include/xapian/enquire.h

Go to the documentation of this file.
00001 
00004 /* Copyright 1999,2000,2001 BrightStation PLC
00005  * Copyright 2001,2002 Ananova Ltd
00006  * Copyright 2002,2003,2004,2005,2006 Olly Betts
00007  *
00008  * This program is free software; you can redistribute it and/or
00009  * modify it under the terms of the GNU General Public License as
00010  * published by the Free Software Foundation; either version 2 of the
00011  * License, or (at your option) any later version.
00012  *
00013  * This program is distributed in the hope that it will be useful,
00014  * but WITHOUT ANY WARRANTY; without even the implied warranty of
00015  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
00016  * GNU General Public License for more details.
00017  *
00018  * You should have received a copy of the GNU General Public License
00019  * along with this program; if not, write to the Free Software
00020  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301
00021  * USA
00022  */
00023 
00024 #ifndef XAPIAN_INCLUDED_ENQUIRE_H
00025 #define XAPIAN_INCLUDED_ENQUIRE_H
00026 
00027 #include <string>
00028 #include <time.h> // for time_t
00029 
00030 #include <xapian/base.h>
00031 #include <xapian/error.h>
00032 #include <xapian/types.h>
00033 #include <xapian/termiterator.h>
00034 
00035 namespace Xapian {
00036 
00037 class Database;
00038 class Document;
00039 class ErrorHandler;
00040 class MSetIterator;
00041 class Query;
00042 class Weight;
00043 
00047 class MSet {
00048     public:
00049         class Internal;
00051         Xapian::Internal::RefCntPtr<Internal> internal;
00052 
00054         explicit MSet(MSet::Internal * internal_);
00055 
00057         MSet();
00058 
00060         ~MSet();
00061 
00063         MSet(const MSet & other);
00064 
00066         void operator=(const MSet &other);
00067 
00083         void fetch(const MSetIterator &begin, const MSetIterator &end) const;
00084 
00087         void fetch(const MSetIterator &item) const;
00088 
00091         void fetch() const;
00092 
00097         Xapian::percent convert_to_percent(Xapian::weight wt) const;
00098 
00100         Xapian::percent convert_to_percent(const MSetIterator &it) const;
00101 
00109         Xapian::doccount get_termfreq(const std::string &tname) const;
00110 
00118         Xapian::weight get_termweight(const std::string &tname) const;
00119 
00127         Xapian::doccount get_firstitem() const;
00128 
00138         Xapian::doccount get_matches_lower_bound() const;
00139 
00152         Xapian::doccount get_matches_estimated() const;
00153 
00163         Xapian::doccount get_matches_upper_bound() const;
00164 
00170         Xapian::weight get_max_possible() const;
00171 
00185         Xapian::weight get_max_attained() const;
00186 
00188         Xapian::doccount size() const;
00189 
00191         Xapian::doccount max_size() const { return size(); }
00192 
00194         bool empty() const;
00195 
00197         void swap(MSet & other);
00198 
00200         MSetIterator begin() const;
00201 
00203         MSetIterator end() const;
00204 
00206         MSetIterator back() const;
00207 
00217         MSetIterator operator[](Xapian::doccount i) const;
00218 
00220 
00221         typedef MSetIterator value_type; // FIXME: not assignable...
