Example usage for org.apache.spark.unsafe.types UTF8String fromString

Introduction

In this page you can find the example usage for org.apache.spark.unsafe.types UTF8String fromString.

Prototype

public static UTF8String fromString(String str) 

Source Link

Document

Creates an UTF8String from String.

Usage

From source file:com.github.sadikovi.riff.tree.FilterApi.java

License:Open Source License

/**
 * Method to convert object into typed expression, throws exception if no match is found
 * @param obj object to convert, must never be null
 * @return typed expression for this object
 *///from  w  ww.  j av a  2  s.  co  m
protected static TypedExpression objToExpression(Object obj) {
    if (obj == null) {
        throw new NullPointerException("Cannot convert null into typed expression");
    }

    if (obj instanceof Integer) {
        return new IntegerExpression((Integer) obj);
    } else if (obj instanceof Long) {
        return new LongExpression((Long) obj);
    } else if (obj instanceof String) {
        return new UTF8StringExpression(UTF8String.fromString((String) obj));
    } else if (obj instanceof UTF8String) {
        return new UTF8StringExpression((UTF8String) obj);
    } else if (obj instanceof Date) {
        return new DateExpression((Date) obj);
    } else if (obj instanceof Timestamp) {
        return new TimestampExpression((Timestamp) obj);
    } else if (obj instanceof Boolean) {
        return new BooleanExpression((Boolean) obj);
    } else if (obj instanceof Short) {
        return new ShortExpression((Short) obj);
    } else if (obj instanceof Byte) {
        return new ByteExpression((Byte) obj);
    } else {
        throw new UnsupportedOperationException("Object " + obj + " of " + obj.getClass());
    }
}

---

From source file:com.hurence.logisland.connect.source.KafkaConnectStreamSource.java

License:Apache License

private UTF8String toUTFString(Object o) {
    if (o != null) {
        return UTF8String.fromString(o.toString());
    }/*from   w  w w .  ja  va 2 s  . c  o m*/
    return UTF8String.EMPTY_UTF8;
}

---

From source file:com.splicemachine.db.iapi.types.SQLChar.java

License:Apache License

/**
 *
 * Write into the Project Tungsten Format (UnsafeRow).
 *
 * @see UnsafeRowWriter#write(int, UTF8String)
 *
 * @param unsafeRowWriter/*from w  w w.j a v a 2  s. c om*/
 * @param ordinal
 */
@Override
public void write(UnsafeRowWriter unsafeRowWriter, int ordinal) {
    if (isNull())
        unsafeRowWriter.setNullAt(ordinal);
    else {
        unsafeRowWriter.write(ordinal, UTF8String.fromString(value));
    }
}

---

From source file:org.apache.carbondata.converter.SparkDataTypeConverterImpl.java

License:Apache License

@Override
public byte[] convertFromStringToByte(Object data) {
    if (null == data) {
        return null;
    }/*w w w. j  a  va 2s .c  o  m*/
    return UTF8String.fromString((String) data).getBytes();
}

---

From source file:org.apache.carbondata.converter.SparkDataTypeConverterImpl.java

License:Apache License

@Override
public Object convertFromStringToUTF8String(Object data) {
    if (null == data) {
        return null;
    }/*  w w  w . ja v a 2s  .  c om*/
    return UTF8String.fromString((String) data);
}

---

From source file:org.apache.carbondata.core.scan.collector.impl.RestructureBasedRawResultCollector.java

License:Apache License

/**
 * This method will fill the no dictionary byte array with newly added no dictionary columns
 *
 * @param noDictionaryKeyArray/*from  w  w w. j a v a 2s .  c o m*/
 * @return
 */
private byte[][] fillNoDictionaryKeyArrayWithLatestSchema(byte[][] noDictionaryKeyArray) {
    QueryDimension[] actualQueryDimensions = tableBlockExecutionInfos.getActualQueryDimensions();
    byte[][] noDictionaryKeyArrayWithNewlyAddedColumns = new byte[noDictionaryKeyArray.length
            + dimensionInfo.getNewNoDictionaryColumnCount()][];
    int existingColumnValueIndex = 0;
    int newKeyArrayIndex = 0;
    for (int i = 0; i < dimensionInfo.getDimensionExists().length; i++) {
        if (!actualQueryDimensions[i].getDimension().hasEncoding(Encoding.DICTIONARY)
                && !actualQueryDimensions[i].getDimension().hasEncoding(Encoding.IMPLICIT)) {
            // if dimension exists then add the byte array value else add the default value
            if (dimensionInfo.getDimensionExists()[i]) {
                noDictionaryKeyArrayWithNewlyAddedColumns[newKeyArrayIndex++] = noDictionaryKeyArray[existingColumnValueIndex++];
            } else {
                byte[] newColumnDefaultValue = null;
                Object defaultValue = dimensionInfo.getDefaultValues()[i];
                if (null != defaultValue) {
                    newColumnDefaultValue = ((UTF8String) defaultValue).getBytes();
                } else {
                    newColumnDefaultValue = UTF8String.fromString(CarbonCommonConstants.MEMBER_DEFAULT_VAL)
                            .getBytes();
                }
                noDictionaryKeyArrayWithNewlyAddedColumns[newKeyArrayIndex++] = newColumnDefaultValue;
            }
        }
    }
    return noDictionaryKeyArrayWithNewlyAddedColumns;
}

