weight.cc

Go to the documentation of this file.

 
/* Copyright (C) 2007,2008,2009,2014,2017,2019,2024 Olly Betts
 * Copyright (C) 2009 Lemur Consulting Ltd
 * Copyright (C) 2017 Vivek Pal
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License as
 * published by the Free Software Foundation; either version 2 of the
 * License, or (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, see
 * <https://www.gnu.org/licenses/>.
 */
 
#include <config.h>
 
#include "xapian/weight.h"
 
#include "backends/leafpostlist.h"
#include "weightinternal.h"
 
#include "omassert.h"
#include "debuglog.h"
 
#include "xapian/error.h"
 
using namespace std;
 
namespace Xapian {
 
void
Weight::init_(const Internal & stats, Xapian::termcount query_length,
              const Xapian::Database::Internal* shard)
{
    LOGCALL_VOID(MATCH, "Weight::init_", stats | query_length | shard);
    collection_size_ = stats.collection_size;
    rset_size_ = stats.rset_size;
    if (stats_needed & AVERAGE_LENGTH)
        average_length_ = stats.get_average_length();
    if (stats_needed & DOC_LENGTH_MAX)
        doclength_upper_bound_ = shard->get_doclength_upper_bound();
    if (stats_needed & DOC_LENGTH_MIN)
        doclength_lower_bound_ = shard->get_doclength_lower_bound();
    if (stats_needed & UNIQUE_TERMS_MAX)
        unique_terms_upper_bound_ = shard->get_unique_terms_upper_bound();
    if (stats_needed & UNIQUE_TERMS_MIN)
        unique_terms_lower_bound_ = shard->get_unique_terms_lower_bound();
    if (stats_needed & TOTAL_LENGTH)
        total_length_ = stats.total_length;
    if (stats_needed & DB_DOC_LENGTH_MAX)
        db_doclength_upper_bound_ = stats.db_doclength_upper_bound;
    if (stats_needed & DB_DOC_LENGTH_MIN)
        db_doclength_lower_bound_ = stats.db_doclength_lower_bound;
    if (stats_needed & DB_UNIQUE_TERMS_MAX)
        db_unique_terms_upper_bound_ = stats.db_unique_terms_upper_bound;
    if (stats_needed & DB_UNIQUE_TERMS_MIN)
        db_unique_terms_lower_bound_ = stats.db_unique_terms_lower_bound;
    collectionfreq_ = 0;
    wdf_upper_bound_ = 0;
    termfreq_ = 0;
    reltermfreq_ = 0;
    query_length_ = query_length;
    wqf_ = 1;
    init(0.0);
}
 
void
Weight::init_(const Internal & stats, Xapian::termcount query_length,
              const string & term, Xapian::termcount wqf, double factor,
              const Xapian::Database::Internal* shard,
              void* postlist_void)
{
    LOGCALL_VOID(MATCH, "Weight::init_", stats | query_length | term | wqf | factor | shard | postlist_void);
    collection_size_ = stats.collection_size;
    rset_size_ = stats.rset_size;
    if (stats_needed & AVERAGE_LENGTH)
        average_length_ = stats.get_average_length();
    if (stats_needed & DOC_LENGTH_MAX)
        doclength_upper_bound_ = shard->get_doclength_upper_bound();
    if (stats_needed & DOC_LENGTH_MIN)
        doclength_lower_bound_ = shard->get_doclength_lower_bound();
    if (stats_needed & UNIQUE_TERMS_MAX)
        unique_terms_upper_bound_ = shard->get_unique_terms_upper_bound();
    if (stats_needed & UNIQUE_TERMS_MIN)
        unique_terms_lower_bound_ = shard->get_unique_terms_lower_bound();
    if (stats_needed & TOTAL_LENGTH)
        total_length_ = stats.total_length;
    if (stats_needed & WDF_MAX) {
        auto postlist = static_cast<LeafPostList*>(postlist_void);
        wdf_upper_bound_ = postlist->get_wdf_upper_bound();
    }
    if (stats_needed & DB_DOC_LENGTH_MAX)
        db_doclength_upper_bound_ = stats.db_doclength_upper_bound;
    if (stats_needed & DB_DOC_LENGTH_MIN)
        db_doclength_lower_bound_ = stats.db_doclength_lower_bound;
    if (stats_needed & DB_UNIQUE_TERMS_MAX)
        db_unique_terms_upper_bound_ = stats.db_unique_terms_upper_bound;
    if (stats_needed & DB_UNIQUE_TERMS_MIN)
        db_unique_terms_lower_bound_ = stats.db_unique_terms_lower_bound;
    if (stats_needed & DB_WDF_MAX) {
        // FIXME: Nothing uses this stat, so for now return a correct but
        // likely fairly loose upper bound.  Once we have something that
        // wants to use this we can implement tracking a per-term wdf_max
        // across the whole database.
        db_wdf_upper_bound_ = stats.db_doclength_upper_bound;
    }
    if (stats_needed & (TERMFREQ | RELTERMFREQ | COLLECTION_FREQ)) {
        bool ok = stats.get_stats(term,
                                  termfreq_, reltermfreq_, collectionfreq_);
        (void)ok;
        Assert(ok);
    }
    query_length_ = query_length;
    wqf_ = wqf;
    init(factor);
}
 