00222         typedef MSetIterator iterator;
00223         typedef MSetIterator const_iterator;
00224         typedef MSetIterator & reference; // Hmm
00225         typedef MSetIterator & const_reference;
00226         typedef MSetIterator * pointer; // Hmm
00227         typedef Xapian::doccount_diff difference_type;
00228         typedef Xapian::doccount size_type;
00230 
00234         std::string get_description() const;
00235 };
00236 
00240 class MSetIterator {
00241     private:
00242         friend class MSet;
00243         friend bool operator==(const MSetIterator &a, const MSetIterator &b);
00244         friend bool operator!=(const MSetIterator &a, const MSetIterator &b);
00245 
00246         MSetIterator(Xapian::doccount index_, const MSet & mset_)
00247             : index(index_), mset(mset_) { }
00248 
00249         Xapian::doccount index;
00250         MSet mset;
00251 
00252     public:
00256         MSetIterator() : index(0), mset() { }
00257 
00258         ~MSetIterator() { }
00259 
00261         MSetIterator(const MSetIterator &other) {
00262             index = other.index;
00263             mset = other.mset;
00264         }
00265 
00267         void operator=(const MSetIterator &other) {
00268             index = other.index;
00269             mset = other.mset;
00270         }
00271 
00273         MSetIterator & operator++() {
00274             ++index;
00275             return *this;
00276         }
00277 
00279         MSetIterator operator++(int) {
00280             MSetIterator tmp = *this;
00281             ++index;
00282             return tmp;
00283         }
00284 
00286         MSetIterator & operator--() {
00287             --index;
00288             return *this;
00289         }
00290 
00292         MSetIterator operator--(int) {
00293             MSetIterator tmp = *this;
00294             --index;
00295             return tmp;
00296         }
00297 
00299         Xapian::docid operator*() const;
00300 
00317         Xapian::Document get_document() const;
00318 
00325         Xapian::doccount get_rank() const {
00326             return mset.get_firstitem() + index;
00327         }
00328 
00330         Xapian::weight get_weight() const;
00331 
00334         std::string get_collapse_key() const;
00335 
00352         Xapian::doccount get_collapse_count() const;
00353 
00359         Xapian::percent get_percent() const;
00360 
00364         std::string get_description() const;
00365 
00367 
00368         typedef std::bidirectional_iterator_tag iterator_category; // FIXME: could enhance to be a randomaccess_iterator
00369         typedef Xapian::docid value_type;
00370         typedef Xapian::doccount_diff difference_type;
00371         typedef Xapian::docid * pointer;
00372         typedef Xapian::docid & reference;
00374 };
00375 
00376 inline bool operator==(const MSetIterator &a, const MSetIterator &b)
00377 {
00378     return (a.index == b.index);
00379 }
00380 
00381 inline bool operator!=(const MSetIterator &a, const MSetIterator &b)
00382 {
00383     return (a.index != b.index);
00384 }
00385 
00386 class ESetIterator;
00387 
00392 class ESet {
00393     public:
00394         class Internal;
00396         Xapian::Internal::RefCntPtr<Internal> internal;
00397 
00399         ESet();
00400 
00402         ~ESet();
00403 
00405         ESet(const ESet & other);
00406 
00408         void operator=(const ESet &other);
00409 
00414         Xapian::termcount get_ebound() const;
00415 
00417         Xapian::termcount size() const;
00418 
00420         Xapian::termcount max_size() const { return size(); }
00421 
00423         bool empty() const;
00424 
00426         void swap(ESet & other);
00427 
00429         ESetIterator begin() const;
00430 
00432         ESetIterator end() const;
00433 
00435         ESetIterator back() const;
00436 
00438         ESetIterator operator[](Xapian::termcount i) const;
00439 
00444         std::string get_description() const;
00445 };
00446 
00448 class ESetIterator {
00449     private:
00450         friend class ESet;
00451         friend bool operator==(const ESetIterator &a, const ESetIterator &b);
00452         friend bool operator!=(const ESetIterator &a, const ESetIterator &b);
00453 
00454         ESetIterator(Xapian::termcount index_, const ESet & eset_)
00455             : index(index_), eset(eset_) { }
00456 
00457         Xapian::termcount index;
00458         ESet eset;
00459 
00460     public:
00464         ESetIterator() : index(0), eset() { }
00465 
00466         ~ESetIterator() { }
00467 
00469         ESetIterator(const ESetIterator &other) {
00470             index = other.index;
00471             eset = other.eset;
00472         }
00473 
00475         void operator=(const ESetIterator &other) {
00476             index = other.index;
00477             eset = other.eset;
00478         }
00479 
00481         ESetIterator & operator++() {
00482             ++index;
00483             return *this;
00484         }
00485 
00487         ESetIterator operator++(int) {
00488             ESetIterator tmp = *this;
00489             ++index;
00490             return tmp;
00491         }
00492 
00494         ESetIterator & operator--() {
00495             --index;
00496             return *this;
00497         }
00498 
00500         ESetIterator operator--(int) {
00501             ESetIterator tmp = *this;
00502             --index;
00503             return tmp;
00504         }
00505 
00507         const std::string & operator *() const;
00508 
00510         Xapian::weight get_weight() const;
00511 
00515         std::string get_description() const;
00516 
00518 
00519         typedef std::bidirectional_iterator_tag iterator_category; // FIXME: go for randomaccess_iterator!
