line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
/* Extended regular expression matching and search library. |
2
|
|
|
|
|
|
|
Copyright (C) 2002-2014 Free Software Foundation, Inc. |
3
|
|
|
|
|
|
|
This file is part of the GNU C Library. |
4
|
|
|
|
|
|
|
Contributed by Isamu Hasegawa . |
5
|
|
|
|
|
|
|
|
6
|
|
|
|
|
|
|
The GNU C Library is free software; you can redistribute it and/or |
7
|
|
|
|
|
|
|
modify it under the terms of the GNU Lesser General Public |
8
|
|
|
|
|
|
|
License as published by the Free Software Foundation; either |
9
|
|
|
|
|
|
|
version 2.1 of the License, or (at your option) any later version. |
10
|
|
|
|
|
|
|
|
11
|
|
|
|
|
|
|
The GNU C Library is distributed in the hope that it will be useful, |
12
|
|
|
|
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of |
13
|
|
|
|
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
14
|
|
|
|
|
|
|
Lesser General Public License for more details. |
15
|
|
|
|
|
|
|
|
16
|
|
|
|
|
|
|
You should have received a copy of the GNU Lesser General Public |
17
|
|
|
|
|
|
|
License along with the GNU C Library; if not, see |
18
|
|
|
|
|
|
|
. */ |
19
|
|
|
|
|
|
|
|
20
|
|
|
|
|
|
|
static void re_string_construct_common (pTHX_ const char *str, Idx len, |
21
|
|
|
|
|
|
|
re_string_t *pstr, |
22
|
|
|
|
|
|
|
RE_TRANSLATE_TYPE trans, bool icase, |
23
|
|
|
|
|
|
|
const re_dfa_t *dfa) internal_function; |
24
|
|
|
|
|
|
|
static re_dfastate_t *create_ci_newstate (pTHX_ const re_dfa_t *dfa, |
25
|
|
|
|
|
|
|
const re_node_set *nodes, |
26
|
|
|
|
|
|
|
re_hashval_t hash) internal_function; |
27
|
|
|
|
|
|
|
static re_dfastate_t *create_cd_newstate (pTHX_ const re_dfa_t *dfa, |
28
|
|
|
|
|
|
|
const re_node_set *nodes, |
29
|
|
|
|
|
|
|
unsigned int context, |
30
|
|
|
|
|
|
|
re_hashval_t hash) internal_function; |
31
|
|
|
|
|
|
|
|
32
|
|
|
|
|
|
|
/* Functions for string operation. */ |
33
|
|
|
|
|
|
|
|
34
|
|
|
|
|
|
|
/* This function allocate the buffers. It is necessary to call |
35
|
|
|
|
|
|
|
re_string_reconstruct before using the object. */ |
36
|
|
|
|
|
|
|
|
37
|
|
|
|
|
|
|
static reg_errcode_t |
38
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
39
|
14
|
|
|
|
|
|
re_string_allocate (pTHX_ re_string_t *pstr, const char *str, Idx len, Idx init_len, |
40
|
|
|
|
|
|
|
RE_TRANSLATE_TYPE trans, bool icase, const re_dfa_t *dfa) |
41
|
|
|
|
|
|
|
{ |
42
|
|
|
|
|
|
|
reg_errcode_t ret; |
43
|
|
|
|
|
|
|
Idx init_buf_len; |
44
|
|
|
|
|
|
|
|
45
|
|
|
|
|
|
|
/* Ensure at least one character fits into the buffers. */ |
46
|
14
|
100
|
|
|
|
|
if (init_len < dfa->mb_cur_max) |
47
|
3
|
|
|
|
|
|
init_len = dfa->mb_cur_max; |
48
|
14
|
|
|
|
|
|
init_buf_len = (len + 1 < init_len) ? len + 1: init_len; |
49
|
14
|
|
|
|
|
|
re_string_construct_common (aTHX_ str, len, pstr, trans, icase, dfa); |
50
|
|
|
|
|
|
|
|
51
|
14
|
|
|
|
|
|
ret = re_string_realloc_buffers (aTHX_ pstr, init_buf_len); |
52
|
14
|
50
|
|
|
|
|
if (BE (ret != REG_NOERROR, 0)) |
