List of usage examples for org.apache.hadoop.mapreduce RecordWriter close
public abstract void close(TaskAttemptContext context) throws IOException, InterruptedException;
RecordWriter to future operations. From source file:org.zuinnote.hadoop.office.format.mapreduce.OfficeFormatHadoopExcelNormalTest.java
License:Apache License
@Test public void writeExcelOutputFormatExcel2013SingleSheetEncryptedPositiveSignedNegative() throws IOException, InterruptedException { // one row string and three columns ("test1","test2","test3") // (String formattedValue, String comment, String formula, String address,String // sheetName) SpreadSheetCellDAO a1 = new SpreadSheetCellDAO("test1", "", "", "A1", "Sheet1"); SpreadSheetCellDAO b1 = new SpreadSheetCellDAO("test2", "", "", "B1", "Sheet1"); SpreadSheetCellDAO c1 = new SpreadSheetCellDAO("test3", "", "", "C1", "Sheet1"); // empty row => nothing todo // one row numbers (1,2,3) SpreadSheetCellDAO a3 = new SpreadSheetCellDAO("", "", "1", "A3", "Sheet1"); SpreadSheetCellDAO b3 = new SpreadSheetCellDAO("", "", "2", "B3", "Sheet1"); SpreadSheetCellDAO c3 = new SpreadSheetCellDAO("", "", "3", "C3", "Sheet1"); // one row formulas (=A3+B3) SpreadSheetCellDAO a4 = new SpreadSheetCellDAO("", "", "A3+B3", "A4", "Sheet1"); // write/*from w w w . ja v a 2 s. c o m*/ Job job = Job.getInstance(); Configuration conf = job.getConfiguration(); String fileName = "excel2013singlesheettestoutencryptedpositivesignednegative"; String tmpDir = tmpPath.toString(); Path outputPath = new Path(tmpDir); conf.set("mapreduce.output.basename", fileName); // set generic outputformat settings // set locale to the one of the test data conf.set("hadoopoffice.read.locale.bcp47", "de"); conf.set("hadoopoffice.write.mimeType", "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"); // new // excel // format // security // for the new Excel format you need to decide on your own which algorithms are // secure conf.set("hadoopoffice.write.security.crypt.encrypt.mode", "agile"); conf.set("hadoopoffice.write.security.crypt.encrypt.algorithm", "aes256"); conf.set("hadoopoffice.write.security.crypt.chain.mode", "cbc"); conf.set("hadoopoffice.write.security.crypt.hash.algorithm", "sha512"); conf.set("hadoopoffice.write.security.crypt.password", "test"); conf.set(MRJobConfig.TASK_ATTEMPT_ID, attempt); conf.setInt(MRJobConfig.APPLICATION_ATTEMPT_ID, 0); conf.setInt(FileOutputCommitter.FILEOUTPUTCOMMITTER_ALGORITHM_VERSION, 1); FileOutputFormat.setOutputPath(job, outputPath); JobContext jContext = new JobContextImpl(conf, taskID.getJobID()); TaskAttemptContext context = new TaskAttemptContextImpl(conf, taskID); FileOutputCommitter committer = new FileOutputCommitter(outputPath, context); // setup committer.setupJob(jContext); committer.setupTask(context); ExcelFileOutputFormat outputFormat = new ExcelFileOutputFormat(); RecordWriter<NullWritable, SpreadSheetCellDAO> writer = outputFormat.getRecordWriter(context); assertNotNull(writer, "Format returned null RecordWriter"); writer.write(null, a1); writer.write(null, b1); writer.write(null, c1); writer.write(null, a3); writer.write(null, b3); writer.write(null, c3); writer.write(null, a4); writer.close(context); committer.commitTask(context); // try to read it again conf = new Configuration(defaultConf); job = Job.getInstance(conf); fileName = fileName + this.outputbaseAppendix; Path inputFile = new Path(tmpDir + File.separator + "_temporary" + File.separator + "0" + File.separator + taskAttempt + File.separator + fileName + ".xlsx"); // set locale to the one of the test data conf.set("hadoopoffice.read.locale.bcp47", "de"); // you just need to provide the password to read encrypted data conf.set("hadoopoffice.read.security.crypt.password", "test"); conf.set("hadoopoffice.read.security.sign.verifysignature", "true"); // will fail no signature provided ExcelFileInputFormat inputFormat = new ExcelFileInputFormat(); FileInputFormat.setInputPaths(job, inputFile); TaskAttemptContext context2 = new TaskAttemptContextImpl(conf, new TaskAttemptID()); List<InputSplit> splits = inputFormat.getSplits(job); assertEquals(1, splits.size(), "Only one split generated for Excel file"); RecordReader<Text, ArrayWritable> reader = inputFormat.createRecordReader(splits.get(0), context2); InterruptedException ex = assertThrows(InterruptedException.class, () -> reader.initialize(splits.get(0), context2), "Exception is thrown in case signature cannot be verified"); }
