Java RecordReader.close方法代码示例

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
/**
 * From each split sampled, take the first numSamples / numSplits records.
 */
@SuppressWarnings("unchecked") // ArrayList::toArray doesn't preserve type
public K[] getSample(InputFormat<K,V> inf, JobConf job) throws IOException {
  InputSplit[] splits = inf.getSplits(job, job.getNumMapTasks());
  ArrayList<K> samples = new ArrayList<K>(numSamples);
  int splitsToSample = Math.min(maxSplitsSampled, splits.length);
  int splitStep = splits.length / splitsToSample;
  int samplesPerSplit = numSamples / splitsToSample;
  long records = 0;
  for (int i = 0; i < splitsToSample; ++i) {
    RecordReader<K,V> reader = inf.getRecordReader(splits[i * splitStep],
        job, Reporter.NULL);
    K key = reader.createKey();
    V value = reader.createValue();
    while (reader.next(key, value)) {
      samples.add(key);
      key = reader.createKey();
      ++records;
      if ((i+1) * samplesPerSplit <= records) {
        break;
      }
    }
    reader.close();
  }
  return (K[])samples.toArray();
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
/**
 * test DBInputFormat class. Class should split result for chunks
 * @throws Exception
 */
@Test(timeout = 10000)
public void testDBInputFormat() throws Exception {
  JobConf configuration = new JobConf();
  setupDriver(configuration);
  
  DBInputFormat<NullDBWritable> format = new DBInputFormat<NullDBWritable>();
  format.setConf(configuration);
  format.setConf(configuration);
  DBInputFormat.DBInputSplit splitter = new DBInputFormat.DBInputSplit(1, 10);
  Reporter reporter = mock(Reporter.class);
  RecordReader<LongWritable, NullDBWritable> reader = format.getRecordReader(
      splitter, configuration, reporter);

  configuration.setInt(MRJobConfig.NUM_MAPS, 3);
  InputSplit[] lSplits = format.getSplits(configuration, 3);
  assertEquals(5, lSplits[0].getLength());
  assertEquals(3, lSplits.length);

  // test reader .Some simple tests
  assertEquals(LongWritable.class, reader.createKey().getClass());
  assertEquals(0, reader.getPos());
  assertEquals(0, reader.getProgress(), 0.001);
  reader.close();
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
@Test
 public void readEthereumBlockInputFormatBlock1346406Bzip2Compressed() throws IOException, EthereumBlockReadException, ParseException, InterruptedException {
	JobConf job = new JobConf(defaultConf);
ClassLoader classLoader = getClass().getClassLoader();
String fileName="eth1346406.bin.bz2";
String fileNameBlock=classLoader.getResource("testdata/"+fileName).getFile();	
Path file = new Path(fileNameBlock);
   FileInputFormat.setInputPaths(job, file);
   EthereumBlockFileInputFormat format = new EthereumBlockFileInputFormat();
   format.configure(job);
   InputSplit[] inputSplits = format.getSplits(job,1);
 
   assertEquals( 1, inputSplits.length,"Only one split generated for genesis block");
   	RecordReader<BytesWritable, EthereumBlock> reader = format.getRecordReader(inputSplits[0], job, reporter);
assertNotNull( reader,"Format returned  null RecordReader");
	
BytesWritable key = new BytesWritable();	
EthereumBlock block = new EthereumBlock();
assertTrue( reader.next(key,block),"Input Split for block 1346406 contains at least one block");
	
assertEquals( 6, block.getEthereumTransactions().size(),"Block 1346406 must have 6 transactions");
   	assertFalse( reader.next(key,block),"No further blocks in block 1346406");
   	reader.close();
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
@Test
 public void readEthereumBlockInputFormatGenesisBlock() throws IOException, EthereumBlockReadException, ParseException, InterruptedException {
JobConf job = new JobConf(defaultConf);
ClassLoader classLoader = getClass().getClassLoader();
String fileName="ethgenesis.bin";
String fileNameBlock=classLoader.getResource("testdata/"+fileName).getFile();	
Path file = new Path(fileNameBlock);
   FileInputFormat.setInputPaths(job, file);
   EthereumBlockFileInputFormat format = new EthereumBlockFileInputFormat();
   format.configure(job);
   InputSplit[] inputSplits = format.getSplits(job,1);
 
