List of usage examples for org.apache.hadoop.mapreduce Job getConfiguration
public Configuration getConfiguration()
From source file:co.nubetech.hiho.job.DBQueryInputJob.java
License:Apache License
public void runJobs(Configuration conf, int jobCounter) throws IOException { try {/*from w ww . j a v a2 s . c om*/ checkMandatoryConfs(conf); } catch (HIHOException e1) { e1.printStackTrace(); throw new IOException(e1); } Job job = new Job(conf); for (Entry<String, String> entry : conf) { logger.warn("key, value " + entry.getKey() + "=" + entry.getValue()); } // logger.debug("Number of maps " + // conf.getInt("mapred.map.tasks", 1)); // conf.setInt(JobContext.NUM_MAPS, // conf.getInt("mapreduce.job.maps", 1)); // job.getConfiguration().setInt("mapred.map.tasks", 4); job.getConfiguration().setInt(MRJobConfig.NUM_MAPS, conf.getInt(HIHOConf.NUMBER_MAPPERS, 1)); logger.warn("Number of maps " + conf.getInt(MRJobConfig.NUM_MAPS, 1)); job.setJobName("Import job"); job.setJarByClass(DBQueryInputJob.class); String strategy = conf.get(HIHOConf.INPUT_OUTPUT_STRATEGY); OutputStrategyEnum os = OutputStrategyEnum.value(strategy); if (os == null) { throw new IllegalArgumentException("Wrong value of output strategy. Please correct"); } if (os != OutputStrategyEnum.AVRO) { switch (os) { case DUMP: { // job.setMapperClass(DBImportMapper.class); break; } /* * case AVRO: { job.setMapperClass(DBInputAvroMapper.class); // * need avro in cp // job.setJarByClass(Schema.class); // need * jackson which is needed by avro - ugly! // * job.setJarByClass(ObjectMapper.class); * job.setMapOutputKeyClass(NullWritable.class); * job.setMapOutputValueClass(AvroValue.class); * job.setOutputKeyClass(NullWritable.class); * job.setOutputValueClass(AvroValue.class); * job.setOutputFormatClass(AvroOutputFormat.class); * * AvroOutputFormat.setOutputPath(job, new * Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH))); break; } */ case DELIMITED: { job.setMapperClass(DBInputDelimMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(Text.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(Text.class); job.setOutputFormatClass(NoKeyOnlyValueOutputFormat.class); NoKeyOnlyValueOutputFormat.setOutputPath(job, new Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH))); } case JSON: { // job.setMapperClass(DBImportJsonMapper.class); // job.setJarByClass(ObjectMapper.class); break; } default: { job.setMapperClass(DBInputDelimMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(Text.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(Text.class); job.setOutputFormatClass(NoKeyOnlyValueOutputFormat.class); NoKeyOnlyValueOutputFormat.setOutputPath(job, new Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH))); break; } } String inputQuery = conf.get(DBConfiguration.INPUT_QUERY); String inputBoundingQuery = conf.get(DBConfiguration.INPUT_BOUNDING_QUERY); logger.debug("About to set the params"); DBQueryInputFormat.setInput(job, inputQuery, inputBoundingQuery, params); logger.debug("Set the params"); job.setNumReduceTasks(0); try { // job.setJarByClass(Class.forName(conf.get( // org.apache.hadoop.mapred.lib.db.DBConfiguration.DRIVER_CLASS_PROPERTY))); logger.debug("OUTPUT format class is " + job.getOutputFormatClass()); /* * org.apache.hadoop.mapreduce.OutputFormat<?, ?> output = * ReflectionUtils.newInstance(job.getOutputFormatClass(), * job.getConfiguration()); output.checkOutputSpecs(job); */ logger.debug("Class is " + ReflectionUtils .newInstance(job.getOutputFormatClass(), job.getConfiguration()).getClass().getName()); job.waitForCompletion(false); if (conf.get(HIHOConf.INPUT_OUTPUT_LOADTO) != null) { generateHiveScript(conf, job, jobCounter); generatePigScript(conf, job); } } /* * catch (HIHOException h) { h.printStackTrace(); } */ catch (Exception e) { e.printStackTrace(); } catch (HIHOException e) { e.printStackTrace(); } } // avro to be handled differently, thanks to all the incompatibilities // in the apis. else { String inputQuery = conf.get(DBConfiguration.INPUT_QUERY); String inputBoundingQuery = conf.get(DBConfiguration.INPUT_BOUNDING_QUERY); logger.debug("About to set the