Package org.apache.pig.impl.logicalLayer.schema.Schema

Examples of org.apache.pig.impl.logicalLayer.schema.Schema.FieldSchema

346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
    @Test
    public void testMergeDifferentSize2() throws Throwable {


        List<FieldSchema> list1 = new ArrayList<FieldSchema>() ;
        list1.add(new FieldSchema("1a", DataType.BYTEARRAY)) ;
        list1.add(new FieldSchema("1b", DataType.BYTEARRAY)) ;
        list1.add(new FieldSchema("1c", DataType.LONG)) ;

        List<FieldSchema> list2 = new ArrayList<FieldSchema>() ;
        list2.add(new FieldSchema("2a", DataType.BYTEARRAY)) ;
        list2.add(new FieldSchema("2b", DataType.BYTEARRAY)) ;
        list2.add(new FieldSchema("2c", DataType.INTEGER)) ;
        list2.add(new FieldSchema("2d", DataType.MAP)) ;

        Schema schema1 = new Schema(list1) ;
        Schema schema2 = new Schema(list2) ;

        // Merge
View Full Code Here
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
420
421
422
423
424
425
426
427
428
429
430
431
432
    @Test
    public void testMergeMismatchType1() throws Throwable {

        // Generate two schemas
        List<FieldSchema> innerList1 = new ArrayList<FieldSchema>() ;
        innerList1.add(new FieldSchema("11a", DataType.CHARARRAY)) ;
        innerList1.add(new FieldSchema("11b", DataType.FLOAT)) ;

        List<FieldSchema> innerList2 = new ArrayList<FieldSchema>() ;
        innerList2.add(new FieldSchema("22a", DataType.DOUBLE)) ;
        innerList2.add(new FieldSchema(null, DataType.LONG)) ;

        Schema innerSchema1 = new Schema(innerList1) ;
        Schema innerSchema2 = new Schema(innerList2) ;

        List<FieldSchema> list1 = new ArrayList<FieldSchema>() ;
        list1.add(new FieldSchema("1a", DataType.BYTEARRAY)) ;
        list1.add(new FieldSchema("1b", innerSchema1)) ;
        list1.add(new FieldSchema("1c", DataType.MAP)) ;

        List<FieldSchema> list2 = new ArrayList<FieldSchema>() ;
        list2.add(new FieldSchema("2a", DataType.BYTEARRAY)) ;
        list2.add(new FieldSchema("2b", innerSchema2)) ;
        list2.add(new FieldSchema("2c", DataType.INTEGER)) ;

        Schema schema1 = new Schema(list1) ;
        Schema schema2 = new Schema(list2) ;

        // Merge
        Schema mergedSchema = Schema.mergeSchema(schema1,
                                                 schema2,
                                                 true,
                                                 false,
                                                 true) ;


        // Generate expected schema
        List<FieldSchema> expectedInnerList = new ArrayList<FieldSchema>() ;
        expectedInnerList.add(new FieldSchema("22a", DataType.BYTEARRAY)) ;
        expectedInnerList.add(new FieldSchema("11b", DataType.FLOAT)) ;

        Schema expectedInner = new Schema(expectedInnerList) ;

        List<FieldSchema> expectedList = new ArrayList<FieldSchema>() ;
        expectedList.add(new FieldSchema("2a", DataType.BYTEARRAY)) ;
        expectedList.add(new FieldSchema("2b", expectedInner)) ;
        expectedList.add(new FieldSchema("2c", DataType.BYTEARRAY)) ;

        Schema expected = new Schema(expectedList) ;

        // Compare
        Assert.assertTrue(Schema.equals(mergedSchema, expected, false, false)) ;
View Full Code Here
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
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
476
    @Test
    public void testMergeMismatchType2() throws Throwable {

        // Generate two schemas
        List<FieldSchema> innerList1 = new ArrayList<FieldSchema>() ;
        innerList1.add(new FieldSchema("11a", DataType.CHARARRAY)) ;
        innerList1.add(new FieldSchema("11b", DataType.FLOAT)) ;