From source file:org.zuinnote.hadoop.office.format.mapreduce.OfficeFormatHadoopExcelNormalTest.java
License:Apache License
@Disabled("This does not work yet due to a bug in Apache POI that prevents writing correct workbooks containing external references: https://bz.apache.org/bugzilla/show_bug.cgi?id=57184") @Test//w w w . j av a 2 s . c o m public void writeExcelOutputFormatExcel2013SingleSheetOneLinkedWorkbook() throws IOException, InterruptedException { // write linkedworkbook1 // one row string and three columns ("test1","test2","test3") // (String formattedValue, String comment, String formula, String address,String // sheetName) SpreadSheetCellDAO wb1a1 = new SpreadSheetCellDAO("test1", "", "", "A1", "Sheet1"); SpreadSheetCellDAO wb1b1 = new SpreadSheetCellDAO("test2", "", "", "B1", "Sheet1"); SpreadSheetCellDAO wb1c1 = new SpreadSheetCellDAO("test3", "", "", "C1", "Sheet1"); // write Job job = Job.getInstance(); Configuration conf = job.getConfiguration(); String linkedWB1FileName = "excel2013linkedwb1"; String tmpDir = tmpPath.toString(); Path outputPath = new Path(tmpDir); FileOutputFormat.setOutputPath(job, outputPath); conf.set("mapreduce.output.basename", linkedWB1FileName); // set locale to the one of the test data conf.set("hadoopoffice.read.locale.bcp47", "de"); conf.set("hadoopoffice.write.mimeType", "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"); // new // Excel // format, // anyway // default, // but // here // for // illustrative // purposes conf.set(MRJobConfig.TASK_ATTEMPT_ID, attempt); conf.setInt(MRJobConfig.APPLICATION_ATTEMPT_ID, 0); conf.setInt(FileOutputCommitter.FILEOUTPUTCOMMITTER_ALGORITHM_VERSION, 1); FileOutputFormat.setOutputPath(job, outputPath); JobContext jContext = new JobContextImpl(conf, taskID.getJobID()); TaskAttemptContext context = new TaskAttemptContextImpl(conf, taskID); FileOutputCommitter committer = new FileOutputCommitter(outputPath, context); // setup committer.setupJob(jContext); committer.setupTask(context); ExcelFileOutputFormat outputFormat = new ExcelFileOutputFormat(); RecordWriter<NullWritable, SpreadSheetCellDAO> writer = outputFormat.getRecordWriter(context); assertNotNull(writer, "Format returned null RecordWriter"); writer.write(null, wb1a1); writer.write(null, wb1b1); writer.write(null, wb1c1); writer.close(context); committer.commitTask(context); committer.commitJob(jContext); // write mainworkbook linkedWB1FileName = linkedWB1FileName + this.outputbaseAppendix; String linkedWorkbookFilename = "[" + tmpDir + File.separator + linkedWB1FileName + ".xlsx]"; SpreadSheetCellDAO a1 = new SpreadSheetCellDAO("test4", "", "", "A1", "Sheet1"); SpreadSheetCellDAO b1 = new SpreadSheetCellDAO("", "", "[" + linkedWB1FileName + ".xlsx]Sheet1!B1", "B1", "Sheet1"); // should be test2 in the end // write job = Job.getInstance(); conf = job.getConfiguration(); String mainWBfileName = "excel2013singlesheetlinkedwbtestout"; outputPath = new Path(tmpDir); FileOutputFormat.setOutputPath(job, outputPath); conf.set("mapreduce.output.basename", mainWBfileName); // set locale to the one of the test data conf.set("hadoopoffice.read.locale.bcp47", "de"); conf.set("hadoopoffice.write.mimeType", "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"); // new // Excel // format, // anyway // default, // but // here // for // illustrative // purposes