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
@@ -0,0 +1,478 @@
|
|
1
|
+
|
2
|
+
#include <stdio.h>
|
3
|
+
#include <stdlib.h>
|
4
|
+
#include <string.h>
|
5
|
+
|
6
|
+
#include "header.h"
|
7
|
+
|
8
|
+
#define unless(C) if(!(C))
|
9
|
+
|
10
|
+
#define CREATE_SIZE 1
|
11
|
+
|
12
|
+
extern symbol * create_s(void) {
|
13
|
+
symbol * p;
|
14
|
+
void * mem = malloc(HEAD + (CREATE_SIZE + 1) * sizeof(symbol));
|
15
|
+
if (mem == NULL) return NULL;
|
16
|
+
p = (symbol *) (HEAD + (char *) mem);
|
17
|
+
CAPACITY(p) = CREATE_SIZE;
|
18
|
+
SET_SIZE(p, CREATE_SIZE);
|
19
|
+
return p;
|
20
|
+
}
|
21
|
+
|
22
|
+
extern void lose_s(symbol * p) {
|
23
|
+
if (p == NULL) return;
|
24
|
+
free((char *) p - HEAD);
|
25
|
+
}
|
26
|
+
|
27
|
+
/*
|
28
|
+
new_p = skip_utf8(p, c, lb, l, n); skips n characters forwards from p + c
|
29
|
+
if n +ve, or n characters backwards from p + c - 1 if n -ve. new_p is the new
|
30
|
+
position, or 0 on failure.
|
31
|
+
|
32
|
+
-- used to implement hop and next in the utf8 case.
|
33
|
+
*/
|
34
|
+
|
35
|
+
extern int skip_utf8(const symbol * p, int c, int lb, int l, int n) {
|
36
|
+
int b;
|
37
|
+
if (n >= 0) {
|
38
|
+
for (; n > 0; n--) {
|
39
|
+
if (c >= l) return -1;
|
40
|
+
b = p[c++];
|
41
|
+
if (b >= 0xC0) { /* 1100 0000 */
|
42
|
+
while (c < l) {
|
43
|
+
b = p[c];
|
44
|
+
if (b >= 0xC0 || b < 0x80) break;
|
45
|
+
/* break unless b is 10------ */
|
46
|
+
c++;
|
47
|
+
}
|
48
|
+
}
|
49
|
+
}
|
50
|
+
} else {
|
51
|
+
for (; n < 0; n++) {
|
52
|
+
if (c <= lb) return -1;
|
53
|
+
b = p[--c];
|
54
|
+
if (b >= 0x80) { /* 1000 0000 */
|
55
|
+
while (c > lb) {
|
56
|
+
b = p[c];
|
57
|
+
if (b >= 0xC0) break; /* 1100 0000 */
|
58
|
+
c--;
|
59
|
+
}
|
60
|
+
}
|
61
|
+
}
|
62
|
+
}
|
63
|
+
return c;
|
64
|
+
}
|
65
|
+
|
66
|
+
/* Code for character groupings: utf8 cases */
|
67
|
+
|
68
|
+
static int get_utf8(const symbol * p, int c, int l, int * slot) {
|
69
|
+
int b0, b1;
|
70
|
+
if (c >= l) return 0;
|
71
|
+
b0 = p[c++];
|
72
|
+
if (b0 < 0xC0 || c == l) { /* 1100 0000 */
|
73
|
+
* slot = b0; return 1;
|
74
|
+
}
|
75
|
+
b1 = p[c++];
|
76
|
+
if (b0 < 0xE0 || c == l) { /* 1110 0000 */
|
77
|
+
* slot = (b0 & 0x1F) << 6 | (b1 & 0x3F); return 2;
|
78
|
+
}
|
79
|
+
* slot = (b0 & 0xF) << 12 | (b1 & 0x3F) << 6 | (p[c] & 0x3F); return 3;
|
80
|
+
}
|
81
|
+
|
82
|
+
static int get_b_utf8(const symbol * p, int c, int lb, int * slot) {
|
83
|
+
int b0, b1;
|
84
|
+
if (c <= lb) return 0;
|
85
|
+
b0 = p[--c];
|
86
|
+
if (b0 < 0x80 || c == lb) { /* 1000 0000 */
|
87
|
+
* slot = b0; return 1;
|
88
|
+
}
|
89
|
+
b1 = p[--c];
|
90
|
+
if (b1 >= 0xC0 || c == lb) { /* 1100 0000 */
|
91
|
+
* slot = (b1 & 0x1F) << 6 | (b0 & 0x3F); return 2;
|
92
|
+
}
|
93
|
+
* slot = (p[c] & 0xF) << 12 | (b1 & 0x3F) << 6 | (b0 & 0x3F); return 3;
|
94
|
+
}
|
95
|
+
|
96
|
+
extern int in_grouping_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
|
97
|
+
do {
|
98
|
+
int ch;
|
99
|
+
int w = get_utf8(z->p, z->c, z->l, & ch);
|
100
|
+
unless (w) return -1;
|
101
|
+
if (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
|
102
|
+
return w;
|
103
|
+
z->c += w;
|
104
|
+
} while (repeat);
|
105
|
+
return 0;
|
106
|
+
}
|
107
|
+
|
108
|
+
extern int in_grouping_b_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
|
109
|
+
do {
|
110
|
+
int ch;
|
111
|