---

From source file:org.apache.carbondata.core.util.DataTypeUtil.java

License:Apache License

/**
 * Below method will be used to convert the data passed to its actual data
 * type//w  w w  .  j  av a  2s . c  o m
 *
 * @param data           data
 * @param actualDataType actual data type
 * @return actual data after conversion
 */
public static Object getDataBasedOnDataType(String data, DataType actualDataType) {

    if (null == data || CarbonCommonConstants.MEMBER_DEFAULT_VAL.equals(data)) {
        return null;
    }
    try {
        switch (actualDataType) {
        case INT:
            if (data.isEmpty()) {
                return null;
            }
            return Integer.parseInt(data);
        case SHORT:
            if (data.isEmpty()) {
                return null;
            }
            return Short.parseShort(data);
        case DOUBLE:
            if (data.isEmpty()) {
                return null;
            }
            return Double.parseDouble(data);
        case LONG:
            if (data.isEmpty()) {
                return null;
            }
            return Long.parseLong(data);
        case TIMESTAMP:
            if (data.isEmpty()) {
                return null;
            }
            SimpleDateFormat parser = new SimpleDateFormat(
                    CarbonProperties.getInstance().getProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT,
                            CarbonCommonConstants.CARBON_TIMESTAMP_DEFAULT_FORMAT));
            Date dateToStr = null;
            try {
                dateToStr = parser.parse(data);
                return dateToStr.getTime() * 1000;
            } catch (ParseException e) {
                LOGGER.error("Cannot convert" + data + " to Time/Long type value" + e.getMessage());
                return null;
            }
        case DECIMAL:
            if (data.isEmpty()) {
                return null;
            }
            java.math.BigDecimal javaDecVal = new java.math.BigDecimal(data);
            scala.math.BigDecimal scalaDecVal = new scala.math.BigDecimal(javaDecVal);
            org.apache.spark.sql.types.Decimal decConverter = new org.apache.spark.sql.types.Decimal();
            return decConverter.set(scalaDecVal);
        default:
            return UTF8String.fromString(data);
        }
    } catch (NumberFormatException ex) {
        LOGGER.error("Problem while converting data type" + data);
        return null;
    }

}

---

From source file:org.apache.carbondata.spark.readsupport.SparkRowReadSupportImpl.java

License:Apache License

@Override
public Row readRow(Object[] data) {
    for (int i = 0; i < dictionaries.length; i++) {
        if (dictionaries[i] != null) {
            data[i] = DataTypeUtil.getDataBasedOnDataType(
                    dictionaries[i].getDictionaryValueForKey((int) data[i]), dataTypes[i]);
            switch (dataTypes[i]) {
            case STRING:
                data[i] = UTF8String.fromString(data[i].toString());
                break;
            case TIMESTAMP:
                data[i] = new Timestamp((long) data[i] / 1000);
                break;
            default:
            }//from www . j av a 2 s .  co m
        } else if (carbonColumns[i].hasEncoding(Encoding.DIRECT_DICTIONARY)) {
            //convert the long to timestamp in case of direct dictionary column
            if (DataType.TIMESTAMP == carbonColumns[i].getDataType()) {
                data[i] = new Timestamp((long) data[i] / 1000);
            }
        }
    }
    return new GenericRow(data);
}

---

From source file:org.carbondata.scan.util.DataTypeUtil.java

License:Apache License

/**
 * Below method will be used to convert the data passed to its actual data
 * type/*from   ww  w.  jav  a 2  s  .  c  o m*/
 *
 * @param data           data
 * @param actualDataType actual data type
 * @return actual data after conversion
 */
public static Object getDataBasedOnDataType(String data, DataType actualDataType) {

    if (null == data || data.isEmpty()) {
        return null;
    }
    try {
        switch (actualDataType) {
        case INT:
            return Integer.parseInt(data);
        case SHORT:
            return Short.parseShort(data);
        case DOUBLE:
            return Double.parseDouble(data);
        case LONG:
            return Long.parseLong(data);
        case TIMESTAMP:
            SimpleDateFormat parser = new SimpleDateFormat(
                    CarbonProperties.getInstance().getProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT,
                            CarbonCommonConstants.CARBON_TIMESTAMP_DEFAULT_FORMAT));
            Date dateToStr = null;
            try {
                dateToStr = parser.parse(data);
                return dateToStr.getTime() * 1000;
            } catch (ParseException e) {
                LOGGER.error("Cannot convert" + data + " to Time/Long type value" + e.getMessage());
                return null;
            }
        case DECIMAL:
            java.math.BigDecimal javaDecVal = new java.math.BigDecimal(data);
            scala.math.BigDecimal scalaDecVal = new scala.math.BigDecimal(javaDecVal);
            org.apache.spark.sql.types.Decimal decConverter = new org.apache.spark.sql.types.Decimal();
            return decConverter.set(scalaDecVal);
        default:
            return UTF8String.fromString(data);
        }
    } catch (NumberFormatException ex) {
        LOGGER.error("Problem while converting data type" + data);
        return null;
    }

}

---