@Override public Plan getPlan(String... args) { // parse job parameters int numSubTasks = (args.length > 0 ? Integer.parseInt(args[0]) : 1); String dataInput = (args.length > 1 ? args[1] : ""); String output = (args.length > 2 ? args[2] : ""); FileDataSource source = new FileDataSource(new TextInputFormat(), dataInput, "Input Lines"); MapOperator mapper = MapOperator.builder(new TokenizeLine()) .input(source) .name("Tokenize Lines") .build(); ReduceOperator reducer = ReduceOperator.builder(CountWords.class, StringValue.class, 0) .input(mapper) .name("Count Words") .build(); @SuppressWarnings("unchecked") FileDataSink out = new FileDataSink(new CsvOutputFormat("\n", " ", StringValue.class, IntValue.class), output, reducer, "Word Counts"); Plan plan = new Plan(out, "WordCount Example"); plan.setDefaultParallelism(numSubTasks); return plan; }
static Plan getTestPlanPlan(int numSubTasks, String input, String output) { FileDataSource source = new FileDataSource(new TextInputFormat(), input, "Input Lines"); source.setParameter(TextInputFormat.CHARSET_NAME, "ASCII"); MapOperator mapper = MapOperator.builder(new TokenizeLine()) .input(source) .name("Tokenize Lines") .build(); ReduceOperator reducer = ReduceOperator.builder(CountWords.class, StringValue.class, 0) .input(mapper) .name("Count Words") .build(); @SuppressWarnings("unchecked") FileDataSink out = new FileDataSink(new CsvOutputFormat("\n"," ", StringValue.class, IntValue.class), output, reducer, "Word Counts"); Plan plan = new Plan(out, "WordCount Example"); plan.setDefaultParallelism(numSubTasks); return plan; }
public Plan getPlan(int numSubTasks, String dataInput, String output) { // input is {word, count} pair FileDataSource source = new FileDataSource(new TextInputFormat(), dataInput, "Input Lines"); //do a selection using cached file MapOperator mapper = MapOperator.builder(new TokenizeLine()) .input(source) .name("Tokenize Lines") .build(); FileDataSink out = new FileDataSink(new CsvOutputFormat(), output, mapper, "Selection"); CsvOutputFormat.configureRecordFormat(out) .recordDelimiter('\n') .fieldDelimiter(' ') .field(StringValue.class, 0) .field(IntValue.class, 1); Plan plan = new Plan(out, "Distributed Cache"); plan.setDefaultParallelism(numSubTasks); return plan; }
@Override public Plan getPlan(String... args) { int numSubTasks = (args.length > 0 ? Integer.parseInt(args[0]) : 1); String dataInput = (args.length > 1 ? args[1] : ""); String output = (args.length > 2 ? args[2] : ""); FileDataSource source = new FileDataSource(new TextInputFormat(), dataInput, "Input Lines"); MapOperator mapper = MapOperator.builder(new TokenizeLine()).input(source).name("Tokenize Lines").build(); ReduceOperator reducer = ReduceOperator.builder(CountWords.class, StringValue.class, 0).input(mapper) .name("Count Words").build(); FileDataSink out = new FileDataSink(new CsvOutputFormat(), output, reducer, "Word Counts"); CsvOutputFormat.configureRecordFormat(out).recordDelimiter('\n') .fieldDelimiter(' ').field(StringValue.class, 0) .field(IntValue.class, 1); Plan plan = new Plan(out, "WordCount Example"); plan.setDefaultParallelism(numSubTasks); return plan; }
private void checkWordCountWithSortedSink(boolean estimates) { try { FileDataSource sourceNode = new FileDataSource(new TextInputFormat(), IN_FILE, "Input Lines"); MapOperator mapNode = MapOperator.builder(new TokenizeLine()) .input(sourceNode)