+
int w = get_b_utf8(z->p, z->c, z->lb, & ch);
|
112
|
+
unless (w) return -1;
|
113
|
+
if (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
|
114
|
+
return w;
|
115
|
+
z->c -= w;
|
116
|
+
} while (repeat);
|
117
|
+
return 0;
|
118
|
+
}
|
119
|
+
|
120
|
+
extern int out_grouping_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
|
121
|
+
do {
|
122
|
+
int ch;
|
123
|
+
int w = get_utf8(z->p, z->c, z->l, & ch);
|
124
|
+
unless (w) return -1;
|
125
|
+
unless (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
|
126
|
+
return w;
|
127
|
+
z->c += w;
|
128
|
+
} while (repeat);
|
129
|
+
return 0;
|
130
|
+
}
|
131
|
+
|
132
|
+
extern int out_grouping_b_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
|
133
|
+
do {
|
134
|
+
int ch;
|
135
|
+
int w = get_b_utf8(z->p, z->c, z->lb, & ch);
|
136
|
+
unless (w) return -1;
|
137
|
+
unless (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
|
138
|
+
return w;
|
139
|
+
z->c -= w;
|
140
|
+
} while (repeat);
|
141
|
+
return 0;
|
142
|
+
}
|
143
|
+
|
144
|
+
/* Code for character groupings: non-utf8 cases */
|
145
|
+
|
146
|
+
extern int in_grouping(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
|
147
|
+
do {
|
148
|
+
int ch;
|
149
|
+
if (z->c >= z->l) return -1;
|
150
|
+
ch = z->p[z->c];
|
151
|
+
if (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
|
152
|
+
return 1;
|
153
|
+
z->c++;
|
154
|
+
} while (repeat);
|
155
|
+
return 0;
|
156
|
+
}
|
157
|
+
|
158
|
+
extern int in_grouping_b(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
|
159
|
+
do {
|
160
|
+
int ch;
|
161
|
+
if (z->c <= z->lb) return -1;
|
162
|
+
ch = z->p[z->c - 1];
|
163
|
+
if (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
|
164
|
+
return 1;
|
165
|
+
z->c--;
|
166
|
+
} while (repeat);
|
167
|
+
return 0;
|
168
|
+
}
|
169
|
+
|
170
|
+
extern int out_grouping(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
|
171
|
+
do {
|
172
|
+
int ch;
|
173
|
+
if (z->c >= z->l) return -1;
|
174
|
+
ch = z->p[z->c];
|
175
|
+
unless (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
|
176
|
+
return 1;
|
177
|
+
z->c++;
|
178
|
+
} while (repeat);
|
179
|
+
return 0;
|
180
|
+
}
|
181
|
+
|
182
|
+
extern int out_grouping_b(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
|
183
|
+
do {
|
184
|
+
int ch;
|
185
|
+
if (z->c <= z->lb) return -1;
|
186
|
+
ch = z->p[z->c - 1];
|
187
|
+
unless (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
|
188
|
+
return 1;
|
189
|
+
z->c--;
|
190
|
+
} while (repeat);
|
191
|
+
return 0;
|
192
|
+
}
|
193
|
+
|
194
|
+
extern int eq_s(struct SN_env * z, int s_size, const symbol * s) {
|
195
|
+
if (z->l - z->c < s_size || memcmp(z->p + z->c, s, s_size * sizeof(symbol)) != 0) return 0;
|
196
|
+
z->c += s_size; return 1;
|
197
|
+
}
|
198
|
+
|
199
|
+
extern int eq_s_b(struct SN_env * z, int s_size, const symbol * s) {
|
200
|
+
if (z->c - z->lb < s_size || memcmp(z->p + z->c - s_size, s, s_size * sizeof(symbol)) != 0) return 0;
|
201
|
+
z->c -= s_size; return 1;
|
202
|
+
}
|
203
|
+
|
204
|
+
extern int eq_v(struct SN_env * z, const symbol * p) {
|
205
|
+
return eq_s(z, SIZE(p), p);
|
206
|
+
}
|
207
|
+
|
208
|
+
extern int eq_v_b(struct SN_env * z, const symbol * p) {
|
209
|
+
return eq_s_b(z, SIZE(p), p);
|
210
|
+
}
|
211
|
+
|
212
|
+
extern int find_among(struct SN_env * z, const struct among * v, int v_size) {