void
Weight::init_(const Internal & stats, Xapian::termcount query_length,
              double factor, Xapian::doccount termfreq,
              Xapian::doccount reltermfreq, Xapian::termcount collection_freq,
              const Xapian::Database::Internal* shard)
{
    LOGCALL_VOID(MATCH, "Weight::init_", stats | query_length | factor | termfreq | reltermfreq | collection_freq | shard);
    // Synonym case.
    collection_size_ = stats.collection_size;
    rset_size_ = stats.rset_size;
    if (stats_needed & AVERAGE_LENGTH)
        average_length_ = stats.get_average_length();
    if (stats_needed & (DOC_LENGTH_MAX | WDF_MAX)) {
        doclength_upper_bound_ = shard->get_doclength_upper_bound();
        // The doclength is an upper bound on the wdf.  This is obviously true
        // for normal terms, but SynonymPostList ensures that it is also true
        // for synonym terms by clamping the wdf values returned to the
        // doclength.
        //
        // (This clamping is only actually necessary in cases where a
        // constituent term of the synonym is repeated.)
        wdf_upper_bound_ = doclength_upper_bound_;
    }
    if (stats_needed & DOC_LENGTH_MIN)
        doclength_lower_bound_ = shard->get_doclength_lower_bound();
    if (stats_needed & UNIQUE_TERMS_MAX)
        unique_terms_upper_bound_ = shard->get_unique_terms_upper_bound();
    if (stats_needed & UNIQUE_TERMS_MIN)
        unique_terms_lower_bound_ = shard->get_unique_terms_lower_bound();
    if (stats_needed & TOTAL_LENGTH)
        total_length_ = stats.total_length;
    if (stats_needed & (DB_DOC_LENGTH_MAX | DB_WDF_MAX)) {
        db_doclength_upper_bound_ = stats.db_doclength_upper_bound;
        // The doclength is an upper bound on the wdf.  This is obviously true
        // for normal terms, but SynonymPostList ensures that it is also true
        // for synonym terms by clamping the wdf values returned to the
        // doclength.
        //
        // (This clamping is only actually necessary in cases where a
        // constituent term of the synonym is repeated.)
        db_wdf_upper_bound_ = db_doclength_upper_bound_;
    }
    if (stats_needed & DB_DOC_LENGTH_MIN)
        db_doclength_lower_bound_ = stats.db_doclength_lower_bound;
    if (stats_needed & DB_UNIQUE_TERMS_MAX)
        db_unique_terms_upper_bound_ = stats.db_unique_terms_upper_bound;
    if (stats_needed & DB_UNIQUE_TERMS_MIN)
        db_unique_terms_lower_bound_ = stats.db_unique_terms_lower_bound;
 
    termfreq_ = termfreq;
    reltermfreq_ = reltermfreq;
    query_length_ = query_length;
    collectionfreq_ = collection_freq;
    wqf_ = 1;
    init(factor);
}
 
Weight::~Weight() { }
 
string
Weight::name() const
{
    return string();
}
 
string
Weight::serialise() const
{
    throw Xapian::UnimplementedError("serialise() not supported for this Xapian::Weight subclass");
}
 
Weight *
Weight::unserialise(const string &) const
{
    throw Xapian::UnimplementedError("unserialise() not supported for this Xapian::Weight subclass");
}
 
double
Weight::get_sumextra(Xapian::termcount,
                     Xapian::termcount,
                     Xapian::termcount) const
{
    return 0.0;
}
 
double
Weight::get_maxextra() const
{
    return 0.0;
}
 
[[noreturn]]
static inline void
parameter_error(const char* message, const string& scheme, const char* params)
{
    Xapian::Weight::Internal::parameter_error(message, scheme, params);
}
 
const Weight *
Weight::create(const string & s, const Registry & reg)
{
    const char *p = s.c_str();
    std::string scheme;
 
    while (*p != ' ') {
        if (*p == '\0') break;
        scheme += *p;
        p++;
    }
 
    if (*p == ' ') p++;
    auto weight = reg.get_weighting_scheme(scheme);
    if (!weight) {
        // Allow "trad" and "trad <k>" to work despite TradWeight now just
        // being a thin subclass of BM25Weight.
        if (scheme == "trad") {
            const char* params = p;
            double k = 1.0;
            if (*p != '\0') {
                if (!Xapian::Weight::Internal::double_param(&p, &k))
                    parameter_error("Parameter is invalid", scheme, params);
                if (*p)
                    parameter_error("Extra data after parameter",
                                    scheme, params);
            }
            return new BM25Weight(k, 0.0, 0.0, 1.0, 0.0);
        }
        throw InvalidArgumentError("Unknown weighting scheme: " + scheme);
    }
    return weight->create_from_parameters(p);
}
 
Weight *
Weight::create_from_parameters(const char *) const
{
    throw Xapian::UnimplementedError("create_from_parameters() not supported for this Xapian::Weight subclass");
}
 
}