xapian-core/matcher/mergepostlist.cc

   1 /* mergepostlist.cc: merge postlists from different databases
   2  *
   3  * Copyright 1999,2000,2001 BrightStation PLC
   4  * Copyright 2002 Ananova Ltd
   5  * Copyright 2002,2003,2004,2006,2008,2009,2011,2015,2016 Olly Betts
   6  * Copyright 2007,2009 Lemur Consulting Ltd
   7  *
   8  * This program is free software; you can redistribute it and/or
   9  * modify it under the terms of the GNU General Public License as
  10  * published by the Free Software Foundation; either version 2 of the
  11  * License, or (at your option) any later version.
  12  *
  13  * This program is distributed in the hope that it will be useful,
  14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  16  * GNU General Public License for more details.
  17  *
  18  * You should have received a copy of the GNU General Public License
  19  * along with this program; if not, write to the Free Software
  20  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301
  21  * USA
  22  */
  23
  24 #include <config.h>
  25 #include "mergepostlist.h"
  26
  27 #include "multimatch.h"
  28 #include "api/emptypostlist.h"
  29 #include "branchpostlist.h"
  30 #include "debuglog.h"
  31 #include "omassert.h"
  32 #include "valuestreamdocument.h"
  33
  34 #include "xapian/error.h"
  35
  36 // NB don't prune - even with one sublist we still translate docids...
  37
  38 MergePostList::~MergePostList()
  39 {
  40     LOGCALL_DTOR(MATCH, "MergePostList");
  41     std::vector<PostList *>::const_iterator i;
  42     for (i = plists.begin(); i != plists.end(); ++i) {
  43         delete *i;
  44     }
  45 }
  46
  47 PostList *
  48 MergePostList::next(double w_min)
  49 {
  50     LOGCALL(MATCH, PostList *, "MergePostList::next", w_min);
  51     LOGVALUE(MATCH, current);
  52     if (current == -1) current = 0;
  53     while (true) {
  54         // FIXME: should skip over Remote matchers which aren't ready yet
  55         // and come back to them later...
  56         next_handling_prune(plists[current], w_min, matcher);
  57         if (!plists[current]->at_end()) break;
  58         ++current;
  59         if (unsigned(current) >= plists.size()) break;
  60         vsdoc.new_subdb(current);
  61         if (matcher) matcher->recalc_maxweight();
  62     }
  63     LOGVALUE(MATCH, current);
  64     RETURN(NULL);
  65 }
  66
  67 PostList *
  68 MergePostList::skip_to(Xapian::docid did, double w_min)
  69 {
  70     LOGCALL(MATCH, PostList *, "MergePostList::skip_to", did | w_min);
  71     (void)did;
  72     (void)w_min;
  73     // MergePostList doesn't return documents in docid order, so skip_to
  74     // isn't a meaningful operation.
  75     throw Xapian::InvalidOperationError("MergePostList doesn't support skip_to");
  76 }
  77
  78 Xapian::termcount
  79 MergePostList::get_wdf() const
  80 {
  81     LOGCALL(MATCH, Xapian::termcount, "MergePostList::get_wdf", NO_ARGS);
  82     RETURN(plists[current]->get_wdf());
  83 }
  84
  85 Xapian::doccount
  86 MergePostList::get_termfreq_max() const
  87 {
  88     LOGCALL(MATCH, Xapian::doccount, "MergePostList::get_termfreq_max", NO_ARGS);
  89     // sum of termfreqs for all children
  90     Xapian::doccount total = 0;
  91     vector<PostList *>::const_iterator i;
  92     for (i = plists.begin(); i != plists.end(); ++i) {
  93         total += (*i)->get_termfreq_max();
  94     }
  95     RETURN(total);
  96 }
  97
  98 Xapian::doccount
  99 MergePostList::get_termfreq_min() const
 100 {
 101     LOGCALL(MATCH, Xapian::doccount, "MergePostList::get_termfreq_min", NO_ARGS);
 102     // sum of termfreqs for all children
 103     Xapian::doccount total = 0;
 104     vector<PostList *>::const_iterator i;
 105     for (i = plists.begin(); i != plists.end(); ++i) {
 106         total += (*i)->get_termfreq_min();
 107     }
 108     RETURN(total);
 109 }
 110
 111 Xapian::doccount
 112 MergePostList::get_termfreq_est() const
 113 {
 114     LOGCALL(MATCH, Xapian::doccount, "MergePostList::get_termfreq_est", NO_ARGS);
 115     // sum of termfreqs for all children
 116     Xapian::doccount total = 0;
 117     vector<PostList *>::const_iterator i;
 118     for (i = plists.begin(); i != plists.end(); ++i) {
 119         total += (*i)->get_termfreq_est();
 120     }
 121     RETURN(total);
 122 }
 123
 124 Xapian::docid
 125 MergePostList::get_docid() const
 126 {
 127     LOGCALL(MATCH, Xapian::docid, "MergePostList::get_docid", NO_ARGS);
 128     Assert(current != -1);
 129     // FIXME: this needs fixing so we can prune plists - see MultiPostlist
 130     // for code which does this...
