List of usage examples for org.apache.hadoop.fs FileSystem get
public static FileSystem get(Configuration conf) throws IOException
From source file:StripesPMI_nocombiner.java
License:Apache License
/** * Runs this tool./*w w w. ja va 2s . c o m*/ */ @SuppressWarnings({ "static-access" }) public int run(String[] args) throws Exception { Options options = new Options(); options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("input path").create(INPUT)); options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("output path").create(OUTPUT)); options.addOption(OptionBuilder.withArgName("num").hasArg().withDescription("number of reducers") .create(NUM_REDUCERS)); 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) || !cmdline.hasOption(OUTPUT)) { System.out.println("args: " + Arrays.toString(args)); HelpFormatter formatter = new HelpFormatter(); formatter.setWidth(120); formatter.printHelp(this.getClass().getName(), options); ToolRunner.printGenericCommandUsage(System.out); return -1; } String inputPath = cmdline.getOptionValue(INPUT); String outputPath = cmdline.getOptionValue(OUTPUT) + "_TMP";// cmdline.getOptionValue(OUTPUT); int reduceTasks = cmdline.hasOption(NUM_REDUCERS) ? Integer.parseInt(cmdline.getOptionValue(NUM_REDUCERS)) : 1; LOG.info("Tool: " + StripesPMI_nocombiner.class.getSimpleName()); LOG.info(" - input path: " + inputPath); LOG.info(" - output path: " + outputPath); LOG.info(" - number of reducers: " + reduceTasks); Job job_first = Job.getInstance(getConf()); job_first.setJobName(StripesPMI_nocombiner.class.getSimpleName()); job_first.setJarByClass(StripesPMI_nocombiner.class); // Delete the output directory if it exists already. Path outputDir = new Path(outputPath); FileSystem.get(getConf()).delete(outputDir, true); job_first.setNumReduceTasks(reduceTasks); FileInputFormat.setInputPaths(job_first, new Path(inputPath)); FileOutputFormat.setOutputPath(job_first, new Path(outputPath)); job_first.setMapOutputKeyClass(Text.class); job_first.setMapOutputValueClass(String2IntOpenHashMapWritable.class); job_first.setOutputKeyClass(PairOfStrings.class);// Text.class);// PairOfStrings.class); job_first.setOutputValueClass(DoubleWritable.class); job_first.setOutputFormatClass(TextOutputFormat.class);// changed job_first.setMapperClass(MyMapper_first.class); // job_first.setCombinerClass(MyCombiner.class); job_first.setReducerClass(MyReducer_first.class); long startTime = System.currentTimeMillis(); job_first.waitForCompletion(true); // ////////////////START.: run the second MR job to just aggregate result//////////////// inputPath = outputPath;// cmdline.getOptionValue(INPUT); outputPath = cmdline.getOptionValue(OUTPUT); Job job_second = Job.getInstance(getConf()); job_second.setJobName(StripesPMI_nocombiner.class.getSimpleName()); job_second.setJarByClass(StripesPMI_nocombiner.class); // Delete the output directory if it exists already. outputDir = new Path(outputPath); FileSystem.get(getConf()).delete(outputDir, true); job_second.setNumReduceTasks(reduceTasks); FileInputFormat.setInputPaths(job_second, new Path(inputPath)); FileOutputFormat.setOutputPath(job_second, new Path(outputPath)); job_second.setMapOutputKeyClass(Text.class); job_second.setMapOutputValueClass(DoubleWritable.class); job_second.setOutputKeyClass(Text.class);// PairOfStrings.class); job_second.setOutputValueClass(DoubleWritable.class); // job_second.setOutputFormatClass(TextOutputFormat.class);// changed job_second.setMapperClass(MyMapper_second.class); // job_second.setCombinerClass(MyCombiner.class); job_second.setReducerClass(MyReducer_second.class); job_second.waitForCompletion(true); // END//////////// System.out.println("Job Finished in " + (System.currentTimeMillis() - startTime) / 1000.0 + " seconds"); return 0; }
From source file:BuildInvertedIndex.java
License:Apache License
