List of usage examples for org.apache.hadoop.mapreduce Job getConfiguration
public Configuration getConfiguration()
From source file:com.blackberry.logdriver.util.FastSearch.java
License:Apache License
@Override public int run(String[] args) throws Exception { Configuration conf = getConf(); // Configuration processed by ToolRunner // If run by Oozie, then load the Oozie conf too if (System.getProperty("oozie.action.conf.xml") != null) { conf.addResource(new URL("file://" + System.getProperty("oozie.action.conf.xml"))); }// www. ja v a2s . c o m FileSystem fs = FileSystem.get(conf); // The command line options String searchString = null; List<Path> paths = new ArrayList<Path>(); Path outputDir = null; // Load input files from the command line if (args.length < 3) { System.out.println("usage: [genericOptions] searchString input [input ...] output"); System.exit(1); } // Get the files we need from the command line. searchString = args[0]; for (int i = 1; i < args.length - 1; i++) { for (FileStatus f : fs.globStatus(new Path(args[i]))) { paths.add(f.getPath()); } } outputDir = new Path(args[args.length - 1]); @SuppressWarnings("deprecation") Job job = new Job(conf); Configuration jobConf = job.getConfiguration(); job.setJarByClass(FastSearch.class); jobConf.setIfUnset("mapred.job.name", "Search Files"); // To propagate credentials within Oozie if (System.getenv("HADOOP_TOKEN_FILE_LOCATION") != null) { jobConf.set("mapreduce.job.credentials.binary", System.getenv("HADOOP_TOKEN_FILE_LOCATION")); } // Good output separators include things that are unsupported by XML. So we // just send the byte value of the character through. The restriction here // is that it can't be more than 1 byte when UTF-8 encoded, since it will be // read by Pig which only deals with single byte separators. { String outputSeparator = jobConf.get("logdriver.output.field.separator", DEFAULT_OUTPUT_SEPARATOR); byte[] bytes = outputSeparator.getBytes(UTF_8); if (bytes.length != 1) { LOG.error("The output separator must be a single byte in UTF-8."); return 1; } jobConf.set("logdriver.output.field.separator", Byte.toString(bytes[0])); } jobConf.set("logdriver.search.string", Base64.encodeBase64String(searchString.getBytes("UTF-8"))); job.setInputFormatClass(AvroBlockInputFormat.class); job.setMapperClass(SearchMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(NullWritable.class); job.setNumReduceTasks(0); // And set the output as usual job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); for (Path path : paths) { AvroBlockInputFormat.addInputPath(job, path); } // Run the job. if (conf.getBoolean("job.wait", DEFAULT_WAIT_JOB)) { return job.waitForCompletion(true) ? 0 : 1; } else { job.submit(); return 0; } }
From source file:com.blackberry.logdriver.util.Grep.java
License:Apache License
@Override public int run(String[] args) throws Exception { Configuration conf = getConf(); // Configuration processed by ToolRunner // If run by Oozie, then load the Oozie conf too if (System.getProperty("oozie.action.conf.xml") != null) { conf.addResource(new URL("file://" + System.getProperty("oozie.action.conf.xml"))); }//from www .ja v a2 s . com FileSystem fs = FileSystem.get(conf); // The command line options String regex = null; List<Path> paths = new ArrayList<Path>(); Path outputDir = null; // Load input files from the command line if (args.length < 3) { System.out.println("usage: [genericOptions] regex input [input ...] output"); System.exit(1); } // Get the files we need from the command line. regex = args[0]; for (int i = 1; i < args.length - 1; i++) { for (FileStatus f : fs.globStatus(new Path(args[i]))) { paths.add(f.getPath()); } } outputDir = new Path(args[args.length - 1]); @SuppressWarnings("deprecation") Job job = new Job(conf); Configuration jobConf = job.getConfiguration(); job.setJarByClass(Grep.class); jobConf.setIfUnset("mapred.job.name", "Grep Files"); // To propagate credentials within Oozie if (System.getenv("HADOOP_TOKEN_FILE_LOCATION") != null) { jobConf.set("mapreduce.job.credentials.binary", System.getenv("HADOOP_TOKEN_FILE_LOCATION")); } // Good output separators include things that are unsupported by XML. So we // just send the byte value of the character through. The restriction here // is that it can't be more than 1 byte when UTF-8 encoded, since it will be // read by Pig which only deals with single byte separators. { String outputSeparator = jobConf.get("logdriver.output.field.separator", DEFAULT_OUTPUT_SEPARATOR); byte[] bytes = outputSeparator.getBytes(UTF_8); if (bytes.length != 1) { LOG.error("The output separator must be a single byte in UTF-8."); return 1; } jobConf.set("logdriver.output.field.separator", Byte.toString(bytes[0])); } jobConf.set("logdriver.grep.regex", Base64.encodeBase64String(regex.getBytes("UTF-8"))); job.setInputFormatClass(BoomInputFormat.class); job.setMapperClass(GrepMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(NullWritable.class); job.setNumReduceTasks(0); // And set the output as usual job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); for (Path path : paths) { BoomInputFormat.addInputPath(job, path); } // Run the job. if (conf.getBoolean("job.wait", DEFAULT_WAIT_JOB)) { return job.waitForCompletion(true) ? 0 : 1; } else { job.submit(); return 0; } }
From source file:com.blackberry.logdriver.util.MultiSearch.java
License:Apache License
@Override public int run(String[] args) throws Exception { Configuration conf = getConf(); // Configuration processed by ToolRunner // If run by Oozie, then load the Oozie conf too if (System.getProperty("oozie.action.conf.xml") != null) { conf.addResource(new URL("file://" + System.getProperty("oozie.action.conf.xml"))); }/*ww w . j a v a2 s . c om*/ FileSystem fs = FileSystem.get(conf); // The command line options String searchStringDir = null; List<Path> paths = new ArrayList<Path>(); Path outputDir = null; // Load input files from the command line if (args.length < 3) { System.out.println("usage: [genericOptions] searchStringDirectory input [input ...] output"); System.exit(1); } // Get the files we need from the command line. searchStringDir = args[0]; // We are going to be reading all the files in this directory a lot. So // let's up the replication factor by a lot so that they're easy to read. for (FileStatus f : fs.listStatus(new Path(searchStringDir))) { fs.setReplication(f.getPath(), (short) 16); } for (int i = 1; i < args.length - 1; i++) { for (FileStatus f : fs.globStatus(new Path(args[i]))) { paths.add(f.getPath()); } } outputDir = new Path(args[args.length - 1]); @SuppressWarnings("deprecation") Job job = new Job(conf); Configuration jobConf = job.getConfiguration(); job.setJarByClass(MultiSearch.class); jobConf.setIfUnset("mapred.job.name", "MultiSearch"); // To propagate credentials within Oozie if (System.getenv("HADOOP_TOKEN_FILE_LOCATION") != null) { jobConf.set("mapreduce.job.credentials.binary", System.getenv("HADOOP_TOKEN_FILE_LOCATION")); } // Good output separators include things that are unsupported by XML. So we // just send the byte value of the character through. The restriction here // is that it can't be more than 1 byte when UTF-8 encoded, since it will be // read by Pig which only deals with single byte separators. { String outputSeparator = jobConf.get("logdriver.output.field.separator", DEFAULT_OUTPUT_SEPARATOR); byte[] bytes = outputSeparator.getBytes(UTF_8); if (bytes.length != 1) { LOG.error("The output separator must be a single byte in UTF-8."); return 1; } jobConf.set("logdriver.output.field.separator", Byte.toString(bytes[0])); } jobConf.set("logdriver.search.string.dir", searchStringDir); // This search is generally too fast to make good use of 128MB blocks, so // let's set the value to 256MB (if it's not set already) if (jobConf.get("mapred.max.split.size") == null) { jobConf.setLong("mapred.max.split.size", 256 * 1024 * 1024); } job.setInputFormatClass(AvroBlockInputFormat.class); job.setMapperClass(SearchMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(NullWritable.class); job.setNumReduceTasks(0); job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); for (Path path : paths) { AvroBlockInputFormat.addInputPath(job, path); } // Run the job. if (conf.getBoolean("job.wait", DEFAULT_WAIT_JOB)) { return job.waitForCompletion(true) ? 0 : 1; } else { job.submit(); return 0; } }
