Examples of DynamicSerDe

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
    // " where key > 10 group by key");
    String sql = "select key, value from " + tableName + " where key > 10";
    client.execute(sql);

    // Instantiate DynamicSerDe
    DynamicSerDe ds = new DynamicSerDe();
    Properties dsp = new Properties();
    dsp.setProperty(Constants.SERIALIZATION_FORMAT,
        org.apache.hadoop.hive.serde2.thrift.TCTLSeparatedProtocol.class
            .getName());
    dsp.setProperty(
        org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME,
        "result");
    String serDDL = new String("struct result { ");
    List<FieldSchema> schema = client.getThriftSchema().getFieldSchemas();
    for (int pos = 0; pos < schema.size(); pos++) {
      if (pos != 0) {
        serDDL = serDDL.concat(",");
      }
      serDDL = serDDL.concat(schema.get(pos).getType());
      serDDL = serDDL.concat(" ");
      serDDL = serDDL.concat(schema.get(pos).getName());
    }
    serDDL = serDDL.concat("}");

    dsp.setProperty(Constants.SERIALIZATION_DDL, serDDL);
    dsp.setProperty(Constants.SERIALIZATION_LIB, ds.getClass().toString());
    dsp.setProperty(Constants.FIELD_DELIM, "9");
    ds.initialize(new Configuration(), dsp);

    String row = client.fetchOne();
    Object o = ds.deserialize(new BytesWritable(row.getBytes()));

    assertEquals(o.getClass().toString(), "class java.util.ArrayList");
    List<?> lst = (List<?>) o;
    assertEquals(lst.get(0), 238);

    // TODO: serde doesn't like underscore -- struct result { string _c0}
    sql = "select count(1) as c from " + tableName;
    client.execute(sql);
    row = client.fetchOne();

    serDDL = new String("struct result { ");
    schema = client.getThriftSchema().getFieldSchemas();
    for (int pos = 0; pos < schema.size(); pos++) {
      if (pos != 0) {
        serDDL = serDDL.concat(",");
      }
      serDDL = serDDL.concat(schema.get(pos).getType());
      serDDL = serDDL.concat(" ");
      serDDL = serDDL.concat(schema.get(pos).getName());
    }
    serDDL = serDDL.concat("}");

    dsp.setProperty(Constants.SERIALIZATION_DDL, serDDL);
    // Need a new DynamicSerDe instance - re-initialization is not supported.
    ds = new DynamicSerDe();
    ds.initialize(new Configuration(), dsp);
    o = ds.deserialize(new BytesWritable(row.getBytes()));
  }
View Full Code Here

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

52
53
54
55
56
57
58
59
60
61
62
  /**
   * Instantiate the dynamic serde used to deserialize the result row
   */
  public void initDynamicSerde() {
    try {
      ds = new DynamicSerDe();
      Properties dsp = new Properties();
      dsp.setProperty(Constants.SERIALIZATION_FORMAT, org.apache.hadoop.hive.serde2.thrift.TCTLSeparatedProtocol.class.getName());
      dsp.setProperty(org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME, "result");
      dsp.setProperty(Constants.SERIALIZATION_DDL, client.getSchema());
      dsp.setProperty(Constants.SERIALIZATION_LIB, ds.getClass().toString());
View Full Code Here

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

106
107
108
109
110
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
        Properties schema = new Properties();
        schema.setProperty(Constants.SERIALIZATION_FORMAT, protocol);
        schema.setProperty(org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME, "test");
        schema.setProperty(Constants.SERIALIZATION_DDL,
        "struct test { i32 _hello, list<string> 2bye, map<string,i32> another, i32 nhello, double d, double nd}");
        schema.setProperty(Constants.SERIALIZATION_LIB, new DynamicSerDe().getClass().toString());
        HashMap<String, String> p = additionalParams.get(pp);
        if (p != null) {
          for(Entry<String, String> e: p.entrySet()) {
            schema.setProperty(e.getKey(), e.getValue());
          }
        }
 
