List of usage examples for org.apache.hadoop.mapred FileOutputFormat setCompressOutput
public static void setCompressOutput(JobConf conf, boolean compress)
From source file:WikipediaForwardIndexBuilder.java
License:Apache License
@SuppressWarnings("static-access") @Override/*from w w w .j a va2 s .c om*/ public int run(String[] args) throws Exception { Options options = new Options(); options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("input").create(INPUT_OPTION)); options.addOption( OptionBuilder.withArgName("path").hasArg().withDescription("index file").create(INDEX_FILE_OPTION)); options.addOption(OptionBuilder.withArgName("en|sv|de|cs|es|zh|ar|tr").hasArg() .withDescription("two-letter language code").create(LANGUAGE_OPTION)); CommandLine cmdline; CommandLineParser parser = new GnuParser(); try { cmdline = parser.parse(options, args); } catch (ParseException exp) { System.err.println("Error parsing command line: " + exp.getMessage()); return -1; } if (!cmdline.hasOption(INPUT_OPTION) || !cmdline.hasOption(INDEX_FILE_OPTION)) { HelpFormatter formatter = new HelpFormatter(); formatter.printHelp(this.getClass().getName(), options); ToolRunner.printGenericCommandUsage(System.out); return -1; } Path inputPath = new Path(cmdline.getOptionValue(INPUT_OPTION)); String indexFile = cmdline.getOptionValue(INDEX_FILE_OPTION); String tmpPath = "tmp-" + WikipediaForwardIndexBuilder.class.getSimpleName() + "-" + RANDOM.nextInt(10000); if (!inputPath.isAbsolute()) { System.err.println("Error: " + INPUT_OPTION + " must be an absolute path!"); return -1; } String language = null; if (cmdline.hasOption(LANGUAGE_OPTION)) { language = cmdline.getOptionValue(LANGUAGE_OPTION); if (language.length() != 2) { System.err.println("Error: \"" + language + "\" unknown language!"); return -1; } } JobConf conf = new JobConf(getConf(), WikipediaForwardIndexBuilder.class); FileSystem fs = FileSystem.get(conf); LOG.info("Tool name: " + this.getClass().getName()); LOG.info(" - input path: " + inputPath); LOG.info(" - index file: " + indexFile); LOG.info(" - language: " + language); LOG.info("Note: This tool only works on block-compressed SequenceFiles!"); conf.setJobName(String.format("BuildWikipediaForwardIndex[%s: %s, %s: %s, %s: %s]", INPUT_OPTION, inputPath, INDEX_FILE_OPTION, indexFile, LANGUAGE_OPTION, language)); conf.setNumReduceTasks(1); FileInputFormat.setInputPaths(conf, inputPath); FileOutputFormat.setOutputPath(conf, new Path(tmpPath)); FileOutputFormat.setCompressOutput(conf, false); if (language != null) { conf.set("wiki.language", language); } conf.setInputFormat(NoSplitSequenceFileInputFormat.class); conf.setOutputKeyClass(IntWritable.class); conf.setOutputValueClass(Text.class); conf.setMapRunnerClass(MyMapRunner.class); conf.setReducerClass(IdentityReducer.class); // Delete the output directory if it exists already. fs.delete(new Path(tmpPath), true); RunningJob job = JobClient.runJob(conf); Counters counters = job.getCounters(); int blocks = (int) counters.getCounter(Blocks.Total); LOG.info("number of blocks: " + blocks); LOG.info("Writing index file..."); LineReader reader = new LineReader(fs.open(new Path(tmpPath + "/part-00000"))); FSDataOutputStream out = fs.create(new Path(indexFile), true); out.writeUTF(edu.umd.cloud9.collection.wikipedia.WikipediaForwardIndex.class.getCanonicalName()); out.writeUTF(inputPath.toString()); out.writeInt(blocks); int cnt = 0; Text line = new Text(); while (reader.readLine(line) > 0) { String[] arr = line.toString().split("\\s+"); int docno = Integer.parseInt(arr[0]); int offset = Integer.parseInt(arr[1]); short fileno = Short.parseShort(arr[2]); out.writeInt(docno); out.writeInt(offset); out.writeShort(fileno); cnt++; if (cnt % 100000 == 0) { LOG.info(cnt + " blocks written"); } } reader.close(); out.close(); if (cnt != blocks) { throw new RuntimeException("Error: mismatch in block count!"); } // Clean up. fs.delete(new Path(tmpPath), true); return 0; }
From source file:WikipediaDocnoMappingBuilder.java