   assertEquals( 1, inputSplits.length,"Only one split generated for genesis block");
   	RecordReader<BytesWritable, EthereumBlock> reader = format.getRecordReader(inputSplits[0], job, reporter);
assertNotNull( reader,"Format returned  null RecordReader");
BytesWritable key = new BytesWritable();	
EthereumBlock block = new EthereumBlock();
assertTrue( reader.next(key,block),"Input Split for genesis block contains at least one block");
assertEquals( 0, block.getEthereumTransactions().size(),"Genesis Block must have 0 transactions");
   	assertFalse( reader.next(key,block),"No further blocks in genesis Block");
   	reader.close();
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
@Test
 public void readEthereumBlockInputFormatBlock1() throws IOException, EthereumBlockReadException, ParseException, InterruptedException {
	JobConf job = new JobConf(defaultConf);
			ClassLoader classLoader = getClass().getClassLoader();
String fileName="eth1.bin";
String fileNameBlock=classLoader.getResource("testdata/"+fileName).getFile();	
Path file = new Path(fileNameBlock);
   FileInputFormat.setInputPaths(job, file);
   EthereumBlockFileInputFormat format = new EthereumBlockFileInputFormat();
   format.configure(job);
   InputSplit[] inputSplits = format.getSplits(job,1);
 
   assertEquals( 1, inputSplits.length,"Only one split generated for genesis block");
   	RecordReader<BytesWritable, EthereumBlock> reader = format.getRecordReader(inputSplits[0], job, reporter);
assertNotNull( reader,"Format returned  null RecordReader");
BytesWritable key = new BytesWritable();	
EthereumBlock block = new EthereumBlock();
assertTrue( reader.next(key,block),"Input Split for block 1 contains at least one block");
assertEquals( 0, block.getEthereumTransactions().size(),"Block 1 must have 0 transactions");
   	assertFalse( reader.next(key,block),"No further blocks in block 1");
   	reader.close();
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
@Test
 public void readEthereumBlockInputFormatBlock1346406() throws IOException, EthereumBlockReadException, ParseException, InterruptedException {
	JobConf job = new JobConf(defaultConf);
				ClassLoader classLoader = getClass().getClassLoader();
String fileName="eth1346406.bin";
String fileNameBlock=classLoader.getResource("testdata/"+fileName).getFile();	
Path file = new Path(fileNameBlock);
   FileInputFormat.setInputPaths(job, file);
   EthereumBlockFileInputFormat format = new EthereumBlockFileInputFormat();
   format.configure(job);
   InputSplit[] inputSplits = format.getSplits(job,1);
 
   assertEquals( 1, inputSplits.length,"Only one split generated for genesis block");
   	RecordReader<BytesWritable, EthereumBlock> reader = format.getRecordReader(inputSplits[0], job, reporter);
assertNotNull( reader,"Format returned  null RecordReader");
		
BytesWritable key = new BytesWritable();	
EthereumBlock block = new EthereumBlock();
assertTrue( reader.next(key,block),"Input Split for block 1346406 contains at least one block");
assertEquals( 6, block.getEthereumTransactions().size(),"Block 1346406 must have 6 transactions");
   	assertFalse( reader.next(key,block),"No further blocks in block 1346406");
   	reader.close();
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
@Test
 public void readEthereumBlockInputFormatBlock3346406() throws IOException, EthereumBlockReadException, ParseException, InterruptedException {
JobConf job = new JobConf(defaultConf);
ClassLoader classLoader = getClass().getClassLoader();
String fileName="eth3346406.bin";
String fileNameBlock=classLoader.getResource("testdata/"+fileName).getFile();	
Path file = new Path(fileNameBlock);
   FileInputFormat.setInputPaths(job, file);
   EthereumBlockFileInputFormat format = new EthereumBlockFileInputFormat();
   format.configure(job);
   InputSplit[] inputSplits = format.getSplits(job,1);
 
   assertEquals( 1, inputSplits.length,"Only one split generated for genesis block");
   	RecordReader<BytesWritable, EthereumBlock> reader = format.getRecordReader(inputSplits[0], job, reporter);
assertNotNull( reader,"Format returned  null RecordReader");