conf.set("hadoopoffice.write.linkedworkbooks", linkedWorkbookFilename); conf.set(MRJobConfig.TASK_ATTEMPT_ID, attempt); conf.setInt(MRJobConfig.APPLICATION_ATTEMPT_ID, 0); conf.setInt(FileOutputCommitter.FILEOUTPUTCOMMITTER_ALGORITHM_VERSION, 1); FileOutputFormat.setOutputPath(job, outputPath); jContext = new JobContextImpl(conf, taskID.getJobID()); context = new TaskAttemptContextImpl(conf, taskID); committer = new FileOutputCommitter(outputPath, context); // setup committer.setupJob(jContext); committer.setupTask(context); outputFormat = new ExcelFileOutputFormat(); RecordWriter<NullWritable, SpreadSheetCellDAO> writerMain = outputFormat.getRecordWriter(context); assertNotNull(writerMain, "Format returned null RecordWriter"); writerMain.write(null, a1); writerMain.write(null, b1); writerMain.close(context); committer.commitTask(context); committer.commitJob(jContext); // try to read it again job = Job.getInstance(conf); mainWBfileName = mainWBfileName + this.outputbaseAppendix; Path inputFile = new Path(tmpDir + File.separator + mainWBfileName + ".xlsx"); FileInputFormat.setInputPaths(job, inputFile); // set locale to the one of the test data conf.set("hadoopoffice.read.locale.bcp47", "de"); // enable option to read linked workbooks conf.setBoolean("hadoopoffice.read.linkedworkbooks", true); conf.setBoolean("hadoopoffice.read.ignoremissinglinkedworkbooks", false); ExcelFileInputFormat inputFormat = new ExcelFileInputFormat(); context = new TaskAttemptContextImpl(conf, new TaskAttemptID()); List<InputSplit> splits = inputFormat.getSplits(job); assertEquals(1, splits.size(), "Only one split generated for Excel file"); RecordReader<Text, ArrayWritable> reader = inputFormat.createRecordReader(splits.get(0), context); assertNotNull(reader, "Format returned null RecordReader"); reader.initialize(splits.get(0), context); Text spreadSheetKey = new Text(); ArrayWritable spreadSheetValue = new ArrayWritable(SpreadSheetCellDAO.class); assertTrue(reader.nextKeyValue(), "Input Split for Excel file contains row 1 Sheet1"); spreadSheetKey = reader.getCurrentKey(); spreadSheetValue = reader.getCurrentValue(); assertEquals("[" + mainWBfileName + ".xlsx]Sheet1!A1", spreadSheetKey.toString(), "Input Split for Excel file has keyname == \"[" + mainWBfileName + ".xlsx]Sheet1!A1\""); assertEquals(2, spreadSheetValue.get().length, "Input Split for Excel file contains row 1 with 2 columns for Sheet1"); assertEquals("test4", ((SpreadSheetCellDAO) spreadSheetValue.get()[0]).getFormattedValue(), "Input Split for Excel file contains row 1 with cell 1 == \"test4\""); // this comes from the external workbook assertEquals("test2", ((SpreadSheetCellDAO) spreadSheetValue.get()[1]).getFormattedValue(), "Input Split for Excel file contains row 1 with cell 2 == \"test2\""); }
From source file:org.zuinnote.hadoop.office.format.mapreduce.OfficeFormatHadoopExcelNormalTest.java
License:Apache License
@Disabled("This does not work yet due to a bug in Apache POI that prevents writing correct workbooks containing external references: https://bz.apache.org/bugzilla/show_bug.cgi?id=57184") @Test/*from w ww. ja v a2 s . co m*/ public void writeExcelOutputFormatExcel2013SingleSheetTwoLinkedWorkbooks() throws IOException, InterruptedException { // write linkedworkbook1 // one row string and three columns ("test1","test2","test3") // (String formattedValue, String comment, String formula, String address,String // sheetName) SpreadSheetCellDAO wb1a1 = new SpreadSheetCellDAO("test1", "", "", "A1", "Sheet1"); SpreadSheetCellDAO wb1b1 = new SpreadSheetCellDAO("test2", "", "", "B1", "Sheet1"); SpreadSheetCellDAO wb1c1 = new SpreadSheetCellDAO("test3", "", "", "C1", "Sheet1"); // write Job job = Job.getInstance(); Configuration conf = job.getConfiguration(); String linkedWB1FileName = "excel2013linkedwb1"; String tmpDir = tmpPath.toString(); Path outputPath = new Path(tmpDir); FileOutputFormat.setOutputPath(job, outputPath); conf.set("mapreduce.output.basename", linkedWB1FileName); // set locale