ferret 0.11.6 → 0.11.8.4
Sign up to get free protection for your applications and to get access to all the features.
- data/README +10 -22
- data/RELEASE_CHANGES +137 -0
- data/RELEASE_NOTES +60 -0
- data/Rakefile +379 -274
- data/TODO +100 -8
- data/bin/ferret-browser +0 -0
- data/ext/BZLIB_blocksort.c +1094 -0
- data/ext/BZLIB_bzlib.c +1578 -0
- data/ext/BZLIB_compress.c +672 -0
- data/ext/BZLIB_crctable.c +104 -0
- data/ext/BZLIB_decompress.c +626 -0
- data/ext/BZLIB_huffman.c +205 -0
- data/ext/BZLIB_randtable.c +84 -0
- data/ext/{api.c → STEMMER_api.c} +7 -10
- data/ext/{libstemmer.c → STEMMER_libstemmer.c} +3 -2
- data/ext/{stem_ISO_8859_1_danish.c → STEMMER_stem_ISO_8859_1_danish.c} +123 -124
- data/ext/{stem_ISO_8859_1_dutch.c → STEMMER_stem_ISO_8859_1_dutch.c} +177 -188
- data/ext/STEMMER_stem_ISO_8859_1_english.c +1117 -0
- data/ext/{stem_ISO_8859_1_finnish.c → STEMMER_stem_ISO_8859_1_finnish.c} +276 -306
- data/ext/STEMMER_stem_ISO_8859_1_french.c +1246 -0
- data/ext/{stem_ISO_8859_1_german.c → STEMMER_stem_ISO_8859_1_german.c} +161 -170
- data/ext/STEMMER_stem_ISO_8859_1_hungarian.c +1230 -0
- data/ext/STEMMER_stem_ISO_8859_1_italian.c +1065 -0
- data/ext/STEMMER_stem_ISO_8859_1_norwegian.c +297 -0
- data/ext/{stem_ISO_8859_1_porter.c → STEMMER_stem_ISO_8859_1_porter.c} +263 -290
- data/ext/{stem_ISO_8859_1_portuguese.c → STEMMER_stem_ISO_8859_1_portuguese.c} +362 -380
- data/ext/STEMMER_stem_ISO_8859_1_spanish.c +1093 -0
- data/ext/STEMMER_stem_ISO_8859_1_swedish.c +307 -0
- data/ext/STEMMER_stem_ISO_8859_2_romanian.c +998 -0
- data/ext/{stem_KOI8_R_russian.c → STEMMER_stem_KOI8_R_russian.c} +244 -245
- data/ext/STEMMER_stem_UTF_8_danish.c +339 -0
- data/ext/{stem_UTF_8_dutch.c → STEMMER_stem_UTF_8_dutch.c} +192 -211
- data/ext/STEMMER_stem_UTF_8_english.c +1125 -0
- data/ext/{stem_UTF_8_finnish.c → STEMMER_stem_UTF_8_finnish.c} +284 -324
- data/ext/STEMMER_stem_UTF_8_french.c +1256 -0
- data/ext/{stem_UTF_8_german.c → STEMMER_stem_UTF_8_german.c} +170 -187
- data/ext/STEMMER_stem_UTF_8_hungarian.c +1234 -0
- data/ext/STEMMER_stem_UTF_8_italian.c +1073 -0
- data/ext/STEMMER_stem_UTF_8_norwegian.c +299 -0
- data/ext/{stem_UTF_8_porter.c → STEMMER_stem_UTF_8_porter.c} +271 -310
- data/ext/STEMMER_stem_UTF_8_portuguese.c +1023 -0
- data/ext/STEMMER_stem_UTF_8_romanian.c +1004 -0
- data/ext/STEMMER_stem_UTF_8_russian.c +694 -0
- data/ext/STEMMER_stem_UTF_8_spanish.c +1097 -0
- data/ext/STEMMER_stem_UTF_8_swedish.c +309 -0
- data/ext/STEMMER_stem_UTF_8_turkish.c +2205 -0
- data/ext/{utilities.c → STEMMER_utilities.c} +100 -68
- data/ext/analysis.c +276 -121
- data/ext/analysis.h +190 -143
- data/ext/api.h +3 -4
- data/ext/array.c +5 -3
- data/ext/array.h +52 -43
- data/ext/bitvector.c +38 -482
- data/ext/bitvector.h +446 -124
- data/ext/bzlib.h +282 -0
