List of usage examples for org.apache.hadoop.io Text Text
public Text(byte[] utf8)
From source file:co.cask.tephra.persist.HDFSTransactionLogTest.java
License:Apache License
private SequenceFile.Writer getSequenceFileWriter(Configuration configuration, FileSystem fs, long timeInMillis, boolean withMarker) throws IOException { String snapshotDir = configuration.get(TxConstants.Manager.CFG_TX_SNAPSHOT_DIR); Path newLog = new Path(snapshotDir, LOG_FILE_PREFIX + timeInMillis); SequenceFile.Metadata metadata = new SequenceFile.Metadata(); if (withMarker) { metadata.set(new Text(TxConstants.TransactionLog.VERSION_KEY), new Text(Byte.toString(TxConstants.TransactionLog.CURRENT_VERSION))); }//from w ww. ja va 2 s .c o m return SequenceFile.createWriter(fs, configuration, newLog, LongWritable.class, TransactionEdit.class, SequenceFile.CompressionType.NONE, null, null, metadata); }
From source file:co.nubetech.hiho.dedup.HihoTuple.java
License:Apache License
public void setKey(K key) throws IOException { this.key = key; this.hash = HashUtility.getMD5Hash(key); this.keyClass = new Text(key.getClass().getName()); logger.debug("Key is: " + this.getKey()); logger.debug("Hash is: " + this.getHash()); }
From source file:co.nubetech.hiho.dedup.TestDedupJob.java
License:Apache License
@Test public void testDedupByIntWritableKeyWithSequenceFileInputFormat() throws Exception { HashMap<IntWritable, Text> inputData1 = new HashMap<IntWritable, Text>(); inputData1.put(new IntWritable(1), new Text("Xavier Wilson,Mason Holloway,Carlos Johnston,Martin Noel,Drake Mckinney")); inputData1.put(new IntWritable(2), new Text("Kennedy Bailey,Jerome Perry,David Cabrera,Edan Fleming,Orlando Tyson")); inputData1.put(new IntWritable(3), new Text("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein")); createSequenceFileInHdfs(inputData1, "/input1", "testFile1.seq"); HashMap<IntWritable, Text> inputData2 = new HashMap<IntWritable, Text>(); inputData2.put(new IntWritable(1), new Text("Zephania Bauer,Jermaine Gordon,Vincent Moon,Steven Pierce,Jasper Campos")); inputData2.put(new IntWritable(2), new Text("Kennedy Bailey,Plato Atkinson,Stuart Guy,Rooney Levy,Judah Benson")); inputData2.put(new IntWritable(4), new Text("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein")); createSequenceFileInHdfs(inputData2, "/input2", "testFile2.seq"); String[] args = new String[] { "-inputFormat", "org.apache.hadoop.mapreduce.lib.input.SequenceFileInputFormat", "-inputPath", "/input1,/input2", "-outputPath", "output", "-inputKeyClassName", "org.apache.hadoop.io.IntWritable", "-inputValueClassName", "org.apache.hadoop.io.Text", "-dedupBy", "key" }; DedupJob job = runDedupJob(args);/*from w w w . j av a 2s . c om*/ assertEquals(6, job.getTotalRecordsRead()); assertEquals(0, job.getBadRecords()); assertEquals(4, job.getOutput()); assertEquals(2, job.getDuplicateRecords()); FileSystem outputFS = getFileSystem(); Path outputPath = new Path(outputFS.getHomeDirectory(), "output/part-r-00000"); Configuration conf = new Configuration(); SequenceFile.Reader reader = new SequenceFile.Reader(outputFS, outputPath, conf); Writable writableKey = (Writable) ReflectionUtils.newInstance(reader.getKeyClass(), conf); Writable writableValue = (Writable) ReflectionUtils.newInstance(reader.getValueClass(), conf); List<IntWritable> expectedOutput = new ArrayList<IntWritable>(); expectedOutput.add(new IntWritable(1)); expectedOutput.add(new IntWritable(2)); expectedOutput.add(new IntWritable(3)); expectedOutput.add(new IntWritable(4)); int count = 0; while (reader.next(writableKey, writableValue)) { logger.debug("key and value is: " + writableKey + ", " + writableValue); assertTrue("Matched output " + writableKey, expectedOutput.contains(writableKey)); count++; } IOUtils.closeStream(reader); assertEquals(4, count); }
