jk-ferret 0.11.8.2
Sign up to get free protection for your applications and to get access to all the features.
- data/CHANGELOG +24 -0
- data/MIT-LICENSE +20 -0
- data/README +90 -0
- data/RELEASE_CHANGES +137 -0
- data/RELEASE_NOTES +60 -0
- data/Rakefile +443 -0
- data/TODO +109 -0
- data/TUTORIAL +231 -0
- data/bin/ferret-browser +79 -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/STEMMER_api.c +66 -0
- data/ext/STEMMER_libstemmer.c +93 -0
- data/ext/STEMMER_stem_ISO_8859_1_danish.c +337 -0
- data/ext/STEMMER_stem_ISO_8859_1_dutch.c +624 -0
- data/ext/STEMMER_stem_ISO_8859_1_english.c +1117 -0
- data/ext/STEMMER_stem_ISO_8859_1_finnish.c +762 -0
- data/ext/STEMMER_stem_ISO_8859_1_french.c +1246 -0
- data/ext/STEMMER_stem_ISO_8859_1_german.c +503 -0
- 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/STEMMER_stem_ISO_8859_1_porter.c +749 -0
- data/ext/STEMMER_stem_ISO_8859_1_portuguese.c +1017 -0
- 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/STEMMER_stem_KOI8_R_russian.c +700 -0
- data/ext/STEMMER_stem_UTF_8_danish.c +339 -0
- data/ext/STEMMER_stem_UTF_8_dutch.c +634 -0
- data/ext/STEMMER_stem_UTF_8_english.c +1125 -0
- data/ext/STEMMER_stem_UTF_8_finnish.c +768 -0
- data/ext/STEMMER_stem_UTF_8_french.c +1256 -0
- data/ext/STEMMER_stem_UTF_8_german.c +509 -0
- 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/STEMMER_stem_UTF_8_porter.c +755 -0
- 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/STEMMER_utilities.c +478 -0
- data/ext/analysis.c +1710 -0
- data/ext/analysis.h +266 -0
- data/ext/api.h +26 -0
- data/ext/array.c +125 -0
- data/ext/array.h +62 -0
- data/ext/bitvector.c +96 -0
- data/ext/bitvector.h +594 -0
- data/ext/bzlib.h +282 -0
- data/ext/bzlib_private.h +503 -0
- data/ext/compound_io.c +384 -0
- data/ext/config.h +52 -0
- data/ext/document.c +159 -0
- data/ext/document.h +63 -0
- data/ext/except.c +102 -0
- data/ext/except.h +176 -0
- data/ext/extconf.rb +15 -0
- data/ext/ferret.c +416 -0
- data/ext/ferret.h +94 -0
- data/ext/field_index.c +262 -0
- data/ext/field_index.h +52 -0
- data/ext/filter.c +157 -0
- data/ext/fs_store.c +493 -0
- data/ext/global.c +458 -0
- data/ext/global.h +302 -0
- data/ext/hash.c +524 -0
- data/ext/hash.h +515 -0
- data/ext/hashset.c +192 -0
- data/ext/hashset.h +215 -0
- data/ext/header.h +58 -0
- data/ext/helper.c +63 -0
- data/ext/helper.h +21 -0
- data/ext/index.c +6804 -0
- data/ext/index.h +935 -0
- data/ext/internal.h +1019 -0
- data/ext/lang.c +10 -0
- data/ext/lang.h +68 -0
- data/ext/libstemmer.h +79 -0
- data/ext/mempool.c +88 -0
- data/ext/mempool.h +43 -0
- data/ext/modules.h +190 -0
- data/ext/multimapper.c +351 -0
- data/ext/multimapper.h +60 -0
- data/ext/posh.c +1006 -0
- data/ext/posh.h +973 -0
- data/ext/priorityqueue.c +149 -0
- data/ext/priorityqueue.h +155 -0
- data/ext/q_boolean.c +1621 -0
- data/ext/q_const_score.c +162 -0
- data/ext/q_filtered_query.c +212 -0
- data/ext/q_fuzzy.c +280 -0
- data/ext/q_match_all.c +149 -0
- data/ext/q_multi_term.c +673 -0
- data/ext/q_parser.c +3103 -0
- data/ext/q_phrase.c +1206 -0
- data/ext/q_prefix.c +98 -0
- data/ext/q_range.c +682 -0
- data/ext/q_span.c +2390 -0
- data/ext/q_term.c +337 -0
