xapian-core/backends/honey/honey_postlisttable.cc

   1 /** @file honey_postlisttable.cc
   2  * @brief Subclass of HoneyTable which holds postlists.
   3  */
   4 /* Copyright (C) 2007,2008,2009,2010,2013,2014,2015,2016,2017,2018 Olly Betts
   5  *
   6  * This program is free software; you can redistribute it and/or modify
   7  * it under the terms of the GNU General Public License as published by
   8  * the Free Software Foundation; either version 2 of the License, or
   9  * (at your option) any later version.
  10  *
  11  * This program is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14  * GNU General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU General Public License
  17  * along with this program; if not, write to the Free Software
  18  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
  19  */
  20
  21 #include <config.h>
  22
  23 #include "honey_postlisttable.h"
  24
  25 #include "honey_alldocspostlist.h"
  26 #include "honey_cursor.h"
  27 #include "honey_database.h"
  28 #include "honey_defs.h"
  29 #include "honey_postlist.h"
  30 #include "honey_postlist_encodings.h"
  31
  32 #include <memory>
  33
  34 using namespace Honey;
  35 using namespace std;
  36
  37 HoneyPostList*
  38 HoneyPostListTable::open_post_list(const HoneyDatabase* db,
  39                                    const std::string& term,
  40                                    bool need_pos) const
  41 {
  42     Assert(!term.empty());
  43     // Try to position cursor first so we avoid creating HoneyPostList objects
  44     // for terms which don't exist.
  45     unique_ptr<HoneyCursor> cursor(cursor_get());
  46     if (!cursor->find_exact(Honey::make_postingchunk_key(term))) {
  47         // FIXME: Return NULL here and handle that in Query::Internal
  48         // postlist() methods as we build the PostList tree.
  49         // return NULL;
  50         return new HoneyPostList(db, term, NULL);
  51     }
  52
  53     if (need_pos)
  54         return new HoneyPosPostList(db, term, cursor.release());
  55     return new HoneyPostList(db, term, cursor.release());
  56 }
  57
  58 void
  59 HoneyPostListTable::get_freqs(const std::string& term,
  60                               Xapian::doccount* termfreq_ptr,
  61                               Xapian::termcount* collfreq_ptr) const
  62 {
  63     string chunk;
  64     if (!get_exact_entry(Honey::make_postingchunk_key(term), chunk)) {
  65         if (termfreq_ptr) *termfreq_ptr = 0;
  66         if (collfreq_ptr) *collfreq_ptr = 0;
  67         return;
  68     }
  69
  70     const char* p = chunk.data();
  71     const char* pend = p + chunk.size();
  72     Xapian::doccount tf;
  73     Xapian::termcount cf;
  74     if (!decode_initial_chunk_header_freqs(&p, pend, tf, cf))
  75         throw Xapian::DatabaseCorruptError("Postlist initial chunk header");
  76     if (termfreq_ptr) *termfreq_ptr = tf;
  77     if (collfreq_ptr) *collfreq_ptr = cf;
  78 }
  79
  80 void
  81 HoneyPostListTable::get_used_docid_range(Xapian::doccount doccount,
  82                                          Xapian::docid& first,
  83                                          Xapian::docid& last) const
  84 {
  85     unique_ptr<HoneyCursor> cursor;
  86     static const char doclen_key_prefix[2] = {
  87         0, char(Honey::KEY_DOCLEN_CHUNK)
  88     };
  89     if (cursor->find_entry_ge(string(doclen_key_prefix, 2))) {
  90         first = 1;
  91     } else {
  92         // doccount == 0 should be handled by our caller.
  93         Assert(!cursor->after_end());
  94         Xapian::docid last_in_first_chunk = docid_from_key(cursor->current_key);
  95         if (last_in_first_chunk == 0) {
  96             // Note that our caller checks for doccount == 0 and handles that.
  97             throw Xapian::DatabaseCorruptError("Bad first doclen chunk key");
  98         }
  99         cursor->read_tag();
 100         Xapian::docid delta;
 101         const char* p = cursor->current_tag.data();
 102         const char* pend = p + cursor->current_tag.size();
 103         if (!unpack_uint(&p, pend, &delta)) {
 104             throw Xapian::DatabaseCorruptError("Bad first doclen chunk delta");
 105         }
 106         first = last_in_first_chunk - delta;
 107     }
 108
 109     // We know the last docid is at least first - 1 + doccount, so seek
 110     // to there and then scan forwards.  If we match exactly, then that
 111     // is exactly the last docid (our caller handles this case when
 112     // first == 1, but not otherwise).
 113     last = first - 1 + doccount;
 114     if (cursor->find_entry_ge(make_doclenchunk_key(last)))
 115         return;
 116
 117     if (cursor->after_end())
 118         throw Xapian::DatabaseCorruptError("Missing doclen chunk");
 119
 120     do {
 121         Xapian::docid new_last = docid_from_key(cursor->current_key);
 122         if (new_last == 0) {
 123             // We've hit a non-doclen item.
 124             return;
 125         }
 126         last = new_last;
 127         cursor->next();
 128     } while (!cursor->after_end());
 129
 130     // We've reached the end of the table (only possible if there are no terms
 131     // at all!)
 132 }
 133
 134 Xapian::termcount
 135 HoneyPostListTable::get_wdf_upper_bound(const std::string& term) const
 136 {
 137     string chunk;
 138     if (!get_exact_entry(Honey::make_postingchunk_key(term), chunk)) {
 139         // Term not present.
 140         return 0;
 141     }
 142
 143     const char* p = chunk.data();
 144     const char* pend = p + chunk.size();
 145     Xapian::doccount tf;
 146     Xapian::termcount cf;
 147     Xapian::docid first;
 148     Xapian::docid last;
 149     Xapian::docid chunk_last;
 150     Xapian::termcount first_wdf;
 151     Xapian::termcount wdf_max;
 152     if (!decode_initial_chunk_header(&p, pend, tf, cf, first, last, chunk_last,
 153                                      first_wdf, wdf_max))
 154         throw Xapian::DatabaseCorruptError("Postlist initial chunk header");
 155     return wdf_max;
 156 }