License:Apache License
@SuppressWarnings("static-access") @Override//w ww . j a v a 2 s. c o m public int run(String[] args) throws Exception { Options options = new Options(); options.addOption( OptionBuilder.withArgName("path").hasArg().withDescription("XML dump file").create(INPUT_OPTION)); options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("output file") .create(OUTPUT_FILE_OPTION)); options.addOption(OptionBuilder.withArgName("en|sv|de|cs|es|zh|ar|tr").hasArg() .withDescription("two-letter language code").create(LANGUAGE_OPTION)); options.addOption(KEEP_ALL_OPTION, false, "keep all pages"); CommandLine cmdline; CommandLineParser parser = new GnuParser(); try { cmdline = parser.parse(options, args); } catch (ParseException exp) { System.err.println("Error parsing command line: " + exp.getMessage()); return -1; } if (!cmdline.hasOption(INPUT_OPTION) || !cmdline.hasOption(OUTPUT_FILE_OPTION)) { HelpFormatter formatter = new HelpFormatter(); formatter.printHelp(this.getClass().getName(), options); ToolRunner.printGenericCommandUsage(System.out); return -1; } String language = null; if (cmdline.hasOption(LANGUAGE_OPTION)) { language = cmdline.getOptionValue(LANGUAGE_OPTION); if (language.length() != 2) { System.err.println("Error: \"" + language + "\" unknown language!"); return -1; } } String inputPath = cmdline.getOptionValue(INPUT_OPTION); String outputFile = cmdline.getOptionValue(OUTPUT_FILE_OPTION); boolean keepAll = cmdline.hasOption(KEEP_ALL_OPTION); String tmpPath = "tmp-" + WikipediaDocnoMappingBuilder.class.getSimpleName() + "-" + RANDOM.nextInt(10000); LOG.info("Tool name: " + this.getClass().getName()); LOG.info(" - input: " + inputPath); LOG.info(" - output file: " + outputFile); LOG.info(" - keep all pages: " + keepAll); LOG.info(" - language: " + language); // Job job = Job.getInstance(getConf()); JobConf conf = new JobConf(WikipediaDocnoMappingBuilder.class); conf.setJarByClass(WikipediaDocnoMappingBuilder.class); conf.setJobName(String.format("BuildWikipediaDocnoMapping[%s: %s, %s: %s, %s: %s]", INPUT_OPTION, inputPath, OUTPUT_FILE_OPTION, outputFile, LANGUAGE_OPTION, language)); conf.setBoolean(KEEP_ALL_OPTION, keepAll); // .getConfiguration().setBoolean(KEEP_ALL_OPTION, keepAll); if (language != null) { conf.set("wiki.language", language); } conf.setNumReduceTasks(1); FileInputFormat.addInputPath(conf, new Path(inputPath)); FileOutputFormat.setOutputPath(conf, new Path(tmpPath)); FileOutputFormat.setCompressOutput(conf, false); conf.setOutputKeyClass(IntWritable.class); conf.setOutputValueClass(IntWritable.class); conf.setInputFormat(WikipediaPageInputFormat.class); conf.setOutputFormat(TextOutputFormat.class); conf.setMapperClass(MyMapper.class); conf.setReducerClass(MyReducer.class); // Delete the output directory if it exists already. FileSystem.get(getConf()).delete(new Path(tmpPath), true); // job.waitForCompletion(true); RunningJob job = JobClient.runJob(conf); job.waitForCompletion(); // JobClient jobClient = new JobClient(conf); long cnt = keepAll ? job.getCounters().findCounter(PageTypes.TOTAL).getValue() : job.getCounters().findCounter(PageTypes.ARTICLE).getValue(); WikipediaDocnoMapping.writeDocnoMappingData(FileSystem.get(getConf()), tmpPath + "/part-00000", (int) cnt, outputFile); FileSystem.get(getConf()).delete(new Path(tmpPath), true); return 0; }
From source file:RepackWikipedia.java
License:Apache License
@SuppressWarnings("static-access") @Override/*from w w w . ja v a 2 s. co m*/ public int run(String[] args) throws Exception { Options options = new Options(); options.addOption( OptionBuilder.withArgName("path").hasArg().withDescription("XML dump file").create(INPUT_OPTION)); options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("output location") .create(OUTPUT_OPTION)); options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("mapping file") .create(MAPPING_FILE_OPTION)); options.addOption(OptionBuilder.withArgName("block|record|none").hasArg() .withDescription("compression