- data/ext/q_wildcard.c +167 -0
- data/ext/r_analysis.c +2626 -0
- data/ext/r_index.c +3468 -0
- data/ext/r_qparser.c +635 -0
- data/ext/r_search.c +4490 -0
- data/ext/r_store.c +513 -0
- data/ext/r_utils.c +1131 -0
- data/ext/ram_store.c +476 -0
- 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 +1864 -0
- data/ext/search.h +953 -0
- data/ext/similarity.c +151 -0
- data/ext/similarity.h +89 -0
- data/ext/sort.c +786 -0
- data/ext/stem_ISO_8859_1_danish.h +16 -0
- data/ext/stem_ISO_8859_1_dutch.h +16 -0
- data/ext/stem_ISO_8859_1_english.h +16 -0
- data/ext/stem_ISO_8859_1_finnish.h +16 -0
- data/ext/stem_ISO_8859_1_french.h +16 -0
- data/ext/stem_ISO_8859_1_german.h +16 -0
- data/ext/stem_ISO_8859_1_hungarian.h +16 -0
- data/ext/stem_ISO_8859_1_italian.h +16 -0
- data/ext/stem_ISO_8859_1_norwegian.h +16 -0
- data/ext/stem_ISO_8859_1_porter.h +16 -0
- data/ext/stem_ISO_8859_1_portuguese.h +16 -0
- data/ext/stem_ISO_8859_1_spanish.h +16 -0
- data/ext/stem_ISO_8859_1_swedish.h +16 -0
- data/ext/stem_ISO_8859_2_romanian.h +16 -0
- data/ext/stem_KOI8_R_russian.h +16 -0
- data/ext/stem_UTF_8_danish.h +16 -0
- data/ext/stem_UTF_8_dutch.h +16 -0
- data/ext/stem_UTF_8_english.h +16 -0
- data/ext/stem_UTF_8_finnish.h +16 -0
- data/ext/stem_UTF_8_french.h +16 -0
- data/ext/stem_UTF_8_german.h +16 -0
- data/ext/stem_UTF_8_hungarian.h +16 -0
- data/ext/stem_UTF_8_italian.h +16 -0
- data/ext/stem_UTF_8_norwegian.h +16 -0
- data/ext/stem_UTF_8_porter.h +16 -0
- data/ext/stem_UTF_8_portuguese.h +16 -0
- data/ext/stem_UTF_8_romanian.h +16 -0
- data/ext/stem_UTF_8_russian.h +16 -0
- data/ext/stem_UTF_8_spanish.h +16 -0
- data/ext/stem_UTF_8_swedish.h +16 -0
- data/ext/stem_UTF_8_turkish.h +16 -0
- data/ext/stopwords.c +410 -0
- data/ext/store.c +698 -0
- data/ext/store.h +799 -0
- data/ext/symbol.c +10 -0
- data/ext/symbol.h +23 -0
- data/ext/term_vectors.c +73 -0
- data/ext/threading.h +31 -0
- data/ext/win32.h +62 -0
- data/lib/ferret.rb +30 -0
- data/lib/ferret/browser.rb +246 -0
- data/lib/ferret/browser/s/global.js +192 -0
- data/lib/ferret/browser/s/style.css +148 -0
- data/lib/ferret/browser/views/document/list.rhtml +49 -0
- data/lib/ferret/browser/views/document/show.rhtml +27 -0
- data/lib/ferret/browser/views/error/index.rhtml +7 -0
- data/lib/ferret/browser/views/help/index.rhtml +8 -0
- data/lib/ferret/browser/views/home/index.rhtml +29 -0
- data/lib/ferret/browser/views/layout.rhtml +22 -0
- data/lib/ferret/browser/views/term-vector/index.rhtml +4 -0
- data/lib/ferret/browser/views/term/index.rhtml +199 -0
- data/lib/ferret/browser/views/term/termdocs.rhtml +1 -0
- data/lib/ferret/browser/webrick.rb +14 -0
- data/lib/ferret/document.rb +130 -0
- data/lib/ferret/field_infos.rb +44 -0
- data/lib/ferret/field_symbol.rb +87 -0
- data/lib/ferret/index.rb +973 -0
- data/lib/ferret/number_tools.rb +157 -0
- data/lib/ferret/version.rb +3 -0
- data/setup.rb +1555 -0
- data/test/long_running/largefile/tc_largefile.rb +46 -0
- data/test/test_all.rb +5 -0
- data/test/test_helper.rb +29 -0
- data/test/test_installed.rb +1 -0
- data/test/threading/number_to_spoken.rb +132 -0
- data/test/threading/thread_safety_index_test.rb +88 -0