00520         typedef std::string value_type;
00521         typedef Xapian::termcount_diff difference_type;
00522         typedef std::string * pointer;
00523         typedef std::string & reference;
00525 };
00526 
00527 inline bool operator==(const ESetIterator &a, const ESetIterator &b)
00528 {
00529     return (a.index == b.index);
00530 }
00531 
00532 inline bool operator!=(const ESetIterator &a, const ESetIterator &b)
00533 {
00534     return (a.index != b.index);
00535 }
00536 
00541 class RSet {
00542     public:
00544         class Internal;
00545 
00547         Xapian::Internal::RefCntPtr<Internal> internal;
00548 
00550         RSet(const RSet &rset);
00551 
00553         void operator=(const RSet &rset);
00554 
00556         RSet();
00557 
00559         ~RSet();
00560 
00562         Xapian::doccount size() const;
00563 
00565         bool empty() const;
00566 
00568         void add_document(Xapian::docid did);
00569 
00571         void add_document(const Xapian::MSetIterator & i) { add_document(*i); }
00572 
00574         void remove_document(Xapian::docid did);
00575 
00577         void remove_document(const Xapian::MSetIterator & i) { remove_document(*i); }
00578 
00580         bool contains(Xapian::docid did) const;
00581 
00583         bool contains(const Xapian::MSetIterator & i) { return contains(*i); }
00584 
00589         std::string get_description() const;
00590 };
00591 
00594 class MatchDecider {
00595     public:
00598         virtual int operator()(const Xapian::Document &doc) const = 0;
00599 
00601         virtual ~MatchDecider() {}
00602 };
00603 
00606 class ExpandDecider {
00607     public:
00610         virtual int operator()(const std::string & tname) const = 0;
00611 
00613         virtual ~ExpandDecider() {}
00614 };
00615 
00626 class Enquire {
00627     private:
00629         Enquire(const Enquire &);
00630 
00632         void operator=(const Enquire &);
00633 
00634     public:
00635         class Internal;
00637         Xapian::Internal::RefCntPtr<Internal> internal;
00638 
00654         Enquire(const Database &databases, ErrorHandler * errorhandler_ = 0);
00655 
00658         ~Enquire();
00659 
00666         void set_query(const Xapian::Query & query, Xapian::termcount qlen = 0);
00667 
00674         const Xapian::Query & get_query();
00675 
00682         void set_weighting_scheme(const Weight &weight_);
00683 
00710         void set_collapse_key(Xapian::valueno collapse_key);
00711 
00712         typedef enum {
00713             ASCENDING = 1,
00714             DESCENDING = 0,
00715             DONT_CARE = 2
00716         } docid_order;
00717 
00741         void set_docid_order(docid_order order);
00742 
00749         XAPIAN_DEPRECATED(void set_sort_forward(bool sort_forward));
00750 
00769         void set_cutoff(Xapian::percent percent_cutoff, Xapian::weight weight_cutoff = 0);
00770 
00787         XAPIAN_DEPRECATED(void set_sorting(Xapian::valueno sort_key, int sort_bands,
00788                           bool sort_by_relevance = false));
00789 
00794         void set_sort_by_relevance();
00795 
00806         void set_sort_by_value(Xapian::valueno sort_key, bool ascending = true);
00807 
00819         void set_sort_by_value_then_relevance(Xapian::valueno sort_key,
00820                                               bool ascending = true);
00821 
00839         void set_sort_by_relevance_then_value(Xapian::valueno sort_key,
00840                                               bool ascending = true);
00841 
00853         void set_bias(Xapian::weight bias_weight, time_t bias_halflife);
00854 
00880         MSet get_mset(Xapian::doccount first, Xapian::doccount maxitems,
00881                       Xapian::doccount checkatleast = 0,