to the one of the test data conf.set("hadoopoffice.read.locale.bcp47", "de"); conf.set("hadoopoffice.write.mimeType", "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"); // new // Excel // format, // anyway // default, // but // here // for // illustrative // purposes conf.set(MRJobConfig.TASK_ATTEMPT_ID, attempt); conf.setInt(MRJobConfig.APPLICATION_ATTEMPT_ID, 0); conf.setInt(FileOutputCommitter.FILEOUTPUTCOMMITTER_ALGORITHM_VERSION, 1); FileOutputFormat.setOutputPath(job, outputPath); JobContext jContext = new JobContextImpl(conf, taskID.getJobID()); TaskAttemptContext context = new TaskAttemptContextImpl(conf, taskID); FileOutputCommitter committer = new FileOutputCommitter(outputPath, context); // setup committer.setupJob(jContext); committer.setupTask(context); ExcelFileOutputFormat outputFormat = new ExcelFileOutputFormat(); RecordWriter<NullWritable, SpreadSheetCellDAO> writer = outputFormat.getRecordWriter(context); assertNotNull(writer, "Format returned null RecordWriter"); writer.write(null, wb1a1); writer.write(null, wb1b1); writer.write(null, wb1c1); writer.close(context); committer.commitTask(context); committer.commitJob(jContext); // write linkedworkbook2 // one row string and three columns ("test1","test2","test3") // (String formattedValue, String comment, String formula, String address,String // sheetName) SpreadSheetCellDAO wb2a1 = new SpreadSheetCellDAO("test4", "", "", "A1", "Sheet1"); SpreadSheetCellDAO wb2b1 = new SpreadSheetCellDAO("test5", "", "", "B1", "Sheet1"); SpreadSheetCellDAO wb2c1 = new SpreadSheetCellDAO("test6", "", "", "C1", "Sheet1"); // write job = Job.getInstance(); conf = job.getConfiguration(); String linkedWB2FileName = "excel2013linkedwb2"; outputPath = new Path(tmpDir); FileOutputFormat.setOutputPath(job, outputPath); conf.set("mapreduce.output.basename", linkedWB2FileName); // set locale to the one of the test data conf.set("hadoopoffice.read.locale.bcp47", "de"); conf.set("hadoopoffice.write.mimeType", "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"); // new // Excel // format, // anyway // default, // but // here // for // illustrative // purposes conf.set(MRJobConfig.TASK_ATTEMPT_ID, attempt); conf.setInt(MRJobConfig.APPLICATION_ATTEMPT_ID, 0); conf.setInt(FileOutputCommitter.FILEOUTPUTCOMMITTER_ALGORITHM_VERSION, 1); FileOutputFormat.setOutputPath(job, outputPath); jContext = new JobContextImpl(conf, taskID.getJobID()); context = new TaskAttemptContextImpl(conf, taskID); committer = new FileOutputCommitter(outputPath, context); // setup committer.commitTask(context); committer.commitJob(jContext); outputFormat = new ExcelFileOutputFormat(); writer = outputFormat.getRecordWriter(context); assertNotNull(writer, "Format returned null RecordWriter"); writer.write(null, wb2a1); writer.write(null, wb2b1); writer.write(null, wb2c1); writer.close(context); committer.commitTask(context); committer.commitJob(jContext); // write mainworkbook linkedWB1FileName = linkedWB1FileName + this.outputbaseAppendix; linkedWB2FileName = linkedWB2FileName + this.outputbaseAppendix; String linkedWorkbookFilename = "[" + tmpDir + File.separator + linkedWB1FileName + ".xlsx]:[" + tmpDir + File.separator + linkedWB2FileName + ".xlsx]"; SpreadSheetCellDAO a1 = new SpreadSheetCellDAO("test7", "", "", "A1", "Sheet1"); SpreadSheetCellDAO b1 = new SpreadSheetCellDAO("", "", "'[" + linkedWB1FileName + ".xlsx]Sheet1'!B1", "B1", "Sheet1"); // should be test2 in the end SpreadSheetCellDAO c1 = new SpreadSheetCellDAO("", "", "'[" + linkedWB2FileName + ".xlsx]Sheet1'!B1", "B1", "Sheet1"); // should be test5 in the end // write job = Job.getInstance(); conf = job.getConfiguration(); String mainWBfileName = "excel2013singlesheetlinkedwbtestout"; outputPath = new Path(tmpDir); FileOutputFormat.setOutputPath(job, outputPath); conf.set("mapreduce.output.basename", mainWBfileName); // set locale to the one of the test data