- data/test/threading/thread_safety_read_write_test.rb +73 -0
- data/test/threading/thread_safety_test.rb +133 -0
- data/test/unit/analysis/tc_analyzer.rb +550 -0
- data/test/unit/analysis/tc_token_stream.rb +653 -0
- data/test/unit/index/tc_index.rb +867 -0
- data/test/unit/index/tc_index_reader.rb +699 -0
- data/test/unit/index/tc_index_writer.rb +447 -0
- data/test/unit/index/th_doc.rb +332 -0
- data/test/unit/query_parser/tc_query_parser.rb +238 -0
- data/test/unit/search/tc_filter.rb +156 -0
- data/test/unit/search/tc_fuzzy_query.rb +147 -0
- data/test/unit/search/tc_index_searcher.rb +67 -0
- data/test/unit/search/tc_multi_searcher.rb +128 -0
- data/test/unit/search/tc_multiple_search_requests.rb +58 -0
- data/test/unit/search/tc_search_and_sort.rb +179 -0
- data/test/unit/search/tc_sort.rb +49 -0
- data/test/unit/search/tc_sort_field.rb +27 -0
- data/test/unit/search/tc_spans.rb +190 -0
- data/test/unit/search/tm_searcher.rb +436 -0
- data/test/unit/store/tc_fs_store.rb +115 -0
- data/test/unit/store/tc_ram_store.rb +35 -0
- data/test/unit/store/tm_store.rb +34 -0
- data/test/unit/store/tm_store_lock.rb +68 -0
- data/test/unit/tc_document.rb +81 -0
- data/test/unit/tc_field_symbol.rb +26 -0
- data/test/unit/ts_analysis.rb +2 -0
- data/test/unit/ts_index.rb +2 -0
- data/test/unit/ts_largefile.rb +4 -0
- data/test/unit/ts_query_parser.rb +2 -0
- data/test/unit/ts_search.rb +2 -0
- data/test/unit/ts_store.rb +2 -0
- data/test/unit/ts_utils.rb +2 -0
- data/test/unit/utils/tc_bit_vector.rb +295 -0
- data/test/unit/utils/tc_number_tools.rb +117 -0
- data/test/unit/utils/tc_priority_queue.rb +106 -0
- data/test/utils/content_generator.rb +226 -0
- metadata +319 -0
data/ext/compound_io.c
ADDED
@@ -0,0 +1,384 @@
|
|
1
|
+
#include "index.h"
|
2
|
+
#include "array.h"
|
3
|
+
#include "internal.h"
|
4
|
+
|
5
|
+
extern void store_destroy(Store *store);
|
6
|
+
extern InStream *is_new();
|
7
|
+
extern Store *store_new();
|
8
|
+
|
9
|
+
/****************************************************************************
|
10
|
+
*
|
11
|
+
* CompoundStore
|
12
|
+
*
|
13
|
+
****************************************************************************/
|
14
|
+
|
15
|
+
typedef struct FileEntry {
|
16
|
+
off_t offset;
|
17
|
+
off_t length;
|
18
|
+
} FileEntry;
|
19
|
+
|
20
|
+
static void cmpd_touch(Store *store, const char *file_name)
|
21
|
+
{
|
22
|
+
store->dir.cmpd->store->touch(store->dir.cmpd->store, file_name);
|
23
|
+
}
|
24
|
+
|
25
|
+
static int cmpd_exists(Store *store, const char *file_name)
|
26
|
+
{
|
27
|
+
if (h_get(store->dir.cmpd->entries, file_name) != NULL) {
|
28
|
+
return true;
|
29
|
+
}
|
30
|
+
else {
|
31
|
+
return false;
|
32
|
+
}
|
33
|
+
}
|
34
|
+
|
35
|
+
/**
|
36
|
+
* @throws UNSUPPORTED_ERROR
|
37
|
+
*/
|
38
|
+
static int cmpd_remove(Store *store, const char *file_name)
|
39
|
+
{
|
40
|
+
(void)store;
|
41
|
+
(void)file_name;
|
42
|
+
RAISE(UNSUPPORTED_ERROR, "%s", UNSUPPORTED_ERROR_MSG);
|
43
|
+
return 0;
|
44
|
+
}
|
45
|
+
|
46
|
+
/**
|
47
|
+
* @throws UNSUPPORTED_ERROR
|
48
|
+
*/
|
49
|
+
static void cmpd_rename(Store *store, const char *from, const char *to)
|
50
|
+
{
|
51
|
+
(void)store;