From source file:co.nubetech.hiho.dedup.TestDedupJob.java
License:Apache License
@Test public void testDedupByValueWithSequenceFileInputFormat() throws Exception { HashMap<IntWritable, Text> inputData1 = new HashMap<IntWritable, Text>(); inputData1.put(new IntWritable(1), new Text("Xavier Wilson,Mason Holloway,Carlos Johnston,Martin Noel,Drake Mckinney")); inputData1.put(new IntWritable(2), new Text("Kennedy Bailey,Jerome Perry,David Cabrera,Edan Fleming,Orlando Tyson")); inputData1.put(new IntWritable(3), new Text("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein")); createSequenceFileInHdfs(inputData1, "/input1", "testFile1.seq"); HashMap<IntWritable, Text> inputData2 = new HashMap<IntWritable, Text>(); inputData2.put(new IntWritable(1), new Text("Zephania Bauer,Jermaine Gordon,Vincent Moon,Steven Pierce,Jasper Campos")); inputData2.put(new IntWritable(2), new Text("Kennedy Bailey,Plato Atkinson,Stuart Guy,Rooney Levy,Judah Benson")); inputData2.put(new IntWritable(4), new Text("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein")); createSequenceFileInHdfs(inputData2, "/input2", "testFile2.seq"); String[] args = new String[] { "-inputFormat", "org.apache.hadoop.mapreduce.lib.input.SequenceFileInputFormat", "-inputPath", "/input1,/input2", "-outputPath", "output", "-inputKeyClassName", "org.apache.hadoop.io.IntWritable", "-inputValueClassName", "org.apache.hadoop.io.Text", "-dedupBy", "value" }; DedupJob job = runDedupJob(args);/* w w w. j a v a 2 s . c o m*/ assertEquals(6, job.getTotalRecordsRead()); assertEquals(0, job.getBadRecords()); assertEquals(5, job.getOutput()); assertEquals(1, job.getDuplicateRecords()); FileSystem outputFS = getFileSystem(); Path outputPath = new Path(outputFS.getHomeDirectory(), "output/part-r-00000"); Configuration conf = new Configuration(); SequenceFile.Reader reader = new SequenceFile.Reader(outputFS, outputPath, conf); Writable writableKey = (Writable) ReflectionUtils.newInstance(reader.getKeyClass(), conf); Writable writableValue = (Writable) ReflectionUtils.newInstance(reader.getValueClass(), conf); List<Text> expectedOutput = new ArrayList<Text>(); expectedOutput.add(new Text("Xavier Wilson,Mason Holloway,Carlos Johnston,Martin Noel,Drake Mckinney")); expectedOutput.add(new Text("Kennedy Bailey,Jerome Perry,David Cabrera,Edan Fleming,Orlando Tyson")); expectedOutput.add(new Text("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein")); expectedOutput.add(new Text("Zephania Bauer,Jermaine Gordon,Vincent Moon,Steven Pierce,Jasper Campos")); expectedOutput.add(new Text("Kennedy Bailey,Plato Atkinson,Stuart Guy,Rooney Levy,Judah Benson")); int count = 0; while (reader.next(writableKey, writableValue)) { logger.debug("key and value is: " + writableKey + ", " + writableValue); assertTrue("Matched output " + writableValue, expectedOutput.contains(writableValue)); count++; } IOUtils.closeStream(reader); assertEquals(5, count); }
From source file:co.nubetech.hiho.dedup.TestDedupJob.java
License:Apache License