53
|
0
|
|
|
|
|
|
return ret; |
54
|
|
|
|
|
|
|
|
55
|
14
|
|
|
|
|
|
pstr->word_char = dfa->word_char; |
56
|
14
|
|
|
|
|
|
pstr->word_ops_used = dfa->word_ops_used; |
57
|
14
|
100
|
|
|
|
|
pstr->mbs = pstr->mbs_allocated ? pstr->mbs : (unsigned char *) str; |
58
|
14
|
100
|
|
|
|
|
pstr->valid_len = (pstr->mbs_allocated || dfa->mb_cur_max > 1) ? 0 : len; |
|
|
50
|
|
|
|
|
|
59
|
14
|
|
|
|
|
|
pstr->valid_raw_len = pstr->valid_len; |
60
|
14
|
|
|
|
|
|
return REG_NOERROR; |
61
|
|
|
|
|
|
|
} |
62
|
|
|
|
|
|
|
|
63
|
|
|
|
|
|
|
/* This function allocate the buffers, and initialize them. */ |
64
|
|
|
|
|
|
|
|
65
|
|
|
|
|
|
|
static reg_errcode_t |
66
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
67
|
6
|
|
|
|
|
|
re_string_construct (pTHX_ re_string_t *pstr, const char *str, Idx len, |
68
|
|
|
|
|
|
|
RE_TRANSLATE_TYPE trans, bool icase, const re_dfa_t *dfa) |
69
|
|
|
|
|
|
|
{ |
70
|
|
|
|
|
|
|
reg_errcode_t ret; |
71
|
6
|
|
|
|
|
|
memset (pstr, '\0', sizeof (re_string_t)); |
72
|
6
|
|
|
|
|
|
re_string_construct_common (aTHX_ str, len, pstr, trans, icase, dfa); |
73
|
|
|
|
|
|
|
|
74
|
6
|
50
|
|
|
|
|
if (len > 0) |
75
|
|
|
|
|
|
|
{ |
76
|
6
|
|
|
|
|
|
ret = re_string_realloc_buffers (aTHX_ pstr, len + 1); |
77
|
6
|
50
|
|
|
|
|
if (BE (ret != REG_NOERROR, 0)) |
78
|
0
|
|
|
|
|
|
return ret; |
79
|
|
|
|
|
|
|
} |
80
|
6
|
100
|
|
|
|
|
pstr->mbs = pstr->mbs_allocated ? pstr->mbs : (unsigned char *) str; |
81
|
|
|
|
|
|
|
|
82
|
6
|
100
|
|
|
|
|
if (icase) |
83
|
|
|
|
|
|
|
{ |
84
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
85
|
1
|
50
|
|
|
|
|
if (dfa->mb_cur_max > 1) |
86
|
|
|
|
|
|
|
{ |
87
|
|
|
|
|
|
|
while (1) |
88
|
|
|
|
|
|
|
{ |
89
|
1
|
|
|
|
|
|
ret = build_wcs_upper_buffer (aTHX_ pstr); |
90
|
1
|
50
|
|
|
|
|
if (BE (ret != REG_NOERROR, 0)) |
91
|
0
|
|
|
|
|
|
return ret; |
92
|
1
|
50
|
|
|
|
|
if (pstr->valid_raw_len >= len) |
93
|
1
|
|
|
|
|
|
break; |
94
|
0
|
0
|
|
|
|
|
if (pstr->bufs_len > pstr->valid_len + dfa->mb_cur_max) |
95
|
0
|
|
|
|
|
|
break; |
96
|
0
|
|
|
|
|
|
ret = re_string_realloc_buffers (aTHX_ pstr, pstr->bufs_len * 2); |
97
|
0
|
0
|
|
|
|
|
if (BE (ret != REG_NOERROR, 0)) |
98
|
0
|
|
|
|
|
|
return ret; |
99
|
1
|
|
|
|
|
|
} |
100
|
|
|
|
|
|
|
} |
101
|
|
|
|
|
|
|
else |
102
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
103
|
1
|
|
|
|
|
|
build_upper_buffer (aTHX_ pstr); |
104
|
|
|
|
|
|
|
} |
105
|
|
|
|
|
|
|
else |
106
|
|
|
|
|
|
|
{ |
107
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
108
|
5
|
50
|
|
|
|
|
if (dfa->mb_cur_max > 1) |
109
|
5
|
|
|
|
|
|
build_wcs_buffer (aTHX_ pstr); |
110
|
|
|
|
|
|
|
else |
111
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
112
|
|
|
|
|
|
|
{ |
113
|
0
|
0
|
|
|
|
|
if (trans != NULL) |
114
|
0
|
|
|
|
|
|
re_string_translate_buffer (aTHX_ pstr); |
115
|
|
|
|
|
|
|
else |
116
|
|
|
|
|
|
|
{ |
117
|
0
|
|
|
|
|
|
pstr->valid_len = pstr->bufs_len; |
118
|
0
|
|
|
|
|
|