- data/ext/bzlib_private.h +503 -0
- data/ext/compound_io.c +23 -22
- data/ext/config.h +21 -11
- data/ext/document.c +43 -40
- data/ext/document.h +31 -21
- data/ext/except.c +20 -38
- data/ext/except.h +89 -76
- data/ext/extconf.rb +3 -2
- data/ext/ferret.c +49 -35
- data/ext/ferret.h +14 -11
- data/ext/field_index.c +262 -0
- data/ext/field_index.h +52 -0
- data/ext/filter.c +11 -10
- data/ext/fs_store.c +65 -47
- data/ext/global.c +245 -165
- data/ext/global.h +252 -54
- data/ext/hash.c +200 -243
- data/ext/hash.h +205 -163
- data/ext/hashset.c +118 -96
- data/ext/hashset.h +110 -82
- data/ext/header.h +19 -19
- data/ext/helper.c +11 -10
- data/ext/helper.h +14 -6
- data/ext/index.c +745 -366
- data/ext/index.h +503 -529
- data/ext/internal.h +1020 -0
- data/ext/lang.c +10 -0
- data/ext/lang.h +35 -15
- data/ext/mempool.c +5 -4
- data/ext/mempool.h +30 -22
- data/ext/modules.h +35 -7
- data/ext/multimapper.c +43 -2
- data/ext/multimapper.h +32 -23
- data/ext/posh.c +0 -0
- data/ext/posh.h +4 -38
- data/ext/priorityqueue.c +10 -12
- data/ext/priorityqueue.h +33 -21
- data/ext/q_boolean.c +22 -9
- data/ext/q_const_score.c +3 -2
- data/ext/q_filtered_query.c +15 -12
- data/ext/q_fuzzy.c +147 -135
- data/ext/q_match_all.c +3 -2
- data/ext/q_multi_term.c +28 -32
- data/ext/q_parser.c +451 -173
- data/ext/q_phrase.c +158 -79
- data/ext/q_prefix.c +16 -18
- data/ext/q_range.c +363 -31
- data/ext/q_span.c +130 -141
- data/ext/q_term.c +21 -21
- data/ext/q_wildcard.c +19 -23
- data/ext/r_analysis.c +369 -242
- data/ext/r_index.c +421 -434
- data/ext/r_qparser.c +142 -92
- data/ext/r_search.c +790 -407
- data/ext/r_store.c +44 -44
- data/ext/r_utils.c +264 -96
- data/ext/ram_store.c +29 -23
- data/ext/scanner.c +895 -0
- data/ext/scanner.h +36 -0
- data/ext/scanner_mb.c +6701 -0
- data/ext/scanner_utf8.c +4415 -0
- data/ext/search.c +210 -87
- data/ext/search.h +556 -488
- data/ext/similarity.c +17 -16
- data/ext/similarity.h +51 -44
- data/ext/sort.c +157 -354
- data/ext/stem_ISO_8859_1_hungarian.h +16 -0
- data/ext/stem_ISO_8859_2_romanian.h +16 -0
- data/ext/stem_UTF_8_hungarian.h +16 -0
- data/ext/stem_UTF_8_romanian.h +16 -0
- data/ext/stem_UTF_8_turkish.h +16 -0
- data/ext/stopwords.c +287 -278
- data/ext/store.c +57 -51
- data/ext/store.h +308 -286
- data/ext/symbol.c +10 -0
- data/ext/symbol.h +23 -0
- data/ext/term_vectors.c +14 -293
- data/ext/threading.h +22 -22
- data/ext/win32.h +12 -4
- data/lib/ferret.rb +2 -1
- data/lib/ferret/browser.rb +1 -1
- data/lib/ferret/field_symbol.rb +94 -0
- data/lib/ferret/index.rb +221 -34
- data/lib/ferret/number_tools.rb +6 -6
- data/lib/ferret/version.rb +3 -0
- data/test/{unit → long_running}/largefile/tc_largefile.rb +1 -1
- data/test/test_helper.rb +7 -2
- data/test/test_installed.rb +1 -0
- data/test/threading/thread_safety_index_test.rb +10 -1
