Examples of it.unimi.dsi.mg4j.search.CachingDocumentIterator

it.unimi.dsi.mg4j.search.CachingDocumentIterator

A decorator that caches the intervals produced by the underlying document iterator.

Often, scores exhaust the intervals produced by a document iterator to compute their result. However, often you also need those intervals for other purposes (maybe just because you are aggregating several interval-based scorers). Decorating a document iterator with an instance of this class you get again a document iterator, but its intervals can be retrieved several times by calling {@link #intervalIterator(Index)}, {@link #intervalIterator()} and{@link #intervalIterators()}. Important: calls are not nestable: when you require again an iterator, the one previously returned is no longer valid, and when the current document changes (e.g., because of a call to {@link #nextDocument()}) the previously returned interval iterators are invalidated. @author Sebastiano Vigna @since 0.9.1

  /** Delegates to the underlying scorers, possibly wrapping the argument in a
   * {@link CachingDocumentIterator}; then, if {@link #samples} is nonzero computes
   * that many document scores and invokes {@link #setupEqualizationFactors()}.
   */
  public void wrap( DocumentIterator documentIterator ) throws IOException {
    if ( needsCaching ) documentIterator = new CachingDocumentIterator( documentIterator );
    for( int i = n; i-- != 0; ) scorer[ i ].wrap( documentIterator );
    if ( samples > 0 ) {
      // Let us prepare a sample.
      int i;
      for( i = 0; i < samples && ( sampleDocument[ i ] = documentIterator.nextDocument() ) != -1; i++ ) {

Related Classes of it.unimi.dsi.mg4j.search.CachingDocumentIterator

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.