Package org.apache.hive.hcatalog.mapreduce.MultiOutputFormat

Examples of org.apache.hive.hcatalog.mapreduce.MultiOutputFormat.JobConfigurer

288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
    job.setMapperClass(MyMapper.class);
    job.setInputFormatClass(TextInputFormat.class);
    job.setOutputFormatClass(MultiOutputFormat.class);
    job.setNumReduceTasks(0);

    JobConfigurer configurer = MultiOutputFormat.createConfigurer(job);

    for (int i = 0; i < tableNames.length; i++) {
      configurer.addOutputFormat(tableNames[i], HCatOutputFormat.class, BytesWritable.class,
        HCatRecord.class);
      HCatOutputFormat.setOutput(configurer.getJob(tableNames[i]), infoList.get(i));
      HCatOutputFormat.setSchema(configurer.getJob(tableNames[i]),
        schemaMap.get(tableNames[i]));
    }
    configurer.configure();

    Path filePath = createInputFile();
    FileInputFormat.addInputPath(job, filePath);
    Assert.assertTrue(job.waitForCompletion(true));
View Full Code Here
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
    job.setJarByClass(this.getClass());
    job.setInputFormatClass(TextInputFormat.class);
    job.setOutputFormatClass(MultiOutputFormat.class);
    job.setNumReduceTasks(0);

    JobConfigurer configurer = MultiOutputFormat.createConfigurer(job);
    configurer.addOutputFormat("out1", TextOutputFormat.class, IntWritable.class, Text.class);
    configurer.addOutputFormat("out2", SequenceFileOutputFormat.class, Text.class,
      IntWritable.class);
    Path outDir = new Path(workDir.getPath(), job.getJobName());
    FileOutputFormat.setOutputPath(configurer.getJob("out1"), new Path(outDir, "out1"));
    FileOutputFormat.setOutputPath(configurer.getJob("out2"), new Path(outDir, "out2"));

    String fileContent = "Hello World";
    String inputFile = createInputFile(fileContent);
    FileInputFormat.setInputPaths(job, new Path(inputFile));

    //Test for merging of configs
    DistributedCache.addFileToClassPath(new Path(inputFile), job.getConfiguration(), fs);
    String dummyFile = createInputFile("dummy file");
    DistributedCache.addFileToClassPath(new Path(dummyFile), configurer.getJob("out1")
      .getConfiguration(), fs);
    // duplicate of the value. Merging should remove duplicates
    DistributedCache.addFileToClassPath(new Path(inputFile), configurer.getJob("out2")
      .getConfiguration(), fs);

    configurer.configure();

    // Verify if the configs are merged
    Path[] fileClassPaths = DistributedCache.getFileClassPaths(job.getConfiguration());
    List<Path> fileClassPathsList = Arrays.asList(fileClassPaths);
    Assert.assertTrue(fileClassPathsList.contains(new Path(inputFile)));
View Full Code Here
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
    job.setInputFormatClass(TextInputFormat.class);
    job.setOutputFormatClass(MultiOutputFormat.class);
    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(IntWritable.class);

    JobConfigurer configurer = MultiOutputFormat.createConfigurer(job);

    configurer.addOutputFormat("out1", TextOutputFormat.class, IntWritable.class, Text.class);
    configurer.addOutputFormat("out2", SequenceFileOutputFormat.class, Text.class,
      IntWritable.class);
    configurer.addOutputFormat("out3", NullOutputFormat.class, Text.class,
      IntWritable.class);
    Path outDir = new Path(workDir.getPath(), job.getJobName());
    FileOutputFormat.setOutputPath(configurer.getJob("out1"), new Path(outDir, "out1"));
    FileOutputFormat.setOutputPath(configurer.getJob("out2"), new Path(outDir, "out2"));

    configurer.configure();

    String fileContent = "Hello World Hello World World";
    String inputFile = createInputFile(fileContent);
    FileInputFormat.setInputPaths(job, new Path(inputFile));
View Full Code Here
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
    job.setMapperClass(MyMapper.class);
    job.setInputFormatClass(TextInputFormat.class);
    job.setOutputFormatClass(MultiOutputFormat.class);
    job.setNumReduceTasks(0);

    JobConfigurer configurer = MultiOutputFormat.createConfigurer(job);

    for (int i = 0; i < tableNames.length; i++) {
      configurer.addOutputFormat(tableNames[i], HCatOutputFormat.class, BytesWritable.class,
        HCatRecord.class);
      HCatOutputFormat.setOutput(configurer.getJob(tableNames[i]), infoList.get(i));
      HCatOutputFormat.setSchema(configurer.getJob(tableNames[i]),
        schemaMap.get(tableNames[i]));
    }
    configurer.configure();