@Test public void testDedupByLongWritableKeyWithSequenceFileInputFormat() throws Exception { HashMap<LongWritable, Text> inputData1 = new HashMap<LongWritable, Text>(); inputData1.put(new LongWritable(1), new Text("Xavier Wilson,Mason Holloway,Carlos Johnston,Martin Noel,Drake Mckinney")); inputData1.put(new LongWritable(2), new Text("Kennedy Bailey,Jerome Perry,David Cabrera,Edan Fleming,Orlando Tyson")); inputData1.put(new LongWritable(3), new Text("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein")); createSequenceFileInHdfs(inputData1, "/input1", "testFile1.seq"); HashMap<LongWritable, Text> inputData2 = new HashMap<LongWritable, Text>(); inputData2.put(new LongWritable(1), new Text("Zephania Bauer,Jermaine Gordon,Vincent Moon,Steven Pierce,Jasper Campos")); inputData2.put(new LongWritable(2), new Text("Kennedy Bailey,Plato Atkinson,Stuart Guy,Rooney Levy,Judah Benson")); inputData2.put(new LongWritable(4), new Text("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein")); createSequenceFileInHdfs(inputData2, "/input2", "testFile2.seq"); String[] args = new String[] { "-inputFormat", "org.apache.hadoop.mapreduce.lib.input.SequenceFileInputFormat", "-inputPath", "/input1,/input2", "-outputPath", "output", "-inputKeyClassName", "org.apache.hadoop.io.LongWritable", "-inputValueClassName", "org.apache.hadoop.io.Text", "-dedupBy", "key" }; DedupJob job = runDedupJob(args);//from w ww .j a v a2 s.c o m assertEquals(6, job.getTotalRecordsRead()); assertEquals(0, job.getBadRecords()); assertEquals(4, job.getOutput()); assertEquals(2, job.getDuplicateRecords()); FileSystem outputFS = getFileSystem(); Path outputPath = new Path(outputFS.getHomeDirectory(), "output/part-r-00000"); Configuration conf = new Configuration(); SequenceFile.Reader reader = new SequenceFile.Reader(outputFS, outputPath, conf); Writable writableKey = (Writable) ReflectionUtils.newInstance(reader.getKeyClass(), conf); Writable writableValue = (Writable) ReflectionUtils.newInstance(reader.getValueClass(), conf); List<LongWritable> expectedOutput = new ArrayList<LongWritable>(); expectedOutput.add(new LongWritable(1)); expectedOutput.add(new LongWritable(2)); expectedOutput.add(new LongWritable(3)); expectedOutput.add(new LongWritable(4)); int count = 0; while (reader.next(writableKey, writableValue)) { logger.debug("key and value is: " + writableKey + ", " + writableValue); assertTrue("Matched output " + writableKey, expectedOutput.contains(writableKey)); count++; } IOUtils.closeStream(reader); assertEquals(4, count); }
From source file:co.nubetech.hiho.dedup.TestDedupJob.java
License:Apache License
@Test public void testDedupByValueWithSequenceFileAsTextInputFormat() throws Exception { HashMap<Text, Text> inputData1 = new HashMap<Text, Text>(); inputData1.put(new Text("1"), new Text("Xavier Wilson,Mason Holloway,Carlos Johnston,Martin Noel,Drake Mckinney")); inputData1.put(new Text("2"), new Text("Kennedy Bailey,Jerome Perry,David Cabrera,Edan Fleming,Orlando Tyson")); inputData1.put(new Text("3"), new Text("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein")); createSequenceFileInHdfs(inputData1, "/input1", "testFile1.seq"); HashMap<Text, Text> inputData2 = new HashMap<Text, Text>(); inputData2.put(new Text("1"), new Text("Zephania Bauer,Jermaine Gordon,Vincent Moon,Steven Pierce,Jasper Campos")); inputData2.put(new Text("2"), new Text("Kennedy Bailey,Plato Atkinson,Stuart Guy,Rooney Levy,Judah Benson")); inputData2.put(new Text("4"), new Text("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein")); createSequenceFileInHdfs(inputData2, "/input2", "testFile2.seq"); String[] args = new String[] { "-inputFormat", "org.apache.hadoop.mapreduce.lib.input.SequenceFileAsTextInputFormat", "-outputFormat", "co.nubetech.hiho.mapreduce.lib.output.NoKeyOnlyValueOutputFormat", "-inputPath", "/input1,/input2", "-outputPath", "output", "-inputKeyClassName", "org.apache.hadoop.io.Text", "-inputValueClassName", "org.apache.hadoop.io.Text", "-dedupBy", "value" }; DedupJob