xapian-core/matcher/andmaybepostlist.h

   1 /** @file
   2  * @brief Merged postlist: items from one list, weights from both
   3  *
   4  * AND MAYBE of two posting lists
   5  * A AND MAYBE B is logically just A, but we keep B around for weight purposes
   6  */
   7 /* Copyright 1999,2000,2001 BrightStation PLC
   8  * Copyright 2002 Ananova Ltd
   9  * Copyright 2003,2004,2009,2011,2017 Olly Betts
  10  * Copyright 2009 Lemur Consulting Ltd
  11  *
  12  * This program is free software; you can redistribute it and/or
  13  * modify it under the terms of the GNU General Public License as
  14  * published by the Free Software Foundation; either version 2 of the
  15  * License, or (at your option) any later version.
  16  *
  17  * This program is distributed in the hope that it will be useful,
  18  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  19  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  20  * GNU General Public License for more details.
  21  *
  22  * You should have received a copy of the GNU General Public License
  23  * along with this program; if not, write to the Free Software
  24  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301
  25  * USA
  26  */
  27
  28
  29 #ifndef OM_HGUARD_ANDMAYBEPOSTLIST_H
  30 #define OM_HGUARD_ANDMAYBEPOSTLIST_H
  31
  32 #include "branchpostlist.h"
  33
  34 /** A postlist with weights modified by another postlist.
  35  *
  36  *  This postlist returns a posting if and only if it is in the left
  37  *  sub-postlist.
  38  *
  39  *  If the posting does not occur in the right postlist, the weight for the
  40  *  posting is simply that in the left postlist.  If the posting occurs in
  41  *  both postlists, the weight for the posting is the sum of the weights in
  42  *  the sub-postlists.
  43  *
  44  *  This type of postlist is useful for specifying a set of terms which
  45  *  must appear in the query result: these terms can be specified as the
  46  *  left hand argument, with the rest of the query being on the right hand
  47  *  side, and having the effect of modifying the weights.
  48  *
  49  *  The postlist is also used as a "decay product" of other postlist types
  50  *  during the match process: when a postlist can no longer cause a
  51  *  document to enter the mset on its own, but can influence relative
  52  *  rankings, it may be combined using one of these.
  53  */
  54 class AndMaybePostList : public BranchPostList {
  55     private:
  56         Xapian::doccount dbsize; // only need in case we decay to an AndPostList
  57         Xapian::docid lhead, rhead;
  58         double lmax, rmax;
  59
  60         PostList * process_next_or_skip_to(double w_min, PostList *ret);
  61     public:
  62         Xapian::doccount get_termfreq_max() const;
  63         Xapian::doccount get_termfreq_min() const;
  64         Xapian::doccount get_termfreq_est() const;
  65
  66         TermFreqs get_termfreq_est_using_stats(
  67             const Xapian::Weight::Internal & stats) const;
  68
  69         Xapian::docid get_docid() const;
  70         double get_weight() const;
  71         double get_maxweight() const;
  72
  73         double recalc_maxweight();
  74
  75         PostList *next(double w_min);
  76         PostList *skip_to(Xapian::docid did, double w_min);
  77         bool at_end() const;
  78
  79         std::string get_description() const;
  80
  81         /** Return the document length of the document the current term
  82          *  comes from.
  83          */
  84         virtual Xapian::termcount get_doclength() const;
  85
  86         virtual Xapian::termcount get_unique_terms() const;
  87
  88         AndMaybePostList(PostList *left_,
  89                          PostList *right_,
  90                          MultiMatch *matcher_,
  91                          Xapian::doccount dbsize_)
  92                 : BranchPostList(left_, right_, matcher_),
  93                   dbsize(dbsize_), lhead(0), rhead(0)
  94         {
  95             // lmax and rmax will get initialised by a recalc_maxweight
  96         }
  97
  98         /// Constructor for use by decomposing OrPostList
  99         AndMaybePostList(PostList *left_,
 100                          PostList *right_,
 101                          MultiMatch *matcher_,
 102                          Xapian::doccount dbsize_,
 103                          Xapian::docid lhead_,
 104                          Xapian::docid rhead_)
 105                 : BranchPostList(left_, right_, matcher_),
 106                   dbsize(dbsize_), lhead(lhead_), rhead(rhead_)
 107         {
 108             // Initialise the maxweights from the kids so we can avoid forcing
 109             // a full maxweight recalc
 110             lmax = l->get_maxweight();
 111             rmax = r->get_maxweight();
 112         }
 113
 114         /** Synchronise the RHS to the LHS after construction.
 115          *  Used after constructing from a decomposing OrPostList
 116          */
 117         PostList * sync_rhs(double w_min);
 118
 119         /** get_wdf() for ANDMAYBE postlists returns the sum of the wdfs of the
 120          *  sub postlists which are at the current document - this is desirable
 121          *  when the ANDMAYBE is part of a synonym.
 122          */
 123         Xapian::termcount get_wdf() const;
 124
 125         Xapian::termcount count_matching_subqs() const;
 126
 127         void gather_position_lists(OrPositionList* orposlist);
 128 };
 129
 130 #endif /* OM_HGUARD_ANDMAYBEPOSTLIST_H */