/** * Runs this tool.// w ww . j a v a2 s . c om */ @SuppressWarnings({ "static-access" }) public int run(String[] args) throws Exception { Options options = new Options(); options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("input path").create(INPUT)); options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("output path").create(OUTPUT)); options.addOption(OptionBuilder.withArgName("num").hasArg().withDescription("number of reducers") .create(NUM_REDUCERS)); 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) || !cmdline.hasOption(OUTPUT)) { System.out.println("args: " + Arrays.toString(args)); HelpFormatter formatter = new HelpFormatter(); formatter.setWidth(120); formatter.printHelp(this.getClass().getName(), options); ToolRunner.printGenericCommandUsage(System.out); return -1; } String inputPath = cmdline.getOptionValue(INPUT); String outputPath = cmdline.getOptionValue(OUTPUT); int reduceTasks = cmdline.hasOption(NUM_REDUCERS) ? Integer.parseInt(cmdline.getOptionValue(NUM_REDUCERS)) : 1; LOG.info("Tool name: " + BuildInvertedIndex.class.getSimpleName()); LOG.info(" - input path: " + inputPath); LOG.info(" - output path: " + outputPath); LOG.info(" - num reducers: " + reduceTasks); Job job = Job.getInstance(getConf()); job.setJobName(BuildInvertedIndex.class.getSimpleName()); job.setJarByClass(BuildInvertedIndex.class); job.setNumReduceTasks(reduceTasks); FileInputFormat.setInputPaths(job, new Path(inputPath)); FileOutputFormat.setOutputPath(job, new Path(outputPath)); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(PairOfInts.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(PairOfWritables.class); job.setOutputFormatClass(TextOutputFormat.class); job.setMapperClass(MyMapper.class); job.setReducerClass(MyReducer.class); // Delete the output directory if it exists already. Path outputDir = new Path(outputPath); FileSystem.get(getConf()).delete(outputDir, true); long startTime = System.currentTimeMillis(); job.waitForCompletion(true); System.out.println("Job Finished in " + (System.currentTimeMillis() - startTime) / 1000.0 + " seconds"); return 0; }
From source file:FDFGenData.java
License:Open Source License
public static void testwritefile(String tabledir, int num) throws Exception { String rawtmp = "/tmp/raw/rawfile"; FileSystem fs = FileSystem.get(new Configuration()); FileStatus[] fss = fs.listStatus(new Path(tabledir)); int x = 0;//from w w w . ja v a 2 s .c om if (fss != null) { x = fss.length; } PT.testgenrawfiler(rawtmp, num); PT.testwritefdf(tabledir + "file" + (x + 1), rawtmp, false, (short) -1); PT.testgenrawfiler(rawtmp, num); PT.testwritefdf(tabledir + "file" + (x + 2), rawtmp, false, (short) -1); PT.testgenrawfiler(rawtmp, num); PT.testwritefdf(tabledir + "file" + (x + 3), rawtmp, false, (short) -1); PT.testgenrawfiler(rawtmp, num); PT.testwritefdf(tabledir + "file" + (x + 4), rawtmp, false, (short) -1); PT.testgenrawfiler(rawtmp, num); PT.testwritefdf(tabledir + "file" + (x + 5), rawtmp, false, (short) -1); }
From source file:FormatStorageBasicTest.java
License:Open Source License
public void testPersistentLineIndexInfo() { try {// ww w.j a v a 2 s . com String fileName = prefix + "testPersistentLineIndexInfo"; Path path = new Path(fileName); FileSystem fs = FileSystem.get(new Configuration()); FSDataOutputStream out = fs.create(path); IndexInfo info = new IndexInfo(); info.beginLine = 11; info.endLine = 22; info.offset = 33; info.len = 44; info.idx = 55; info.persistentLineIndexInfo(out); out.close(); FSDataInputStream in = fs.open(path); int beginLine = in.readInt(); int endLine = in.readInt(); long offset = in.readLong(); long len = in.readLong(); int idx = in.readInt(); in.close(); if (beginLine != 11) { fail("beginLine fail:" + beginLine); } if (endLine != 22) { fail("endLine fail:" + endLine); } if (offset != 33) { fail("offset fail:" + offset); } if (len != 44) { fail("len fail:" + len); } if (idx != 55) { fail("idx fail:" + idx); } } catch (IOException e) { fail(e.getMessage()); } }
From source file:FormatStorageBasicTest.java
License:Open Source License
public void testPersistentKeyIndexInfo() { try {// w ww . j a v a 2s. c om String fileName = prefix + "testPersistentKeyIndexInfo"; Path path = new Path(fileName); FileSystem fs = FileSystem.get(new Configuration()); FSDataOutputStream out = fs.create(path); IndexInfo info = new IndexInfo(); info.beginKey = 111; info.endKey = 222; info.persistentKeyIndexInfo(out); out.close(); FSDataInputStream in = fs.open(path); int beginKey = in.readInt(); int endKey = in.readInt(); in.close(); if (beginKey != 111) { fail("beginKey fail:" + beginKey); } if (endKey != 222) { fail("beginKey fail:" + beginKey); } } catch (IOException e) { fail(e.getMessage()); } }