        Schema innerSchema1 = new Schema(innerList1) ;

        List<FieldSchema> list1 = new ArrayList<FieldSchema>() ;
        list1.add(new FieldSchema("1a", DataType.BYTEARRAY)) ;
        list1.add(new FieldSchema("1b", innerSchema1)) ;
        list1.add(new FieldSchema("1c", DataType.MAP)) ;

        List<FieldSchema> list2 = new ArrayList<FieldSchema>() ;
        list2.add(new FieldSchema("2a", DataType.BYTEARRAY)) ;
        list2.add(new FieldSchema("2b", DataType.BYTEARRAY)) ;
        list2.add(new FieldSchema("2c", DataType.INTEGER)) ;

        Schema schema1 = new Schema(list1) ;
        Schema schema2 = new Schema(list2) ;

        // Merge
        Schema mergedSchema = Schema.mergeSchema(schema1,
                                                 schema2,
                                                 true,
                                                 false,
                                                 true) ;


        // Generate expected schema
        List<FieldSchema> expectedList = new ArrayList<FieldSchema>() ;
        expectedList.add(new FieldSchema("2a", DataType.BYTEARRAY)) ;
        expectedList.add(new FieldSchema("2b", DataType.TUPLE)) ;
        expectedList.add(new FieldSchema("2c", DataType.BYTEARRAY)) ;

        Schema expected = new Schema(expectedList) ;

        // Compare
        Assert.assertTrue(Schema.equals(mergedSchema, expected, false, false)) ;
View Full Code Here
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
    @Test
    public void testMergeMismatchType3()  {

      // Generate two schemas
        List<FieldSchema> innerList1 = new ArrayList<FieldSchema>() ;
        innerList1.add(new FieldSchema("11a", DataType.CHARARRAY)) ;
        innerList1.add(new FieldSchema("11b", DataType.FLOAT)) ;

        List<FieldSchema> innerList2 = new ArrayList<FieldSchema>() ;
        innerList2.add(new FieldSchema("22a", DataType.DOUBLE)) ;
        innerList2.add(new FieldSchema(null, DataType.LONG)) ;

        Schema innerSchema1 = new Schema(innerList1) ;
        Schema innerSchema2 = new Schema(innerList2) ;

        List<FieldSchema> list1 = new ArrayList<FieldSchema>() ;
        list1.add(new FieldSchema("1a", DataType.BYTEARRAY)) ;
        list1.add(new FieldSchema("1b", innerSchema1)) ;
        list1.add(new FieldSchema("1c", DataType.MAP)) ;

        List<FieldSchema> list2 = new ArrayList<FieldSchema>() ;
        list2.add(new FieldSchema("2a", DataType.BYTEARRAY)) ;
        list2.add(new FieldSchema("2b", innerSchema2)) ;
        list2.add(new FieldSchema("2c", DataType.MAP)) ;

        Schema schema1 = new Schema(list1) ;
        Schema schema2 = new Schema(list2) ;

        // Merge
View Full Code Here
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
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
573
574
575
576
577
    @Test
    public void testMergeDifferentSizeAndTypeMismatch1() throws Throwable {

        // Generate two schemas
        List<FieldSchema> innerList1 = new ArrayList<FieldSchema>() ;
        innerList1.add(new FieldSchema("11a", DataType.INTEGER)) ;
        innerList1.add(new FieldSchema("11b", DataType.FLOAT)) ;
        innerList1.add(new FieldSchema("11c", DataType.CHARARRAY)) ;

        List<FieldSchema> innerList2 = new ArrayList<FieldSchema>() ;
        innerList2.add(new FieldSchema("22a", DataType.CHARARRAY)) ;
        innerList2.add(new FieldSchema(null, DataType.LONG)) ;

        Schema innerSchema1 = new Schema(innerList1) ;
        Schema innerSchema2 = new Schema(innerList2) ;

