOpenThread(fileSplit, this.openTimeout);
	isot.start();
	
	try {
		this.stream = isot.waitForCompletion();
		this.stream = decorateInputStream(this.stream, fileSplit);
	}
	catch (Throwable t) {
		throw new IOException("Error opening the Input Split " + fileSplit.getPath() + 
				" [" + splitStart + "," + splitLength + "]: " + t.getMessage(), t);
	}
	
	// get FSDataInputStream
	if (this.splitStart != 0) {
		this.stream.seek(this.splitStart);
	}
}
 

@Test
public void testReadRecordsLargerThanBuffer() throws IOException {
	final String myString = "aaaaaaaaaaaaaaaaaaaaa\n" +
							"bbbbbbbbbbbbbbbbbbbbbbbbb\n" +
							"ccccccccccccccccccc\n" +
							"ddddddddddddddddddddddddddddddddddd\n";

	final FileInputSplit split = createTempFile(myString);
	FileInputSplit split1 = new FileInputSplit(0, split.getPath(), 0, split.getLength() / 2, split.getHostnames());
	FileInputSplit split2 = new FileInputSplit(1, split.getPath(), split1.getLength(), split.getLength(), split.getHostnames());

	final Configuration parameters = new Configuration();

	format.setBufferSize(8);
	format.configure(parameters);

	String next;
	List<String> result = new ArrayList<String>();


	format.open(split1);
	while ((next = format.nextRecord(null)) != null) {
		result.add(next);
	}
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());
	format.close();

	format.open(split2);
	while ((next = format.nextRecord(null)) != null) {
		result.add(next);
	}
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());
	format.close();

	assertEquals(4, result.size());
	assertEquals(Arrays.asList(myString.split("\n")), result);
}
 

/**
 * Tests that the records are read correctly when the split boundary is in the middle of a record.
 */
@Test
public void testReadOverSplitBoundariesUnaligned() throws IOException {
	final String myString = "value1\nvalue2\nvalue3";
	final FileInputSplit split = createTempFile(myString);

	FileInputSplit split1 = new FileInputSplit(0, split.getPath(), 0, split.getLength() / 2, split.getHostnames());
	FileInputSplit split2 = new FileInputSplit(1, split.getPath(), split1.getLength(), split.getLength(), split.getHostnames());

	final Configuration parameters = new Configuration();

	format.configure(parameters);
	format.open(split1);

	assertEquals("value1", format.nextRecord(null));
	assertEquals("value2", format.nextRecord(null));
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());

	format.close();
	format.open(split2);

	assertEquals("value3", format.nextRecord(null));
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());

	format.close();
}
 

@PublicEvolving
@Override
public void reopen(FileInputSplit split, Long state) throws IOException {
	Preconditions.checkNotNull(split, "reopen() cannot be called on a null split.");
	Preconditions.checkNotNull(state, "reopen() cannot be called with a null initial state.");
	Preconditions.checkArgument(state == -1 || state >= split.getStart(),
		" Illegal offset "+ state +", smaller than the splits start=" + split.getStart());

	try {
		this.open(split);
	} finally {
		this.offset = state;
	}

	if (state > this.splitStart + split.getLength()) {
		this.end = true;
	} else if (state > split.getStart()) {
		initBuffers();

		this.stream.seek(this.offset);
		if (split.getLength() == -1) {
			// this is the case for unsplittable files
			fillBuffer(0);
		} else {
			this.splitLength = this.splitStart + split.getLength() - this.offset;
			if (splitLength <= 0) {
				this.end = true;
			}
		}
	}
}
 

/**
 * Opens an input stream to the file defined in the input format.
 * The stream is positioned at the beginning of the given split.
 * <p>
 * The stream is actually opened in an asynchronous thread to make sure any interruptions to the thread 
 * working on the input format do not reach the file system.
 */
@Override
public void open(FileInputSplit fileSplit) throws IOException {

	this.currentSplit = fileSplit;
	this.splitStart = fileSplit.getStart();
	this.splitLength = fileSplit.getLength();

	if (LOG.isDebugEnabled()) {
		LOG.debug("Opening input split " + fileSplit.getPath() + " [" + this.splitStart + "," + this.splitLength + "]");
	}