From source file:FormatStorageBasicTest.java
License:Open Source License
public void testUnpersistentLineIndexInfo() { try {// ww w. j a v a 2s. com String fileName = prefix + "testPersistentLineIndexInfo"; Path path = new Path(fileName); FileSystem fs = FileSystem.get(new Configuration()); FSDataInputStream in = fs.open(path); IndexInfo info = new IndexInfo(); info.unpersistentLineIndexInfo(in); in.close(); if (info.beginLine != 11) { fail("beginLine fail:" + info.beginLine); } if (info.endLine != 22) { fail("endLine fail:" + info.endLine); } if (info.offset != 33) { fail("offset fail:" + info.offset); } if (info.len != 44) { fail("len fail:" + info.len); } if (info.idx != 55) { fail("idx fail:" + info.idx); } fs.close(); } catch (IOException e) { fail(e.getMessage()); } }
From source file:FormatStorageBasicTest.java
License:Open Source License
public void testUnpersistentKeyIndexInfo() { try {/*ww w . j av a 2 s. co m*/ String fileName = prefix + "testPersistentKeyIndexInfo"; Path path = new Path(fileName); FileSystem fs = FileSystem.get(new Configuration()); FSDataInputStream in = fs.open(path); IndexInfo info = new IndexInfo(); info.unpersistentKeyIndexInfo(in); in.close(); if (info.beginKey != 111) { fail("beginKey fail:" + info.beginKey); } if (info.endKey != 222) { fail("beginKey fail:" + info.beginKey); } fs.close(); } catch (IOException e) { fail(e.getMessage()); } }
From source file:FormatStorageBasicTest.java
License:Open Source License
public void testPersistentLineIndex() { try {/* w w w .java2 s . c o m*/ BlockIndex index = new BlockIndex(); for (int i = 0; i < 10; i++) { IndexInfo info = new IndexInfo(); info.beginLine = i; info.endLine = i + 10; info.offset = i; info.len = i + 20; info.idx = i; index.addIndexInfo(info, ConstVar.LineMode); } String fileName = prefix + "testPersistentLineIndex"; Path path = new Path(fileName); FileSystem fs = FileSystem.get(new Configuration()); FSDataOutputStream out = fs.create(path); index.persistent(out); out.close(); FSDataInputStream in = fs.open(path); for (int i = 0; i < 10; i++) { int beginLine = in.readInt(); int endLine = in.readInt(); long offset = in.readLong(); long len = in.readLong(); int idx = in.readInt(); if (beginLine != i) { fail("beginLine fail:" + beginLine); } if (endLine != i + 10) { fail("endLine fail:" + endLine); } if (offset != i) { fail("offset fail:" + offset); } if (len != i + 20) { fail("len fail:" + len); } if (idx != i) { fail("idx fail:" + idx); } } } catch (IOException e) { fail(e.getMessage()); } }
From source file:FormatStorageBasicTest.java
License:Open Source License
public void testPersistentKeyIndex() { try {// www . j a va 2 s . co m BlockIndex index = new BlockIndex(); for (int i = 0; i < 10; i++) { IndexInfo info = new IndexInfo(); info.beginKey = i; info.endKey = i + 10; index.addIndexInfo(info, ConstVar.KeyMode); } String fileName = prefix + "testPersistentKeyIndex"; Path path = new Path(fileName); FileSystem fs = FileSystem.get(new Configuration()); FSDataOutputStream out = fs.create(path); index.persistent(out); out.close(); FSDataInputStream in = fs.open(path); for (int i = 0; i < 10; i++) { int beginKey = in.readInt(); int endKey = in.readInt(); if (beginKey != i) { fail("beginKey fail:" + beginKey); } if (endKey != i + 10) { fail("endKey fail:" + endKey); } } } catch (IOException e) { fail(e.getMessage()); } }
From source file:FormatStorageBasicTest.java
License:Open Source License
public void testPersistentField() throws IOException { {/*from w w w . j a v a2 s.com*/ String file = prefix + "testPersistentField"; Path path = new Path(file); FileSystem fs = FileSystem.get(new Configuration()); FSDataOutputStream out = fs.create(path); Field field = new Field(ConstVar.FieldType_Byte, ConstVar.Sizeof_Byte, (short) 1); field.persistent(out); if (out.getPos() != 7) { fail("error out.pos:" + out.getPos()); } out.close(); FSDataInputStream in = fs.open(path); byte type = in.readByte(); int len = in.readInt(); short idx = in.readShort(); if (type != ConstVar.FieldType_Byte) { fail("fail type:" + type); } if (len != ConstVar.Sizeof_Byte) { fail("fail len:" + len); } if (idx != 1) { fail("fail idx:" + idx); } } { } { } }