xapian-core/backends/glass/glass_termlist.cc

   1 /* glass_termlist.cc: Termlists in a glass database
   2  *
   3  * Copyright 1999,2000,2001 BrightStation PLC
   4  * Copyright 2002 Ananova Ltd
   5  * Copyright 2002,2003,2004,2006,2007,2008,2009,2010,2014 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 #include "glass_termlist.h"
  25
  26 #include "xapian/error.h"
  27
  28 #include "expand/expandweight.h"
  29 #include "glass_positionlist.h"
  30 #include "debuglog.h"
  31 #include "omassert.h"
  32 #include "pack.h"
  33 #include "str.h"
  34
  35 using namespace std;
  36 using Xapian::Internal::intrusive_ptr;
  37
  38 GlassTermList::GlassTermList(intrusive_ptr<const GlassDatabase> db_,
  39                              Xapian::docid did_)
  40         : db(db_), did(did_), current_wdf(0), current_termfreq(0)
  41 {
  42     LOGCALL_CTOR(DB, "GlassTermList", db_ | did_);
  43
  44     if (!db->termlist_table.get_exact_entry(GlassTermListTable::make_key(did),
  45                                             data))
  46         throw Xapian::DocNotFoundError("No termlist for document " + str(did));
  47
  48     pos = data.data();
  49     end = pos + data.size();
  50
  51     if (pos == end) {
  52         doclen = 0;
  53         termlist_size = 0;
  54         return;
  55     }
  56
  57     // Read doclen
  58     if (!unpack_uint(&pos, end, &doclen)) {
  59         const char *msg;
  60         if (pos == 0) {
  61             msg = "Too little data for doclen in termlist";
  62         } else {
  63             msg = "Overflowed value for doclen in termlist";
  64         }
  65         throw Xapian::DatabaseCorruptError(msg);
  66     }
  67
  68     // Read termlist_size
  69     if (!unpack_uint(&pos, end, &termlist_size)) {
  70         const char *msg;
  71         if (pos == 0) {
  72             msg = "Too little data for list size in termlist";
  73         } else {
  74             msg = "Overflowed value for list size in termlist";
  75         }
  76         throw Xapian::DatabaseCorruptError(msg);
  77     }
  78 }
  79
  80 Xapian::termcount
  81 GlassTermList::get_doclength() const
  82 {
  83     LOGCALL(DB, Xapian::termcount, "GlassTermList::get_doclength", NO_ARGS);
  84     RETURN(doclen);
  85 }
  86
  87 Xapian::termcount
  88 GlassTermList::get_unique_terms() const
  89 {
  90     LOGCALL(DB, Xapian::termcount, "GlassTermList::get_unique_terms", NO_ARGS);
  91     // get_unique_terms() really ought to only count terms with wdf > 0, but
  92     // that's expensive to calculate on demand, so for now let's just ensure
  93     // unique_terms <= doclen.
  94     RETURN(min(termlist_size, doclen));
  95 }
  96
  97 Xapian::termcount
  98 GlassTermList::get_approx_size() const
  99 {
 100     LOGCALL(DB, Xapian::termcount, "GlassTermList::get_approx_size", NO_ARGS);
 101     RETURN(termlist_size);
 102 }
 103
 104 void
 105 GlassTermList::accumulate_stats(Xapian::Internal::ExpandStats & stats) const
 106 {
 107     LOGCALL_VOID(DB, "GlassTermList::accumulate_stats", stats);
 108     Assert(!at_end());
 109     stats.accumulate(current_wdf, doclen, get_termfreq(), db->get_doccount());
 110 }
 111
 112 string
 113 GlassTermList::get_termname() const
 114 {
 115     LOGCALL(DB, string, "GlassTermList::get_termname", NO_ARGS);
 116     RETURN(current_term);
 117 }
 118
 119 Xapian::termcount
 120 GlassTermList::get_wdf() const
 121 {
 122     LOGCALL(DB, Xapian::termcount, "GlassTermList::get_wdf", NO_ARGS);
 123     RETURN(current_wdf);
 124 }
 125
 126 Xapian::doccount
 127 GlassTermList::get_termfreq() const
 128 {
 129     LOGCALL(DB, Xapian::doccount, "GlassTermList::get_termfreq", NO_ARGS);
 130     if (current_termfreq == 0)
 131         db->get_freqs(current_term, &current_termfreq, NULL);
 132     RETURN(current_termfreq);
 133 }
 134
 135 TermList *
 136 GlassTermList::next()
 137 {
 138     LOGCALL(DB, TermList *, "GlassTermList::next", NO_ARGS);
 139     Assert(!at_end());
 140     if (pos == end) {
 141         pos = NULL;
 142         RETURN(NULL);
 143     }
 144
 145     // Reset to 0 to indicate that the termfreq needs to be read.
 146     current_termfreq = 0;
 147
 148     bool wdf_in_reuse = false;
 149     if (!current_term.empty()) {
 150         // Find out how much of the previous term to reuse.
 151         size_t len = static_cast<unsigned char>(*pos++);
 152         if (len > current_term.size()) {
 153             // The wdf is also stored in the "reuse" byte.
 154             wdf_in_reuse = true;
 155             size_t divisor = current_term.size() + 1;
 156             current_wdf = len / divisor - 1;
 157             len %= divisor;
 158         }
 159         current_term.resize(len);
 160     }
 161
 162     // Append the new tail to form the next term.
 163     size_t append_len = static_cast<unsigned char>(*pos++);
 164     current_term.append(pos, append_len);
 165     pos += append_len;
 166
 167     // Read the wdf if it wasn't packed into the reuse byte.
 168     if (!wdf_in_reuse && !unpack_uint(&pos, end, &current_wdf)) {
 169         const char *msg;
 170         if (pos == 0) {
 171             msg = "Too little data for wdf in termlist";
 172         } else {
 173             msg = "Overflowed value for wdf in termlist";
 174         }
 175         throw Xapian::DatabaseCorruptError(msg);
 176     }
 177
 178     RETURN(NULL);
 179 }
 180
 181 TermList *
 182 GlassTermList::skip_to(const string & term)
 183 {
 184     LOGCALL(API, TermList *, "GlassTermList::skip_to", term);
 185     while (pos != NULL && current_term < term) {
 186         (void)GlassTermList::next();
 187     }
 188     RETURN(NULL);
 189 }
 190
 191 bool
 192 GlassTermList::at_end() const
 193 {
 194     LOGCALL(DB, bool, "GlassTermList::at_end", NO_ARGS);
 195     RETURN(pos == NULL);
 196 }
 197
 198 Xapian::termcount
 199 GlassTermList::positionlist_count() const
 200 {
 201     LOGCALL(DB, Xapian::termcount, "GlassTermList::positionlist_count", NO_ARGS);
 202     RETURN(db->position_table.positionlist_count(did, current_term));
 203 }
 204
 205 PositionList*
 206 GlassTermList::positionlist_begin() const
 207 {
 208     LOGCALL(DB, PositionList*, "GlassTermList::positionlist_begin", NO_ARGS);
 209     RETURN(new GlassPositionList(&db->position_table, did, current_term));
 210 }
 211
 212 #ifdef DISABLE_GPL_LIBXAPIAN
 213 # error GPL source we cannot relicense included in libxapian
 214 #endif