pstr->valid_raw_len = pstr->bufs_len; |
119
|
|
|
|
|
|
|
} |
120
|
|
|
|
|
|
|
} |
121
|
|
|
|
|
|
|
} |
122
|
|
|
|
|
|
|
|
123
|
6
|
|
|
|
|
|
return REG_NOERROR; |
124
|
|
|
|
|
|
|
} |
125
|
|
|
|
|
|
|
|
126
|
|
|
|
|
|
|
/* Helper functions for re_string_allocate, and re_string_construct. */ |
127
|
|
|
|
|
|
|
|
128
|
|
|
|
|
|
|
static reg_errcode_t |
129
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
130
|
20
|
|
|
|
|
|
re_string_realloc_buffers (pTHX_ re_string_t *pstr, Idx new_buf_len) |
131
|
|
|
|
|
|
|
{ |
132
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
133
|
20
|
50
|
|
|
|
|
if (pstr->mb_cur_max > 1) |
134
|
|
|
|
|
|
|
{ |
135
|
|
|
|
|
|
|
/* Avoid overflow in realloc. */ |
136
|
20
|
|
|
|
|
|
const size_t max_object_size = MAX (sizeof (rpl__wint_t), sizeof (Idx)); |
137
|
20
|
50
|
|
|
|
|
if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) < new_buf_len, 0)) |
138
|
0
|
|
|
|
|
|
return REG_ESPACE; |
139
|
|
|
|
|
|
|
|
140
|
20
|
50
|
|
|
|
|
re_realloc (pstr->wcs, rpl__wint_t, new_buf_len); |
141
|
20
|
50
|
|
|
|
|
if (pstr->offsets != NULL) |
142
|
|
|
|
|
|
|
{ |
143
|
0
|
0
|
|
|
|
|
re_realloc (pstr->offsets, Idx, new_buf_len); |
144
|
|
|
|
|
|
|
} |
145
|
|
|
|
|
|
|
} |
146
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
147
|
20
|
100
|
|
|
|
|
if (pstr->mbs_allocated) |
148
|
|
|
|
|
|
|
{ |
149
|
2
|
|
|
|
|
|
re_realloc (pstr->mbs, unsigned char, new_buf_len); |
150
|
|
|
|
|
|
|
} |
151
|
20
|
|
|
|
|
|
pstr->bufs_len = new_buf_len; |
152
|
20
|
|
|
|
|
|
return REG_NOERROR; |
153
|
|
|
|
|
|
|
} |
154
|
|
|
|
|
|
|
|
155
|
|
|
|
|
|
|
|
156
|
|
|
|
|
|
|
static void |
157
|
|
|
|
|
|
|
internal_function |
158
|
20
|
|
|
|
|
|
re_string_construct_common (pTHX_ const char *str, Idx len, re_string_t *pstr, |
159
|
|
|
|
|
|
|
RE_TRANSLATE_TYPE trans, bool icase, |
160
|
|
|
|
|
|
|
const re_dfa_t *dfa) |
161
|
|
|
|
|
|
|
{ |
162
|
20
|
|
|
|
|
|
pstr->raw_mbs = (const unsigned char *) str; |
163
|
20
|
|
|
|
|
|
pstr->len = len; |
164
|
20
|
|
|
|
|
|
pstr->raw_len = len; |
165
|
20
|
|
|
|
|
|
pstr->trans = trans; |
166
|
20
|
|
|
|
|
|
pstr->icase = icase; |
167
|
20
|
50
|
|
|
|
|
pstr->mbs_allocated = (trans != NULL || icase); |
|
|
100
|
|
|
|
|
|
168
|
20
|
|
|
|
|
|
pstr->mb_cur_max = dfa->mb_cur_max; |
169
|
20
|
|
|
|
|
|
pstr->is_utf8 = dfa->is_utf8; |
170
|
20
|
|
|
|
|
|
pstr->map_notascii = dfa->map_notascii; |
171
|
20
|
|
|
|
|
|
pstr->stop = pstr->len; |
172
|
20
|
|
|
|
|
|
pstr->raw_stop = pstr->stop; |
173
|
20
|
|
|
|
|
|
} |
174
|
|
|
|
|
|
|
|
175
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
176
|
|
|
|
|
|
|
|
177
|
|
|
|
|
|
|
/* Build wide character buffer PSTR->WCS. |
178
|
|
|
|
|
|
|
If the byte sequence of the string are: |
179
|
|
|
|
|
|
|
(0), (1), (0), (1), |
180
|
|
|
|
|
|
|
Then wide character buffer will be: |
181
|
|
|
|
|
|
|
, rpl__WEOF , , rpl__WEOF , |
182
|
|
|
|
|
|
|
We use rpl__WEOF for padding, they indicate that the position isn't |
183
|
|
|
|
|
|