	
	// open the split in an asynchronous thread
	final InputSplitOpenThread isot = new InputSplitOpenThread(fileSplit, this.openTimeout);
	isot.start();
	
	try {
		this.stream = isot.waitForCompletion();
		this.stream = decorateInputStream(this.stream, fileSplit);
	}
	catch (Throwable t) {
		throw new IOException("Error opening the Input Split " + fileSplit.getPath() + 
				" [" + splitStart + "," + splitLength + "]: " + t.getMessage(), t);
	}
	
	// get FSDataInputStream
	if (this.splitStart != 0) {
		this.stream.seek(this.splitStart);
	}
}
 

private TimestampedFileInputSplit getTimestampedSplit(long modTime, FileInputSplit split) {
	Preconditions.checkNotNull(split);
	return new TimestampedFileInputSplit(
		modTime,
		split.getSplitNumber(),
		split.getPath(),
		split.getStart(),
		split.getLength(),
		split.getHostnames());
}
 

/**
 * Tests that the correct number of records is read when the split boundary is exact at the record boundary.
 */
@Test
public void testReadWithBufferSizeIsMultiple() throws IOException {
	final String myString = "aaaaaaa\nbbbbbbb\nccccccc\nddddddd\n";
	final FileInputSplit split = createTempFile(myString);

	FileInputSplit split1 = new FileInputSplit(0, split.getPath(), 0, split.getLength() / 2, split.getHostnames());
	FileInputSplit split2 = new FileInputSplit(1, split.getPath(), split1.getLength(), split.getLength(), split.getHostnames());

	final Configuration parameters = new Configuration();

	format.setBufferSize(2 * ((int) split1.getLength()));
	format.configure(parameters);

	String next;
	int count = 0;

	// read split 1
	format.open(split1);
	while ((next = format.nextRecord(null)) != null) {
		assertEquals(7, next.length());
		count++;
	}
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());
	format.close();

	// this one must have read one too many, because the next split will skipp the trailing remainder
	// which happens to be one full record
	assertEquals(3, count);

	// read split 2
	format.open(split2);
	while ((next = format.nextRecord(null)) != null) {
		assertEquals(7, next.length());
		count++;
	}
	format.close();

	assertEquals(4, count);
}
 

/**
 * Tests that the correct number of records is read when the split boundary is exact at the record boundary.
 */
@Test
public void testReadWithBufferSizeIsMultiple() throws IOException {
	final String myString = "aaaaaaa\nbbbbbbb\nccccccc\nddddddd\n";
	final FileInputSplit split = createTempFile(myString);

	FileInputSplit split1 = new FileInputSplit(0, split.getPath(), 0, split.getLength() / 2, split.getHostnames());
	FileInputSplit split2 = new FileInputSplit(1, split.getPath(), split1.getLength(), split.getLength(), split.getHostnames());

	final Configuration parameters = new Configuration();

	format.setBufferSize(2 * ((int) split1.getLength()));
	format.configure(parameters);

	String next;
	int count = 0;

	// read split 1
	format.open(split1);
	while ((next = format.nextRecord(null)) != null) {
		assertEquals(7, next.length());
		count++;
	}
	assertNull(format.nextRecord(null));
	assertTrue(format.reachedEnd());
	format.close();

	// this one must have read one too many, because the next split will skipp the trailing remainder
	// which happens to be one full record
	assertEquals(3, count);

	// read split 2
	format.open(split2);
	while ((next = format.nextRecord(null)) != null) {
		assertEquals(7, next.length());
		count++;
	}
	format.close();

	assertEquals(4, count);
}