From source file:com.blackberry.logdriver.util.Search.java
License:Apache License
@Override public int run(String[] args) throws Exception { Configuration conf = getConf(); // Configuration processed by ToolRunner // If run by Oozie, then load the Oozie conf too if (System.getProperty("oozie.action.conf.xml") != null) { conf.addResource(new URL("file://" + System.getProperty("oozie.action.conf.xml"))); }// w w w. j a v a 2 s .c o m FileSystem fs = FileSystem.get(conf); // The command line options String searchString = null; List<Path> paths = new ArrayList<Path>(); Path outputDir = null; // Load input files from the command line if (args.length < 3) { System.out.println("usage: [genericOptions] searchString input [input ...] output"); System.exit(1); } // Get the files we need from the command line. searchString = args[0]; for (int i = 1; i < args.length - 1; i++) { for (FileStatus f : fs.globStatus(new Path(args[i]))) { paths.add(f.getPath()); } } outputDir = new Path(args[args.length - 1]); @SuppressWarnings("deprecation") Job job = new Job(conf); Configuration jobConf = job.getConfiguration(); job.setJarByClass(Search.class); jobConf.setIfUnset("mapred.job.name", "Search Files"); // To propagate credentials within Oozie if (System.getenv("HADOOP_TOKEN_FILE_LOCATION") != null) { jobConf.set("mapreduce.job.credentials.binary", System.getenv("HADOOP_TOKEN_FILE_LOCATION")); } // Good output separators include things that are unsupported by XML. So we // just send the byte value of the character through. The restriction here // is that it can't be more than 1 byte when UTF-8 encoded, since it will be // read by Pig which only deals with single byte separators. { String outputSeparator = jobConf.get("logdriver.output.field.separator", DEFAULT_OUTPUT_SEPARATOR); byte[] bytes = outputSeparator.getBytes(UTF_8); if (bytes.length != 1) { LOG.error("The output separator must be a single byte in UTF-8."); return 1; } jobConf.set("logdriver.output.field.separator", Byte.toString(bytes[0])); } jobConf.set("logdriver.search.string", searchString); job.setInputFormatClass(BoomInputFormat.class); job.setMapperClass(SearchMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(NullWritable.class); job.setNumReduceTasks(0); // And set the output as usual job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); for (Path path : paths) { BoomInputFormat.addInputPath(job, path); } // Run the job. if (conf.getBoolean("job.wait", DEFAULT_WAIT_JOB)) { return job.waitForCompletion(true) ? 0 : 1; } else { job.submit(); return 0; } }
From source file:com.bol.crazypigs.HBaseStorage15.java
License:Apache License
@Override public void setLocation(String location, Job job) throws IOException { Properties udfProps = getUDFProperties(); job.getConfiguration().setBoolean("pig.noSplitCombination", true); m_conf = initializeLocalJobConfig(job); String delegationTokenSet = udfProps.getProperty(HBASE_TOKEN_SET); if (delegationTokenSet == null) { addHBaseDelegationToken(m_conf, job); udfProps.setProperty(HBASE_TOKEN_SET, "true"); }//from w w w . ja v a 2 s. c o m String tablename = location; if (location.startsWith("hbase://")) { tablename = location.substring(8); } m_conf.set(TableInputFormat.INPUT_TABLE, tablename); String projectedFields = udfProps.getProperty(projectedFieldsName()); if (projectedFields != null) { // update columnInfo_ pushProjection((RequiredFieldList) ObjectSerializer.deserialize(projectedFields)); } addFiltersWithoutColumnPrefix(columnInfo_); if (requiredFieldList != null) { Properties p = UDFContext.getUDFContext().getUDFProperties(this.getClass(), new String[] { contextSignature }); p.setProperty(contextSignature + "_projectedFields", ObjectSerializer.serialize(requiredFieldList)); } }