        List<FieldSchema> list1 = new ArrayList<FieldSchema>() ;
        list1.add(new FieldSchema("1a", DataType.INTEGER)) ;
        list1.add(new FieldSchema("1b", innerSchema1)) ;
        list1.add(new FieldSchema("1c", DataType.LONG)) ;

        List<FieldSchema> list2 = new ArrayList<FieldSchema>() ;
        list2.add(new FieldSchema("2a", DataType.CHARARRAY)) ;
        list2.add(new FieldSchema("2b", innerSchema2)) ;
        list2.add(new FieldSchema("2c", DataType.INTEGER)) ;
        list2.add(new FieldSchema("2d", DataType.MAP)) ;

        Schema schema1 = new Schema(list1) ;
        Schema schema2 = new Schema(list2) ;

        // Merge
        Schema mergedSchema = Schema.mergeSchema(schema1,
                                                 schema2,
                                                 true,
                                                 true,
                                                 true) ;


        // Generate expected schema
        List<FieldSchema> expectedInnerList = new ArrayList<FieldSchema>() ;
        expectedInnerList.add(new FieldSchema("22a", DataType.BYTEARRAY)) ;
        expectedInnerList.add(new FieldSchema("11b", DataType.FLOAT)) ;
        expectedInnerList.add(new FieldSchema("11c", DataType.CHARARRAY)) ;

        Schema expectedInner = new Schema(expectedInnerList) ;

        List<FieldSchema> expectedList = new ArrayList<FieldSchema>() ;
        expectedList.add(new FieldSchema("2a", DataType.BYTEARRAY)) ;
        expectedList.add(new FieldSchema("2b", expectedInner)) ;
        expectedList.add(new FieldSchema("2c", DataType.LONG)) ;
        expectedList.add(new FieldSchema("2d", DataType.MAP)) ;

        Schema expected = new Schema(expectedList) ;

        // Compare
        Assert.assertTrue(Schema.equals(mergedSchema, expected, false, false)) ;
View Full Code Here
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
   
    @Test
    public void testSchemaEqualTwoLevelAccess() throws Exception {
       
        List<FieldSchema> innerList1 = new ArrayList<FieldSchema>() ;
        innerList1.add(new FieldSchema("11a", DataType.INTEGER)) ;
        innerList1.add(new FieldSchema("11b", DataType.LONG)) ;
       
        List<FieldSchema> innerList2 = new ArrayList<FieldSchema>() ;
        innerList2.add(new FieldSchema("11a", DataType.INTEGER)) ;
        innerList2.add(new FieldSchema("11b", DataType.LONG)) ;
       
        Schema innerSchema1 = new Schema(innerList1) ;
        Schema innerSchema2 = new Schema(innerList2) ;
               
        List<FieldSchema> list1 = new ArrayList<FieldSchema>() ;
        list1.add(new FieldSchema("1a", DataType.BYTEARRAY)) ;
        list1.add(new FieldSchema("1b", innerSchema1)) ;
        list1.add(new FieldSchema("1c", DataType.INTEGER)) ;
       
        List<FieldSchema> list2 = new ArrayList<FieldSchema>() ;
        list2.add(new FieldSchema("1a", DataType.BYTEARRAY)) ;
        list2.add(new FieldSchema("1b", innerSchema2)) ;
        list2.add(new FieldSchema("1c", DataType.INTEGER)) ;
       
        Schema schema1 = new Schema(list1) ;
        Schema schema2 = new Schema(list2) ;       
     
        Schema.FieldSchema bagFs1 = new Schema.FieldSchema("b", schema1, DataType.BAG);
View Full Code Here
75
76
77
78
79
80
81
82
83
84
85
            description = "autogenerated from Pig Field Schema";
            Schema inner = fieldSchema.schema;
            if (type == DataType.BAG && fieldSchema.schema != null
                    && !fieldSchema.schema.isTwoLevelAccessRequired()) {
                log.info("Insert two-level access to Resource Schema");
                FieldSchema fs = new FieldSchema("t", fieldSchema.schema);
                inner = new Schema(fs);               
            }
           