    Path filePath = createInputFile();
    FileInputFormat.addInputPath(job, filePath);
    Assert.assertTrue(job.waitForCompletion(true));
View Full Code Here
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
    job.setMapperClass(MyMapper.class);
    job.setInputFormatClass(TextInputFormat.class);
    job.setOutputFormatClass(MultiOutputFormat.class);
    job.setNumReduceTasks(0);

    JobConfigurer configurer = MultiOutputFormat.createConfigurer(job);

    for (int i = 0; i < tableNames.length; i++) {
      configurer.addOutputFormat(tableNames[i], HCatOutputFormat.class, BytesWritable.class,
        HCatRecord.class);
      HCatOutputFormat.setOutput(configurer.getJob(tableNames[i]), infoList.get(i));
      HCatOutputFormat.setSchema(configurer.getJob(tableNames[i]),
        schemaMap.get(tableNames[i]));
    }
    configurer.configure();

    Path filePath = createInputFile();
    FileInputFormat.addInputPath(job, filePath);
    Assert.assertTrue(job.waitForCompletion(true));
View Full Code Here
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
    job.setJarByClass(this.getClass());
    job.setInputFormatClass(TextInputFormat.class);
    job.setOutputFormatClass(MultiOutputFormat.class);
    job.setNumReduceTasks(0);

    JobConfigurer configurer = MultiOutputFormat.createConfigurer(job);
    configurer.addOutputFormat("out1", TextOutputFormat.class, IntWritable.class, Text.class);
    configurer.addOutputFormat("out2", SequenceFileOutputFormat.class, Text.class,
      IntWritable.class);
    Path outDir = new Path(workDir.getPath(), job.getJobName());
    FileOutputFormat.setOutputPath(configurer.getJob("out1"), new Path(outDir, "out1"));
    FileOutputFormat.setOutputPath(configurer.getJob("out2"), new Path(outDir, "out2"));

    String fileContent = "Hello World";
    String inputFile = createInputFile(fileContent);
    FileInputFormat.setInputPaths(job, new Path(inputFile));

    //Test for merging of configs
    DistributedCache.addFileToClassPath(new Path(inputFile), job.getConfiguration(), fs);
    String dummyFile = createInputFile("dummy file");
    DistributedCache.addFileToClassPath(new Path(dummyFile), configurer.getJob("out1")
      .getConfiguration(), fs);
    // duplicate of the value. Merging should remove duplicates
    DistributedCache.addFileToClassPath(new Path(inputFile), configurer.getJob("out2")
      .getConfiguration(), fs);

    configurer.configure();

    // Verify if the configs are merged
    Path[] fileClassPaths = DistributedCache.getFileClassPaths(job.getConfiguration());
    List<Path> fileClassPathsList = Arrays.asList(fileClassPaths);
    Assert.assertTrue(fileClassPathsList.contains(new Path(inputFile)));
View Full Code Here
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
    job.setInputFormatClass(TextInputFormat.class);
    job.setOutputFormatClass(MultiOutputFormat.class);
    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(IntWritable.class);

    JobConfigurer configurer = MultiOutputFormat.createConfigurer(job);

    configurer.addOutputFormat("out1", TextOutputFormat.class, IntWritable.class, Text.class);
    configurer.addOutputFormat("out2", SequenceFileOutputFormat.class, Text.class,
      IntWritable.class);
    configurer.addOutputFormat("out3", NullOutputFormat.class, Text.class,
      IntWritable.class);
    Path outDir = new Path(workDir.getPath(), job.getJobName());
    FileOutputFormat.setOutputPath(configurer.getJob("out1"), new Path(outDir, "out1"));
    FileOutputFormat.setOutputPath(configurer.getJob("out2"), new Path(outDir, "out2"));

    configurer.configure();

    String fileContent = "Hello World Hello World World";
    String inputFile = createInputFile(fileContent);
    FileInputFormat.setInputPaths(job, new Path(inputFile));
View Full Code Here
TOP

Related Classes of org.apache.hive.hcatalog.mapreduce.MultiOutputFormat.JobConfigurer

  • org.apache.hive.hcatalog.mapreduce.TestHCatMultiOutputFormat
  • org.apache.hive.hcatalog.mapreduce.TestMultiOutputFormat
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.