Examples of org.apache.hadoop.hive.serde2.objectinspector.primitive.StringObjectInspector

Package org.apache.hadoop.hive.serde2.objectinspector.primitive

Examples of org.apache.hadoop.hive.serde2.objectinspector.primitive.StringObjectInspector

org.apache.hadoop.hive.serde2.objectinspector.primitive.StringObjectInspector
A StringObjectInspector inspects an Object representing a String.

        DoubleObjectInspector doi = (DoubleObjectInspector) poi;
        LazyBinaryUtils.writeDouble(byteStream, doi.get(obj));
        return;
      }
      case STRING: {
        StringObjectInspector soi = (StringObjectInspector) poi;
        Text t = soi.getPrimitiveWritableObject(obj);
        serializeText(byteStream, t, skipLengthPrefix);
        return;
      }
      case CHAR: {
        HiveCharObjectInspector hcoi = (HiveCharObjectInspector) poi;

View Full Code Here

        r.set(doi.get(o));
        tbOut.write(r);
        return;
      }
      case STRING: {
        StringObjectInspector soi = (StringObjectInspector) poi;
        Text t = soi.getPrimitiveWritableObject(o);
        tbOut.write(t);
        return;
      }
      default: {
        throw new RuntimeException("Unrecognized type: "

View Full Code Here

      Object field = outputRowOI
          .getStructFieldData(obj, outputFieldRefs.get(c));
      ObjectInspector fieldOI = outputFieldRefs.get(c)
          .getFieldObjectInspector();
      // The data must be of type String
      StringObjectInspector fieldStringOI = (StringObjectInspector) fieldOI;
      // Convert the field to Java class String, because objects of String type
      // can be
      // stored in String, Text, or some other classes.
      outputFields[c] = fieldStringOI.getPrimitiveJavaObject(field);
    }


    // Format the String
    String outputRowString = null;
    try {

View Full Code Here

        writeByte(buffer, (byte) (v >> 8), invert);
        writeByte(buffer, (byte) v, invert);
        return;
      }
      case STRING: {
        StringObjectInspector soi = (StringObjectInspector) poi;
        Text t = soi.getPrimitiveWritableObject(o);
        serializeBytes(buffer, t.getBytes(), t.getLength(), invert);
        return;
      }


      case CHAR: {
        HiveCharObjectInspector hcoi = (HiveCharObjectInspector) poi;
        HiveCharWritable hc = hcoi.getPrimitiveWritableObject(o);
        // Trailing space should ignored for char comparisons.
        // So write stripped values for this SerDe.
        Text t = hc.getStrippedValue();
        serializeBytes(buffer, t.getBytes(), t.getLength(), invert);
        return;
      }
      case VARCHAR: {
        HiveVarcharObjectInspector hcoi = (HiveVarcharObjectInspector)poi;
        HiveVarcharWritable hc = hcoi.getPrimitiveWritableObject(o);
        // use varchar's text field directly
        Text t = hc.getTextValue();
        serializeBytes(buffer, t.getBytes(), t.getLength(), invert);
        return;
      }


      case BINARY: {
        BinaryObjectInspector baoi = (BinaryObjectInspector) poi;
        BytesWritable ba = baoi.getPrimitiveWritableObject(o);
        byte[] toSer = new byte[ba.getLength()];
        System.arraycopy(ba.getBytes(), 0, toSer, 0, ba.getLength());
        serializeBytes(buffer, toSer, ba.getLength(), invert);
        return;
      }
      case  DATE: {
        DateObjectInspector doi = (DateObjectInspector) poi;
        int v = doi.getPrimitiveWritableObject(o).getDays();
        serializeInt(buffer, v, invert);
        return;
      }
      case TIMESTAMP: {
        TimestampObjectInspector toi = (TimestampObjectInspector) poi;
        TimestampWritable t = toi.getPrimitiveWritableObject(o);
        byte[] data = t.getBinarySortable();
        for (int i = 0; i < data.length; i++) {
          writeByte(buffer, data[i], invert);
        }
        return;
      }
      case DECIMAL: {
        // decimals are encoded in three pieces:
        // sign: 1, 2 or 3 for smaller, equal or larger than 0 respectively
        // factor: Number that indicates the amount of digits you have to move
        // the decimal point left or right until the resulting number is smaller
        // than zero but has something other than 0 as the first digit.
        // digits: which is a string of all the digits in the decimal. If the number
        // is negative the binary string will be inverted to get the correct ordering.
        // Example: 0.00123
        // Sign is 3 (bigger than 0)
        // Factor is -2 (move decimal point 2 positions right)
        // Digits are: 123


