Examples of org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector

Package org.apache.hadoop.hive.ql.exec.vector

Examples of org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector

org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector
This class supports string and binary data by value reference -- i.e. each field is explicitly present, as opposed to provided by a dictionary reference. In some cases, all the values will be in the same byte array to begin with, but this need not be the case. If each value is in a separate byte array to start with, or not all of the values are in the same original byte array, you can still assign data by reference into this column vector. This gives flexibility to use this in multiple situations.
When setting data by reference, the caller is responsible for allocating the byte arrays used to hold the data. You can also set data by value, as long as you call the initBuffer() method first. You can mix "by value" and "by reference" in the same column vector, though that use is probably not typical.

    return b;
  }


  private VectorizedRowBatch getBatchStrDblLongWithStrOut() {
    VectorizedRowBatch b = new VectorizedRowBatch(4);
    BytesColumnVector strCol = new BytesColumnVector();
    LongColumnVector longCol = new LongColumnVector();
    DoubleColumnVector dblCol = new DoubleColumnVector();
    BytesColumnVector outCol = new BytesColumnVector();
    b.cols[0] = strCol;
    b.cols[1] = longCol;
    b.cols[2] = dblCol;
    b.cols[3] = outCol;


    strCol.initBuffer();
    strCol.setVal(0, blue, 0, blue.length);
    strCol.setVal(1, red, 0, red.length);
    longCol.vector[0] = 0;
    longCol.vector[1] = 1;
    dblCol.vector[0] = 0.0;
    dblCol.vector[1] = 1.0;


    // set one null value for possible later use
    longCol.isNull[1] = true;


    // but have no nulls initially
    longCol.noNulls = true;
    strCol.noNulls = true;
    dblCol.noNulls = true;
    outCol.initBuffer();
    b.size = 2;
    return b;
  }

View Full Code Here

      red = "red".getBytes("UTF-8");
      unknown = "UNKNOWN".getBytes("UTF-8");
    } catch (Exception e) {
      ;
    }
    BytesColumnVector out;


    // with nulls
    b = getBatchStrDblLongWithStrOut();
    b.cols[0].noNulls = false;
    b.cols[0].isNull[0] = true; // set 1st entry to null

View Full Code Here

            lcv.vector[row] = TimestampUtils.getTimeNanoSec(t);
          }
        };


      } else if (types[i].equalsIgnoreCase("string")) {
        batch.cols[i] = new BytesColumnVector(batchSize);
        columnAssign[i] = new ColumnVectorAssign() {
          @Override
          public void assign(
              ColumnVector columnVector,
              int row,
              Object value) {
            BytesColumnVector bcv = (BytesColumnVector) columnVector;
            String s = (String) value;
            byte[] bytes = s.getBytes();
            bcv.vector[row] = bytes;
            bcv.start[row] = 0;
            bcv.length[row] = bytes.length;

View Full Code Here

    int size = 20;
    VectorizedRowBatch vrg = VectorizedRowGroupGenUtil.getVectorizedRowBatch(size, 4, 0);


    LongColumnVector lcv = (LongColumnVector) vrg.cols[0];
    DoubleColumnVector dcv = new DoubleColumnVector(size);
    BytesColumnVector bcv = new BytesColumnVector(size);
    DecimalColumnVector dv = new DecimalColumnVector(5, 1);
    vrg.cols[1] = dcv;
    vrg.cols[2] = bcv;
    vrg.cols[3] = dv;

View Full Code Here

    lv.vector[2] = 1;
    lv.vector[3] = 1;
    batch.cols[0] = lv;


    // set second argument to IF
    BytesColumnVector v = new BytesColumnVector();
    v.initBuffer();
    setString(v, 0, "arg2_0");
    setString(v, 1, "arg2_1");
    setString(v, 2, "arg2_2");
    setString(v, 3, "arg2_3");


    batch.cols[1] = v;


    // set third argument to IF
    v = new BytesColumnVector();
    v.initBuffer();
    setString(v, 0, "arg3_0");
    setString(v, 1, "arg3_1");
    setString(v, 2, "arg3_2");
    setString(v, 3, "arg3_3");
    batch.cols[2] = v;


    // set output column
    v = new BytesColumnVector();
    v.initBuffer();
    batch.cols[3] = v;
    batch.size = 4;
    return batch;
  }

View Full Code Here


  @Test
  public void testIfExprStringColumnStringColumn() {
    VectorizedRowBatch batch = getBatch1Long3BytesVectors();
    VectorExpression expr = new IfExprStringColumnStringColumn(0, 1, 2, 3);
    BytesColumnVector r = (BytesColumnVector) batch.cols[3];
    expr.evaluate(batch);
    assertTrue(getString(r, 0).equals("arg3_0"));
    assertTrue(getString(r, 1).equals("arg3_1"));
    assertTrue(getString(r, 2).equals("arg2_2"));
    assertTrue(getString(r, 3).equals("arg2_3"));