|
52
|
+
(void)from;
|
53
|
+
(void)to;
|
54
|
+
RAISE(UNSUPPORTED_ERROR, "%s", UNSUPPORTED_ERROR_MSG);
|
55
|
+
}
|
56
|
+
|
57
|
+
static int cmpd_count(Store *store)
|
58
|
+
{
|
59
|
+
return store->dir.cmpd->entries->size;
|
60
|
+
}
|
61
|
+
|
62
|
+
static void cmpd_each(Store *store,
|
63
|
+
void (*func)(const char *fname, void *arg), void *arg)
|
64
|
+
{
|
65
|
+
Hash *ht = store->dir.cmpd->entries;
|
66
|
+
int i;
|
67
|
+
for (i = 0; i <= ht->mask; i++) {
|
68
|
+
char *fn = (char *)ht->table[i].key;
|
69
|
+
if (fn) {
|
70
|
+
func(fn, arg);
|
71
|
+
}
|
72
|
+
}
|
73
|
+
}
|
74
|
+
|
75
|
+
|
76
|
+
/**
|
77
|
+
* @throws UNSUPPORTED_ERROR
|
78
|
+
*/
|
79
|
+
static void cmpd_clear(Store *store)
|
80
|
+
{
|
81
|
+
(void)store;
|
82
|
+
RAISE(UNSUPPORTED_ERROR, "%s", UNSUPPORTED_ERROR_MSG);
|
83
|
+
}
|
84
|
+
|
85
|
+
static void cmpd_close_i(Store *store)
|
86
|
+
{
|
87
|
+
CompoundStore *cmpd = store->dir.cmpd;
|
88
|
+
if (cmpd->stream == NULL) {
|
89
|
+
RAISE(IO_ERROR, "Tried to close already closed compound store");
|
90
|
+
}
|
91
|
+
|
92
|
+
h_destroy(cmpd->entries);
|
93
|
+
|
94
|
+
is_close(cmpd->stream);
|
95
|
+
cmpd->stream = NULL;
|
96
|
+
free(store->dir.cmpd);
|
97
|
+
store_destroy(store);
|
98
|
+
}
|
99
|
+
|
100
|
+
static off_t cmpd_length(Store *store, const char *file_name)
|
101
|
+
{
|
102
|
+
FileEntry *fe = (FileEntry *)h_get(store->dir.cmpd->entries, file_name);
|
103
|
+
if (fe != NULL) {
|
104
|
+
return fe->length;
|
105
|
+
}
|
106
|
+
else {
|
107
|
+
return 0;
|
108
|
+
}
|
109
|
+
}
|
110
|
+
|
111
|
+
static void cmpdi_seek_i(InStream *is, off_t pos)
|
112
|
+
{
|
113
|
+
(void)is;
|
114
|
+
(void)pos;
|
115
|
+
}
|
116
|
+
|
117
|
+
static void cmpdi_close_i(InStream *is)
|
118
|
+
{
|
119
|
+
free(is->d.cis);
|
120
|
+
}
|
121
|
+
|
122
|
+
static off_t cmpdi_length_i(InStream *is)
|
123
|
+
{
|
124
|
+
return (is->d.cis->length);
|
125
|
+
}
|
126
|
+
|
127
|
+
/*
|
128
|
+
* raises: EOF_ERROR
|
129
|
+
*/
|
130
|
+
static void cmpdi_read_i(InStream *is, uchar *b, int len)
|
131
|
+
{
|
132
|
+
CompoundInStream *cis = is->d.cis;
|
133
|
+
off_t start = is_pos(is);
|
134
|
+
|
135
|
+
if ((start + len) > cis->length) {
|
136
|
+
RAISE(EOF_ERROR, "Tried to read past end of file. File length is "
|
137
|
+
"<%"OFF_T_PFX"d> and tried to read to <%"OFF_T_PFX"d>",
|
138
|
+
cis->length, start + len);
|
139
|
+
}
|
140
|
+
|
141
|
+
is_seek(cis->sub, cis->offset + start);
|
142
|
+
is_read_bytes(cis->sub, b, len);
|
143
|
+
}
|
144
|
+
|
145
|
+
static const struct InStreamMethods CMPD_IN_STREAM_METHODS = {
|
146
|
+
cmpdi_read_i,
|
147
|
+
cmpdi_seek_i,
|
148
|
+
cmpdi_length_i,
|
149
|
+
cmpdi_close_i
|
150
|
+
};
|
151
|
+
|
152
|
+
static InStream *cmpd_create_input(InStream *sub_is, off_t offset, off_t length)
|
153
|
+
{
|
154
|
+
InStream *is = is_new();
|
155
|
+
CompoundInStream *cis = ALLOC(CompoundInStream);
|
156
|
+
|
157
|
+
cis->sub = sub_is;
|
158
|
+
cis->offset = offset;
|
159
|
+
cis->length = length;
|
160
|
+
is->d.cis = cis;
|
161
|
+
is->m = &CMPD_IN_STREAM_METHODS;
|
162
|
+
|
163
|
+
return is;
|
164
|
+
}
|
165
|
+
|
166
|
+