|
213
|
+
|
214
|
+
int i = 0;
|
215
|
+
int j = v_size;
|
216
|
+
|
217
|
+
int c = z->c; int l = z->l;
|
218
|
+
symbol * q = z->p + c;
|
219
|
+
|
220
|
+
const struct among * w;
|
221
|
+
|
222
|
+
int common_i = 0;
|
223
|
+
int common_j = 0;
|
224
|
+
|
225
|
+
int first_key_inspected = 0;
|
226
|
+
|
227
|
+
while(1) {
|
228
|
+
int k = i + ((j - i) >> 1);
|
229
|
+
int diff = 0;
|
230
|
+
int common = common_i < common_j ? common_i : common_j; /* smaller */
|
231
|
+
w = v + k;
|
232
|
+
{
|
233
|
+
int i2; for (i2 = common; i2 < w->s_size; i2++) {
|
234
|
+
if (c + common == l) { diff = -1; break; }
|
235
|
+
diff = q[common] - w->s[i2];
|
236
|
+
if (diff != 0) break;
|
237
|
+
common++;
|
238
|
+
}
|
239
|
+
}
|
240
|
+
if (diff < 0) { j = k; common_j = common; }
|
241
|
+
else { i = k; common_i = common; }
|
242
|
+
if (j - i <= 1) {
|
243
|
+
if (i > 0) break; /* v->s has been inspected */
|
244
|
+
if (j == i) break; /* only one item in v */
|
245
|
+
|
246
|
+
/* - but now we need to go round once more to get
|
247
|
+
v->s inspected. This looks messy, but is actually
|
248
|
+
the optimal approach. */
|
249
|
+
|
250
|
+
if (first_key_inspected) break;
|
251
|
+
first_key_inspected = 1;
|
252
|
+
}
|
253
|
+
}
|
254
|
+
while(1) {
|
255
|
+
w = v + i;
|
256
|
+
if (common_i >= w->s_size) {
|
257
|
+
z->c = c + w->s_size;
|
258
|
+
if (w->function == 0) return w->result;
|
259
|
+
{
|
260
|
+
int res = w->function(z);
|
261
|
+
z->c = c + w->s_size;
|
262
|
+
if (res) return w->result;
|
263
|
+
}
|
264
|
+
}
|
265
|
+
i = w->substring_i;
|
266
|
+
if (i < 0) return 0;
|
267
|
+
}
|
268
|
+
}
|
269
|
+
|
270
|
+
/* find_among_b is for backwards processing. Same comments apply */
|
271
|
+
|
272
|
+
extern int find_among_b(struct SN_env * z, const struct among * v, int v_size) {
|
273
|
+
|
274
|
+
int i = 0;
|
275
|
+
int j = v_size;
|
276
|
+
|
277
|
+
int c = z->c; int lb = z->lb;
|
278
|
+
symbol * q = z->p + c - 1;
|
279
|
+
|
280
|
+
const struct among * w;
|
281
|
+
|
282
|
+
int common_i = 0;
|
283
|
+
int common_j = 0;
|
284
|
+
|
285
|
+
int first_key_inspected = 0;
|
286
|
+
|
287
|
+
while(1) {
|
288
|
+
int k = i + ((j - i) >> 1);
|
289
|
+
int diff = 0;
|
290
|
+
int common = common_i < common_j ? common_i : common_j;
|
291
|
+
w = v + k;
|
292
|
+
{
|
293
|
+
int i2; for (i2 = w->s_size - 1 - common; i2 >= 0; i2--) {
|
294
|
+
if (c - common == lb) { diff = -1; break; }
|
295
|
+
diff = q[- common] - w->s[i2];
|
296
|
+
if (diff != 0) break;
|
297
|
+
common++;
|
298
|
+
}
|
299
|
+
}
|
300
|
+
if (diff < 0) { j = k; common_j = common; }
|
301
|
+
else { i = k; common_i = common; }
|
302
|
+
if (j - i <= 1) {
|
303
|
+
if (i > 0) break;
|
304
|
+
if (j == i) break;
|
305
|
+
if (first_key_inspected) break;
|
306
|
+
first_key_inspected = 1;
|
307
|
+
}
|
308
|
+
}
|
309
|
+
while(1) {
|
310
|
+
w = v + i;
|
311
|
+
if (common_i >= w->s_size) {
|
312
|
+
z->c = c - w->s_size;
|
313
|
+
if (w->function == 0) return w->result;
|
314
|
+
{
|
315
|
+
int res = w->function(z);
|
316
|
+
z->c = c - w->s_size;
|
317
|
+
if (res) return w->result;
|
318
|
+
}
|
319
|
+
}
|
320
|
+
i = w->substring_i;
|
321
|
+
if (i < 0) return 0;
|
322
|
+
}
|
323
|
+
}
|
324
|
+
|
325
|
+
|
326
|
+
/* Increase the size of the buffer pointed to by p to at least n symbols.