        HiveDecimalObjectInspector boi = (HiveDecimalObjectInspector) poi;
        HiveDecimal dec = boi.getPrimitiveJavaObject(o);


        // get the sign of the big decimal
        int sign = dec.compareTo(HiveDecimal.ZERO);


        // we'll encode the absolute value (sign is separate)
        dec = dec.abs();


        // get the scale factor to turn big decimal into a decimal < 1
        int factor = dec.precision() - dec.scale();
        factor = sign == 1 ? factor : -factor;


        // convert the absolute big decimal to string
        dec.scaleByPowerOfTen(Math.abs(dec.scale()));
        String digits = dec.unscaledValue().toString();


        // finally write out the pieces (sign, scale, digits)
        writeByte(buffer, (byte) ( sign + 1), invert);
        writeByte(buffer, (byte) ((factor >> 24) ^ 0x80), invert);
        writeByte(buffer, (byte) ( factor >> 16), invert);
        writeByte(buffer, (byte) ( factor >> 8), invert);
        writeByte(buffer, (byte)   factor, invert);
        serializeBytes(buffer, digits.getBytes(decimalCharSet),
            digits.length(), sign == -1 ? !invert : invert);
        return;
      }


      default: {
        throw new RuntimeException("Unrecognized type: "
            + poi.getPrimitiveCategory());
      }
      }
    }
    case LIST: {
      ListObjectInspector loi = (ListObjectInspector) oi;
      ObjectInspector eoi = loi.getListElementObjectInspector();


      // \1 followed by each element
      int size = loi.getListLength(o);
      for (int eid = 0; eid < size; eid++) {
        writeByte(buffer, (byte) 1, invert);
        serialize(buffer, loi.getListElement(o, eid), eoi, invert);
      }
      // and \0 to terminate
      writeByte(buffer, (byte) 0, invert);
      return;
    }
    case MAP: {
      MapObjectInspector moi = (MapObjectInspector) oi;
      ObjectInspector koi = moi.getMapKeyObjectInspector();
      ObjectInspector voi = moi.getMapValueObjectInspector();


      // \1 followed by each key and then each value
      Map<?, ?> map = moi.getMap(o);
      for (Map.Entry<?, ?> entry : map.entrySet()) {
        writeByte(buffer, (byte) 1, invert);
        serialize(buffer, entry.getKey(), koi, invert);
        serialize(buffer, entry.getValue(), voi, invert);
      }
      // and \0 to terminate
      writeByte(buffer, (byte) 0, invert);
      return;
    }
    case STRUCT: {
      StructObjectInspector soi = (StructObjectInspector) oi;
      List<? extends StructField> fields = soi.getAllStructFieldRefs();


      for (int i = 0; i < fields.size(); i++) {
        serialize(buffer, soi.getStructFieldData(o, fields.get(i)), fields.get(
            i).getFieldObjectInspector(), invert);
      }
      return;
    }
    case UNION: {

View Full Code Here

    for (int c = 0; c < numCols; c++) {
      final Object field = outputRowOI.getStructFieldData(obj, outputFieldRefs.get(c));
      final ObjectInspector fieldOI = outputFieldRefs.get(c).getFieldObjectInspector();


      // The data must be of type String
      final StringObjectInspector fieldStringOI = (StringObjectInspector) fieldOI;


      // Convert the field to Java class String, because objects of String type
      // can be stored in String, Text, or some other classes.
      outputFields[c] = fieldStringOI.getPrimitiveJavaObject(field);
    }


    final StringWriter writer = new StringWriter();
    final CSVWriter csv = newWriter(writer, separatorChar, quoteChar, escapeChar);