From source file:com.bol.crazypigs.HBaseStorage15.java
License:Apache License
private JobConf initializeLocalJobConfig(Job job) { Properties udfProps = getUDFProperties(); Configuration jobConf = job.getConfiguration(); JobConf localConf = new JobConf(jobConf); if (udfProps.containsKey(HBASE_CONFIG_SET)) { for (Entry<Object, Object> entry : udfProps.entrySet()) { localConf.set((String) entry.getKey(), (String) entry.getValue()); }/*ww w .j a va 2s. c o m*/ } else { Configuration hbaseConf = HBaseConfiguration.create(); for (Entry<String, String> entry : hbaseConf) { // JobConf may have some conf overriding ones in hbase-site.xml // So only copy hbase config not in job config to UDFContext // Also avoids copying core-default.xml and core-site.xml // props in hbaseConf to UDFContext which would be redundant. if (jobConf.get(entry.getKey()) == null) { udfProps.setProperty(entry.getKey(), entry.getValue()); localConf.set(entry.getKey(), entry.getValue()); } } udfProps.setProperty(HBASE_CONFIG_SET, "true"); } return localConf; }
From source file:com.bol.crazypigs.HBaseStorage15.java
License:Apache License
@Override public void setStoreLocation(String location, Job job) throws IOException { if (location.startsWith("hbase://")) { job.getConfiguration().set(TableOutputFormat.OUTPUT_TABLE, location.substring(8)); } else {// ww w . j a v a 2s.c o m job.getConfiguration().set(TableOutputFormat.OUTPUT_TABLE, location); } String serializedSchema = getUDFProperties().getProperty(contextSignature + "_schema"); if (serializedSchema != null) { schema_ = (ResourceSchema) ObjectSerializer.deserialize(serializedSchema); } m_conf = initializeLocalJobConfig(job); // Not setting a udf property and getting the hbase delegation token // only once like in setLocation as setStoreLocation gets different Job // objects for each call and the last Job passed is the one that is // launched. So we end up getting multiple hbase delegation tokens. addHBaseDelegationToken(m_conf, job); }
From source file:com.bonc.mr_roamRecognition_hjpt.comm.NewFileOutputFormat.java
License:Apache License
/** * Set whether the output of the job is compressed. * /*ww w . java 2 s . co m*/ * @param job * the job to modify * @param compress * should the output of the job be compressed? */ public static void setCompressOutput(Job job, boolean compress) { job.getConfiguration().setBoolean(FileOutputFormat.COMPRESS, compress); }
From source file:com.bonc.mr_roamRecognition_hjpt.comm.NewFileOutputFormat.java
License:Apache License
/** * Set the {@link CompressionCodec} to be used to compress job outputs. * //from w w w .j a va 2 s . co m * @param job * the job to modify * @param codecClass * the {@link CompressionCodec} to be used to compress the job * outputs */ public static void setOutputCompressorClass(Job job, Class<? extends CompressionCodec> codecClass) { setCompressOutput(job, true); job.getConfiguration().setClass(FileOutputFormat.COMPRESS_CODEC, codecClass, CompressionCodec.class); }
From source file:com.bonc.mr_roamRecognition_hjpt.comm.NewFileOutputFormat.java
License:Apache License
/** * Set the {@link Path} of the output directory for the map-reduce job. * * @param job/*from w w w .j a va 2s .c om*/ * The job to modify * @param outputDir * the {@link Path} of the output directory for the map-reduce * job. */ public static void setOutputPath(Job job, Path outputDir) { try { outputDir = outputDir.getFileSystem(job.getConfiguration()).makeQualified(outputDir); } catch (IOException e) { // Throw the IOException as a RuntimeException to be compatible with // MR1 throw new RuntimeException(e); } job.getConfiguration().set(FileOutputFormat.OUTDIR, outputDir.toString()); }