|
327
|
+
* If insufficient memory, returns NULL and frees the old buffer.
|
328
|
+
*/
|
329
|
+
static symbol * increase_size(symbol * p, int n) {
|
330
|
+
symbol * q;
|
331
|
+
int new_size = n + 20;
|
332
|
+
void * mem = realloc((char *) p - HEAD,
|
333
|
+
HEAD + (new_size + 1) * sizeof(symbol));
|
334
|
+
if (mem == NULL) {
|
335
|
+
lose_s(p);
|
336
|
+
return NULL;
|
337
|
+
}
|
338
|
+
q = (symbol *) (HEAD + (char *)mem);
|
339
|
+
CAPACITY(q) = new_size;
|
340
|
+
return q;
|
341
|
+
}
|
342
|
+
|
343
|
+
/* to replace symbols between c_bra and c_ket in z->p by the
|
344
|
+
s_size symbols at s.
|
345
|
+
Returns 0 on success, -1 on error.
|
346
|
+
Also, frees z->p (and sets it to NULL) on error.
|
347
|
+
*/
|
348
|
+
extern int replace_s(struct SN_env * z, int c_bra, int c_ket, int s_size, const symbol * s, int * adjptr)
|
349
|
+
{
|
350
|
+
int adjustment;
|
351
|
+
int len;
|
352
|
+
if (z->p == NULL) {
|
353
|
+
z->p = create_s();
|
354
|
+
if (z->p == NULL) return -1;
|
355
|
+
}
|
356
|
+
adjustment = s_size - (c_ket - c_bra);
|
357
|
+
len = SIZE(z->p);
|
358
|
+
if (adjustment != 0) {
|
359
|
+
if (adjustment + len > CAPACITY(z->p)) {
|
360
|
+
z->p = increase_size(z->p, adjustment + len);
|
361
|
+
if (z->p == NULL) return -1;
|
362
|
+
}
|
363
|
+
memmove(z->p + c_ket + adjustment,
|
364
|
+
z->p + c_ket,
|
365
|
+
(len - c_ket) * sizeof(symbol));
|
366
|
+
SET_SIZE(z->p, adjustment + len);
|
367
|
+
z->l += adjustment;
|
368
|
+
if (z->c >= c_ket)
|
369
|
+
z->c += adjustment;
|
370
|
+
else
|
371
|
+
if (z->c > c_bra)
|
372
|
+
z->c = c_bra;
|
373
|
+
}
|
374
|
+
unless (s_size == 0) memmove(z->p + c_bra, s, s_size * sizeof(symbol));
|
375
|
+
if (adjptr != NULL)
|
376
|
+
*adjptr = adjustment;
|
377
|
+
return 0;
|
378
|
+
}
|
379
|
+
|
380
|
+
static int slice_check(struct SN_env * z) {
|
381
|
+
|
382
|
+
if (z->bra < 0 ||
|
383
|
+
z->bra > z->ket ||
|
384
|
+
z->ket > z->l ||
|
385
|
+
z->p == NULL ||
|
386
|
+
z->l > SIZE(z->p)) /* this line could be removed */
|
387
|
+
{
|
388
|
+
#if 0
|
389
|
+
fprintf(stderr, "faulty slice operation:\n");
|
390
|
+
debug(z, -1, 0);
|
391
|
+
#endif
|
392
|
+
return -1;
|
393
|
+
}
|
394
|
+
return 0;
|
395
|
+
}
|
396
|
+
|
397
|
+
extern int slice_from_s(struct SN_env * z, int s_size, const symbol * s) {
|
398
|
+
if (slice_check(z)) return -1;
|
399
|
+
return replace_s(z, z->bra, z->ket, s_size, s, NULL);
|
400
|
+
}
|
401
|
+
|
402
|
+