- data/test/threading/thread_safety_read_write_test.rb +4 -7
- data/test/threading/thread_safety_test.rb +0 -0
- data/test/unit/analysis/tc_analyzer.rb +29 -27
- data/test/unit/analysis/tc_token_stream.rb +23 -16
- data/test/unit/index/tc_index.rb +116 -11
- data/test/unit/index/tc_index_reader.rb +27 -27
- data/test/unit/index/tc_index_writer.rb +10 -0
- data/test/unit/index/th_doc.rb +38 -21
- data/test/unit/search/tc_filter.rb +31 -10
- data/test/unit/search/tc_index_searcher.rb +6 -0
- data/test/unit/search/tm_searcher.rb +53 -1
- data/test/unit/store/tc_fs_store.rb +40 -2
- data/test/unit/store/tc_ram_store.rb +0 -0
- data/test/unit/store/tm_store.rb +0 -0
- data/test/unit/store/tm_store_lock.rb +7 -6
- data/test/unit/tc_field_symbol.rb +26 -0
- data/test/unit/ts_analysis.rb +0 -0
- data/test/unit/ts_index.rb +0 -0
- data/test/unit/ts_store.rb +0 -0
- data/test/unit/ts_utils.rb +0 -0
- data/test/unit/utils/tc_number_tools.rb +0 -0
- data/test/utils/content_generator.rb +226 -0
- metadata +262 -221
- data/ext/inc/lang.h +0 -48
- data/ext/inc/threading.h +0 -31
- data/ext/stem_ISO_8859_1_english.c +0 -1156
- data/ext/stem_ISO_8859_1_french.c +0 -1276
- data/ext/stem_ISO_8859_1_italian.c +0 -1091
- data/ext/stem_ISO_8859_1_norwegian.c +0 -296
- data/ext/stem_ISO_8859_1_spanish.c +0 -1119
- data/ext/stem_ISO_8859_1_swedish.c +0 -307
- data/ext/stem_UTF_8_danish.c +0 -344
- data/ext/stem_UTF_8_english.c +0 -1176
- data/ext/stem_UTF_8_french.c +0 -1296
- data/ext/stem_UTF_8_italian.c +0 -1113
- data/ext/stem_UTF_8_norwegian.c +0 -302
- data/ext/stem_UTF_8_portuguese.c +0 -1055
- data/ext/stem_UTF_8_russian.c +0 -709
- data/ext/stem_UTF_8_spanish.c +0 -1137
- data/ext/stem_UTF_8_swedish.c +0 -313
- data/lib/ferret_version.rb +0 -3
data/ext/symbol.c
ADDED
data/ext/symbol.h
ADDED
@@ -0,0 +1,23 @@
|
|
1
|
+
#ifndef FRT_SYMBOL_H
|
2
|
+
#define FRT_SYMBOL_H
|
3
|
+
|
4
|
+
#include <ruby.h>
|
5
|
+
|
6
|
+
typedef void *FrtSymbol;
|
7
|
+
|
8
|
+
VALUE rb_id2str(ID id);
|
9
|
+
#define frt_symbol_init()
|
10
|
+
#define frt_intern(str) (FrtSymbol)rb_intern(str)
|
11
|
+
extern FrtSymbol frt_intern_and_free(char *str);
|
12
|
+
|
13
|
+
#define FRT_I frt_intern
|
14
|
+
#define FRT_InF frt_intern_and_free
|
15
|
+
#define FRT_S(_sym) rb_id2name((ID)_sym)
|
16
|
+
|
17
|
+
#define FSYM2SYM(_sym) ID2SYM((ID)_sym)
|
18
|
+
#define SYM2FSYM(_sym) (FrtSymbol)SYM2ID(_sym)
|
19
|
+
|
20
|
+
#define frt_sym_hash(sym) frt_str_hash(rb_id2name((ID)sym))
|
21
|
+
#define frt_sym_len(sym) strlen(rb_id2name((ID)sym))
|
22
|
+
|
23
|
+
#endif
|
data/ext/term_vectors.c
CHANGED
@@ -2,6 +2,8 @@
|
|
2
2
|
#include "index.h"
|
3
3
|
#include "array.h"
|
4
4
|
#include "helper.h"
|
5
|
+
#include "symbol.h"
|
6
|
+
#include "internal.h"
|
5
7
|
|
6
8
|