params"); // co.nubetech.apache.hadoop.mapred.DBQueryInputFormat.setInput(job, // inputQuery, inputBoundingQuery, params); logger.debug("Set the params"); JobConf jobConf = new JobConf(conf); try { GenericDBWritable queryWritable = getDBWritable(jobConf); Schema pair = DBMapper.getPairSchema(queryWritable.getColumns()); AvroJob.setMapOutputSchema(jobConf, pair); GenericRecordAvroOutputFormat.setOutputPath(jobConf, new Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH))); co.nubetech.apache.hadoop.mapred.DBQueryInputFormat.setInput(jobConf, inputQuery, inputBoundingQuery, params); jobConf.setInputFormat(co.nubetech.apache.hadoop.mapred.DBQueryInputFormat.class); jobConf.setMapperClass(DBInputAvroMapper.class); jobConf.setMapOutputKeyClass(NullWritable.class); jobConf.setMapOutputValueClass(AvroValue.class); jobConf.setOutputKeyClass(NullWritable.class); jobConf.setOutputValueClass(Text.class); jobConf.setOutputFormat(GenericRecordAvroOutputFormat.class); jobConf.setJarByClass(DBQueryInputJob.class); jobConf.setStrings("io.serializations", "org.apache.hadoop.io.serializer.JavaSerialization,org.apache.hadoop.io.serializer.WritableSerialization,org.apache.avro.mapred.AvroSerialization"); jobConf.setNumReduceTasks(0); /* * jobConf.setOutputFormat(org.apache.hadoop.mapred. * SequenceFileOutputFormat.class); * org.apache.hadoop.mapred.SequenceFileOutputFormat * .setOutputPath(jobConf, new * Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH))); */ JobClient.runJob(jobConf); } catch (Throwable e) { e.printStackTrace(); } } }
From source file:co.nubetech.hiho.job.ExportToDB.java
License:Apache License
public int run(String[] args) throws Exception { Configuration conf = getConf(); populateConfiguration(args, conf);/*from w w w .j a va 2 s .c om*/ try { checkMandatoryConfs(conf); } catch (HIHOException e1) { e1.printStackTrace(); throw new Exception(e1); } Job job = new Job(conf); job.getConfiguration().setInt(MRJobConfig.NUM_MAPS, conf.getInt(HIHOConf.NUMBER_MAPPERS, 1)); job.setJobName("HihoDBExport"); job.setMapperClass(GenericDBLoadDataMapper.class); job.setJarByClass(ExportToDB.class); job.setNumReduceTasks(0); job.setInputFormatClass(TextInputFormat.class); TextInputFormat.addInputPath(job, new Path(inputPath)); GenericDBOutputFormat.setOutput(job, tableName, columnNames); int ret = 0; try { ret = job.waitForCompletion(true) ? 0 : 1; } catch (Exception e) { e.printStackTrace(); } return ret; }
From source file:co.nubetech.hiho.job.ExportToOracleDb.java
License:Apache License
@Override public int run(String[] args) throws IOException { Configuration conf = getConf(); for (Entry<String, String> entry : conf) { logger.debug("key, value " + entry.getKey() + "=" + entry.getValue()); }//from www. jav a 2s. c o m for (int i = 0; i < args.length; i++) { logger.debug("Remaining arguments are" + " " + args[i]); } populateConfiguration(args, conf); try { checkMandatoryConfs(conf); } catch (HIHOException e1) { e1.printStackTrace(); throw new IOException(e1); } Job job = new Job(conf); job.setJobName("OracleLoading"); job.setMapperClass(OracleLoadMapper.class); job.setJarByClass(ExportToOracleDb.class); job.getConfiguration().setInt(MRJobConfig.NUM_MAPS, conf.getInt(HIHOConf.NUMBER_MAPPERS, 1)); try { // we first create the external table definition String query = conf.get(HIHOConf.EXTERNAL_TABLE_DML); // create table if user has specified if (query != null) { this.runQuery(query, conf); } } catch (HIHOException e1) { e1.printStackTrace(); } // verify required properties are loaded job.setNumReduceTasks(0); job.setInputFormatClass(FileStreamInputFormat.class); FileStreamInputFormat.addInputPath(job, new Path(inputPath)); job.setMapOutputKeyClass(NullWritable.class); job.setMapOutputValueClass(NullWritable.class); // job.setJarByClass(com.mysql.jdbc.Driver.class); job.setOutputFormatClass(NullOutputFormat.class); int ret = 0; try { ret = job.waitForCompletion(true) ? 0 : 1; } catch (Exception e) { e.printStackTrace(); } // run alter table query and add locations try { this.runQuery(getAlterTableDML(new Path(inputPath), conf), conf); } catch (HIHOException e1) { e1.printStackTrace(); } return ret; }