extern int slice_from_v(struct SN_env * z, const symbol * p) {
|
403
|
+
return slice_from_s(z, SIZE(p), p);
|
404
|
+
}
|
405
|
+
|
406
|
+
extern int slice_del(struct SN_env * z) {
|
407
|
+
return slice_from_s(z, 0, 0);
|
408
|
+
}
|
409
|
+
|
410
|
+
extern int insert_s(struct SN_env * z, int bra, int ket, int s_size, const symbol * s) {
|
411
|
+
int adjustment;
|
412
|
+
if (replace_s(z, bra, ket, s_size, s, &adjustment))
|
413
|
+
return -1;
|
414
|
+
if (bra <= z->bra) z->bra += adjustment;
|
415
|
+
if (bra <= z->ket) z->ket += adjustment;
|
416
|
+
return 0;
|
417
|
+
}
|
418
|
+
|
419
|
+
extern int insert_v(struct SN_env * z, int bra, int ket, const symbol * p) {
|
420
|
+
int adjustment;
|
421
|
+
if (replace_s(z, bra, ket, SIZE(p), p, &adjustment))
|
422
|
+
return -1;
|
423
|
+
if (bra <= z->bra) z->bra += adjustment;
|
424
|
+
if (bra <= z->ket) z->ket += adjustment;
|
425
|
+
return 0;
|
426
|
+
}
|
427
|
+
|
428
|
+
extern symbol * slice_to(struct SN_env * z, symbol * p) {
|
429
|
+
if (slice_check(z)) {
|
430
|
+
lose_s(p);
|
431
|
+
return NULL;
|
432
|
+
}
|
433
|
+
{
|
434
|
+
int len = z->ket - z->bra;
|
435
|
+
if (CAPACITY(p) < len) {
|
436
|
+
p = increase_size(p, len);
|
437
|
+
if (p == NULL)
|
438
|
+
return NULL;
|
439
|
+
}
|
440
|
+
memmove(p, z->p + z->bra, len * sizeof(symbol));
|
441
|
+
SET_SIZE(p, len);
|
442
|
+
}
|
443
|
+
return p;
|
444
|
+
}
|
445
|
+
|
446
|
+
extern symbol * assign_to(struct SN_env * z, symbol * p) {
|
447
|
+
int len = z->l;
|
448
|
+
if (CAPACITY(p) < len) {
|
449
|
+
p = increase_size(p, len);
|
450
|
+
if (p == NULL)
|
451
|
+
return NULL;
|
452
|
+
}
|
453
|
+
memmove(p, z->p, len * sizeof(symbol));
|
454
|
+
SET_SIZE(p, len);
|
455
|
+
return p;
|
456
|
+
}
|
457
|
+
|
458
|
+
#if 0
|
459
|
+
extern void debug(struct SN_env * z, int number, int line_count) {
|
460
|
+
int i;
|
461
|
+
int limit = SIZE(z->p);
|
462
|
+
/*if (number >= 0) printf("%3d (line %4d): '", number, line_count);*/
|
463
|
+
if (number >= 0) printf("%3d (line %4d): [%d]'", number, line_count,limit);
|
464
|
+
for (i = 0; i <= limit; i++) {
|
465
|
+
if (z->lb == i) printf("{");
|
466
|
+
if (z->bra == i) printf("[");
|
467
|
+
if (z->c == i) printf("|");
|
468
|
+
if (z->ket == i) printf("]");
|
469
|
+
if (z->l == i) printf("}");
|
470
|
+
if (i < limit)
|
471
|
+
{ int ch = z->p[i];
|
472
|
+
if (ch == 0) ch = '#';
|
473
|
+
printf("%c", ch);
|
474
|
+
}
|
475
|
+
}
|
476
|
+
printf("'\n");
|
477
|
+
}
|
478
|
+
#endif
|