        DynamicSerDe serde = new DynamicSerDe();
        serde.initialize(new Configuration(), schema);
       
        // Try getObjectInspector
        ObjectInspector oi = serde.getObjectInspector();
        System.out.println("TypeName = " + oi.getTypeName());

       
        // Try to serialize
        BytesWritable bytes = (BytesWritable) serde.serialize(struct, oi);
        System.out.println("bytes =" + hexString(bytes));
      
        if (!isBinary) {
          System.out.println("bytes in text =" + new String(bytes.get(), 0, bytes.getSize()));
        }
       
        // Try to deserialize
        Object o = serde.deserialize(bytes);
        System.out.println("o class = " + o.getClass());
        List<?> olist = (List<?>)o;
        System.out.println("o size = " + olist.size());
        System.out.println("o[0] class = " + olist.get(0).getClass());
        System.out.println("o[1] class = " + olist.get(1).getClass());
View Full Code Here

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
    schema.setProperty(org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME, "test");
    schema.setProperty(Constants.SERIALIZATION_DDL, ddl);
    schema.setProperty(Constants.SERIALIZATION_LIB, DynamicSerDe.class.getName());
    schema.setProperty(Constants.SERIALIZATION_SORT_ORDER, order);

    DynamicSerDe serde = new DynamicSerDe();
    serde.initialize(new Configuration(), schema);

    ObjectInspector oi = serde.getObjectInspector();

    // Try to serialize
    BytesWritable bytes[] = new BytesWritable[structs.length];
    for (int i=0; i<structs.length; i++) {
      bytes[i] = new BytesWritable();
      BytesWritable s = (BytesWritable)serde.serialize(structs[i], oi);
      bytes[i].set(s);
      if (i>0) {
        int compareResult = bytes[i-1].compareTo(bytes[i]);
        if ( (compareResult<0 && !ascending) || (compareResult>0 && ascending) ) {
          System.out.println("Test failed in " + (ascending ? "ascending" : "descending") + " order.");
          System.out.println("serialized data of " + structs[i-1] + " = " + hexString(bytes[i-1]));
          System.out.println("serialized data of " + structs[i] + " = " + hexString(bytes[i]));
          fail("Sort order of serialized " + structs[i-1] + " and " + structs[i] + " are reversed!");           
        }
      }
    }

    // Try to deserialize
    Object[] deserialized = new Object[structs.length];
    for (int i=0; i<structs.length; i++) {
      deserialized[i] = serde.deserialize(bytes[i]);
      if (!structs[i].equals(deserialized[i])) {
        System.out.println("structs[i] = " + structs[i]);
        System.out.println("deserialized[i] = " + deserialized[i]);
        System.out.println("serialized[i] = " + hexString(bytes[i]));
        assertEquals(structs[i], deserialized[i]);
View Full Code Here

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
      Properties schema = new Properties();
      schema.setProperty(Constants.SERIALIZATION_FORMAT, org.apache.hadoop.hive.serde2.thrift.TCTLSeparatedProtocol.class.getName());
      schema.setProperty(org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME, "test");
      schema.setProperty(Constants.SERIALIZATION_DDL,
                         "struct test { i32 hello, list<string> bye, map<string,i32> another}");
      schema.setProperty(Constants.SERIALIZATION_LIB, new DynamicSerDe().getClass().toString());

      schema.setProperty(Constants.FIELD_DELIM, "9");
      schema.setProperty(Constants.COLLECTION_DELIM, "1");
      schema.setProperty(Constants.LINE_DELIM, "2");
      schema.setProperty(Constants.MAPKEY_DELIM, "4");

      DynamicSerDe serde = new DynamicSerDe();
      serde.initialize(new Configuration(), schema);