/****************************************************************************
|
7
9
|
*
|
@@ -18,12 +20,11 @@ void tv_destroy(TermVector *tv)
|
|
18
20
|
free(tv->terms[i].positions);
|
19
21
|
}
|
20
22
|
free(tv->offsets);
|
21
|
-
free(tv->field);
|
22
23
|
free(tv->terms);
|
23
24
|
free(tv);
|
24
25
|
}
|
25
26
|
|
26
|
-
int
|
27
|
+
int tv_scan_to_term_index(TermVector *tv, const char *term)
|
27
28
|
{
|
28
29
|
int lo = 0; /* search starts array */
|
29
30
|
int hi = tv->term_cnt - 1; /* for 1st element < n, return its index */
|
@@ -45,18 +46,24 @@ int tv_get_tv_term_index(TermVector *tv, const char *term)
|
|
45
46
|
return mid;
|
46
47
|
}
|
47
48
|
}
|
48
|
-
|
49
|
-
|
49
|
+
return lo;
|
50
|
+
}
|
51
|
+
|
52
|
+
int tv_get_term_index(TermVector *tv, const char *term)
|
53
|
+
{
|
54
|
+
int index = tv_scan_to_term_index(tv, term);
|
55
|
+
if (index < tv->term_cnt && (0 == strcmp(term, tv->terms[index].text))) {
|
56
|
+
/* found term */
|
57
|
+
return index;
|
50
58
|
}
|
51
59
|
else {
|
52
60
|
return -1;
|
53
61
|
}
|
54
|
-
return hi;
|
55
62
|
}
|
56
63
|
|
57
|
-
|
64
|
+
TVTerm *tv_get_tv_term(TermVector *tv, const char *term)
|
58
65
|
{
|
59
|
-
int index =
|
66
|
+
int index = tv_get_term_index(tv, term);
|
60
67
|
if (index >= 0) {
|
61
68
|
return &(tv->terms[index]);
|
62
69
|
}
|
@@ -64,289 +71,3 @@ extern TVTerm *tv_get_tv_term(TermVector *tv, const char *term)
|
|
64
71
|
return NULL;
|
65
72
|
}
|
66
73
|
}
|
67
|
-
|
68
|
-
/****************************************************************************
|
69
|
-
*
|
70
|
-
* TermVectorsReader
|
71
|
-
*
|
72
|
-
****************************************************************************/
|
73
|
-
|
74
|
-
TermVectorsReader *tvr_open(Store *store,
|
75
|
-
const char *segment,
|
76
|
-
FieldInfos *fis)
|
77
|
-
{
|
78
|
-
TermVectorsReader *tvr = ALLOC(TermVectorsReader);
|
79
|
-
char file_name[SEGMENT_NAME_MAX_LENGTH];
|
80
|
-
|
81
|
-
tvr->fis = fis;
|
82
|
-
sprintf(file_name, "%s.tvx", segment);
|
83
|
-
tvr->tvx_in = store->open_input(store, file_name);
|
84
|
-
tvr->size = is_length(tvr->tvx_in) / 12;
|
85
|
-
|
86
|
-
sprintf(file_name, "%s.tvd", segment);
|
87
|
-
tvr->tvd_in = store->open_input(store, file_name);
|
88
|
-
return tvr;
|
89
|
-
}
|
90
|
-
|
91
|
-
TermVectorsReader *tvr_clone(TermVectorsReader *orig)
|
92
|
-
{
|
93
|
-
TermVectorsReader *tvr = ALLOC(TermVectorsReader);
|
94
|
-
|
95
|
-
memcpy(tvr, orig, sizeof(TermVectorsReader));
|
96
|
-
tvr->tvx_in = is_clone(orig->tvx_in);
|
97
|
-
tvr->tvd_in = is_clone(orig->tvd_in);
|
98
|
-
|
99
|
-
return tvr;
|
100
|
-
}
|
101
|
-
|
102
|
-
void tvr_close(TermVectorsReader *tvr)
|
103
|
-
{
|
104
|
-
is_close(tvr->tvx_in);
|
105
|
-
is_close(tvr->tvd_in);
|
106
|
-
free(tvr);
|
107
|
-
}
|
108
|
-
|
109
|
-