|
a first byte of a multibyte character. |
184
|
|
|
|
|
|
|
|
185
|
|
|
|
|
|
|
Note that this function assumes PSTR->VALID_LEN elements are already |
186
|
|
|
|
|
|
|
built and starts from PSTR->VALID_LEN. */ |
187
|
|
|
|
|
|
|
|
188
|
|
|
|
|
|
|
static void |
189
|
|
|
|
|
|
|
internal_function |
190
|
40
|
|
|
|
|
|
build_wcs_buffer (pTHX_ re_string_t *pstr) |
191
|
|
|
|
|
|
|
{ |
192
|
|
|
|
|
|
|
rpl__mbstate_t prev_st; |
193
|
|
|
|
|
|
|
Idx byte_idx, end_idx, remain_len; |
194
|
|
|
|
|
|
|
size_t mbclen; |
195
|
|
|
|
|
|
|
#if (defined(_LIBC) || defined(_PERL_I18N)) |
196
|
|
|
|
|
|
|
unsigned char buf[rpl__MB_LEN_MAX]; |
197
|
|
|
|
|
|
|
assert (rpl__MB_LEN_MAX >= pstr->mb_cur_max); |
198
|
|
|
|
|
|
|
#else |
199
|
|
|
|
|
|
|
unsigned char buf[64]; |
200
|
|
|
|
|
|
|
#endif |
201
|
|
|
|
|
|
|
|
202
|
|
|
|
|
|
|
/* Build the buffers from pstr->valid_len to either pstr->len or |
203
|
|
|
|
|
|
|
pstr->bufs_len. */ |
204
|
40
|
|
|
|
|
|
end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len; |
205
|
152
|
100
|
|
|
|
|
for (byte_idx = pstr->valid_len; byte_idx < end_idx;) |
206
|
|
|
|
|
|
|
{ |
207
|
|
|
|
|
|
|
rpl__wchar_t wc; |
208
|
|
|
|
|
|
|
const char *p; |
209
|
|
|
|
|
|
|
|
210
|
112
|
|
|
|
|
|
remain_len = end_idx - byte_idx; |
211
|
112
|
|
|
|
|
|
prev_st = pstr->cur_state; |
212
|
|
|
|
|
|
|
/* Apply the translation if we need. */ |
213
|
112
|
50
|
|
|
|
|
if (BE (pstr->trans != NULL, 0)) |
214
|
|
|
|
|
|
|
{ |
215
|
|
|
|
|
|
|
int i, ch; |
216
|
|
|
|
|
|
|
|
217
|
0
|
0
|
|
|
|
|
for (i = 0; i < pstr->mb_cur_max && i < remain_len; ++i) |
|
|
0
|
|
|
|
|
|
218
|
|
|
|
|
|
|
{ |
219
|
0
|
|
|
|
|
|
ch = pstr->raw_mbs [pstr->raw_mbs_idx + byte_idx + i]; |
220
|
0
|
|
|
|
|
|
buf[i] = pstr->mbs[byte_idx + i] = pstr->trans[ch]; |
221
|
|
|
|
|
|
|
} |
222
|
0
|
|
|
|
|
|
p = (const char *) buf; |
223
|
|
|
|
|
|
|
} |
224
|
|
|
|
|
|
|
else |
225
|
112
|
|
|
|
|
|
p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + byte_idx; |
226
|
112
|
|
|
|
|
|
mbclen = rpl__mbrtowc (&wc, p, remain_len, &pstr->cur_state); |
227
|
112
|
50
|
|
|
|
|
if (BE (mbclen == (size_t) -1 || mbclen == 0 |
|
|
50
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
228
|
|
|
|
|
|
|
|| (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len), 0)) |
229
|
|
|
|
|
|
|
{ |
230
|
|
|
|
|
|
|
/* We treat these cases as a singlebyte character. */ |
231
|
0
|
|
|
|
|
|
mbclen = 1; |
232
|
0
|
|
|
|
|
|
wc = (rpl__wchar_t) pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]; |
233
|
0
|
0
|
|
|
|
|
if (BE (pstr->trans != NULL, 0)) |
234
|
0
|
|
|
|
|
|
wc = pstr->trans[wc]; |
235
|
0
|
|
|
|
|
|
pstr->cur_state = prev_st; |
236
|
|
|
|
|
|
|
} |
237
|
112
|
50
|
|
|
|
|
else if (BE (mbclen == (size_t) -2, 0)) |
238
|
|
|
|
|
|
|
{ |
239
|
|
|
|
|
|
|
/* The buffer doesn't have enough space, finish to build. */ |
240
|
0
|
|
|
|
|
|
pstr->cur_state = prev_st; |
241
|
0
|
|
|
|
|
|
break; |
242
|
|
|
|
|
|
|
} |
243
|
|
|
|
|
|
|