View Full Code Here


    assertEquals("baddies", fieldRef.getFieldName());


    Object theStringObject = oi.getStructFieldData(row, fieldRef);
    assertTrue(fieldRef.getFieldObjectInspector() instanceof StringObjectInspector);
    StringObjectInspector soi = (StringObjectInspector)fieldRef.getFieldObjectInspector();


    String finalValue = soi.getPrimitiveJavaObject(theStringObject);
    assertEquals("DALEKS", finalValue);
  }

View Full Code Here

    StandardStructObjectInspector oi = (StandardStructObjectInspector)aoig.getObjectInspector();
    List<Object> fieldsDataAsList = oi.getStructFieldsDataAsList(row);
    assertEquals(1, fieldsDataAsList.size());
    StructField fieldRef = oi.getStructFieldRef(fieldName);
    ObjectInspector fieldObjectInspector = fieldRef.getFieldObjectInspector();
    StringObjectInspector soi = (StringObjectInspector)fieldObjectInspector;


    if(expected == null) {
      assertNull(soi.getPrimitiveJavaObject(rowElement));
    } else {
      assertEquals(expected, soi.getPrimitiveJavaObject(rowElement));
    }
  }

View Full Code Here

    }


    charsetOI = (StringObjectInspector) arguments[1];


    // If the character set for encoding is constant, we can optimize that
    StringObjectInspector charSetOI = (StringObjectInspector) arguments[1];
    if (charSetOI instanceof ConstantObjectInspector){
      String charSetName = ((Text) ((ConstantObjectInspector) charSetOI).getWritableConstantValue()).toString();
      decoder = Charset.forName(charSetName).newDecoder().onMalformedInput(CodingErrorAction.REPORT).onUnmappableCharacter(CodingErrorAction.REPORT);
    }

View Full Code Here

    org.apache.hadoop.mapred.RecordReader reader =
        in.getRecordReader(splits[0], conf, Reporter.NULL);
    Object key = reader.createKey();
    Writable value = (Writable) reader.createValue();
    List<? extends StructField> fields =inspector.getAllStructFieldRefs();
    StringObjectInspector strInspector = (StringObjectInspector)
        fields.get(0).getFieldObjectInspector();
    assertEquals(true, reader.next(key, value));
    assertEquals("owen", strInspector.getPrimitiveJavaObject(inspector.
        getStructFieldData(value, fields.get(0))));
    assertEquals(true, reader.next(key, value));
    assertEquals("beth", strInspector.getPrimitiveJavaObject(inspector.
        getStructFieldData(value, fields.get(0))));
    assertEquals(true, reader.next(key, value));
    assertEquals("laurel", strInspector.getPrimitiveJavaObject(inspector.
        getStructFieldData(value, fields.get(0))));
    assertEquals(true, reader.next(key, value));
    assertEquals("hazen", strInspector.getPrimitiveJavaObject(inspector.
        getStructFieldData(value, fields.get(0))));
    assertEquals(true, reader.next(key, value));
    assertEquals("colin", strInspector.getPrimitiveJavaObject(inspector.
        getStructFieldData(value, fields.get(0))));
    assertEquals(true, reader.next(key, value));
    assertEquals("miles", strInspector.getPrimitiveJavaObject(inspector.
        getStructFieldData(value, fields.get(0))));
    assertEquals(false, reader.next(key, value));
    reader.close();
  }

