Class ScoredDocIdsUtils


  • public class ScoredDocIdsUtils
    extends Object
    Utility methods for Scored Doc IDs.
    WARNING: This API is experimental and might change in incompatible ways in the next release.
    • Constructor Detail

      • ScoredDocIdsUtils

        public ScoredDocIdsUtils()
    • Method Detail

      • getComplementSet

        public static final ScoredDocIDs getComplementSet​(ScoredDocIDs docids,
                                                          IndexReader reader)
                                                   throws IOException
        Create a complement of the input set. The returned ScoredDocIDs does not contain any scores, which makes sense given that the complementing documents were not scored. Note: the complement set does NOT contain doc ids which are noted as deleted by the given reader
        Parameters:
        docids - to be complemented.
        reader - holding the number of documents & information about deletions.
        Throws:
        IOException
      • createScoredDocIDsSubset

        public static final ScoredDocIDs createScoredDocIDsSubset​(ScoredDocIDs allDocIds,
                                                                  int[] sampleSet)
                                                           throws IOException
        Create a subset of an existing ScoredDocIDs object.
        Parameters:
        allDocIds - orginal set
        sampleSet - Doc Ids of the subset.
        Throws:
        IOException
      • createAllDocsScoredDocIDs

        public static final ScoredDocIDs createAllDocsScoredDocIDs​(IndexReader reader)
        Creates a ScoredDocIDs which returns document IDs all non-deleted doc ids according to the given reader. The returned set contains the range of [0 .. reader.maxDoc ) doc ids
      • createScoredDocIds

        public static final ScoredDocIDs createScoredDocIds​(DocIdSet docIdSet,
                                                            int maxDoc)
        Create a ScoredDocIDs out of a given docIdSet and the total number of documents in an index