job = runDedupJob(args);//from www. j av a2 s . c o m assertEquals(6, job.getTotalRecordsRead()); assertEquals(0, job.getBadRecords()); assertEquals(5, job.getOutput()); assertEquals(1, job.getDuplicateRecords()); FileSystem outputFS = getFileSystem(); Path outputPath = new Path(outputFS.getHomeDirectory(), "output"); FileStatus[] status = outputFS.listStatus(outputPath, getOutputPathFilter()); assertTrue(outputFS.exists(outputPath)); List<String> expectedOutput = new ArrayList<String>(); expectedOutput.add("Xavier Wilson,Mason Holloway,Carlos Johnston,Martin Noel,Drake Mckinney"); expectedOutput.add("Kennedy Bailey,Jerome Perry,David Cabrera,Edan Fleming,Orlando Tyson"); expectedOutput.add("Drake Mckinney,Murphy Baird,Theodore Lindsey,Nehru Wilcox,Harper Klein"); expectedOutput.add("Zephania Bauer,Jermaine Gordon,Vincent Moon,Steven Pierce,Jasper Campos"); expectedOutput.add("Kennedy Bailey,Plato Atkinson,Stuart Guy,Rooney Levy,Judah Benson"); int count = 0; for (FileStatus fileStat : status) { logger.debug("File status is " + fileStat.getPath() + " and is it a dir? " + fileStat.isDirectory()); FSDataInputStream in = outputFS.open(fileStat.getPath()); String line = null; while ((line = in.readLine()) != null) { logger.debug("Output is " + line); assertTrue("Matched output " + line, expectedOutput.contains(line)); expectedOutput.remove(line); count++; } in.close(); } assertEquals(5, count); }
From source file:co.nubetech.hiho.dedup.TestDedupJob.java
License:Apache License
@Test public void testDedupByCustomObjectKeyWithSequenceFileInputFormat() throws Exception { Student student1 = setStudent(new Text("Sam"), new Text("US"), new IntWritable(1), new LongWritable(9999999998l), new DoubleWritable(99.12)); Student student2 = setStudent(new Text("John"), new Text("AUS"), new IntWritable(2), new LongWritable(9999999999l), new DoubleWritable(90.12)); Student student3 = setStudent(new Text("Mary"), new Text("UK"), new IntWritable(3), new LongWritable(9999999988l), new DoubleWritable(69.12)); Student student4 = setStudent(new Text("Kelvin"), new Text("UK"), new IntWritable(4), new LongWritable(9999998888l), new DoubleWritable(59.12)); HashMap<Student, Text> inputData1 = new HashMap<Student, Text>(); inputData1.put(student1, new Text("Macon Kent,6269 Aenean St.,1-247-399-1051,08253")); inputData1.put(student2, new Text("Dale Zamora,521-7792 Mauris Rd.,1-214-625-6970,90510")); inputData1.put(student3, new Text("Charles Wood,525-9709 In Rd.,1-370-528-4758,62714")); createSequenceFileInHdfs(inputData1, "/input1", "testFile1.seq"); HashMap<Student, Text> inputData2 = new HashMap<Student, Text>(); inputData2.put(student2, new Text("Austin Farley,4794 Donec Ave,1-230-823-8164,13508")); inputData2.put(student3, new Text("Macaulay Jackson,5435 Dui. Avenue,1-770-395-6446,31584")); inputData2.put(student4, new Text("Timon Leonard,716 Ac Ave,1-857-935-3882,62240")); createSequenceFileInHdfs(inputData2, "/input2", "testFile2.seq"); String[] args = new String[] { "-inputPath", "/input1,/input2", "-outputPath", "output", "-dedupBy", "key", "-inputFormat", "org.apache.hadoop.mapreduce.lib.input.SequenceFileInputFormat", "-inputKeyClassName", "co.nubetech.hiho.testdata.Student", "-inputValueClassName", "org.apache.hadoop.io.Text" }; DedupJob job = runDedupJob(args);// w w w . j av a 2 s. c o m assertEquals(6, job.getTotalRecordsRead()); assertEquals(0, job.getBadRecords()); assertEquals(4, job.getOutput()); assertEquals(2, job.getDuplicateRecords()); FileSystem