 131     RETURN((plists[current]->get_docid() - 1) * plists.size() + current + 1);
 132 }
 133
 134 double
 135 MergePostList::get_weight() const
 136 {
 137     LOGCALL(MATCH, double, "MergePostList::get_weight", NO_ARGS);
 138     Assert(current != -1);
 139     RETURN(plists[current]->get_weight());
 140 }
 141
 142 const string *
 143 MergePostList::get_sort_key() const
 144 {
 145     LOGCALL(MATCH, const string *, "MergePostList::get_sort_key", NO_ARGS);
 146     Assert(current != -1);
 147     RETURN(plists[current]->get_sort_key());
 148 }
 149
 150 const string *
 151 MergePostList::get_collapse_key() const
 152 {
 153     LOGCALL(MATCH, const string *, "MergePostList::get_collapse_key", NO_ARGS);
 154     Assert(current != -1);
 155     RETURN(plists[current]->get_collapse_key());
 156 }
 157
 158 double
 159 MergePostList::get_maxweight() const
 160 {
 161     LOGCALL(MATCH, double, "MergePostList::get_maxweight", NO_ARGS);
 162     RETURN(w_max);
 163 }
 164
 165 double
 166 MergePostList::recalc_maxweight()
 167 {
 168     LOGCALL(MATCH, double, "MergePostList::recalc_maxweight", NO_ARGS);
 169     w_max = 0;
 170     vector<PostList *>::iterator i;
 171     for (i = plists.begin(); i != plists.end(); ++i) {
 172         double w = (*i)->recalc_maxweight();
 173         if (w > w_max) w_max = w;
 174     }
 175     RETURN(w_max);
 176 }
 177
 178 bool
 179 MergePostList::at_end() const
 180 {
 181     LOGCALL(MATCH, bool, "MergePostList::at_end", NO_ARGS);
 182     Assert(current != -1);
 183     RETURN(unsigned(current) >= plists.size());
 184 }
 185
 186 string
 187 MergePostList::get_description() const
 188 {
 189     string desc = "( Merge ";
 190     vector<PostList *>::const_iterator i;
 191     for (i = plists.begin(); i != plists.end(); ++i) {
 192         desc += (*i)->get_description() + " ";
 193     }
 194     return desc + ")";
 195 }
 196
 197 Xapian::termcount
 198 MergePostList::get_doclength() const
 199 {
 200     LOGCALL(MATCH, Xapian::termcount, "MergePostList::get_doclength", NO_ARGS);
 201     Assert(current != -1);
 202     RETURN(plists[current]->get_doclength());
 203 }
 204
 205 Xapian::termcount
 206 MergePostList::get_unique_terms() const
 207 {
 208     LOGCALL(MATCH, Xapian::termcount, "MergePostList::get_unique_terms", NO_ARGS);
 209     Assert(current != -1);
 210     RETURN(plists[current]->get_unique_terms());
 211 }
 212
 213 Xapian::termcount
 214 MergePostList::count_matching_subqs() const
 215 {
 216     LOGCALL(MATCH, Xapian::termcount, "MergePostList::count_matching_subqs", NO_ARGS);
 217     RETURN(plists[current]->count_matching_subqs());
 218 }