View Full Code Here

        .getStructFieldRef("float1").getFieldObjectInspector();
    DoubleObjectInspector dbl = (DoubleObjectInspector) readerInspector
        .getStructFieldRef("double1").getFieldObjectInspector();
    BinaryObjectInspector bi = (BinaryObjectInspector) readerInspector
        .getStructFieldRef("bytes1").getFieldObjectInspector();
    StringObjectInspector st = (StringObjectInspector) readerInspector
        .getStructFieldRef("string1").getFieldObjectInspector();
    StructObjectInspector mid = (StructObjectInspector) readerInspector
        .getStructFieldRef("middle").getFieldObjectInspector();
    List<? extends StructField> midFields = mid.getAllStructFieldRefs();
    ListObjectInspector midli = (ListObjectInspector) midFields.get(0)
        .getFieldObjectInspector();
    StructObjectInspector inner = (StructObjectInspector) midli
        .getListElementObjectInspector();
    List<? extends StructField> inFields = inner.getAllStructFieldRefs();
    ListObjectInspector li = (ListObjectInspector) readerInspector
        .getStructFieldRef("list").getFieldObjectInspector();
    MapObjectInspector ma = (MapObjectInspector) readerInspector
        .getStructFieldRef("map").getFieldObjectInspector();
    TimestampObjectInspector tso = (TimestampObjectInspector) readerInspector
        .getStructFieldRef("ts").getFieldObjectInspector();
    HiveDecimalObjectInspector dco = (HiveDecimalObjectInspector) readerInspector
        .getStructFieldRef("decimal1").getFieldObjectInspector();
    StringObjectInspector mk = (StringObjectInspector) ma
        .getMapKeyObjectInspector();
    RecordReader rows = reader.rows();
    Object row = rows.next(null);
    assertNotNull(row);
    // check the contents of the first row
    assertEquals(false,
        bo.get(readerInspector.getStructFieldData(row, fields.get(0))));
    assertEquals(1,
        by.get(readerInspector.getStructFieldData(row, fields.get(1))));
    assertEquals(1024,
        sh.get(readerInspector.getStructFieldData(row, fields.get(2))));
    assertEquals(65536,
        in.get(readerInspector.getStructFieldData(row, fields.get(3))));
    assertEquals(Long.MAX_VALUE,
        lo.get(readerInspector.getStructFieldData(row, fields.get(4))));
    assertEquals(1.0,
        fl.get(readerInspector.getStructFieldData(row, fields.get(5))), 0.00001);
    assertEquals(-15.0,
        dbl.get(readerInspector.getStructFieldData(row, fields.get(6))),
        0.00001);
    assertEquals(bytes(0, 1, 2, 3, 4),
        bi.getPrimitiveWritableObject(readerInspector.getStructFieldData(row,
            fields.get(7))));
    assertEquals("hi", st.getPrimitiveJavaObject(readerInspector
        .getStructFieldData(row, fields.get(8))));
    List<?> midRow = midli.getList(mid.getStructFieldData(
        readerInspector.getStructFieldData(row, fields.get(9)),
        midFields.get(0)));
    assertNotNull(midRow);
    assertEquals(2, midRow.size());
    assertEquals(1,
        in.get(inner.getStructFieldData(midRow.get(0), inFields.get(0))));
    assertEquals("bye", st.getPrimitiveJavaObject(inner.getStructFieldData(
        midRow.get(0), inFields.get(1))));
    assertEquals(2,
        in.get(inner.getStructFieldData(midRow.get(1), inFields.get(0))));
    assertEquals("sigh", st.getPrimitiveJavaObject(inner.getStructFieldData(
        midRow.get(1), inFields.get(1))));
    List<?> list = li.getList(readerInspector.getStructFieldData(row,
        fields.get(10)));
    assertEquals(2, list.size());
    assertEquals(3,
        in.get(inner.getStructFieldData(list.get(0), inFields.get(0))));
    assertEquals("good", st.getPrimitiveJavaObject(inner.getStructFieldData(
        list.get(0), inFields.get(1))));
    assertEquals(4,
        in.get(inner.getStructFieldData(list.get(1), inFields.get(0))));
    assertEquals("bad", st.getPrimitiveJavaObject(inner.getStructFieldData(
        list.get(1), inFields.get(1))));
    Map<?, ?> map = ma.getMap(readerInspector.getStructFieldData(row,
        fields.get(11)));
    assertEquals(0, map.size());
    assertEquals(Timestamp.valueOf("2000-03-12 15:00:00"),
        tso.getPrimitiveJavaObject(readerInspector.getStructFieldData(row,
            fields.get(12))));
    assertEquals(HiveDecimal.create("12345678.6547456"),
        dco.getPrimitiveJavaObject(readerInspector.getStructFieldData(row,
            fields.get(13))));