|
244
|
|
|
|
|
|
|
/* Write wide character and padding. */ |
245
|
112
|
|
|
|
|
|
pstr->wcs[byte_idx++] = wc; |
246
|
|
|
|
|
|
|
/* Write paddings. */ |
247
|
200
|
100
|
|
|
|
|
for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;) |
248
|
88
|
|
|
|
|
|
pstr->wcs[byte_idx++] = rpl__WEOF; |
249
|
|
|
|
|
|
|
} |
250
|
40
|
|
|
|
|
|
pstr->valid_len = byte_idx; |
251
|
40
|
|
|
|
|
|
pstr->valid_raw_len = byte_idx; |
252
|
40
|
|
|
|
|
|
} |
253
|
|
|
|
|
|
|
|
254
|
|
|
|
|
|
|
/* Build wide character buffer PSTR->WCS like build_wcs_buffer, |
255
|
|
|
|
|
|
|
but for REG_ICASE. */ |
256
|
|
|
|
|
|
|
|
257
|
|
|
|
|
|
|
static reg_errcode_t |
258
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
259
|
17
|
|
|
|
|
|
build_wcs_upper_buffer (pTHX_ re_string_t *pstr) |
260
|
|
|
|
|
|
|
{ |
261
|
|
|
|
|
|
|
rpl__mbstate_t prev_st; |
262
|
|
|
|
|
|
|
Idx src_idx, byte_idx, end_idx, remain_len; |
263
|
|
|
|
|
|
|
size_t mbclen; |
264
|
|
|
|
|
|
|
#if (defined(_LIBC) || defined(_PERL_I18N)) |
265
|
|
|
|
|
|
|
char buf[rpl__MB_LEN_MAX]; |
266
|
|
|
|
|
|
|
assert (rpl__MB_LEN_MAX >= pstr->mb_cur_max); |
267
|
|
|
|
|
|
|
#else |
268
|
|
|
|
|
|
|
char buf[64]; |
269
|
|
|
|
|
|
|
#endif |
270
|
|
|
|
|
|
|
|
271
|
17
|
|
|
|
|
|
byte_idx = pstr->valid_len; |
272
|
17
|
|
|
|
|
|
end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len; |
273
|
|
|
|
|
|
|
|
274
|
|
|
|
|
|
|
/* The following optimization assumes that ASCII characters can be |
275
|
|
|
|
|
|
|
mapped to wide characters with a simple cast. */ |
276
|
17
|
50
|
|
|
|
|
if (! pstr->map_notascii && pstr->trans == NULL && !pstr->offsets_needed) |
|
|
50
|
|
|
|
|
|
|
|
100
|
|
|
|
|
|
277
|
|
|
|
|
|
|
{ |
278
|
30
|
100
|
|
|
|
|
while (byte_idx < end_idx) |
279
|
|
|
|
|
|
|
{ |
280
|
|
|
|
|
|
|
rpl__wchar_t wc; |
281
|
|
|
|
|
|
|
|
282
|
24
|
100
|
|
|
|
|
if (rpl__isascii (pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]) |
283
|
18
|
50
|
|
|
|
|
&& rpl__mbsinit (&pstr->cur_state)) |
284
|
|
|
|
|
|
|
{ |
285
|
|
|
|
|
|
|
/* In case of a singlebyte character. */ |
286
|
18
|
|
|
|
|
|
pstr->mbs[byte_idx] |
287
|
18
|
|
|
|
|
|
= rpl__toupper (pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]); |
288
|
|
|
|
|
|
|
/* The next step uses the assumption that rpl__wchar_t is encoded |
289
|
|
|
|
|
|
|
ASCII-safe: all ASCII values can be converted like this. */ |
290
|
18
|
|
|
|
|
|
pstr->wcs[byte_idx] = (rpl__wchar_t) pstr->mbs[byte_idx]; |
291
|
18
|
|
|
|
|
|
++byte_idx; |
292
|
18
|
|
|
|
|
|
continue; |
293
|
|
|
|
|
|
|
} |
294
|
|
|
|
|
|
|
|
295
|
6
|
|
|
|
|
|
remain_len = end_idx - byte_idx; |
296
|
6
|
|
|
|
|
|
prev_st = pstr->cur_state; |
297
|
6
|
|
|
|
|
|
mbclen = rpl__mbrtowc (&wc, |
298
|
|
|
|
|
|
|
((const char *) pstr->raw_mbs + pstr->raw_mbs_idx |
299
|
|
|
|
|
|
|
+ byte_idx), remain_len, &pstr->cur_state); |
300
|
6
|
100
|
|
|
|
|
if (BE (mbclen < (size_t) -2, 1)) |
301
|
|
|
|
|
|
|
{ |
302
|
2
|
|
|
|
|
|
rpl__wchar_t wcu = wc; |
303