00882                       const RSet * omrset = 0,
00883                       const MatchDecider * mdecider = 0) const;
00884         MSet get_mset(Xapian::doccount first, Xapian::doccount maxitems,
00885                       const RSet * omrset,
00886                       const MatchDecider * mdecider = 0) const {
00887             return get_mset(first, maxitems, 0, omrset, mdecider);
00888         }
00889 
00890         static const int include_query_terms = 1;
00891         static const int use_exact_termfreq = 2;
00914         ESet get_eset(Xapian::termcount maxitems,
00915                         const RSet & omrset,
00916                         int flags = 0,
00917                         double k = 1.0,
00918                         const Xapian::ExpandDecider * edecider = 0) const;
00919 
00933         inline ESet get_eset(Xapian::termcount maxitems, const RSet & omrset,
00934                                const Xapian::ExpandDecider * edecider) const {
00935             return get_eset(maxitems, omrset, 0, 1.0, edecider);
00936         }
00937 
00966         TermIterator get_matching_terms_begin(Xapian::docid did) const;
00967 
00969         TermIterator get_matching_terms_end(Xapian::docid /*did*/) const {
00970             return TermIterator(NULL);
00971         }
00972 
00995         TermIterator get_matching_terms_begin(const MSetIterator &it) const;
00996 
00998         TermIterator get_matching_terms_end(const MSetIterator &/*it*/) const {
00999             return TermIterator(NULL);
01000         }
01001 
01008         void register_match_decider(const std::string &name,
01009                                     const MatchDecider *mdecider = NULL);
01010 
01014         std::string get_description() const;
01015 };
01016 
01017 }
01018 
01019 class RemoteServer;
01020 
01021 namespace Xapian {
01022 
01024 class Weight {
01025     friend class Enquire; // So Enquire can clone us
01026     friend class ::RemoteServer; // So RemoteServer can clone us - FIXME
01027     public:
01028         class Internal;
01029     protected:
01030         Weight(const Weight &);
01031     private:
01032         void operator=(Weight &);
01033 
01043         virtual Weight * clone() const = 0;
01044 
01045     protected:
01046         const Internal * internal; // Weight::Internal == StatsSource
01047         Xapian::doclength querysize;
01048         Xapian::termcount wqf;
01049         std::string tname;
01050 
01051     public:
01052         Weight() { }
01053         virtual ~Weight() { }
01054 
01067         Weight * create(const Internal * internal_, Xapian::doclength querysize_,
01068                           Xapian::termcount wqf_, std::string tname_) const {
01069             Weight * wt = clone();
01070             wt->internal = internal_;
01071             wt->querysize = querysize_;
01072             wt->wqf = wqf_;
01073             wt->tname = tname_;
01074             return wt;
01075         }
01076 
01081         virtual std::string name() const = 0;
01082 
01084         virtual std::string serialise() const = 0;
01085 
01087         virtual Weight * unserialise(const std::string &s) const = 0;
01088 
01096         virtual Xapian::weight get_sumpart(Xapian::termcount wdf,
01097                                       Xapian::doclength len) const = 0;
01098 
01104         virtual Xapian::weight get_maxpart() const = 0;
01105 
01114         virtual Xapian::weight get_sumextra(Xapian::doclength len) const = 0;
01115 
01119         virtual Xapian::weight get_maxextra() const = 0;
01120 
01122         virtual bool get_sumpart_needs_doclength() const { return true; }
01123 };
01124 
01126 class BoolWeight : public Weight {
