1 /** @file chert_values.h
2 * @brief ChertValueManager class
3 */
4 /* Copyright (C) 2008 Olly Betts
5 * Copyright (C) 2008 Lemur Consulting Ltd
6 *
7 * This program is free software; you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation; either version 2 of the License, or
10 * (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 USA
20 */
21
22 #ifndef XAPIAN_INCLUDED_CHERT_VALUES_H
23 #define XAPIAN_INCLUDED_CHERT_VALUES_H
24
25 #include "pack.h"
26 #include "valuestats.h"
27
28 #include "xapian/error.h"
29 #include "xapian/types.h"
30
31 #include "autoptr.h"
32 #include <map>
33 #include <string>
34
35 class ChertCursor;
36
37 /** Generate a key for a value stream chunk. */
38 inline std::string
make_valuechunk_key(Xapian::valueno slot,Xapian::docid did)39 make_valuechunk_key(Xapian::valueno slot, Xapian::docid did)
40 {
41 std::string key("\0\xd8", 2);
42 pack_uint(key, slot);
43 pack_uint_preserving_sort(key, did);
44 return key;
45 }
46
47 inline Xapian::docid
docid_from_key(Xapian::valueno required_slot,const std::string & key)48 docid_from_key(Xapian::valueno required_slot, const std::string & key)
49 {
50 const char * p = key.data();
51 const char * end = p + key.length();
52 // Fail if not a value chunk key.
53 if (end - p < 2 || *p++ != '\0' || *p++ != '\xd8') return 0;
54 Xapian::valueno slot;
55 if (!unpack_uint(&p, end, &slot))
56 throw Xapian::DatabaseCorruptError("bad value key");
57 // Fail if for a different slot.
58 if (slot != required_slot) return 0;
59 Xapian::docid did;
60 if (!unpack_uint_preserving_sort(&p, end, &did))
61 throw Xapian::DatabaseCorruptError("bad value key");
62 return did;
63 }
64
65 namespace Xapian {
66 class Document;
67 }
68
69 class ChertPostListTable;
70 class ChertTermListTable;
71 struct ValueStats;
72
73 class ChertValueManager {
74 /** The value number for the most recently used value statistics.
75 *
76 * Set to Xapian::BAD_VALUENO if no value statistics are currently
77 * cached.
78 */
79 mutable Xapian::valueno mru_slot;
80
81 /** The most recently used value statistics. */
82 mutable ValueStats mru_valstats;
83
84 ChertPostListTable * postlist_table;
85
86 ChertTermListTable * termlist_table;
87
88 std::map<Xapian::docid, std::string> slots;
89
90 std::map<Xapian::valueno, std::map<Xapian::docid, std::string> > changes;
91
92 mutable AutoPtr<ChertCursor> cursor;
93
94 void add_value(Xapian::docid did, Xapian::valueno slot,
95 const std::string & val);
96
97 void remove_value(Xapian::docid did, Xapian::valueno slot);
98
99 Xapian::docid get_chunk_containing_did(Xapian::valueno slot,
100 Xapian::docid did,
101 std::string &chunk) const;
102
103 /** Get the statistics for value slot @a slot. */
104 void get_value_stats(Xapian::valueno slot) const;
105
106 void get_value_stats(Xapian::valueno slot, ValueStats & stats) const;
107
108 public:
109 /** Create a new ChertValueManager object. */
ChertValueManager(ChertPostListTable * postlist_table_,ChertTermListTable * termlist_table_)110 ChertValueManager(ChertPostListTable * postlist_table_,
111 ChertTermListTable * termlist_table_)
112 : mru_slot(Xapian::BAD_VALUENO),
113 postlist_table(postlist_table_),
114 termlist_table(termlist_table_) { }
115
116 // Merge in batched-up changes.
117 void merge_changes();
118
119 void add_document(Xapian::docid did, const Xapian::Document &doc,
120 std::map<Xapian::valueno, ValueStats> & value_stats);
121
122 void delete_document(Xapian::docid did,
123 std::map<Xapian::valueno, ValueStats> & value_stats);
124
125 void replace_document(Xapian::docid did, const Xapian::Document &doc,
126 std::map<Xapian::valueno, ValueStats> & value_stats);
127
128 std::string get_value(Xapian::docid did, Xapian::valueno slot) const;
129
130 void get_all_values(std::map<Xapian::valueno, std::string> & values,
131 Xapian::docid did) const;
132
get_value_freq(Xapian::valueno slot)133 Xapian::doccount get_value_freq(Xapian::valueno slot) const {
134 if (mru_slot != slot) get_value_stats(slot);
135 return mru_valstats.freq;
136 }
137
get_value_lower_bound(Xapian::valueno slot)138 std::string get_value_lower_bound(Xapian::valueno slot) const {
139 if (mru_slot != slot) get_value_stats(slot);
140 return mru_valstats.lower_bound;
141 }
142
get_value_upper_bound(Xapian::valueno slot)143 std::string get_value_upper_bound(Xapian::valueno slot) const {
144 if (mru_slot != slot) get_value_stats(slot);
145 return mru_valstats.upper_bound;
146 }
147
148 /** Write the updated statistics to the table.
149 *
150 * If the @a freq member of the statistics for a particular slot is 0, the
151 * statistics for that slot will be cleared.
152 *
153 * @param value_stats The statistics to set.
154 */
155 void set_value_stats(std::map<Xapian::valueno, ValueStats> & value_stats);
156
reset()157 void reset() {
158 /// Ignore any old cached valuestats.
159 mru_slot = Xapian::BAD_VALUENO;
160 }
161
is_modified()162 bool is_modified() const {
163 return !changes.empty();
164 }
165
cancel()166 void cancel() {
167 // Discard batched-up changes.
168 slots.clear();
169 changes.clear();
170 }
171 };
172
173 class ValueChunkReader {
174 const char *p;
175 const char *end;
176
177 Xapian::docid did;
178
179 std::string value;
180
181 public:
182 /// Create a ValueChunkReader which is already at_end().
ValueChunkReader()183 ValueChunkReader() : p(NULL) { }
184
ValueChunkReader(const char * p_,size_t len,Xapian::docid did_)185 ValueChunkReader(const char * p_, size_t len, Xapian::docid did_) {
186 assign(p_, len, did_);
187 }
188
189 void assign(const char * p_, size_t len, Xapian::docid did_);
190
at_end()191 bool at_end() const { return p == NULL; }
192
get_docid()193 Xapian::docid get_docid() const { return did; }
194
get_value()195 const std::string & get_value() const { return value; }
196
197 void next();
198
199 void skip_to(Xapian::docid target);
200 };
201
202 #endif // XAPIAN_INCLUDED_CHERT_VALUES_H
203