type").create(COMPRESSION_TYPE_OPTION)); options.addOption(OptionBuilder.withArgName("en|sv|de").hasArg().withDescription("two-letter language code") .create(LANGUAGE_OPTION)); CommandLine cmdline; CommandLineParser parser = new GnuParser(); try { cmdline = parser.parse(options, args); } catch (ParseException exp) { System.err.println("Error parsing command line: " + exp.getMessage()); return -1; } if (!cmdline.hasOption(INPUT_OPTION) || !cmdline.hasOption(OUTPUT_OPTION) || !cmdline.hasOption(MAPPING_FILE_OPTION) || !cmdline.hasOption(COMPRESSION_TYPE_OPTION)) { HelpFormatter formatter = new HelpFormatter(); formatter.printHelp(this.getClass().getName(), options); ToolRunner.printGenericCommandUsage(System.out); return -1; } String inputPath = cmdline.getOptionValue(INPUT_OPTION); String outputPath = cmdline.getOptionValue(OUTPUT_OPTION); String mappingFile = cmdline.getOptionValue(MAPPING_FILE_OPTION); String compressionType = cmdline.getOptionValue(COMPRESSION_TYPE_OPTION); if (!"block".equals(compressionType) && !"record".equals(compressionType) && !"none".equals(compressionType)) { System.err.println("Error: \"" + compressionType + "\" unknown compression type!"); return -1; } String language = null; if (cmdline.hasOption(LANGUAGE_OPTION)) { language = cmdline.getOptionValue(LANGUAGE_OPTION); if (language.length() != 2) { System.err.println("Error: \"" + language + "\" unknown language!"); return -1; } } // this is the default block size int blocksize = 1000000; //Job job = Job.getInstance(getConf()); JobConf conf = new JobConf(RepackWikipedia.class); conf.setJarByClass(RepackWikipedia.class); conf.setJobName(String.format("RepackWikipedia[%s: %s, %s: %s, %s: %s, %s: %s]", INPUT_OPTION, inputPath, OUTPUT_OPTION, outputPath, COMPRESSION_TYPE_OPTION, compressionType, LANGUAGE_OPTION, language)); conf.set(DOCNO_MAPPING_FIELD, mappingFile); LOG.info("Tool name: " + this.getClass().getName()); LOG.info(" - XML dump file: " + inputPath); LOG.info(" - output path: " + outputPath); LOG.info(" - docno mapping data file: " + mappingFile); LOG.info(" - compression type: " + compressionType); LOG.info(" - language: " + language); if ("block".equals(compressionType)) { LOG.info(" - block size: " + blocksize); } conf.setNumReduceTasks(0); FileInputFormat.addInputPath(conf, new Path(inputPath)); FileOutputFormat.setOutputPath(conf, new Path(outputPath)); if ("none".equals(compressionType)) { FileOutputFormat.setCompressOutput(conf, false); } else { FileOutputFormat.setCompressOutput(conf, true); if ("record".equals(compressionType)) { SequenceFileOutputFormat.setOutputCompressionType(conf, SequenceFile.CompressionType.RECORD); } else { SequenceFileOutputFormat.setOutputCompressionType(conf, SequenceFile.CompressionType.BLOCK); conf.setInt("io.seqfile.compress.blocksize", blocksize); } } if (language != null) { conf.set("wiki.language", language); } conf.setInputFormat(WikipediaPageInputFormat.class); conf.setOutputFormat(SequenceFileOutputFormat.class); conf.setOutputKeyClass(IntWritable.class); conf.setOutputValueClass(WikipediaPage.class); conf.setMapperClass(MyMapper.class); // Delete the output directory if it exists already. FileSystem.get(getConf()).delete(new Path(outputPath), true); //job.waitForCompletion(true); JobClient.runJob(conf); return 0; }
From source file:com.alexholmes.hadooputils.combine.avro.mapred.CombineAvroInputFormatTest.java
License:Apache License
@SuppressWarnings("deprecation") public void testJob() throws Exception { JobConf job = new JobConf(); String dir = System.getProperty("test.dir", ".") + "/mapred"; Path outputPath = new Path(dir + "/out"); outputPath.getFileSystem(job).delete(outputPath); WordCountUtil.writeLinesFile();/* ww w.ja v a 2s . co m*/ job.setJobName("wordcount"); AvroJob.setInputSchema(job, Schema.create(Schema.Type.STRING)); AvroJob.setOutputSchema(job, new Pair<Utf8, Long>(new Utf8(""), 0L).getSchema()); AvroJob.setMapperClass(job, MapImpl.class); AvroJob.setCombinerClass(job, ReduceImpl.class); AvroJob.setReducerClass(job, ReduceImpl.class); job.setInputFormat(CombineAvroInputFormat.class); FileInputFormat.setInputPaths(job, new Path(dir + "/in")); FileOutputFormat.setOutputPath(job, outputPath); FileOutputFormat.setCompressOutput(job, true); WordCountUtil.setMeta(job); JobClient.runJob(job); WordCountUtil.validateCountsFile(); }