outputFS = getFileSystem(); Path outputPath = new Path(outputFS.getHomeDirectory(), "output/part-r-00000"); Configuration conf = new Configuration(); SequenceFile.Reader reader = new SequenceFile.Reader(outputFS, outputPath, conf); Writable writableKey = (Writable) ReflectionUtils.newInstance(reader.getKeyClass(), conf); Writable writableValue = (Writable) ReflectionUtils.newInstance(reader.getValueClass(), conf); List<Student> expectedOutput = new ArrayList<Student>(); expectedOutput.add(student1); expectedOutput.add(student2); expectedOutput.add(student3); expectedOutput.add(student4); int count = 0; while (reader.next(writableKey, writableValue)) { logger.debug("key and value is: " + writableKey + ", " + writableValue); assertTrue("Matched output " + writableKey, expectedOutput.contains(writableKey)); count++; } IOUtils.closeStream(reader); assertEquals(4, count); }
From source file:co.nubetech.hiho.dedup.TestDedupKeyMapper.java
License:Apache License
@Test public final void testMapperValidValues() throws IOException, InterruptedException { Mapper.Context context = mock(Mapper.Context.class); Counters counters = new Counters(); Counter counter = counters.findCounter(DedupRecordCounter.TOTAL_RECORDS_READ); when(context.getCounter(DedupRecordCounter.TOTAL_RECORDS_READ)).thenReturn(counter); DedupKeyMapper<Text, String> mapper = new DedupKeyMapper<Text, String>(); Text key = new Text("abc123"); String val = "valueOfKey"; mapper.map(key, val, context); HihoTuple<Text> hihoTuple = new HihoTuple<Text>(); hihoTuple.setKey(key);//w ww .j a va 2 s . c o m verify(context).write(hihoTuple, val); assertEquals(1, context.getCounter(DedupRecordCounter.TOTAL_RECORDS_READ).getValue()); }
From source file:co.nubetech.hiho.dedup.TestDedupKeyReducer.java
License:Apache License
@Test public void testReducerValidValues() throws IOException, InterruptedException { Text key = new Text("key123"); HihoTuple hihoTuple = new HihoTuple(); hihoTuple.setKey(key);//from w w w .jav a 2 s.c o m String value1 = new String("value1"); String value2 = new String("value2"); String value3 = new String("value3"); ArrayList<String> values = new ArrayList<String>(); values.add(value1); values.add(value2); values.add(value3); Reducer.Context context = mock(Reducer.Context.class); Counters counters = new Counters(); Counter counter = counters.findCounter(DedupRecordCounter.OUTPUT); when(context.getCounter(DedupRecordCounter.OUTPUT)).thenReturn(counter); DedupKeyReducer dedupReducer = new DedupKeyReducer(); dedupReducer.reduce(hihoTuple, values, context); verify(context).write(key, value1); assertEquals(1, context.getCounter(DedupRecordCounter.OUTPUT).getValue()); }
From source file:co.nubetech.hiho.dedup.TestDedupKeyReducer.java
License:Apache License
@Test public void testReducerForLongWritableKey() throws IOException, InterruptedException { LongWritable key = new LongWritable(Long.parseLong("123")); HihoTuple hihoTuple = new HihoTuple(); hihoTuple.setKey(key);//from w ww.j ava2 s .c om Text value1 = new Text("value1"); ArrayList<Text> values = new ArrayList<Text>(); values.add(value1); Reducer.Context context = mock(Reducer.Context.class); Counters counters = new Counters(); Counter counter = counters.findCounter(DedupRecordCounter.OUTPUT); when(context.getCounter(DedupRecordCounter.OUTPUT)).thenReturn(counter); DedupKeyReducer dedupReducer = new DedupKeyReducer(); dedupReducer.reduce(hihoTuple, values, context); verify(context).write(key, value1); assertEquals(1, context.getCounter(DedupRecordCounter.OUTPUT).getValue()); }