|
2
|
50
|
|
|
|
|
if (rpl__iswlower (wc)) |
304
|
|
|
|
|
|
|
{ |
305
|
|
|
|
|
|
|
size_t mbcdlen; |
306
|
|
|
|
|
|
|
|
307
|
2
|
|
|
|
|
|
wcu = rpl__towupper (wc); |
308
|
2
|
|
|
|
|
|
mbcdlen = rpl__wcrtomb (buf, wcu, &prev_st); |
309
|
2
|
50
|
|
|
|
|
if (BE (mbclen == mbcdlen, 1)) |
310
|
0
|
|
|
|
|
|
Copy (buf, pstr->mbs + byte_idx, mbclen, char); |
311
|
|
|
|
|
|
|
else |
312
|
|
|
|
|
|
|
{ |
313
|
2
|
|
|
|
|
|
src_idx = byte_idx; |
314
|
2
|
|
|
|
|
|
goto offsets_needed; |
315
|
|
|
|
|
|
|
} |
316
|
|
|
|
|
|
|
} |
317
|
|
|
|
|
|
|
else |
318
|
0
|
|
|
|
|
|
Copy (pstr->raw_mbs + pstr->raw_mbs_idx + byte_idx, pstr->mbs + byte_idx, mbclen, char); |
319
|
0
|
|
|
|
|
|
pstr->wcs[byte_idx++] = wcu; |
320
|
|
|
|
|
|
|
/* Write paddings. */ |
321
|
0
|
0
|
|
|
|
|
for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;) |
322
|
0
|
|
|
|
|
|
pstr->wcs[byte_idx++] = rpl__WEOF; |
323
|
|
|
|
|
|
|
} |
324
|
4
|
50
|
|
|
|
|
else if (mbclen == (size_t) -1 || mbclen == 0 |
|
|
0
|
|
|
|
|
|
325
|
0
|
0
|
|
|
|
|
|| (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len)) |
|
|
0
|
|
|
|
|
|
326
|
4
|
|
|
|
|
|
{ |
327
|
|
|
|
|
|
|
/* It is an invalid character, an incomplete character |
328
|
|
|
|
|
|
|
at the end of the string, or '\0'. Just use the byte. */ |
329
|
4
|
|
|
|
|
|
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]; |
330
|
4
|
|
|
|
|
|
pstr->mbs[byte_idx] = ch; |
331
|
|
|
|
|
|
|
/* And also cast it to wide char. */ |
332
|
4
|
|
|
|
|
|
pstr->wcs[byte_idx++] = (rpl__wchar_t) ch; |
333
|
4
|
50
|
|
|
|
|
if (BE (mbclen == (size_t) -1, 0)) |
334
|
4
|
|
|
|
|
|
pstr->cur_state = prev_st; |
335
|
|
|
|
|
|
|
} |
336
|
|
|
|
|
|
|
else |
337
|
|
|
|
|
|
|
{ |
338
|
|
|
|
|
|
|
/* The buffer doesn't have enough space, finish to build. */ |
339
|
0
|
|
|
|
|
|
pstr->cur_state = prev_st; |
340
|
4
|
|
|
|
|
|
break; |
341
|
|
|
|
|
|
|
} |
342
|
|
|
|
|
|
|
} |
343
|
6
|
|
|
|
|
|
pstr->valid_len = byte_idx; |
344
|
6
|
|
|
|
|
|
pstr->valid_raw_len = byte_idx; |
345
|
6
|
|
|
|
|
|
return REG_NOERROR; |
346
|
|
|
|
|
|
|
} |
347
|
|
|
|
|
|
|
else |
348
|
21
|
100
|
|
|
|
|
for (src_idx = pstr->valid_raw_len; byte_idx < end_idx;) |
349
|
|
|
|
|
|
|
{ |
350
|
|
|
|
|
|
|
rpl__wchar_t wc; |
351
|
|
|
|
|
|
|
const char *p; |
352
|
|
|
|
|
|
|
offsets_needed: |
353
|
12
|
|
|
|
|
|
remain_len = end_idx - byte_idx; |
354
|
12
|
|
|
|
|
|
prev_st = pstr->cur_state; |
355
|
12
|
50
|
|
|
|
|
if (BE (pstr->trans != NULL, 0)) |
356
|
|
|
|
|
|
|
{ |
357
|
|
|
|
|
|
|
int i, ch; |
358
|
|
|
|
|
|
|
|
359
|
0
|
0
|
|
|
|
|
for (i = 0; i < pstr->mb_cur_max && i < remain_len; ++i) |
|
|
0
|
|
|
|
|
|
360
|
|
|
|
|
|
|
{ |
361
|
0
|
|
|
|
|
|
ch = pstr->raw_mbs [pstr->raw_mbs_idx + src_idx + i]; |
362
|
0
|
|
|
|
|
|
buf[i] = pstr->trans[ch]; |
363
|
|
|
|
|
|
|
} |
364
|
0
|
|
|
|
|
|
p = (const char *) buf; |
365
|
|
|
|
|
|
|
} |
366
|
|
|
|
|
|
|
else |
367
|
12
|
|
|
|
|
|
p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + src_idx; |