TermVector *tvr_read_term_vector(TermVectorsReader *tvr, int field_num)
|
110
|
-
{
|
111
|
-
TermVector *tv = ALLOC_AND_ZERO(TermVector);
|
112
|
-
InStream *tvd_in = tvr->tvd_in;
|
113
|
-
FieldInfo *fi = tvr->fis->fields[field_num];
|
114
|
-
const int num_terms = is_read_vint(tvd_in);
|
115
|
-
|
116
|
-
tv->field_num = field_num;
|
117
|
-
tv->field = estrdup(fi->name);
|
118
|
-
|
119
|
-
if (num_terms > 0) {
|
120
|
-
int i, j, delta_start, delta_len, total_len, freq;
|
121
|
-
int store_positions = fi_store_positions(fi);
|
122
|
-
int store_offsets = fi_store_offsets(fi);
|
123
|
-
uchar buffer[MAX_WORD_SIZE];
|
124
|
-
TVTerm *term;
|
125
|
-
|
126
|
-
tv->term_cnt = num_terms;
|
127
|
-
tv->terms = ALLOC_AND_ZERO_N(TVTerm, num_terms);
|
128
|
-
|
129
|
-
for (i = 0; i < num_terms; i++) {
|
130
|
-
term = &(tv->terms[i]);
|
131
|
-
/* read delta encoded term */
|
132
|
-
delta_start = is_read_vint(tvd_in);
|
133
|
-
delta_len = is_read_vint(tvd_in);
|
134
|
-
total_len = delta_start + delta_len;
|
135
|
-
is_read_bytes(tvd_in, buffer + delta_start, delta_len);
|
136
|
-
buffer[total_len++] = '\0';
|
137
|
-
term->text = memcpy(ALLOC_N(char, total_len), buffer, total_len);
|
138
|
-
|
139
|
-
/* read freq */
|
140
|
-
freq = term->freq = is_read_vint(tvd_in);
|
141
|
-
|
142
|
-
/* read positions if necessary */
|
143
|
-
if (store_positions) {
|
144
|
-
int *positions = term->positions = ALLOC_N(int, freq);
|
145
|
-
int pos = 0;
|
146
|
-
for (j = 0; j < freq; j++) {
|
147
|
-
positions[j] = pos += is_read_vint(tvd_in);
|
148
|
-
}
|
149
|
-
}
|
150
|
-
|
151
|
-
/* read offsets if necessary */
|
152
|
-
}
|
153
|
-
if (store_offsets) {
|
154
|
-
int num_positions = tv->offset_cnt = is_read_vint(tvd_in);
|
155
|
-
Offset *offsets = tv->offsets = ALLOC_N(Offset, num_positions);
|
156
|
-
int offset = 0;
|
157
|
-
for (i = 0; i < num_positions; i++) {
|
158
|
-
offsets[i].start = offset += is_read_vint(tvd_in);
|
159
|
-
offsets[i].end = offset += is_read_vint(tvd_in);
|
160
|
-
}
|
161
|
-
}
|
162
|
-
}
|
163
|
-
return tv;
|
164
|
-
}
|
165
|
-
|
166
|
-
HashTable *tvr_get_tv(TermVectorsReader *tvr, int doc_num)
|
167
|
-
{
|
168
|
-
HashTable *term_vectors = h_new_str((free_ft)NULL, (free_ft)&tv_destroy);
|
169
|
-
int i;
|
170
|
-
InStream *tvx_in = tvr->tvx_in;
|
171
|
-
InStream *tvd_in = tvr->tvd_in;
|
172
|
-
off_t data_ptr, field_index_ptr;
|
173
|
-
int field_cnt;
|
174
|
-
int *field_nums;
|
175
|
-
|
176
|
-
if (doc_num >= 0 && doc_num < tvr->size) {
|
177
|
-
is_seek(tvx_in, 12 * doc_num);
|
178
|
-
|
179
|
-
data_ptr = (off_t)is_read_u64(tvx_in);
|
180
|
-
field_index_ptr = data_ptr + (off_t)is_read_u32(tvx_in);
|
181
|
-
|
182
|
-
/* scan fields to get position of field_num's term vector */
|
183
|
-
is_seek(tvd_in, field_index_ptr);
|
184
|
-
|
185
|
-
