Class ReadOnceBag

src/org/apache/pig/data/ReadOnceBag.java:37–255

This bag does not store the tuples in memory, but has access to an iterator typically provided by Hadoop. Use this when you already have an iterator over tuples and do not want to copy over again to a new bag.

 * tuples and do not want to copy over again to a new bag.
 */
public class ReadOnceBag implements DataBag {

    // The Packager that created this
    protected Packager pkgr;

    //The iterator of Tuples. Marked transient because we will never serialize this.
    protected transient Iterator<NullableTuple> tupIter;

    // The key being worked on
    protected PigNullableWritable keyWritable;

    /**
     *
     */
    private static final long serialVersionUID = 2L;

    public ReadOnceBag() {
    }

    /**
     * This constructor creates a bag out of an existing iterator
     * of tuples by taking ownership of the iterator and NOT
     * copying the elements of the iterator.
     * @param pkg POPackageLite
     * @param tupIter Iterator<NullableTuple>
     * @param key Object
     */
    public ReadOnceBag(Packager pkgr, Iterator<NullableTuple> tupIter,
            PigNullableWritable keyWritable) {
        this.pkgr = pkgr;
        this.tupIter = tupIter;
        this.keyWritable = keyWritable;
    }

    /* (non-Javadoc)
     * @see org.apache.pig.impl.util.Spillable#getMemorySize()
     */
    @Override
    public long getMemorySize() {
        return 0;
    }

    /* (non-Javadoc)
     * @see org.apache.pig.impl.util.Spillable#spill()

     */
    @Override
    public long spill() {
        throw new RuntimeException("ReadOnceBag does not support spill operation");
    }

    /* (non-Javadoc)
     * @see org.apache.pig.data.DataBag#add(org.apache.pig.data.Tuple)
     */
    @Override
    public void add(Tuple t) {
        throw new RuntimeException("ReadOnceBag does not support add operation");
    }