List of usage examples for org.apache.hadoop.fs FileSystem getHomeDirectory
public Path getHomeDirectory()
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);// ww w . ja v a 2 s . 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);//from ww w.j a va 2 s. c om 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);// w w w .j av a 2 s .co 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 w w w . ja va2 s .c om*/ 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);/*from w w w . j a va 2 s . co 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.job.TestDBQueryInputJobWithCluster.java
License:Apache License
@Test public void testBasicTableImport() throws Exception { DBQueryInputJob job = new DBQueryInputJob(); String[] args = new String[] { "-jdbcDriver", "org.hsqldb.jdbcDriver", "-jdbcUrl", "jdbc:hsqldb:hsql://localhost/URLAccess", // "-jdbcUsername", "", // "-jdbcPassword", "", "-outputPath", "testBasicTableImport", "-outputStrategy", "delimited", "-delimiter", "DELIM", "-numberOfMappers", "2", "-inputTableName", "Pageview", "-inputOrderBy", "pageview" }; int res = ToolRunner.run(createJobConf(), job, args); assertEquals(0, res);/*www. java2 s.c o m*/ //lets verify the result now FileSystem outputFS = getFileSystem(); //Path outputPath = getOutputDir(); Path outputPath = new Path(outputFS.getHomeDirectory(), "testBasicTableImport"); FileStatus[] status = outputFS.listStatus(outputPath, getOutputPathFilter()); assertTrue(outputFS.exists(outputPath)); List<String> expectedOutput = new ArrayList<String>(); expectedOutput.add("/aDELIM1000"); expectedOutput.add("/bDELIM2000"); expectedOutput.add("/cDELIM3000"); expectedOutput.add("/dDELIM4000"); expectedOutput.add("/eDELIM5000"); expectedOutput.add("/fDELIM6000"); expectedOutput.add("/gDELIM7000"); expectedOutput.add("/hDELIM8000"); expectedOutput.add("/iDELIM9000"); expectedOutput.add("/jDELIM10000"); 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(10, count); }
From source file:co.nubetech.hiho.job.TestDBQueryInputJobWithCluster.java
License:Apache License
@Test public void testBasicAvroTableImport() throws Exception { DBQueryInputJob job = new DBQueryInputJob(); String[] args = new String[] { "-jdbcDriver", "org.hsqldb.jdbcDriver", "-jdbcUrl", "jdbc:hsqldb:hsql://localhost/URLAccess", "-outputPath", "testQueryBasedImport", "-inputQuery", "select url,pageview,commentCount from Pageview, PageComment where Pageview.url = PageComment.url", "-inputBoundingQuery", "select min(commentCount), max(commentCount) from PageComment", "-outputStrategy", "AVRO", "-delimiter", "DELIM", "-numberOfMappers", "2", "-inputOrderBy", "Pageview.pageview" }; int res = ToolRunner.run(createJobConf(), job, args); assertEquals(0, res);//from ww w . j av a2 s. com //lets verify the result now FileSystem outputFS = getFileSystem(); //Path outputPath = getOutputDir(); Path outputPath = new Path(outputFS.getHomeDirectory(), "testBasicTableImport"); FileStatus[] status = outputFS.listStatus(outputPath, getOutputPathFilter()); assertTrue(outputFS.exists(outputPath)); /* List<String> expectedOutput = new ArrayList<String>(); expectedOutput.add("/aDELIM1000"); expectedOutput.add("/bDELIM2000"); expectedOutput.add("/cDELIM3000"); expectedOutput.add("/dDELIM4000"); expectedOutput.add("/eDELIM5000"); expectedOutput.add("/fDELIM6000"); expectedOutput.add("/gDELIM7000"); expectedOutput.add("/hDELIM8000"); expectedOutput.add("/iDELIM9000"); expectedOutput.add("/jDELIM10000"); 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(10, count); */ }
From source file:co.nubetech.hiho.job.TestDBQueryInputJobWithCluster.java
License:Apache License