field_cnt = is_read_vint(tvd_in);
|
186
|
-
field_nums = ALLOC_N(int, field_cnt);
|
187
|
-
|
188
|
-
for (i = 0; i < field_cnt; i++) {
|
189
|
-
field_nums[i] = is_read_vint(tvd_in);
|
190
|
-
is_read_vint(tvd_in); /* skip space, we don't need it */
|
191
|
-
}
|
192
|
-
is_seek(tvd_in, data_ptr);
|
193
|
-
|
194
|
-
for (i = 0; i < field_cnt; i++) {
|
195
|
-
TermVector *tv = tvr_read_term_vector(tvr, field_nums[i]);
|
196
|
-
h_set(term_vectors, tv->field, tv);
|
197
|
-
}
|
198
|
-
free(field_nums);
|
199
|
-
}
|
200
|
-
return term_vectors;
|
201
|
-
}
|
202
|
-
|
203
|
-
TermVector *tvr_get_field_tv(TermVectorsReader *tvr,
|
204
|
-
int doc_num,
|
205
|
-
int field_num)
|
206
|
-
{
|
207
|
-
int i;
|
208
|
-
InStream *tvx_in = tvr->tvx_in;
|
209
|
-
InStream *tvd_in = tvr->tvd_in;
|
210
|
-
off_t data_ptr, field_index_ptr;
|
211
|
-
int field_cnt;
|
212
|
-
int offset = 0;
|
213
|
-
TermVector *tv = NULL;
|
214
|
-
|
215
|
-
if (doc_num >= 0 && doc_num < tvr->size) {
|
216
|
-
is_seek(tvx_in, 12 * doc_num);
|
217
|
-
|
218
|
-
data_ptr = (off_t)is_read_u64(tvx_in);
|
219
|
-
field_index_ptr = data_ptr + (off_t)is_read_u32(tvx_in);
|
220
|
-
|
221
|
-
/* scan fields to get position of field_num's term vector */
|
222
|
-
is_seek(tvd_in, field_index_ptr);
|
223
|
-
|
224
|
-
field_cnt = is_read_vint(tvd_in);
|
225
|
-
for (i = 0; i < field_cnt; i++) {
|
226
|
-
if ((int)is_read_vint(tvd_in) == field_num) {
|
227
|
-
break;
|
228
|
-
}
|
229
|
-
offset += is_read_vint(tvd_in); /* space taken by field */
|
230
|
-
}
|
231
|
-
if (i < field_cnt) {
|
232
|
-
/* field was found */
|
233
|
-
is_seek(tvd_in, data_ptr + offset);
|
234
|
-
tv = tvr_read_term_vector(tvr, field_num);
|
235
|
-
}
|
236
|
-
}
|
237
|
-
return tv;
|
238
|
-
}
|
239
|
-
|
240
|
-
/****************************************************************************
|
241
|
-
*
|
242
|
-
* TermVectorsWriter
|
243
|
-
*
|
244
|
-
****************************************************************************/
|
245
|
-
|
246
|
-
TermVectorsWriter *tvw_open(Store *store, const char *segment, FieldInfos *fis)
|
247
|
-
{
|
248
|
-
TermVectorsWriter *tvw = ALLOC(TermVectorsWriter);
|
249
|
-
char file_name[SEGMENT_NAME_MAX_LENGTH];
|
250
|
-
tvw->fis = fis;
|
251
|
-
tvw->fields = ary_new_type_capa(TVField, TV_FIELD_INIT_CAPA);
|
252
|
-
|
253
|
-
snprintf(file_name, SEGMENT_NAME_MAX_LENGTH, "%s.tvx", segment);
|
254
|
-
tvw->tvx_out = store->new_output(store, file_name);
|
255
|
-
|
256
|
-
snprintf(file_name, SEGMENT_NAME_MAX_LENGTH, "%s.tvd", segment);
|
257
|
-
tvw->tvd_out = store->new_output(store, file_name);
|
258
|
-
|
259
|
-
return tvw;
|
260
|
-
}
|
261
|
-
|
262
|
-
void tvw_close(TermVectorsWriter *tvw)
|
263
|
-
{
|
264
|
-
os_close(tvw->tvx_out);
|
265
|
-
os_close(tvw->tvd_out);
|
266
|
-
ary_free(tvw->fields);