      TCTLSeparatedProtocol prot = (TCTLSeparatedProtocol)serde.oprot_;
      assertTrue(prot.getPrimarySeparator().equals("\u0009"));
     
      ObjectInspector oi = serde.getObjectInspector();

      // Try to serialize
      BytesWritable bytes = (BytesWritable) serde.serialize(struct, oi);
       
      hexString(bytes);

      String compare = "234" + "\u0009" + "firstString" + "\u0001" + "secondString" + "\u0009" + "firstKey" + "\u0004" + "1" + "\u0001" + "secondKey" + "\u0004" + "2";

      System.out.println("bytes in text =" + new String(bytes.get(), 0, bytes.getSize()) + ">");
      System.out.println("compare to    =" + compare + ">");
       
      assertTrue(compare.equals( new String(bytes.get(), 0, bytes.getSize())));
     
      // Try to deserialize
      Object o = serde.deserialize(bytes);
      System.out.println("o class = " + o.getClass());
      List<?> olist = (List<?>)o;
      System.out.println("o size = " + olist.size());
      System.out.println("o[0] class = " + olist.get(0).getClass());
      System.out.println("o[1] class = " + olist.get(1).getClass());
View Full Code Here

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
      Properties schema = new Properties();
      schema.setProperty(Constants.SERIALIZATION_FORMAT, org.apache.hadoop.hive.serde2.thrift.TCTLSeparatedProtocol.class.getName());
      schema.setProperty(org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME, "test");
      schema.setProperty(Constants.SERIALIZATION_DDL,
                         "struct test { i32 hello, list<string> bye, map<string,i32> another}");
      schema.setProperty(Constants.SERIALIZATION_LIB, new DynamicSerDe().getClass().toString());
      schema.setProperty(TCTLSeparatedProtocol.ReturnNullsKey, "true");

      DynamicSerDe serde = new DynamicSerDe();
      serde.initialize(new Configuration(), schema);

      ObjectInspector oi = serde.getObjectInspector();
     
      // Try to serialize
      BytesWritable bytes = (BytesWritable) serde.serialize(struct, oi);
       
      hexString(bytes);

      // Try to deserialize
      Object o = serde.deserialize(bytes);
      assertEquals(struct, o);
     
    } catch (Throwable e) {
      e.printStackTrace();
      throw e;
View Full Code Here

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
      Properties schema = new Properties();
      schema.setProperty(Constants.SERIALIZATION_FORMAT, org.apache.hadoop.hive.serde2.thrift.TCTLSeparatedProtocol.class.getName());
      schema.setProperty(org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME, "test");
      schema.setProperty(Constants.SERIALIZATION_DDL,
                         "struct test { i32 hello, list<string> bye, map<string,i32> another}");
      schema.setProperty(Constants.SERIALIZATION_LIB, new DynamicSerDe().getClass().toString());
      schema.setProperty(TCTLSeparatedProtocol.ReturnNullsKey, "true");

      DynamicSerDe serde = new DynamicSerDe();
      serde.initialize(new Configuration(), schema);

      ObjectInspector oi = serde.getObjectInspector();

      // Try to serialize
      BytesWritable bytes = (BytesWritable) serde.serialize(struct, oi);
       
      hexString(bytes);

      // Try to deserialize
      Object o = serde.deserialize(bytes);
      List<?> olist = (List<?>)o;

      assertTrue(olist.size() == 3);
      assertEquals(null, olist.get(0));
      assertEquals(null, olist.get(1));
View Full Code Here

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
      Properties schema = new Properties();
      schema.setProperty(Constants.SERIALIZATION_FORMAT, org.apache.hadoop.hive.serde2.thrift.TCTLSeparatedProtocol.class.getName());
      schema.setProperty(org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME, "test");
      schema.setProperty(Constants.SERIALIZATION_DDL,
                         "struct test { i32 hello, list<string> bye, map<string,i32> another}");
      schema.setProperty(Constants.SERIALIZATION_LIB, new DynamicSerDe().getClass().toString());

      schema.setProperty(TCTLSeparatedProtocol.ReturnNullsKey, "true");
      DynamicSerDe serde = new DynamicSerDe();
      serde.initialize(new Configuration(), schema);