static InStream *cmpd_open_input(Store *store, const char *file_name)
|
167
|
+
{
|
168
|
+
FileEntry *entry;
|
169
|
+
CompoundStore *cmpd = store->dir.cmpd;
|
170
|
+
InStream *is;
|
171
|
+
|
172
|
+
mutex_lock(&store->mutex);
|
173
|
+
if (cmpd->stream == NULL) {
|
174
|
+
mutex_unlock(&store->mutex);
|
175
|
+
RAISE(IO_ERROR, "Can't open compound file input stream. Parent "
|
176
|
+
"stream is closed.");
|
177
|
+
}
|
178
|
+
|
179
|
+
entry = (FileEntry *)h_get(cmpd->entries, file_name);
|
180
|
+
if (entry == NULL) {
|
181
|
+
mutex_unlock(&store->mutex);
|
182
|
+
RAISE(IO_ERROR, "File %s does not exist: ", file_name);
|
183
|
+
}
|
184
|
+
|
185
|
+
is = cmpd_create_input(cmpd->stream, entry->offset, entry->length);
|
186
|
+
mutex_unlock(&store->mutex);
|
187
|
+
|
188
|
+
return is;
|
189
|
+
}
|
190
|
+
|
191
|
+
static OutStream *cmpd_new_output(Store *store, const char *file_name)
|
192
|
+
{
|
193
|
+
(void)store;
|
194
|
+
(void)file_name;
|
195
|
+
RAISE(UNSUPPORTED_ERROR, "%s", UNSUPPORTED_ERROR_MSG);
|
196
|
+
return NULL;
|
197
|
+
}
|
198
|
+
|
199
|
+
static Lock *cmpd_open_lock_i(Store *store, const char *lock_name)
|
200
|
+
{
|
201
|
+
(void)store;
|
202
|
+
(void)lock_name;
|
203
|
+
RAISE(UNSUPPORTED_ERROR, "%s", UNSUPPORTED_ERROR_MSG);
|
204
|
+
return NULL;
|
205
|
+
}
|
206
|
+
|
207
|
+
static void cmpd_close_lock_i(Lock *lock)
|
208
|
+
{
|
209
|
+
(void)lock;
|
210
|
+
RAISE(UNSUPPORTED_ERROR, "%s", UNSUPPORTED_ERROR_MSG);
|
211
|
+
}
|
212
|
+
|
213
|
+
Store *open_cmpd_store(Store *store, const char *name)
|
214
|
+
{
|
215
|
+
int count, i;
|
216
|
+
off_t offset;
|
217
|
+
char *fname;
|
218
|
+
FileEntry *entry = NULL;
|
219
|
+
Store *new_store = NULL;
|
220
|
+
CompoundStore *volatile cmpd = NULL;
|
221
|
+
InStream *volatile is = NULL;
|
222
|
+
|
223
|
+
TRY
|
224
|
+
cmpd = ALLOC_AND_ZERO(CompoundStore);
|
225
|
+
|
226
|
+
cmpd->store = store;
|
227
|
+
cmpd->name = name;
|
228
|
+
cmpd->entries = h_new_str(&free, &free);
|
229
|
+
is = cmpd->stream = store->open_input(store, cmpd->name);
|
230
|
+
|
231
|
+
/* read the directory and init files */
|
232
|
+
count = is_read_vint(is);
|
233
|
+
entry = NULL;
|
234
|
+
for (i = 0; i < count; i++) {
|
235
|
+
offset = (off_t)is_read_i64(is);
|
236
|
+
fname = is_read_string(is);
|
237
|
+
|
238
|
+
if (entry != NULL) {
|
239
|
+
/* set length of the previous entry */
|
240
|
+
entry->length = offset - entry->offset;
|
241
|
+
}
|
242
|
+
|
243
|
+
entry = ALLOC(FileEntry);
|
244
|
+
entry->offset = offset;
|
245
|
+
h_set(cmpd->entries, fname, entry);
|
246
|
+
}
|
247
|
+
XCATCHALL
|
248
|
+
if (is) is_close(is);
|
249
|
+
if (cmpd->entries) h_destroy(cmpd->entries);
|
250
|
+
free(cmpd);
|
251
|
+
XENDTRY
|
252
|
+
|
253
|
+
/* set the length of the final entry */
|
254
|
+
if (entry != NULL) {
|
255
|
+
entry->length = is_length(is) - entry->offset;
|
256
|
+
}
|
257
|
+
|
258
|
+
new_store = store_new();
|
259
|
+
new_store->dir.cmpd = cmpd;
|
260
|
+
new_store->touch = &cmpd_touch;
|
261
|
+
new_store->exists = &cmpd_exists;
|
262
|
+
new_store->remove = &cmpd_remove;
|
263
|
+
new_store->rename = &cmpd_rename;
|
264
|
+
new_store->count = &cmpd_count;
|
265
|
+