01127     public:
01128         BoolWeight * clone() const {
01129             return new BoolWeight;
01130         }
01131         BoolWeight() { }
01132         ~BoolWeight() { }
01133         std::string name() const { return "Bool"; }
01134         std::string serialise() const { return ""; }
01135         BoolWeight * unserialise(const std::string & /*s*/) const {
01136             return new BoolWeight;
01137         }
01138         Xapian::weight get_sumpart(Xapian::termcount /*wdf*/, Xapian::doclength /*len*/) const { return 0; }
01139         Xapian::weight get_maxpart() const { return 0; }
01140 
01141         Xapian::weight get_sumextra(Xapian::doclength /*len*/) const { return 0; }
01142         Xapian::weight get_maxextra() const { return 0; }
01143 
01144         bool get_sumpart_needs_doclength() const { return false; }
01145 };
01146 
01159 class BM25Weight : public Weight {
01160     private:
01161         mutable Xapian::weight termweight;
01162         mutable Xapian::doclength lenpart;
01163 
01164         double k1, k2, k3, b;
01165         Xapian::doclength min_normlen;
01166 
01167         mutable bool weight_calculated;
01168 
01169         void calc_termweight() const;
01170 
01171     public:
01190         BM25Weight(double k1_, double k2_, double k3_, double b_,
01191                    double min_normlen_)
01192                 : k1(k1_), k2(k2_), k3(k3_), b(b_), min_normlen(min_normlen_),
01193                   weight_calculated(false)
01194         {
01195             if (k1 < 0) k1 = 0;
01196             if (k2 < 0) k2 = 0;
01197             if (k3 < 0) k3 = 0;
01198             if (b < 0) b = 0; else if (b > 1) b = 1;
01199         }
01200         BM25Weight() : k1(1), k2(0), k3(1), b(0.5), min_normlen(0.5),
01201                        weight_calculated(false) { }
01202 
01203         BM25Weight * clone() const;
01204         ~BM25Weight() { }
01205         std::string name() const;
01206         std::string serialise() const;
01207         BM25Weight * unserialise(const std::string & s) const;
01208         Xapian::weight get_sumpart(Xapian::termcount wdf, Xapian::doclength len) const;
01209         Xapian::weight get_maxpart() const;
01210 
01211         Xapian::weight get_sumextra(Xapian::doclength len) const;
01212         Xapian::weight get_maxextra() const;
01213 
01214         bool get_sumpart_needs_doclength() const;
01215 };
01216 
01230 class TradWeight : public Weight {
01231     private:
01232         mutable Xapian::weight termweight;
01233         mutable Xapian::doclength lenpart;
01234 
01235         double param_k;
01236 
01237         mutable bool weight_calculated;
01238 
01239         void calc_termweight() const;
01240 
01241     public:
01249         explicit TradWeight(double k) : param_k(k), weight_calculated(false) {
01250             if (param_k < 0) param_k = 0;
01251         }
01252 
01253         TradWeight() : param_k(1.0), weight_calculated(false) { }
01254 
01255         TradWeight * clone() const;
01256         ~TradWeight() { }
01257         std::string name() const;
01258         std::string serialise() const;
01259         TradWeight * unserialise(const std::string & s) const;
01260 
01261         Xapian::weight get_sumpart(Xapian::termcount wdf, Xapian::doclength len) const;
01262         Xapian::weight get_maxpart() const;
01263 
01264         Xapian::weight get_sumextra(Xapian::doclength len) const;
01265         Xapian::weight get_maxextra() const;
01266 
01267         bool get_sumpart_needs_doclength() const;
01268 };
01269 
01270 }
01271 
01272 #endif /* XAPIAN_INCLUDED_ENQUIRE_H */

Documentation for Xapian (version 0.9.9).
Generated on 9 Nov 2006 by Doxygen 1.4.6.