            // allow partial schema
            if ((type == DataType.BAG || type == DataType.TUPLE)
View Full Code Here
5738
5739
5740
5741
5742
5743
5744
5745
5746
5747
5748
5749
5750
            return null;
        }
       
        @Override
        public Schema outputSchema(Schema input) {
            Schema.FieldSchema charFs = new FieldSchema(null, DataType.CHARARRAY);
            Schema.FieldSchema intFs = new FieldSchema(null, DataType.INTEGER);
            Schema.FieldSchema floatFs = new FieldSchema(null, DataType.FLOAT);
            Schema bagSchema = new Schema();
            bagSchema.add(charFs);
            bagSchema.add(intFs);
            bagSchema.add(floatFs);
            Schema.FieldSchema bagFs;
View Full Code Here
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
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
        if (columnsToRead == null) {

            fieldschema = new ArrayList<FieldSchema>(cols.size());

            for(int i = 0; i < cols.size(); i++){
                fieldschema.add(new FieldSchema(cols.get(i), HiveRCSchemaUtil.findPigDataType(types.get(i))));
            }


        } else {
            //compile list for column filtering
            Set<String> columnToReadList = HiveRCSchemaUtil.compileSet(columnsToRead);

            if(columnToReadList.size() < 1)
                throw new RuntimeException("Error parsing columns: " + columnsToRead);

            applyColumnRead = true;
            int columnToReadLen = columnToReadList.size();

            fieldschema = new ArrayList<FieldSchema>(columnToReadLen);


            //--- create Pig Schema and add columnToReadPositions.
            columnToReadPositions = new int[columnToReadLen];

            int len = cols.size();
            String columnName = null;
            int colArrayPosindex = 0;

            for(int i = 0; i < len; i++){
                //i is the column position
                columnName = cols.get(i);
                if(columnToReadList.contains(columnName)){
                    //if the column is contained in the columnList then add its position to the columnPositions array and to the pig schema
                    columnToReadPositions[colArrayPosindex++] = i;

                    fieldschema.add(new FieldSchema(columnName, HiveRCSchemaUtil.findPigDataType(types.get(i))));
                }

            }


            //sort column positions
            Arrays.sort(columnToReadPositions);

        }

        if(includeDayDateColumn){
            fieldschema.add(new FieldSchema(DAY_DATE_COLUMN, DataType.CHARARRAY));
        }

        pigSchema = new ResourceSchema(new Schema(fieldschema));

        props =  new Properties();
View Full Code Here
553
554
555
556
557
558
559
560
561
562
563
        constant1.setFieldSchema(new Schema.FieldSchema(null, innerTupleSchema, DataType.TUPLE));
        LOConst constant2 =  new LOConst(plan, genNewOperatorKey(), tuple) ;
        constant2.setType(DataType.TUPLE) ;
        constant2.setFieldSchema(new Schema.FieldSchema(null, tupleSchema, DataType.TUPLE));
        LOCast cast1 = new LOCast(plan, genNewOperatorKey(), DataType.TUPLE) ;
        cast1.setFieldSchema(new FieldSchema(null, castSchema, DataType.TUPLE));
       
        LOEqual equal1 = new LOEqual(plan, genNewOperatorKey()) ;
       
        plan.add(constant1) ;
        plan.add(constant2) ;
View Full Code Here
TOP

Related Classes of org.apache.pig.impl.logicalLayer.schema.Schema.FieldSchema

  • com.acme.math.Pow
  • com.cloudera.science.pig.Bin
  • com.cloudera.science.pig.Combinatorial
  • com.cloudera.science.pig.Quantile
  • com.mozilla.pig.eval.Coalesce
  • com.twitter.elephantbird.pig.load.LocationAsTuple
  • com.twitter.elephantbird.pig.load.LzoW3CLogLoader
  • com.twitter.elephantbird.pig.mahout.VectorWritableConverter
  • com.twitter.elephantbird.pig.piggybank.GenericInvoker
  • com.twitter.elephantbird.pig.util.ProtobufToPig
    • 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.