conf.set("hadoopoffice.read.locale.bcp47", "de"); conf.set("hadoopoffice.write.mimeType", "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"); // new // Excel // format, // anyway // default, // but // here // for // illustrative // purposes conf.set("hadoopoffice.write.linkedworkbooks", linkedWorkbookFilename); conf.set(MRJobConfig.TASK_ATTEMPT_ID, attempt); conf.setInt(MRJobConfig.APPLICATION_ATTEMPT_ID, 0); conf.setInt(FileOutputCommitter.FILEOUTPUTCOMMITTER_ALGORITHM_VERSION, 1); FileOutputFormat.setOutputPath(job, outputPath); jContext = new JobContextImpl(conf, taskID.getJobID()); context = new TaskAttemptContextImpl(conf, taskID); committer = new FileOutputCommitter(outputPath, context); // setup committer.setupJob(jContext); committer.setupTask(context); outputFormat = new ExcelFileOutputFormat(); RecordWriter<NullWritable, SpreadSheetCellDAO> writerMain = outputFormat.getRecordWriter(context); assertNotNull(writerMain, "Format returned null RecordWriter"); writerMain.write(null, a1); writerMain.write(null, b1); writerMain.write(null, c1); writerMain.close(context); committer.commitTask(context); committer.commitJob(jContext); // try to read it again job = Job.getInstance(conf); mainWBfileName = mainWBfileName + this.outputbaseAppendix; Path inputFile = new Path(tmpDir + File.separator + mainWBfileName + ".xlsx"); FileInputFormat.setInputPaths(job, inputFile); // set locale to the one of the test data conf.set("hadoopoffice.read.locale.bcp47", "de"); // enable option to read linked workbooks conf.setBoolean("hadoopoffice.read.linkedworkbooks", true); conf.setBoolean("hadoopoffice.read.ignoremissinglinkedworkbooks", false); ExcelFileInputFormat inputFormat = new ExcelFileInputFormat(); context = new TaskAttemptContextImpl(conf, new TaskAttemptID()); List<InputSplit> splits = inputFormat.getSplits(job); assertEquals(1, splits.size(), "Only one split generated for Excel file"); RecordReader<Text, ArrayWritable> reader = inputFormat.createRecordReader(splits.get(0), context); assertNotNull(reader, "Format returned null RecordReader"); reader.initialize(splits.get(0), context); Text spreadSheetKey = new Text(); ArrayWritable spreadSheetValue = new ArrayWritable(SpreadSheetCellDAO.class); assertTrue(reader.nextKeyValue(), "Input Split for Excel file contains row 1 Sheet1"); spreadSheetKey = reader.getCurrentKey(); spreadSheetValue = reader.getCurrentValue(); assertEquals("[" + mainWBfileName + ".xlsx]Sheet1!A1", spreadSheetKey.toString(), "Input Split for Excel file has keyname == \"[" + mainWBfileName + ".xlsx]Sheet1!A1\""); assertEquals(2, spreadSheetValue.get().length, "Input Split for Excel file contains row 1 with 2 columns for Sheet1"); assertEquals("test7", ((SpreadSheetCellDAO) spreadSheetValue.get()[0]).getFormattedValue(), "Input Split for Excel file contains row 1 with cell 1 == \"test7\""); // this comes from the external workbook assertEquals("test2", ((SpreadSheetCellDAO) spreadSheetValue.get()[1]).getFormattedValue(), "Input Split for Excel file contains row 1 with cell 2 == \"test2\""); assertEquals("test5", ((SpreadSheetCellDAO) spreadSheetValue.get()[2]).getFormattedValue(), "Input Split for Excel file contains row 1 with cell 3 == \"test5\""); }
From source file:parquet.hadoop.TestMemoryManager.java
License:Apache License
@Test public void testMemoryManager() throws Exception { //Verify the adjusted rowGroupSize of writers RecordWriter writer1 = createWriter(1); verifyRowGroupSize(rowGroupSize);// w w w .j ava 2 s.c o m RecordWriter writer2 = createWriter(2); verifyRowGroupSize(rowGroupSize); RecordWriter writer3 = createWriter(3); verifyRowGroupSize((int) Math.floor(expectPoolSize / 3)); writer1.close(null); verifyRowGroupSize(rowGroupSize); writer2.close(null); verifyRowGroupSize(rowGroupSize); writer3.close(null); //Verify the memory pool Assert.assertEquals("memory pool size is incorrect.", expectPoolSize, parquetOutputFormat.getMemoryManager().getTotalMemoryPool()); }