|
267
|
-
free(tvw);
|
268
|
-
}
|
269
|
-
|
270
|
-
void tvw_open_doc(TermVectorsWriter *tvw)
|
271
|
-
{
|
272
|
-
ary_size(tvw->fields) = 0;
|
273
|
-
tvw->tvd_ptr = os_pos(tvw->tvd_out);
|
274
|
-
os_write_u64(tvw->tvx_out, tvw->tvd_ptr);
|
275
|
-
}
|
276
|
-
|
277
|
-
void tvw_close_doc(TermVectorsWriter *tvw)
|
278
|
-
{
|
279
|
-
int i;
|
280
|
-
OutStream *tvd_out = tvw->tvd_out;
|
281
|
-
os_write_u32(tvw->tvx_out, (f_u32)(os_pos(tvw->tvd_out) - tvw->tvd_ptr));
|
282
|
-
os_write_vint(tvd_out, ary_size(tvw->fields));
|
283
|
-
for (i = 0; i < ary_size(tvw->fields); i++) {
|
284
|
-
os_write_vint(tvd_out, tvw->fields[i].field_num);
|
285
|
-
os_write_vint(tvd_out, tvw->fields[i].size);
|
286
|
-
}
|
287
|
-
}
|
288
|
-
|
289
|
-
void tvw_add_postings(TermVectorsWriter *tvw,
|
290
|
-
int field_num,
|
291
|
-
PostingList **plists,
|
292
|
-
int posting_count,
|
293
|
-
Offset *offsets,
|
294
|
-
int offset_count)
|
295
|
-
{
|
296
|
-
int i, delta_start, delta_length;
|
297
|
-
const char *last_term = EMPTY_STRING;
|
298
|
-
off_t tvd_start_pos = os_pos(tvw->tvd_out);
|
299
|
-
OutStream *tvd_out = tvw->tvd_out;
|
300
|
-
PostingList *plist;
|
301
|
-
Posting *posting;
|
302
|
-
Occurence *occ;
|
303
|
-
FieldInfo *fi = tvw->fis->fields[field_num];
|
304
|
-
int store_positions = fi_store_positions(fi);
|
305
|
-
|
306
|
-
ary_grow(tvw->fields);
|
307
|
-
ary_last(tvw->fields).field_num = field_num;
|
308
|
-
|
309
|
-
os_write_vint(tvd_out, posting_count);
|
310
|
-
for (i = 0; i < posting_count; i++) {
|
311
|
-
plist = plists[i];
|
312
|
-
posting = plist->last;
|
313
|
-
delta_start = hlp_string_diff(last_term, plist->term);
|
314
|
-
delta_length = plist->term_len - delta_start;
|
315
|
-
|
316
|
-
os_write_vint(tvd_out, delta_start); /* write shared prefix length */
|
317
|
-
os_write_vint(tvd_out, delta_length); /* write delta length */
|
318
|
-
/* write delta chars */
|
319
|
-
os_write_bytes(tvd_out,
|
320
|
-
(uchar *)(plist->term + delta_start),
|
321
|
-
delta_length);
|
322
|
-
os_write_vint(tvd_out, posting->freq);
|
323
|
-
last_term = plist->term;
|
324
|
-
|
325
|
-
if (store_positions) {
|
326
|
-
/* use delta encoding for positions */
|
327
|
-
int last_pos = 0;
|
328
|
-
for (occ = posting->first_occ; occ; occ = occ->next) {
|
329
|
-
os_write_vint(tvd_out, occ->pos - last_pos);
|
330
|
-
last_pos = occ->pos;
|
331
|
-
}
|
332
|
-
}
|
333
|
-
|
334
|
-
}
|
335
|
-
|
336
|
-
if (fi_store_offsets(fi)) {
|
337
|
-
/* use delta encoding for offsets */
|
338
|
-
int last_end = 0;
|
339
|
-
os_write_vint(tvd_out, offset_count); /* write shared prefix length */
|
340
|
-
for (i = 0; i < offset_count; i++) {
|
341
|
-
int start = offsets[i].start;
|
342
|
-
int end = offsets[i].end;
|
343
|
-
os_write_vint(tvd_out, start - last_end);