    // check the contents of second row
    assertEquals(true, rows.hasNext());
    rows.seekToRow(7499);
    row = rows.next(null);
    assertEquals(true,
        bo.get(readerInspector.getStructFieldData(row, fields.get(0))));
    assertEquals(100,
        by.get(readerInspector.getStructFieldData(row, fields.get(1))));
    assertEquals(2048,
        sh.get(readerInspector.getStructFieldData(row, fields.get(2))));
    assertEquals(65536,
        in.get(readerInspector.getStructFieldData(row, fields.get(3))));
    assertEquals(Long.MAX_VALUE,
        lo.get(readerInspector.getStructFieldData(row, fields.get(4))));
    assertEquals(2.0,
        fl.get(readerInspector.getStructFieldData(row, fields.get(5))), 0.00001);
    assertEquals(-5.0,
        dbl.get(readerInspector.getStructFieldData(row, fields.get(6))),
        0.00001);
    assertEquals(bytes(), bi.getPrimitiveWritableObject(readerInspector
        .getStructFieldData(row, fields.get(7))));
    assertEquals("bye", st.getPrimitiveJavaObject(readerInspector
        .getStructFieldData(row, fields.get(8))));
    midRow = midli.getList(mid.getStructFieldData(
        readerInspector.getStructFieldData(row, fields.get(9)),
        midFields.get(0)));
    assertNotNull(midRow);
    assertEquals(2, midRow.size());
    assertEquals(1,
        in.get(inner.getStructFieldData(midRow.get(0), inFields.get(0))));
    assertEquals("bye", st.getPrimitiveJavaObject(inner.getStructFieldData(
        midRow.get(0), inFields.get(1))));
    assertEquals(2,
        in.get(inner.getStructFieldData(midRow.get(1), inFields.get(0))));
    assertEquals("sigh", st.getPrimitiveJavaObject(inner.getStructFieldData(
        midRow.get(1), inFields.get(1))));
    list = li.getList(readerInspector.getStructFieldData(row, fields.get(10)));
    assertEquals(3, list.size());
    assertEquals(100000000,
        in.get(inner.getStructFieldData(list.get(0), inFields.get(0))));
    assertEquals("cat", st.getPrimitiveJavaObject(inner.getStructFieldData(
        list.get(0), inFields.get(1))));
    assertEquals(-100000,
        in.get(inner.getStructFieldData(list.get(1), inFields.get(0))));
    assertEquals("in", st.getPrimitiveJavaObject(inner.getStructFieldData(
        list.get(1), inFields.get(1))));
    assertEquals(1234,
        in.get(inner.getStructFieldData(list.get(2), inFields.get(0))));
    assertEquals("hat", st.getPrimitiveJavaObject(inner.getStructFieldData(
        list.get(2), inFields.get(1))));
    map = ma.getMap(readerInspector.getStructFieldData(row, fields.get(11)));
    assertEquals(2, map.size());
    boolean[] found = new boolean[2];
    for(Object key : map.keySet()) {
      String str = mk.getPrimitiveJavaObject(key);
      if (str.equals("chani")) {
        assertEquals(false, found[0]);
        assertEquals(5,
            in.get(inner.getStructFieldData(map.get(key), inFields.get(0))));
        assertEquals(str, st.getPrimitiveJavaObject(inner.getStructFieldData(

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.hadoop.hive.serde2.objectinspector.primitive.StringObjectInspector

com.asakusafw.directio.hive.orc.OrcFileInput

com.asakusafw.directio.hive.serde.DataModelDriver

com.asakusafw.directio.hive.serde.DataModelDriverTest

com.asakusafw.directio.hive.serde.ValueSerdeFactoryTest

com.facebook.hive.orc.lazy.LazyDoubleTreeReader

com.facebook.hive.orc.lazy.LazyShortDictionaryTreeReader

com.facebook.hive.orc.lazy.LazyShortDirectTreeReader

com.facebook.hive.orc.lazy.OrcLazyStructObjectInspector

com.facebook.hive.orc.OrcOutputFormat$OrcRecordWriter

com.facebook.hive.orc.OrcStruct$OrcStructInspector

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.