From source file:co.nubetech.hiho.mapreduce.lib.db.apache.DBOutputFormat.java
License:Apache License
private static DBConfiguration setOutput(Job job, JobConf jobConf, String tableName) throws IOException { job.setOutputFormatClass(DBOutputFormat.class); jobConf.setReduceSpeculativeExecution(false); DBConfiguration dbConf = new DBConfiguration(job.getConfiguration()); dbConf.setOutputTableName(tableName); return dbConf; }
From source file:co.nubetech.hiho.mapreduce.lib.db.DBQueryInputFormat.java
License:Apache License
/** * setInput() takes a custom query and a separate "bounding query" to use * instead of the custom "count query" used by DBInputFormat. *//* w ww. j a v a 2s . c o m*/ public static void setInput(Job job, String inputQuery, String inputBoundingQuery, ArrayList params) throws IOException { DBInputFormat.setInput(job, GenericDBWritable.class, inputQuery, ""); if (inputBoundingQuery != null) { job.getConfiguration().set(DBConfiguration.INPUT_BOUNDING_QUERY, inputBoundingQuery); } if (params != null) { DefaultStringifier<ArrayList> stringifier = new DefaultStringifier<ArrayList>(job.getConfiguration(), ArrayList.class); job.getConfiguration().set(HIHOConf.QUERY_PARAMS, stringifier.toString(params)); logger.debug("Converted params and saved them into config"); } job.setInputFormatClass(DBQueryInputFormat.class); }
From source file:co.nubetech.hiho.mapreduce.lib.db.GenericDBOutputFormat.java
License:Apache License
public static void setOutput(Job job, String tableName, String columnNames) throws IOException { job.setOutputFormatClass(GenericDBOutputFormat.class); DBConfiguration dbConf = new DBConfiguration(job.getConfiguration()); dbConf.setOutputTableName(tableName); dbConf.setOutputFieldNames(columnNames); String dbDriver = job.getConfiguration().get(DBConfiguration.DRIVER_CLASS_PROPERTY); String connString = job.getConfiguration().get(DBConfiguration.URL_PROPERTY); String username = job.getConfiguration().get(DBConfiguration.USERNAME_PROPERTY); String password = job.getConfiguration().get(DBConfiguration.PASSWORD_PROPERTY); Connection conn;//from ww w . jav a 2s.c om PreparedStatement stmt; try { Class.forName(dbDriver).newInstance(); conn = DriverManager.getConnection(connString, username, password); String query = "select " + columnNames + " from " + tableName; stmt = conn.prepareStatement(query); ResultSetMetaData meta = stmt.getMetaData(); ArrayList<ColumnInfo> columnInfo = populateColumnInfo(meta); String jsonString = getJsonStringOfColumnInfo(columnInfo); job.getConfiguration().set(HIHOConf.COLUMN_INFO, jsonString); logger.debug("columnInfo is: " + job.getConfiguration().get(HIHOConf.COLUMN_INFO)); stmt.close(); conn.close(); } catch (Exception e) { e.printStackTrace(); throw new IOException(e); } }
From source file:co.nubetech.hiho.merge.MergeJob.java
License:Apache License
@Override public int run(String[] args) throws Exception { populateConfiguration(args);//www .j a v a 2s . c o m try { checkMandatoryConfs(); } catch (HIHOException e1) { e1.printStackTrace(); throw new Exception(e1); } Class inputFormatClass = Class.forName(inputFormat); Class outputFormatClass = Class.forName(outputFormat); Class inputKeyClass = Class.forName(inputKeyClassName); Class inputValueClass = Class.forName(inputValueClassName); Configuration conf = getConf(); conf.set(HIHOConf.MERGE_OLD_PATH, oldPath); conf.set(HIHOConf.MERGE_NEW_PATH, newPath); Job job = new Job(conf); job.setJobName("Merge job"); job.setJarByClass(MergeJob.class); if (mergeBy.equals("key")) { job.setMapperClass(MergeKeyMapper.class); job.setReducerClass(MergeKeyReducer.class); } else if (mergeBy.equals("value")) { job.setMapperClass(MergeValueMapper.class); job.setReducerClass(MergeValueReducer.class); } job.setInputFormatClass(inputFormatClass); DelimitedTextInputFormat.setProperties(job, delimiter, column); job.setMapOutputKeyClass(HihoTuple.class); job.setMapOutputValueClass(HihoValue.class); job.setOutputKeyClass(inputKeyClass); job.setOutputValueClass(inputValueClass); FileInputFormat.setInputPaths(job, oldPath + "," + newPath); job.setOutputFormatClass(outputFormatClass); FileOutputFormat.setOutputPath(job, new Path(outputPath)); try { logger.debug("Output format class is " + job.getOutputFormatClass()); logger.debug("Class is " + ReflectionUtils .newInstance(job.getOutputFormatClass(), job.getConfiguration()).getClass().getName()); job.waitForCompletion(false); if (job.isComplete()) { Counters counters = job.getCounters(); totalRecordsOld = counters.findCounter(MergeRecordCounter.TOTAL_RECORDS_OLD).getValue(); totalRecordsNew = counters.findCounter(MergeRecordCounter.TOTAL_RECORDS_NEW).getValue(); badRecords = counters.findCounter(MergeRecordCounter.BAD_RECORD).getValue(); output = counters.findCounter(MergeRecordCounter.OUTPUT).getValue(); logger.info("Total old records read are: " + totalRecordsOld); logger.info("Total new records read are: " + totalRecordsNew); logger.info("Bad Records are: " + badRecords); logger.info("Output records are: " + output); } } catch (Exception e) { e.printStackTrace(); } return 0; }
From source file:com.abel.hwfs.custom.output.SetSizeDBOutputFormat.java
License:Apache License
private static MyDBConfiguration setOutput(Job job, String tableName) throws IOException { job.setOutputFormatClass(SetSizeDBOutputFormat.class); job.setReduceSpeculativeExecution(false); MyDBConfiguration dbConf = new MyDBConfiguration(job.getConfiguration()); dbConf.setOutputTableName(tableName); return dbConf; }
From source file:com.acme.io.JsonLoader.java
License:Apache License
/** * Get a schema for the data to be loaded. * @param location Location as returned by * {@link LoadFunc#relativeToAbsolutePath(String, org.apache.hadoop.fs.Path)} * @param job The {@link Job} object - this should be used only to obtain * cluster properties through {@link Job#getConfiguration()} and not to * set/query any runtime job information. * @return schema for the data to be loaded. This schema should represent * all tuples of the returned data. If the schema is unknown or it is * not possible to return a schema that represents all returned data, * then null should be returned. The schema should not be affected by * pushProjection, ie. getSchema should always return the original schema * even after pushProjection// w w w . j a v a 2 s . com * @throws IOException if an exception occurs while determining the schema */ public ResourceSchema getSchema(String location, Job job) throws IOException { // Open the schema file and read the schema // Get an HDFS handle. FileSystem fs = FileSystem.get(job.getConfiguration()); DataInputStream in = fs.open(new Path(location + "/_schema")); String line = in.readLine(); in.close(); // Parse the schema ResourceSchema s = new ResourceSchema(Utils.getSchemaFromString(line)); if (s == null) { throw new IOException("Unable to parse schema found in file " + location + "/_schema"); } // Now that we have determined the schema, store it in our // UDFContext properties object so we have it when we need it on the // backend UDFContext udfc = UDFContext.getUDFContext(); Properties p = udfc.getUDFProperties(this.getClass(), new String[] { udfcSignature }); p.setProperty("pig.jsonloader.schema", line); return s; }
From source file:com.acme.io.JsonStorage.java
License:Apache License
/** * Store schema of the data being written * @param schema Schema to be recorded//from w w w . java 2s . com * @param location Location as returned by * {@link LoadFunc#relativeToAbsolutePath(String, org.apache.hadoop.fs.Path)} * @param job The {@link Job} object - this should be used only to obtain * cluster properties through {@link Job#getConfiguration()} and not to * set/query any runtime job information. * @throws IOException */ public void storeSchema(ResourceSchema schema, String location, Job job) throws IOException { // Store the schema in a side file in the same directory. MapReduce // does not include files starting with "_" when reading data for a job. FileSystem fs = FileSystem.get(job.getConfiguration()); DataOutputStream out = fs.create(new Path(location + "/_schema")); out.writeBytes(schema.toString()); out.writeByte('\n'); out.close(); }