new_store->clear = &cmpd_clear;
|
266
|
+
new_store->length = &cmpd_length;
|
267
|
+
new_store->each = &cmpd_each;
|
268
|
+
new_store->close_i = &cmpd_close_i;
|
269
|
+
new_store->new_output = &cmpd_new_output;
|
270
|
+
new_store->open_input = &cmpd_open_input;
|
271
|
+
new_store->open_lock_i = &cmpd_open_lock_i;
|
272
|
+
new_store->close_lock_i = &cmpd_close_lock_i;
|
273
|
+
|
274
|
+
return new_store;
|
275
|
+
}
|
276
|
+
|
277
|
+
/****************************************************************************
|
278
|
+
*
|
279
|
+
* CompoundWriter
|
280
|
+
*
|
281
|
+
****************************************************************************/
|
282
|
+
|
283
|
+
CompoundWriter *open_cw(Store *store, char *name)
|
284
|
+
{
|
285
|
+
CompoundWriter *cw = ALLOC(CompoundWriter);
|
286
|
+
cw->store = store;
|
287
|
+
cw->name = name;
|
288
|
+
cw->ids = hs_new_str(&free);
|
289
|
+
cw->file_entries = ary_new_type_capa(CWFileEntry, CW_INIT_CAPA);
|
290
|
+
return cw;
|
291
|
+
}
|
292
|
+
|
293
|
+
void cw_add_file(CompoundWriter *cw, char *id)
|
294
|
+
{
|
295
|
+
id = estrdup(id);
|
296
|
+
if (hs_add(cw->ids, id) != HASH_KEY_DOES_NOT_EXIST) {
|
297
|
+
RAISE(IO_ERROR, "Tried to add file \"%s\" which has already been "
|
298
|
+
"added to the compound store", id);
|
299
|
+
}
|
300
|
+
|
301
|
+
ary_grow(cw->file_entries);
|
302
|
+
ary_last(cw->file_entries).name = id;
|
303
|
+
}
|
304
|
+
|
305
|
+
static void cw_copy_file(CompoundWriter *cw, CWFileEntry *src, OutStream *os)
|
306
|
+
{
|
307
|
+
off_t start_ptr = os_pos(os);
|
308
|
+
off_t end_ptr;
|
309
|
+
off_t remainder, length, len;
|
310
|
+
uchar buffer[BUFFER_SIZE];
|
311
|
+
|
312
|
+
InStream *is = cw->store->open_input(cw->store, src->name);
|
313
|
+
|
314
|
+
remainder = length = is_length(is);
|
315
|
+
|
316
|
+
while (remainder > 0) {
|
317
|
+
len = MIN(remainder, BUFFER_SIZE);
|
318
|
+
is_read_bytes(is, buffer, len);
|
319
|
+
os_write_bytes(os, buffer, len);
|
320
|
+
remainder -= len;
|
321
|
+
}
|
322
|
+
|
323
|
+
/* Verify that remainder is 0 */
|
324
|
+
if (remainder != 0) {
|
325
|
+
RAISE(IO_ERROR, "There seems to be an error in the compound file "
|
326
|
+
"should have read to the end but there are <%"OFF_T_PFX"d> "
|
327
|
+
"bytes left", remainder);
|
328
|
+
}
|
329
|
+
|
330
|
+
/* Verify that the output length diff is equal to original file */
|
331
|
+
end_ptr = os_pos(os);
|
332
|
+
len = end_ptr - start_ptr;
|
333
|
+
if (len != length) {
|
334
|
+
RAISE(IO_ERROR, "Difference in compound file output file offsets "
|
335
|
+
"<%"OFF_T_PFX"d> does not match the original file lenght "
|
336
|
+
"<%"OFF_T_PFX"d>", len, length);
|
337
|
+
}
|
338
|
+
|
339
|
+
is_close(is);
|
340
|
+
}
|
341
|
+
|
342
|
+
void cw_close(CompoundWriter *cw)
|
343
|
+
{
|
344
|
+
OutStream *os = NULL;
|
345
|
+
int i;
|
346
|
+
|
347
|
+
if (cw->ids->size <= 0) {
|
348
|
+
RAISE(STATE_ERROR, "Tried to merge compound file with no entries");
|
349
|
+
}
|
350
|
+
|
351
|
+
os = cw->store->new_output(cw->store, cw->name);
|
352
|
+
|
353
|
+
os_write_vint(os, ary_size(cw->file_entries));
|
354
|
+
|
355
|
+
/* Write the directory with all offsets at 0.