      ObjectInspector oi = serde.getObjectInspector();

      // Try to serialize
      BytesWritable bytes = (BytesWritable) serde.serialize(struct, oi);
       
      hexString(bytes);

      // Try to deserialize
      Object o = serde.deserialize(bytes);
      List<?> olist = (List<?>)o;

      assertTrue(olist.size() == 3);
      assertEquals(null, olist.get(0));
      assertEquals(0, ((List<?>)olist.get(1)).size());
View Full Code Here

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
      Properties schema = new Properties();
      schema.setProperty(Constants.SERIALIZATION_FORMAT, org.apache.hadoop.hive.serde2.thrift.TCTLSeparatedProtocol.class.getName());
      schema.setProperty(org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME, "test");
      schema.setProperty(Constants.SERIALIZATION_DDL,
                         "struct test { i32 hello, list<string> bye, map<string,i32> another}");
      schema.setProperty(Constants.SERIALIZATION_LIB, new DynamicSerDe().getClass().toString());

      schema.setProperty(TCTLSeparatedProtocol.ReturnNullsKey, "false");
      DynamicSerDe serde = new DynamicSerDe();
      serde.initialize(new Configuration(), schema);

      ObjectInspector oi = serde.getObjectInspector();

      // Try to serialize
      BytesWritable bytes = (BytesWritable) serde.serialize(struct, oi);
       
      hexString(bytes);

      // Try to deserialize
      Object o = serde.deserialize(bytes);
      List<?> olist = (List<?>)o;

      assertTrue(olist.size() == 3);
      assertEquals(new Integer(0), (Integer)olist.get(0));
      List<?> num1 = (List<?>)olist.get(1);
View Full Code Here

Examples of org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe

649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
      //      schema.setProperty(Constants.SERIALIZATION_FORMAT, com.facebook.thrift.protocol.TJSONProtocol.class.getName());
      schema.setProperty(Constants.SERIALIZATION_FORMAT, com.facebook.thrift.protocol.TBinaryProtocol.class.getName());
      schema.setProperty(org.apache.hadoop.hive.metastore.api.Constants.META_TABLE_NAME, "test");
      schema.setProperty(Constants.SERIALIZATION_DDL,
                         "struct inner { i32 field1, string field2 },struct  test {inner foo,  i32 hello, list<string> bye, map<string,i32> another}");
      schema.setProperty(Constants.SERIALIZATION_LIB, new DynamicSerDe().getClass().toString());

     
      //
      // construct object of above type
      //

      // construct the inner struct
      ArrayList<Object> innerStruct = new ArrayList<Object>();
      innerStruct.add(new Integer(22));
      innerStruct.add(new String("hello world"));

      // construct outer struct
      ArrayList<String> bye = new ArrayList<String> ();
      bye.add("firstString");
      bye.add("secondString");
      HashMap<String, Integer> another = new HashMap<String, Integer>(); 
      another.put("firstKey", 1);
      another.put("secondKey", 2);

      ArrayList<Object> struct = new ArrayList<Object>();

      struct.add(innerStruct);
      struct.add(Integer.valueOf(234));
      struct.add(bye);
      struct.add(another);

      DynamicSerDe serde = new DynamicSerDe();
      serde.initialize(new Configuration(), schema);

      ObjectInspector oi = serde.getObjectInspector();

      // Try to serialize
      BytesWritable bytes = (BytesWritable) serde.serialize(struct, oi);

      // Try to deserialize
      Object o = serde.deserialize(bytes);
      List<?> olist = (List<?>)o;


      assertEquals(4, olist.size());
      assertEquals(innerStruct, olist.get(0));
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.