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

Introduction

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

Prototype

public byte[] getBytes() 

Source Link

Document

Returns the underline bytes, will be a copy of it if it's part of another array.

Usage

From source file:com.github.sadikovi.riff.column.UTF8ColumnFilter.java

License:Open Source License

@Override
public boolean mightContain(UTF8String value) {
    if (value == null)
        return hasNulls();
    return filter.mightContainBinary(value.getBytes());
}

---

From source file:com.github.sadikovi.riff.stats.UTF8Statistics.java

License:Open Source License

/**
 * Return deep copy of UTF8String, this method forcefully copies `getBytes()` bytes in
 * UTF8String, since it does not return copy when backed by single array.
 * Clone is null safe, and would return null for null input.
 * @param str UTF8 string to clone/*from w w w. j a  va 2  s. com*/
 * @return copy
 */
private UTF8String clone(UTF8String str) {
    if (str == null)
        return null;
    byte[] bytes = new byte[str.numBytes()];
    System.arraycopy(str.getBytes(), 0, bytes, 0, bytes.length);
    return UTF8String.fromBytes(bytes);
}

---

From source file:org.apache.carbondata.streaming.CarbonStreamRecordReader.java

License:Apache License

private void putRowToColumnBatch(int rowId) {
    for (int i = 0; i < projection.length; i++) {
        Object value = outputValues[i];
        ColumnVector col = columnarBatch.column(i);
        org.apache.spark.sql.types.DataType t = col.dataType();
        if (null == value) {
            col.putNull(rowId);//from   w w  w.j  av a2  s.  co m
        } else {
            if (t == org.apache.spark.sql.types.DataTypes.BooleanType) {
                col.putBoolean(rowId, (boolean) value);
            } else if (t == org.apache.spark.sql.types.DataTypes.ByteType) {
                col.putByte(rowId, (byte) value);
            } else if (t == org.apache.spark.sql.types.DataTypes.ShortType) {
                col.putShort(rowId, (short) value);
            } else if (t == org.apache.spark.sql.types.DataTypes.IntegerType) {
                col.putInt(rowId, (int) value);
            } else if (t == org.apache.spark.sql.types.DataTypes.LongType) {
                col.putLong(rowId, (long) value);
            } else if (t == org.apache.spark.sql.types.DataTypes.FloatType) {
                col.putFloat(rowId, (float) value);
            } else if (t == org.apache.spark.sql.types.DataTypes.DoubleType) {
                col.putDouble(rowId, (double) value);
            } else if (t == org.apache.spark.sql.types.DataTypes.StringType) {
                UTF8String v = (UTF8String) value;
                col.putByteArray(rowId, v.getBytes());
            } else if (t instanceof org.apache.spark.sql.types.DecimalType) {
                DecimalType dt = (DecimalType) t;
                Decimal d = Decimal.fromDecimal(value);
                if (dt.precision() <= Decimal.MAX_INT_DIGITS()) {
                    col.putInt(rowId, (int) d.toUnscaledLong());
                } else if (dt.precision() <= Decimal.MAX_LONG_DIGITS()) {
                    col.putLong(rowId, d.toUnscaledLong());
                } else {
                    final BigInteger integer = d.toJavaBigDecimal().unscaledValue();
                    byte[] bytes = integer.toByteArray();
                    col.putByteArray(rowId, bytes, 0, bytes.length);
                }
            } else if (t instanceof CalendarIntervalType) {
                CalendarInterval c = (CalendarInterval) value;
                col.getChildColumn(0).putInt(rowId, c.months);
                col.getChildColumn(1).putLong(rowId, c.microseconds);
            } else if (t instanceof org.apache.spark.sql.types.DateType) {
                col.putInt(rowId, (int) value);
            } else if (t instanceof org.apache.spark.sql.types.TimestampType) {
                col.putLong(rowId, (long) value);
            }
        }
    }
}

---