|
356
|
+
* Remember the positions of directory entries so that we can adjust the
|
357
|
+
* offsets later */
|
358
|
+
for (i = 0; i < ary_size(cw->file_entries); i++) {
|
359
|
+
cw->file_entries[i].dir_offset = os_pos(os);
|
360
|
+
os_write_u64(os, 0); /* for now */
|
361
|
+
os_write_string(os, cw->file_entries[i].name);
|
362
|
+
}
|
363
|
+
|
364
|
+
/* Open the files and copy their data into the stream. Remember the
|
365
|
+
* locations of each file's data section. */
|
366
|
+
for (i = 0; i < ary_size(cw->file_entries); i++) {
|
367
|
+
cw->file_entries[i].data_offset = os_pos(os);
|
368
|
+
cw_copy_file(cw, &cw->file_entries[i], os);
|
369
|
+
}
|
370
|
+
|
371
|
+
/* Write the data offsets into the directory of the compound stream */
|
372
|
+
for (i = 0; i < ary_size(cw->file_entries); i++) {
|
373
|
+
os_seek(os, cw->file_entries[i].dir_offset);
|
374
|
+
os_write_u64(os, cw->file_entries[i].data_offset);
|
375
|
+
}
|
376
|
+
|
377
|
+
if (os) {
|
378
|
+
os_close(os);
|
379
|
+
}
|
380
|
+
|
381
|
+
hs_destroy(cw->ids);
|
382
|
+
ary_free(cw->file_entries);
|
383
|
+
free(cw);
|
384
|
+
}
|
data/ext/config.h
ADDED
@@ -0,0 +1,52 @@
|
|
1
|
+
#ifndef FRT_DEFINES_H
|
2
|
+
#define FRT_DEFINES_H
|
3
|
+
|
4
|
+
#ifdef __cplusplus
|
5
|
+
extern "C" {
|
6
|
+
#endif
|
7
|
+
|
8
|
+
#include <sys/types.h>
|
9
|
+
#include <limits.h>
|
10
|
+
#include "posh.h"
|
11
|
+
|
12
|
+
#ifndef false
|
13
|
+
#define false 0
|
14
|
+
#endif
|
15
|
+
#ifndef true
|
16
|
+
#define true 1
|
17
|
+
#endif
|
18
|
+
|
19
|
+
#ifndef __cplusplus
|
20
|
+
typedef unsigned int bool;
|
21
|
+
#endif
|
22
|
+
typedef unsigned char frt_uchar;
|
23
|
+
|
24
|
+
typedef posh_u16_t frt_u16;
|
25
|
+
typedef posh_i16_t frt_i16;
|
26
|
+
typedef posh_u32_t frt_u32;
|
27
|
+
typedef posh_i32_t frt_i32;
|
28
|
+
typedef posh_u64_t frt_u64;
|
29
|
+
typedef posh_i64_t frt_i64;
|
30
|
+
|
31
|
+
#if ( LONG_MAX == 2147483647 ) && defined(_FILE_OFFSET_BITS) && (_FILE_OFFSET_BITS == 64)
|
32
|
+
#define FRT_OFF_T_PFX "ll"
|
33
|
+
#else
|
34
|
+
#define FRT_OFF_T_PFX "l"
|
35
|
+
#endif
|
36
|
+
|
37
|
+
#if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) && !defined(__cplusplus)
|
38
|
+
#define FRT_IS_C99
|
39
|
+
#define FRT_HAS_ISO_VARARGS
|
40
|
+
#define FRT_HAS_VARARGS
|
41
|
+
#endif
|
42
|
+
|
43
|
+
#if defined(__GNUC__) && !defined(__STRICT_ANSI__) && !defined(__cplusplus)
|
44
|
+
#define FRT_HAS_GNUC_VARARGS
|
45
|
+
#define FRT_HAS_VARARGS
|
46
|
+
#endif
|
47
|
+
|
48
|
+
#ifdef __cplusplus
|
49
|
+
} // extern "C"
|
50
|
+
#endif
|
51
|
+
|
52
|
+
#endif
|
data/ext/document.c
ADDED
@@ -0,0 +1,159 @@
|
|
1
|
+
#include "document.h"
|
2
|
+
#include "symbol.h"
|
3
|
+
#include <string.h>
|
4
|
+
#include "internal.h"
|
5
|
+
|
6
|
+
/****************************************************************************
|
7
|
+
*
|
8
|
+
* DocField
|
9
|
+
*
|
10
|
+
****************************************************************************/
|
11
|
+
|
12
|
+
DocField *df_new(Symbol name)
|
13
|
+
{
|
14
|
+
DocField *df = ALLOC(DocField);
|
15
|
+
df->name = name;
|
16
|
+
df->size = 0;
|
17
|
+
df->capa = DF_INIT_CAPA;
|
18
|
+
df->data = ALLOC_N(char *, df->capa);
|
19
|
+
df->lengths = ALLOC_N(int, df->capa);
|
20
|
+
df->destroy_data = false;
|
21
|
+
df->boost = 1.0;
|
22
|
+
return df;
|
23
|
+
}
|
24
|
+
|
25
|
+
DocField *df_add_data_len(DocField *df, char *data, int len)
|
26
|
+
{
|
27
|
+
if (df->size >= df->capa) {
|
28
|
+
df->capa <<= 2;
|
29
|
+
REALLOC_N(df->data, char *, df->capa);
|
30
|
+
REALLOC_N(df->lengths, int, df->capa);
|
31
|
+
}
|
32
|
+
df->data[df->size] = data;