|
344
|
-
os_write_vint(tvd_out, end - start);
|
345
|
-
last_end = end;
|
346
|
-
}
|
347
|
-
}
|
348
|
-
|
349
|
-
ary_last(tvw->fields).size = os_pos(tvd_out) - tvd_start_pos;
|
350
|
-
}
|
351
|
-
|
352
|
-
|
data/ext/threading.h
CHANGED
@@ -4,28 +4,28 @@
|
|
4
4
|
#include "hash.h"
|
5
5
|
#define UNTHREADED 1
|
6
6
|
|
7
|
-
typedef void *
|
8
|
-
typedef struct
|
9
|
-
typedef int
|
10
|
-
#define
|
11
|
-
#define
|
12
|
-
#define
|
13
|
-
#define
|
14
|
-
#define
|
15
|
-
#define
|
16
|
-
#define
|
17
|
-
#define
|
18
|
-
#define
|
19
|
-
#define
|
20
|
-
#define
|
21
|
-
#define
|
22
|
-
#define
|
23
|
-
#define
|
7
|
+
typedef void * frt_mutex_t;
|
8
|
+
typedef struct FrtHash *frt_thread_key_t;
|
9
|
+
typedef int frt_thread_once_t;
|
10
|
+
#define FRT_MUTEX_INITIALIZER NULL
|
11
|
+
#define FRT_MUTEX_RECURSIVE_INITIALIZER NULL
|
12
|
+
#define FRT_THREAD_ONCE_INIT 1;
|
13
|
+
#define frt_mutex_init(a, b)
|
14
|
+
#define frt_mutex_lock(a)
|
15
|
+
#define frt_mutex_trylock(a)
|
16
|
+
#define frt_mutex_unlock(a)
|
17
|
+
#define frt_mutex_destroy(a)
|
18
|
+
#define frt_thread_key_create(a, b) frb_thread_key_create(a, b)
|
19
|
+
#define frt_thread_key_delete(a) frb_thread_key_delete(a)
|
20
|
+
#define frt_thread_setspecific(a, b) frb_thread_setspecific(a, b)
|
21
|
+
#define frt_thread_getspecific(a) frb_thread_getspecific(a)
|
22
|
+
#define frt_thread_exit(a)
|
23
|
+
#define frt_thread_once(a, b) frb_thread_once(a, b)
|
24
24
|
|
25
|
-
void
|
26
|
-
void
|
27
|
-
void
|
28
|
-
void
|
29
|
-
void *
|
25
|
+
void frb_thread_once(int *once_control, void (*init_routine)(void));
|
26
|
+
void frb_thread_key_create(frt_thread_key_t *key, frt_free_ft destroy);
|
27
|
+
void frb_thread_key_delete(frt_thread_key_t key);
|
28
|
+
void frb_thread_setspecific(frt_thread_key_t key, const void *pointer);
|
29
|
+
void *frb_thread_getspecific(frt_thread_key_t key);
|
30
30
|
|
31
31
|
#endif
|
data/ext/win32.h
CHANGED
@@ -1,8 +1,11 @@
|
|
1
|
-
#include "global.h"
|
2
|
-
|
3
1
|
#ifndef FRT_WIN32_H
|
4
2
|
#define FRT_WIN32_H
|
5
3
|
|
4
|
+
#ifdef __cplusplus
|
5
|
+
extern "C" {
|
6
|
+
#endif
|
7
|
+
|
8
|
+
#include "global.h"
|
6
9
|
#include <io.h>
|
7
10
|
|
8
11
|
struct dirent
|
@@ -19,8 +22,8 @@ typedef struct DIR
|
|
19
22
|
|
20
23
|
DIR *opendir(const char *dirname)
|
21
24
|
{
|
22
|
-
DIR *d =
|
23
|
-
char dirname_buf[
|
25
|
+
DIR *d = FRT_ALLOC_AND_ZERO(DIR);
|
26
|
+
char dirname_buf[FRT_MAX_FILE_PATH];
|
24
27
|
long ff_res;
|
25
28
|
sprintf(dirname_buf, "%s\\*", dirname);
|
26
29
|
ff_res = _findfirst(dirname_buf, &d->find_data);
|
@@ -51,4 +54,9 @@ void closedir(DIR *d)
|
|
51
54
|
_findclose(d->handle);
|
52
55
|
free(d);
|
53
56
|
}
|
57
|
+
|
58
|
+
#ifdef __cplusplus
|
59
|
+
} // extern "C"
|
60
|
+
#endif
|
61
|
+
|
54
62
|
#endif
|