368
|
12
|
|
|
|
|
|
mbclen = rpl__mbrtowc (&wc, p, remain_len, &pstr->cur_state); |
369
|
12
|
100
|
|
|
|
|
if (BE (mbclen < (size_t) -2, 1)) |
370
|
|
|
|
|
|
|
{ |
371
|
10
|
|
|
|
|
|
rpl__wchar_t wcu = wc; |
372
|
10
|
100
|
|
|
|
|
if (rpl__iswlower (wc)) |
373
|
|
|
|
|
|
|
{ |
374
|
|
|
|
|
|
|
size_t mbcdlen; |
375
|
|
|
|
|
|
|
|
376
|
3
|
|
|
|
|
|
wcu = rpl__towupper (wc); |
377
|
3
|
|
|
|
|
|
mbcdlen = rpl__wcrtomb ((char *) buf, wcu, &prev_st); |
378
|
3
|
50
|
|
|
|
|
if (BE (mbclen == mbcdlen, 1)) |
379
|
0
|
|
|
|
|
|
Copy (buf, pstr->mbs + byte_idx, mbclen, char); |
380
|
3
|
50
|
|
|
|
|
else if (mbcdlen != (size_t) -1) |
381
|
|
|
|
|
|
|
{ |
382
|
|
|
|
|
|
|
size_t i; |
383
|
|
|
|
|
|
|
|
384
|
3
|
50
|
|
|
|
|
if (byte_idx + mbcdlen > pstr->bufs_len) |
385
|
|
|
|
|
|
|
{ |
386
|
0
|
|
|
|
|
|
pstr->cur_state = prev_st; |
387
|
0
|
|
|
|
|
|
break; |
388
|
|
|
|
|
|
|
} |
389
|
|
|
|
|
|
|
|
390
|
3
|
100
|
|
|
|
|
if (pstr->offsets == NULL) |
391
|
|
|
|
|
|
|
{ |
392
|
1
|
50
|
|
|
|
|
re_malloc (pstr->offsets, Idx, pstr->bufs_len); |
393
|
|
|
|
|
|
|
} |
394
|
3
|
100
|
|
|
|
|
if (!pstr->offsets_needed) |
395
|
|
|
|
|
|
|
{ |
396
|
11
|
100
|
|
|
|
|
for (i = 0; i < (size_t) byte_idx; ++i) |
397
|
9
|
|
|
|
|
|
pstr->offsets[i] = i; |
398
|
2
|
|
|
|
|
|
pstr->offsets_needed = 1; |
399
|
|
|
|
|
|
|
} |
400
|
|
|
|
|
|
|
|
401
|
3
|
|
|
|
|
|
Copy (buf, pstr->mbs + byte_idx, mbcdlen, char); |
402
|
3
|
|
|
|
|
|
pstr->wcs[byte_idx] = wcu; |
403
|
3
|
|
|
|
|
|
pstr->offsets[byte_idx] = src_idx; |
404
|
3
|
50
|
|
|
|
|
for (i = 1; i < mbcdlen; ++i) |
405
|
|
|
|
|
|
|
{ |
406
|
0
|
|
|
|
|
|
pstr->offsets[byte_idx + i] |
407
|
0
|
0
|
|
|
|
|
= src_idx + (i < mbclen ? i : mbclen - 1); |
408
|
0
|
|
|
|
|
|
pstr->wcs[byte_idx + i] = rpl__WEOF; |
409
|
|
|
|
|
|
|
} |
410
|
3
|
|
|
|
|
|
pstr->len += mbcdlen - mbclen; |
411
|
3
|
50
|
|
|
|
|
if (pstr->raw_stop > src_idx) |
412
|
3
|
|
|
|
|
|
pstr->stop += mbcdlen - mbclen; |
413
|
3
|
|
|
|
|
|
end_idx = (pstr->bufs_len > pstr->len) |
414
|
3
|
|
|
|
|
|
? pstr->len : pstr->bufs_len; |
415
|
3
|
|
|
|
|
|
byte_idx += mbcdlen; |
416
|
3
|
|
|
|
|
|
src_idx += mbclen; |
417
|
3
|
|
|
|
|
|
continue; |
418
|
|
|
|
|
|
|
} |
419
|
|
|
|
|
|
|
else |
420
|
0
|
|
|
|
|
|
Copy (p, pstr->mbs + byte_idx, mbclen, char); |
421
|
|
|
|
|
|
|
} |
422
|
|
|
|
|
|
|
else |
423
|
7
|
|
|
|
|
|
Copy (p, pstr->mbs + byte_idx, mbclen, char); |
424
|
|
|
|
|
|
|
|
425
|
7
|
50
|
|
|
|
|
if (BE (pstr->offsets_needed != 0, 0)) |
426
|
|
|
|
|
|
|
{ |
427
|
|
|
|
|
|
|
size_t i; |
428
|
14
|
100
|
|
|
|
|
for (i = 0; i < mbclen; ++i) |
429
|
7
|
|
|
|
|
|
pstr->offsets[byte_idx + i] = src_idx + i; |
430
|
|
|
|
|
|
|
} |
431
|
7
|
|
|
|
|
|
src_idx += mbclen; |
432
|
|
|
|
|
|
|
|
433
|
7
|
|
|
|
|
|
pstr->wcs[byte_idx++] = wcu; |
434
|
|
|
|
|
|
|
/* Write paddings. */ |
435
|
7
|
50
|
|
|
|
|
for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;) |
436
|
0
|
|
|
|
|
|