BytesWritable key = new BytesWritable();	
EthereumBlock block = new EthereumBlock();
assertTrue( reader.next(key,block),"Input Split for block 3346406 contains at least one block");
assertEquals( 7, block.getEthereumTransactions().size(),"Block 3346406 must have 7 transactions");
   	assertFalse( reader.next(key,block),"No further blocks in block 3346406");
   	reader.close();
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
/**
 * Close all child RRs.
 */
public void close() throws IOException {
  if (kids != null) {
    for (RecordReader<K,? extends Writable> rr : kids) {
      rr.close();
    }
  }
  if (jc != null) {
    jc.close();
  }
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
/**
 * Dump given list of files to standard output as typed bytes.
 */
@SuppressWarnings("unchecked")
private int dumpTypedBytes(List<FileStatus> files) throws IOException {
  JobConf job = new JobConf(getConf()); 
  DataOutputStream dout = new DataOutputStream(System.out);
  AutoInputFormat autoInputFormat = new AutoInputFormat();
  for (FileStatus fileStatus : files) {
    FileSplit split = new FileSplit(fileStatus.getPath(), 0,
      fileStatus.getLen() * fileStatus.getBlockSize(),
      (String[]) null);
    RecordReader recReader = null;
    try {
      recReader = autoInputFormat.getRecordReader(split, job, Reporter.NULL);
      Object key = recReader.createKey();
      Object value = recReader.createValue();
      while (recReader.next(key, value)) {
        if (key instanceof Writable) {
          TypedBytesWritableOutput.get(dout).write((Writable) key);
        } else {
          TypedBytesOutput.get(dout).write(key);
        }
        if (value instanceof Writable) {
          TypedBytesWritableOutput.get(dout).write((Writable) value);
        } else {
          TypedBytesOutput.get(dout).write(value);
        }
      }
    } finally {
      if (recReader != null) {
        recReader.close();
      }
    }
  }
  dout.flush();
  return 0;
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
private void verifyWithMockedMapReduce(JobConf job, int numRegions, int expectedNumSplits,
    byte[] startRow, byte[] stopRow) throws IOException, InterruptedException {
  TableSnapshotInputFormat tsif = new TableSnapshotInputFormat();
  InputSplit[] splits = tsif.getSplits(job, 0);

  Assert.assertEquals(expectedNumSplits, splits.length);

  HBaseTestingUtility.SeenRowTracker rowTracker =
    new HBaseTestingUtility.SeenRowTracker(startRow, stopRow);

  for (int i = 0; i < splits.length; i++) {
    // validate input split
    InputSplit split = splits[i];
    Assert.assertTrue(split instanceof TableSnapshotInputFormat.TableSnapshotRegionSplit);

    // validate record reader
    OutputCollector collector = mock(OutputCollector.class);
    Reporter reporter = mock(Reporter.class);
    RecordReader<ImmutableBytesWritable, Result> rr = tsif.getRecordReader(split, job, reporter);

    // validate we can read all the data back
    ImmutableBytesWritable key = rr.createKey();
    Result value = rr.createValue();
    while (rr.next(key, value)) {
      verifyRowFromMap(key, value);
      rowTracker.addRow(key.copyBytes());
    }

    rr.close();
  }

  // validate all rows are seen
  rowTracker.validate();
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
@Test
public void readEthereumBlockInputFormatBlock403419() throws IOException, EthereumBlockReadException, ParseException, InterruptedException {
	JobConf job = new JobConf(defaultConf);
	ClassLoader classLoader = getClass().getClassLoader();
	String fileName="block403419.bin";
	String fileNameBlock=classLoader.getResource("testdata/"+fileName).getFile();	
	Path file = new Path(fileNameBlock);
    FileInputFormat.setInputPaths(job, file);
    EthereumBlockFileInputFormat format = new EthereumBlockFileInputFormat();
    format.configure(job);
    InputSplit[] inputSplits = format.getSplits(job,1);
  
    assertEquals( 1, inputSplits.length,"Only one split generated for block 403419");
    	RecordReader<BytesWritable, EthereumBlock> reader = format.getRecordReader(inputSplits[0], job, reporter);
	assertNotNull( reader,"Format returned  null RecordReader");
	BytesWritable key = new BytesWritable();	
	EthereumBlock block = new EthereumBlock();
	assertTrue( reader.next(key,block),"Input Split for block 403419 contains at least one block");
	assertEquals( 2, block.getEthereumTransactions().size(),"Block 403419 must have 2 transactions");
	EthereumBlockHeader ethereumBlockHeader = block.getEthereumBlockHeader();
	assertEquals(
			"f8b483dba2c3b7176a3da549ad41a48bb3121069",
			bytesToHex(ethereumBlockHeader.getCoinBase()).toLowerCase(),
			"Block 403419 was mined by f8b483dba2c3b7176a3da549ad41a48bb3121069"
	);
	assertEquals(
			"08741fa532c05804d9c1086a311e47cc024bbc43980f561041ad1fbb3c223322",
			bytesToHex(ethereumBlockHeader.getParentHash()).toLowerCase(),
			"The parent of block 403419 has hash 08741fa532c05804d9c1086a311e47cc024bbc43980f561041ad1fbb3c223322"
	);
    	assertFalse( reader.next(key,block),"No further lock 403419  in genesis Block");
    	
    	reader.close();
	