From source file:com.tomslabs.grid.avro.TextTypedBytesToAvroOutputFormat.java
License:Apache License
/** Enable output compression using the deflate codec and specify its level. */ public static void setDeflateLevel(JobConf job, int level) { FileOutputFormat.setCompressOutput(job, true); job.setInt(DEFLATE_LEVEL_KEY, level); }
From source file:de.tudarmstadt.lt.n2n.hadoop.FlipJoBims.java
License:Apache License
public static void main(String[] args) throws Exception { JobConf conf = new JobConf(FlipJoBims.class); /* begin necessary for UKP cluster */ conf.setMemoryForMapTask(1000L); // 1 GB /* necessary for UKP cdh3 */ conf.setMemoryForReduceTask(1000L); // 1 GB /* necessary for UKP cdh3 */ FileOutputFormat.setCompressOutput(conf, true); // compress output FileOutputFormat.setOutputCompressorClass(conf, org.apache.hadoop.io.compress.BZip2Codec.class); /* use the bzip2 codec for compression */ conf.setCompressMapOutput(true); // compress mapper output /* end necessary for UKP cluster */ conf.setJobName(FlipJoBims.class.getSimpleName()); args = new GenericOptionsParser(conf, args).getRemainingArgs(); conf.setInputFormat(KeyValueTextInputFormat.class); conf.setOutputFormat(TextOutputFormat.class); conf.setMapperClass(FlipJoBims.Map.class); conf.setNumReduceTasks(0);/* www. j a v a2 s. co m*/ // conf.setReducerClass(IdentityReducer.class); conf.setMapOutputKeyClass(Text.class); conf.setOutputKeyClass(Text.class); conf.setMapOutputValueClass(Text.class); conf.setOutputValueClass(Text.class); FileInputFormat.setInputPaths(conf, new Path(args[0])); FileOutputFormat.setOutputPath(conf, new Path(args[1])); JobClient.runJob(conf); }
From source file:de.tudarmstadt.ukp.dkpro.bigdata.io.hadoop.XCASSequenceFileWriter.java
License:Apache License
@SuppressWarnings("deprecation") @Override/*from w ww . j av a 2 s.c o m*/ public void initialize(org.apache.uima.UimaContext context) throws org.apache.uima.resource.ResourceInitializationException { super.initialize(context); final JobConf conf = new JobConf(); this.path = new File((String) context.getConfigParameterValue(PARAM_PATH)); conf.set("fs.default.name", this.fileSystemName); // Compress Map output if (this.compress) { System.out.println("compressing"); conf.set("mapred.compress.map.output", "true"); conf.set("mapred.map.output.compression.codec", "org.apache.hadoop.io.compress.SnappyCodec"); } // Compress MapReduce output conf.set("mapred.output.compress", "true"); conf.set("mapred.output.compression", "org.apache.hadoop.io.compress.SnappyCodec"); FileOutputFormat.setCompressOutput(conf, true); // FileOutputFormat.setOutputCompressorClass(conf, SnappyCodec.class); final String filename = this.path + "/" + "part-00000"; try { final FileSystem fs = FileSystem.get(URI.create(filename), conf); final Path path = new Path(URI.create(filename).toString()); this.writer = SequenceFile.createWriter(fs, conf, path, Text.class, Text.class); } catch (final IOException e) { throw new ResourceInitializationException(); } }
From source file:edu.ub.ahstfg.indexer.Indexer.java
License:Open Source License
@Override public int run(String[] arg0) throws Exception { LOG.info("Creating Hadoop job for Indexer."); JobConf job = new JobConf(getConf()); job.setJarByClass(Indexer.class); LOG.info("Setting input path to '" + INPUT_PATH + "'"); FileInputFormat.setInputPaths(job, new Path(INPUT_PATH)); // Set filters if it's necessary. LOG.info("Clearing the output path at '" + OUTPUT_PATH + "'"); // Change URI to Path if it's necessary. FileSystem fs = FileSystem.get(new URI(OUTPUT_PATH), job); if (fs.exists(new Path(OUTPUT_PATH))) { fs.delete(new Path(OUTPUT_PATH), true); }//from w w w . ja v a 2 s . com