|
33
|
+
df->lengths[df->size] = len;
|
34
|
+
df->size++;
|
35
|
+
return df;
|
36
|
+
}
|
37
|
+
|
38
|
+
DocField *df_add_data(DocField *df, char *data)
|
39
|
+
{
|
40
|
+
return df_add_data_len(df, data, strlen(data));
|
41
|
+
}
|
42
|
+
|
43
|
+
void df_destroy(DocField *df)
|
44
|
+
{
|
45
|
+
if (df->destroy_data) {
|
46
|
+
int i;
|
47
|
+
for (i = 0; i < df->size; i++) {
|
48
|
+
free(df->data[i]);
|
49
|
+
}
|
50
|
+
}
|
51
|
+
free(df->data);
|
52
|
+
free(df->lengths);
|
53
|
+
free(df);
|
54
|
+
}
|
55
|
+
|
56
|
+
/*
|
57
|
+
* Format for one item is: name: "data"
|
58
|
+
* for more items : name: ["data", "data", "data"]
|
59
|
+
*/
|
60
|
+
char *df_to_s(DocField *df)
|
61
|
+
{
|
62
|
+
#define APPEND(dst, src) ((dst)[0] = (src)[0], 1)
|
63
|
+
#define APPEND2(dst, src) (APPEND(dst, src), APPEND(dst+1, src+1), 2)
|
64
|
+
|
65
|
+
int i, len = 0, namelen = sym_len(df->name);
|
66
|
+
char *str, *s;
|
67
|
+
for (i = 0; i < df->size; i++) {
|
68
|
+
len += df->lengths[i] + 4;
|
69
|
+
}
|
70
|
+
s = str = ALLOC_N(char, namelen + len + 5);
|
71
|
+
memcpy(s, df->name, namelen);
|
72
|
+
s += namelen;
|
73
|
+
s += APPEND2(s, ": ");
|
74
|
+
|
75
|
+
if (df->size > 1) {
|
76
|
+
s += APPEND(s, "[");
|
77
|
+
}
|
78
|
+
for (i = 0; i < df->size; i++) {
|
79
|
+
if (i != 0) {
|
80
|
+
s += APPEND2(s, ", ");
|
81
|
+
}
|
82
|
+
s += APPEND(s, "\"");
|
83
|
+
memcpy(s, df->data[i], df->lengths[i]);
|
84
|
+
s += df->lengths[i];
|
85
|
+
s += APPEND(s, "\"");
|
86
|
+
}
|
87
|
+
|
88
|
+
if (df->size > 1) {
|
89
|
+
s += APPEND(s, "]");
|
90
|
+
}
|
91
|
+
*s = 0;
|
92
|
+
return str;
|
93
|
+
}
|
94
|
+
|
95
|
+
/****************************************************************************
|
96
|
+
*
|
97
|
+
* Document
|
98
|
+
*
|
99
|
+
****************************************************************************/
|
100
|
+
|
101
|
+
Document *doc_new()
|
102
|
+
{
|
103
|
+
Document *doc = ALLOC(Document);
|
104
|
+
doc->field_dict = h_new_ptr((free_ft)&df_destroy);
|
105
|
+
doc->size = 0;
|
106
|
+
doc->capa = DOC_INIT_CAPA;
|
107
|
+
doc->fields = ALLOC_N(DocField *, doc->capa);
|
108
|
+
doc->boost = 1.0;
|
109
|
+
return doc;
|
110
|
+
}
|
111
|
+
|
112
|
+
DocField *doc_add_field(Document *doc, DocField *df)
|
113
|
+
{
|
114
|
+
if (!h_set_safe(doc->field_dict, df->name, df)) {
|
115
|
+
RAISE(EXCEPTION, "tried to add %s field which alread existed\n",
|
116
|
+
S(df->name));
|
117
|
+
}
|
118
|
+
if (doc->size >= doc->capa) {
|
119
|
+
doc->capa <<= 1;
|
120
|
+
REALLOC_N(doc->fields, DocField *, doc->capa);
|
121
|
+
}
|
122
|
+
doc->fields[doc->size] = df;
|
123
|
+
doc->size++;
|
124
|
+
return df;
|
125
|
+
}
|
126
|
+
|
127
|
+
DocField *doc_get_field(Document *doc, Symbol name)
|
128
|
+
{
|
129
|
+
return (DocField *)h_get(doc->field_dict, name);
|
130
|
+
}
|
131
|
+
|
132
|
+
char *doc_to_s(Document *doc)
|
133
|
+
{
|
134
|
+
int i;
|
135
|
+
int len = 0;
|
136
|
+
char **fields = ALLOC_N(char *, doc->size);
|
137
|
+
char *buf, *s;
|
138
|
+
|
139
|
+
for (i = 0; i < doc->size; i++) {
|
140
|
+
fields[i] = df_to_s(doc->fields[i]);
|
141
|
+
len += strlen(fields[i]) + 5;
|
142
|
+
}
|
143
|
+
s = buf = ALLOC_N(char, len + 12);
|
144
|
+
s += sprintf(buf, "Document [\n");
|
145
|
+
for (i = 0; i < doc->size; i++) {
|
146
|
+
s += sprintf(s, " =>%s\n", fields[i]);
|
147
|
+
free(fields[i]);
|
148
|
+
}
|
149
|
+
free(fields);
|
150
|
+
return buf;
|
151
|
+
}
|
152
|
+
|
153
|
+
void doc_destroy(Document *doc)
|
154
|
+
{
|
155
|
+
h_destroy(doc->field_dict);
|
156
|
+
free(doc->fields);
|
157
|
+
free(doc);
|
158
|
+
}
|
159
|
+
|