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
@Test
public void readEthereumBlockInputFormatBlock447533() throws IOException, EthereumBlockReadException, ParseException, InterruptedException {
	JobConf job = new JobConf(defaultConf);
	ClassLoader classLoader = getClass().getClassLoader();
	String fileName="block447533.bin";
	String fileNameBlock=classLoader.getResource("testdata/"+fileName).getFile();	
	Path file = new Path(fileNameBlock);
    FileInputFormat.setInputPaths(job, file);
    EthereumBlockFileInputFormat format = new EthereumBlockFileInputFormat();
    format.configure(job);
    InputSplit[] inputSplits = format.getSplits(job,1);
  
    assertEquals( 1, inputSplits.length,"Only one split generated for block 447533");
    	RecordReader<BytesWritable, EthereumBlock> reader = format.getRecordReader(inputSplits[0], job, reporter);
	assertNotNull( reader,"Format returned  null RecordReader");
	BytesWritable key = new BytesWritable();	
	EthereumBlock block = new EthereumBlock();
	assertTrue( reader.next(key,block),"Input Split for block 447533 contains at least one block");
	assertEquals( 2, block.getEthereumTransactions().size(),"Block 447533 must have 2 transactions");
	EthereumBlockHeader ethereumBlockHeader = block.getEthereumBlockHeader();
	assertEquals(
			"a027231f42c80ca4125b5cb962a21cd4f812e88f",
			bytesToHex(ethereumBlockHeader.getCoinBase()).toLowerCase(),
			"Block 447533 was mined by a027231f42c80ca4125b5cb962a21cd4f812e88f"
	);
	assertEquals(
			"043559b70c54f0eea6a90b384286d7ab312129603e750075d09fd35e66f8068a",
			bytesToHex(ethereumBlockHeader.getParentHash()).toLowerCase(),
			"The parent of block 447533 has hash 043559b70c54f0eea6a90b384286d7ab312129603e750075d09fd35e66f8068a"
	);
    	assertFalse( reader.next(key,block),"No further block  in  block 447533");
    	
    	reader.close();
	
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
@Test
 public void readEthereumBlockInputFormatBlock0to10() throws IOException, EthereumBlockReadException, ParseException, InterruptedException {
JobConf job = new JobConf(defaultConf);
ClassLoader classLoader = getClass().getClassLoader();
String fileName="eth0to10.bin";
String fileNameBlock=classLoader.getResource("testdata/"+fileName).getFile();	
Path file = new Path(fileNameBlock);
   FileInputFormat.setInputPaths(job, file);
   EthereumBlockFileInputFormat format = new EthereumBlockFileInputFormat();
   format.configure(job);
   InputSplit[] inputSplits = format.getSplits(job,1);
 
   assertEquals( 1, inputSplits.length,"Only one split generated for genesis block");
   	RecordReader<BytesWritable, EthereumBlock> reader = format.getRecordReader(inputSplits[0], job, reporter);
assertNotNull( reader,"Format returned  null RecordReader");
BytesWritable key = new BytesWritable();	
EthereumBlock block = new EthereumBlock();
int count=0;
while (count<11) {
	if (reader.next(key,block)) {
		count++;
	}
}
assertEquals(11,count,"Block 0..10 contains 11 blocks");

   	assertFalse( reader.next(key,block),"No further blocks in block 0..10");
   	reader.close();
}
 

import org.apache.hadoop.mapred.RecordReader; //导入方法依赖的package包/类
@Test
 public void readEthereumBlockInputFormatBlock3510000to3510010() throws IOException, EthereumBlockReadException, ParseException, InterruptedException {
	JobConf job = new JobConf(defaultConf);
				ClassLoader classLoader = getClass().getClassLoader();
String fileName="eth351000to3510010.bin";
String fileNameBlock=classLoader.getResource("testdata/"+fileName).getFile();	
Path file = new Path(fileNameBlock);
   FileInputFormat.setInputPaths(job, file);
   EthereumBlockFileInputFormat format = new EthereumBlockFileInputFormat();
   format.configure(job);
   InputSplit[] inputSplits = format.getSplits(job,1);
 
   assertEquals( 1, inputSplits.length,"Only one split generated for genesis block");
   	RecordReader<BytesWritable, EthereumBlock> reader = format.getRecordReader(inputSplits[0], job, reporter);
assertNotNull( reader,"Format returned  null RecordReader");
BytesWritable key = new BytesWritable();	
EthereumBlock block = new EthereumBlock();
int count=0;
while (count<11) {
	if (reader.next(key,block)) {
		count++;
	}
}
assertEquals(11,count,"Block 3510000 .. 3510010 contains 11 blocks");

   	assertFalse( reader.next(key,block),"No further blocks in block 3510000 .. 3510010");
   	reader.close();
}