@Test public void testQueryBasedImport() throws Exception { DBQueryInputJob job = new DBQueryInputJob(); String[] args = new String[] { "-jdbcDriver", "org.hsqldb.jdbcDriver", "-jdbcUrl", "jdbc:hsqldb:hsql://localhost/URLAccess", "-outputPath", "testQueryBasedImport", "-inputQuery", "select url,pageview,commentCount from Pageview, PageComment where Pageview.url = PageComment.url", "-inputBoundingQuery", "select min(commentCount), max(commentCount) from PageComment", "-outputStrategy", "delimited", "-delimiter", "DELIM", "-numberOfMappers", "2", "-inputOrderBy", "Pageview.pageview" }; int res = ToolRunner.run(createJobConf(), job, args); assertEquals(0, res);//from w w w. j a v a 2s. c o m //lets verify the result now FileSystem outputFS = getFileSystem(); Path outputPath = new Path(outputFS.getHomeDirectory(), "testQueryBasedImport"); FileStatus[] status = outputFS.listStatus(outputPath, getOutputPathFilter()); assertTrue(outputFS.exists(outputPath)); List<String> expectedOutput = new ArrayList<String>(); expectedOutput.add("/aDELIM1000DELIM10"); expectedOutput.add("/bDELIM2000DELIM10"); expectedOutput.add("/cDELIM3000DELIM10"); expectedOutput.add("/dDELIM4000DELIM10"); expectedOutput.add("/eDELIM5000DELIM10"); expectedOutput.add("/fDELIM6000DELIM10"); expectedOutput.add("/gDELIM7000DELIM10"); expectedOutput.add("/hDELIM8000DELIM10"); expectedOutput.add("/iDELIM9000DELIM10"); expectedOutput.add("/jDELIM10000DELIM10"); 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(10, count); }
From source file:co.nubetech.hiho.merge.TestMergeJob.java
License:Apache License
@Test public void testMergeByCustomObjectKeyWithSequenceFileInputFormat() 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[] { "-newPath", "/input1", "-oldPath", "/input2", "-mergeBy", "key", "-outputPath", "output", "-inputFormat", "org.apache.hadoop.mapreduce.lib.input.SequenceFileInputFormat", "-inputKeyClassName", "co.nubetech.hiho.testdata.Student", "-inputValueClassName", "org.apache.hadoop.io.Text" }; MergeJob job = runMergeJobs(args);/*from w ww . j a va 2s. co m*/ assertEquals(3, job.getTotalRecordsNew()); assertEquals(3, job.getTotalRecordsOld()); assertEquals(0, job.getBadRecords()); assertEquals(4, job.getOutput()); 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.merge.TestMergeJob.java
License:Apache License
@Test public void testMergeByIntWritableKeyWithSequenceFileInputFormat() throws Exception { HashMap<IntWritable, Text> inputData1 = new HashMap<IntWritable, Text>(); inputData1.put(new IntWritable(1), new Text("Macon Kent,6269 Aenean St.,1-247-399-1051,08253")); inputData1.put(new IntWritable(2), new Text("Dale Zamora,521-7792 Mauris Rd.,1-214-625-6970,90510")); inputData1.put(new IntWritable(3), new Text("Charles Wood,525-9709 In Rd.,1-370-528-4758,62714")); createSequenceFileInHdfs(inputData1, "/input1", "testFile1.seq"); HashMap<IntWritable, Text> inputData2 = new HashMap<IntWritable, Text>(); inputData2.put(new IntWritable(1), new Text("Macaulay Jackson,5435 Dui. Avenue,1-770-395-6446,31584")); inputData2.put(new IntWritable(2), new Text("Timon Leonard,716 Ac Ave,1-857-935-3882,62240")); inputData2.put(new IntWritable(4), new Text("Charles Wood,525-9709 In Rd.,1-370-528-4758,62714")); createSequenceFileInHdfs(inputData2, "/input2", "testFile2.seq"); String[] args = new String[] { "-newPath", "/input1", "-oldPath", "/input2", "-mergeBy", "key", "-outputPath", "output", "-inputFormat", "org.apache.hadoop.mapreduce.lib.input.SequenceFileInputFormat", "-inputKeyClassName", "org.apache.hadoop.io.IntWritable", "-inputValueClassName", "org.apache.hadoop.io.Text" }; MergeJob job = runMergeJobs(args);//from w w w . j ava2 s. c o m assertEquals(3, job.getTotalRecordsNew()); assertEquals(3, job.getTotalRecordsOld()); assertEquals(0, job.getBadRecords()); assertEquals(4, job.getOutput()); 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); }