LOG.info("Setting output path to '" + OUTPUT_PATH + "'"); FileOutputFormat.setOutputPath(job, new Path(OUTPUT_PATH)); FileOutputFormat.setCompressOutput(job, false); LOG.info("Setting input format."); job.setInputFormat(ArcInputFormat.class); LOG.info("Setting output format."); job.setOutputFormat(IndexOutputFormat.class); LOG.info("Setting output data types."); job.setOutputKeyClass(Text.class); job.setOutputValueClass(IndexRecord.class); LOG.info("Setting mapper and reducer."); job.setMapperClass(IndexerMapper.class); job.setMapOutputValueClass(ParsedDocument.class); job.setReducerClass(IndexerReducer.class); if (JobClient.runJob(job).isSuccessful()) { return 0; } else { return 1; } }
From source file:edu.ub.ahstfg.indexer.wordcount.WordCount.java
License:Open Source License
@Override public int run(String[] args) throws Exception { LOG.info("Creating Hadoop job for ARC input files word count."); JobConf job = new JobConf(getConf()); job.setJarByClass(WordCount.class); LOG.info("Setting input path to '" + inputPath + "'"); FileInputFormat.setInputPaths(job, new Path(inputPath)); // Set filters if it's necessary. LOG.info("Clearing the output path at '" + outputPath + "'"); // Change URI to Path if it's necessary. FileSystem fs = FileSystem.get(new URI(outputPath), job); if (fs.exists(new Path(outputPath))) { fs.delete(new Path(outputPath), true); }/* w ww .ja va 2s . co m*/ LOG.info("Setting output path to '" + outputPath + "'"); FileOutputFormat.setOutputPath(job, new Path(outputPath)); FileOutputFormat.setCompressOutput(job, false); LOG.info("Setting input format."); // job.setInputFormat(TextInputFormat.class); job.setInputFormat(ArcInputFormat.class); LOG.info("Setting output format."); job.setOutputFormat(TextOutputFormat.class); LOG.info("Setting output data types."); job.setOutputKeyClass(Text.class); job.setOutputValueClass(LongWritable.class); LOG.info("Setting mapper and reducer."); // job.setMapperClass(WordCountTextInputMapper.class); job.setMapperClass(WordCountArcInputMapper.class); job.setReducerClass(LongSumReducer.class); if (JobClient.runJob(job).isSuccessful()) { return 0; } else { return 1; } }
From source file:edu.ub.ahstfg.kmeans.KmeansIteration.java
License:Open Source License
@Override public int run(String[] args) throws IOException, URISyntaxException { job = new JobConf(getConf()); params.toJobConf(job);//from w w w . j a va 2 s. com job.setInt(ParamSet.N_ITERATION, nIter); LOG.info("Iteration " + nIter + " > Iniciating Kmeans iteration " + nIter); job.setJarByClass(KmeansIteration.class); // TODO may change for Clusterizer.class LOG.info("Iteration " + nIter + " > Setting input path to '" + inputPath + "'"); FileInputFormat.setInputPaths(job, new Path(inputPath)); FileInputFormat.setInputPathFilter(job, SampleFilter.class); LOG.info("Iteration " + nIter + " > Clearing the output path at '" + outputPath + "'"); FileSystem fs = FileSystem.get(new URI(outputPath), job); if (fs.exists(new Path(outputPath))) { fs.delete(new Path(outputPath), true); } LOG.info("Iteration " + nIter + " > Setting output path to '" + outputPath + "'"); FileOutputFormat.setOutputPath(job, new Path(outputPath)); FileOutputFormat.setCompressOutput(job, false); LOG.info("Iteration " + nIter + " > Setting input format."); job.setInputFormat(IndexInputFormat.class); LOG.info("Iteration " + nIter + " > Setting output format."); job.setOutputFormat(TextOutputFormat.class); LOG.info("Iteration " + nIter + " > Setting output data types."); job.setOutputKeyClass(Text.class); job.setOutputValueClass(IntWritable.class); LOG.info("Iteration " + nIter + " > Setting mapper and reducer."); job.setMapperClass(KmeansMapper.class); job.setMapOutputKeyClass(IntWritable.class); job.setMapOutputValueClass(DocumentDistance.class); job.setReducerClass(KmeansReducer.class); LOG.info("Iteration " + nIter + " > Running job..."); boolean done = JobClient.runJob(job).isSuccessful(); LOG.info("Iteration " + nIter + " > Job done."); if (done) { return 0; } else { return 1; } }