xapian-core/matcher/synonympostlist.cc

   1 /** @file
   2  * @brief Combine subqueries, weighting as if they are synonyms
   3  */
   4 /* Copyright 2007,2009 Lemur Consulting Ltd
   5  * Copyright 2009,2011,2014,2016,2018 Olly Betts
   6  *
   7  * This program is free software; you can redistribute it and/or
   8  * modify it under the terms of the GNU General Public License as
   9  * published by the Free Software Foundation; either version 2 of the
  10  * License, or (at your option) any later version.
  11  *
  12  * This program is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15  * GNU General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU General Public License
  18  * along with this program; if not, write to the Free Software
  19  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301
  20  * USA
  21  */
  22
  23 #include <config.h>
  24
  25 #include "synonympostlist.h"
  26
  27 #include "branchpostlist.h"
  28 #include "debuglog.h"
  29 #include "omassert.h"
  30
  31 SynonymPostList::~SynonymPostList()
  32 {
  33     delete wt;
  34     delete subtree;
  35 }
  36
  37 void
  38 SynonymPostList::set_weight(const Xapian::Weight * wt_)
  39 {
  40     delete wt;
  41     wt = wt_;
  42     want_doclength = wt->get_sumpart_needs_doclength_();
  43     want_wdf = wt->get_sumpart_needs_wdf_();
  44     want_unique_terms = wt->get_sumpart_needs_uniqueterms_();
  45 }
  46
  47 PostList *
  48 SynonymPostList::next(double w_min)
  49 {
  50     LOGCALL(MATCH, PostList *, "SynonymPostList::next", w_min);
  51     (void)w_min;
  52     next_handling_prune(subtree, 0, matcher);
  53     RETURN(NULL);
  54 }
  55
  56 PostList *
  57 SynonymPostList::skip_to(Xapian::docid did, double w_min)
  58 {
  59     LOGCALL(MATCH, PostList *, "SynonymPostList::skip_to", did | w_min);
  60     (void)w_min;
  61     skip_to_handling_prune(subtree, did, 0, matcher);
  62     RETURN(NULL);
  63 }
  64
  65 double
  66 SynonymPostList::get_weight() const
  67 {
  68     LOGCALL(MATCH, double, "SynonymPostList::get_weight", NO_ARGS);
  69     // The wdf returned can be higher than the doclength.  In particular, this
  70     // can currently occur if the query contains a term more than once; the wdf
  71     // of each occurrence is added up.
  72     //
  73     // However, it's reasonable for weighting algorithms to optimise by
  74     // assuming that get_wdf() will never return more than get_doclength(),
  75     // since the doclength is the sum of the wdfs.
  76     //
  77     // Therefore, we simply clamp the wdf value to the doclength, to ensure
  78     // that this is true.  Note that this requires the doclength to be
  79     // calculated even if the weight object doesn't want it.
  80
  81     Xapian::termcount unique_terms = 0;
  82     if (want_unique_terms)
  83         unique_terms = get_unique_terms();
  84     if (want_wdf) {
  85         Xapian::termcount wdf = get_wdf();
  86         Xapian::termcount doclen = 0;
  87         if (want_doclength || (!wdf_disjoint && wdf > doclen_lower_bound)) {
  88             doclen = get_doclength();
  89             if (wdf > doclen) wdf = doclen;
  90         }
  91         double sumpart = wt->get_sumpart(wdf, doclen, unique_terms);
  92         AssertRel(sumpart, <=, wt->get_maxpart());
  93         RETURN(sumpart);
  94     }
  95     Xapian::termcount doclen = want_doclength ? get_doclength() : 0;
  96     RETURN(wt->get_sumpart(0, doclen, unique_terms));
  97 }
  98
  99 double
 100 SynonymPostList::get_maxweight() const
 101 {
 102     LOGCALL(MATCH, double, "SynonymPostList::get_maxweight", NO_ARGS);
 103     RETURN(wt->get_maxpart());
 104 }
 105
 106 double
 107 SynonymPostList::recalc_maxweight()
 108 {
 109     LOGCALL(MATCH, double, "SynonymPostList::recalc_maxweight", NO_ARGS);
 110     RETURN(SynonymPostList::get_maxweight());
 111 }
 112
 113 Xapian::termcount
 114 SynonymPostList::get_wdf() const {
 115     LOGCALL(MATCH, Xapian::termcount, "SynonymPostList::get_wdf", NO_ARGS);
 116     RETURN(subtree->get_wdf());
 117 }
 118
 119 Xapian::doccount
 120 SynonymPostList::get_termfreq_min() const {
 121     LOGCALL(MATCH, Xapian::doccount, "SynonymPostList::get_termfreq_min", NO_ARGS);
 122     RETURN(subtree->get_termfreq_min());
 123 }
 124
 125 Xapian::doccount
 126 SynonymPostList::get_termfreq_est() const {
 127     LOGCALL(MATCH, Xapian::doccount, "SynonymPostList::get_termfreq_est", NO_ARGS);
 128     RETURN(subtree->get_termfreq_est());
 129 }
 130
 131 Xapian::doccount
 132 SynonymPostList::get_termfreq_max() const {
 133     LOGCALL(MATCH, Xapian::doccount, "SynonymPostList::get_termfreq_max", NO_ARGS);
 134     RETURN(subtree->get_termfreq_max());
 135 }
 136
 137 Xapian::docid
 138 SynonymPostList::get_docid() const {
 139     LOGCALL(MATCH, Xapian::docid, "SynonymPostList::get_docid", NO_ARGS);
 140     RETURN(subtree->get_docid());
 141 }
 142
 143 Xapian::termcount
 144 SynonymPostList::get_doclength() const {
 145     LOGCALL(MATCH, Xapian::termcount, "SynonymPostList::get_doclength", NO_ARGS);
 146     RETURN(subtree->get_doclength());
 147 }
 148
 149 Xapian::termcount
 150 SynonymPostList::get_unique_terms() const {
 151     LOGCALL(MATCH, Xapian::termcount, "SynonymPostList::get_unique_terms", NO_ARGS);
 152     RETURN(subtree->get_unique_terms());
 153 }
 154
 155 bool
 156 SynonymPostList::at_end() const {
 157     LOGCALL(MATCH, bool, "SynonymPostList::at_end", NO_ARGS);
 158     RETURN(subtree->at_end());
 159 }
 160
 161 Xapian::termcount
 162 SynonymPostList::count_matching_subqs() const
 163 {
 164     return 1;
 165 }
 166
 167 std::string
 168 SynonymPostList::get_description() const
 169 {
 170     return "(Synonym " + subtree->get_description() + ")";
 171 }