View Full Code Here

  @Test
  public void testIfExprStringColumnStringScalar() {
    VectorizedRowBatch batch = getBatch1Long3BytesVectors();
    byte[] scalar = getUTF8Bytes("scalar");
    VectorExpression expr = new IfExprStringColumnStringScalar(0, 1, scalar, 3);
    BytesColumnVector r = (BytesColumnVector) batch.cols[3];
    expr.evaluate(batch);
    assertTrue(getString(r, 0).equals("scalar"));
    assertTrue(getString(r, 1).equals("scalar"));
    assertTrue(getString(r, 2).equals("arg2_2"));
    assertTrue(getString(r, 3).equals("arg2_3"));
    assertTrue(r.noNulls);


    // test for null input strings
    batch = getBatch1Long3BytesVectors();
    BytesColumnVector arg2 = (BytesColumnVector) batch.cols[1];
    arg2.noNulls = false;
    arg2.isNull[2] = true;
    arg2.vector[2] = null;
    expr.evaluate(batch);
    r = (BytesColumnVector) batch.cols[3];

View Full Code Here

  @Test
  public void testIfExprStringScalarStringColumn() {
    VectorizedRowBatch batch = getBatch1Long3BytesVectors();
    byte[] scalar = getUTF8Bytes("scalar");
    VectorExpression expr = new IfExprStringScalarStringColumn(0,scalar, 2, 3);
    BytesColumnVector r = (BytesColumnVector) batch.cols[3];
    expr.evaluate(batch);
    assertTrue(getString(r, 0).equals("arg3_0"));
    assertTrue(getString(r, 1).equals("arg3_1"));
    assertTrue(getString(r, 2).equals("scalar"));
    assertTrue(getString(r, 3).equals("scalar"));
    assertTrue(r.noNulls);


    // test for null input strings
    batch = getBatch1Long3BytesVectors();
    BytesColumnVector arg3 = (BytesColumnVector) batch.cols[2];
    arg3.noNulls = false;
    arg3.isNull[1] = true;
    arg3.vector[1] = null;
    expr.evaluate(batch);
    r = (BytesColumnVector) batch.cols[3];

View Full Code Here

    // standard case
    VectorizedRowBatch batch = getBatch1Long3BytesVectors();
    byte[] scalar1 = getUTF8Bytes("scalar1");
    byte[] scalar2 = getUTF8Bytes("scalar2");
    VectorExpression expr = new IfExprStringScalarStringScalar(0,scalar1, scalar2, 3);
    BytesColumnVector r = (BytesColumnVector) batch.cols[3];
    expr.evaluate(batch);
    assertTrue(getString(r, 0).equals("scalar2"));
    assertTrue(getString(r, 1).equals("scalar2"));
    assertTrue(getString(r, 2).equals("scalar1"));
    assertTrue(getString(r, 3).equals("scalar1"));

View Full Code Here

    return batch;
  }


  private VectorizedRowBatch getVectorizedRandomRowBatchStringLong(int seed, int size) {
    VectorizedRowBatch batch = new VectorizedRowBatch(2, size);
    BytesColumnVector bcv = new BytesColumnVector(size);
    Random rand = new Random(seed);
    for (int i = 0; i < size; i++) {
      /* all 32 bit numbers qualify & multiply up to get nano-seconds */
      byte[] encoded = encodeTime(1000 * 1000 * 1000 * rand.nextInt());
      bcv.vector[i] = encoded;

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector

org.apache.hadoop.hive.ql.exec.vector.expressions.AbstractFilterStringColLikeStringScalar

org.apache.hadoop.hive.ql.exec.vector.expressions.aggregates.gen.VectorUDAFMaxString

org.apache.hadoop.hive.ql.exec.vector.expressions.aggregates.gen.VectorUDAFMinString

org.apache.hadoop.hive.ql.exec.vector.expressions.CastStringToDate

org.apache.hadoop.hive.ql.exec.vector.expressions.CastStringToDecimal

org.apache.hadoop.hive.ql.exec.vector.expressions.DecimalToStringUnaryUDF

org.apache.hadoop.hive.ql.exec.vector.expressions.FilterStringColumnInList

org.apache.hadoop.hive.ql.exec.vector.expressions.FuncLongToString

org.apache.hadoop.hive.ql.exec.vector.expressions.gen.FilterStringColEqualStringColumn

org.apache.hadoop.hive.ql.exec.vector.expressions.gen.FilterStringColEqualStringScalar

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.