pstr->wcs[byte_idx++] = rpl__WEOF; |
437
|
|
|
|
|
|
|
} |
438
|
2
|
50
|
|
|
|
|
else if (mbclen == (size_t) -1 || mbclen == 0 |
|
|
0
|
|
|
|
|
|
439
|
0
|
0
|
|
|
|
|
|| (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len)) |
|
|
0
|
|
|
|
|
|
440
|
2
|
|
|
|
|
|
{ |
441
|
|
|
|
|
|
|
/* It is an invalid character or '\0'. Just use the byte. */ |
442
|
2
|
|
|
|
|
|
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + src_idx]; |
443
|
|
|
|
|
|
|
|
444
|
2
|
50
|
|
|
|
|
if (BE (pstr->trans != NULL, 0)) |
445
|
0
|
|
|
|
|
|
ch = pstr->trans [ch]; |
446
|
2
|
|
|
|
|
|
pstr->mbs[byte_idx] = ch; |
447
|
|
|
|
|
|
|
|
448
|
2
|
50
|
|
|
|
|
if (BE (pstr->offsets_needed != 0, 0)) |
449
|
2
|
|
|
|
|
|
pstr->offsets[byte_idx] = src_idx; |
450
|
2
|
|
|
|
|
|
++src_idx; |
451
|
|
|
|
|
|
|
|
452
|
|
|
|
|
|
|
/* And also cast it to wide char. */ |
453
|
2
|
|
|
|
|
|
pstr->wcs[byte_idx++] = (rpl__wchar_t) ch; |
454
|
2
|
50
|
|
|
|
|
if (BE (mbclen == (size_t) -1, 0)) |
455
|
2
|
|
|
|
|
|
pstr->cur_state = prev_st; |
456
|
|
|
|
|
|
|
} |
457
|
|
|
|
|
|
|
else |
458
|
|
|
|
|
|
|
{ |
459
|
|
|
|
|
|
|
/* The buffer doesn't have enough space, finish to build. */ |
460
|
0
|
|
|
|
|
|
pstr->cur_state = prev_st; |
461
|
9
|
|
|
|
|
|
break; |
462
|
|
|
|
|
|
|
} |
463
|
|
|
|
|
|
|
} |
464
|
11
|
|
|
|
|
|
pstr->valid_len = byte_idx; |
465
|
11
|
|
|
|
|
|
pstr->valid_raw_len = src_idx; |
466
|
17
|
|
|
|
|
|
return REG_NOERROR; |
467
|
|
|
|
|
|
|
} |
468
|
|
|
|
|
|
|
|
469
|
|
|
|
|
|
|
/* Skip characters until the index becomes greater than NEW_RAW_IDX. |
470
|
|
|
|
|
|
|
Return the index. */ |
471
|
|
|
|
|
|
|
|
472
|
|
|
|
|
|
|
static Idx |
473
|
|
|
|
|
|
|
internal_function |
474
|
10
|
|
|
|
|
|
re_string_skip_chars (pTHX_ re_string_t *pstr, Idx new_raw_idx, rpl__wint_t *last_wc) |
475
|
|
|
|
|
|
|
{ |
476
|
|
|
|
|
|
|
rpl__mbstate_t prev_st; |
477
|
|
|
|
|
|
|
Idx rawbuf_idx; |
478
|
|
|
|
|
|
|
size_t mbclen; |
479
|
10
|
|
|
|
|
|
rpl__wint_t wc = rpl__WEOF; |
480
|
|
|
|
|
|
|
|
481
|
|
|
|
|
|
|
/* Skip the characters which are not necessary to check. */ |
482
|
46
|
100
|
|
|
|
|
for (rawbuf_idx = pstr->raw_mbs_idx + pstr->valid_raw_len; |
483
|
|
|
|
|
|
|
rawbuf_idx < new_raw_idx;) |
484
|
|
|
|
|
|
|
{ |
485
|
|
|
|
|
|
|
rpl__wchar_t wc2; |
486
|
36
|
|
|
|
|
|
Idx remain_len = pstr->raw_len - rawbuf_idx; |
487
|
36
|
|
|
|
|
|
prev_st = pstr->cur_state; |
488
|
36
|
|
|
|
|
|
mbclen = rpl__mbrtowc (&wc2, (const char *) pstr->raw_mbs + rawbuf_idx, |
489
|
|
|
|
|
|
|
remain_len, &pstr->cur_state); |
490
|
36
|
50
|
|
|
|
|
if (BE (mbclen == (size_t) -2 || mbclen == (size_t) -1 || mbclen == 0, 0)) |
|
|
50
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
491
|
|
|
|
|
|
|
{ |
492
|
|
|
|
|
|
|
/* We treat these cases as a single byte character. */ |
493
|
0
|
0
|
|
|
|
|
if (mbclen == 0 || remain_len == 0) |
|
|
0
|
|
|
|
|
|
494
|
0
|
|
|
|
|
|
wc = L'\0'; |
495
|
|
|
|
|
|
|
else |
496
|
0
|
|
|
|
|
|
wc = *(unsigned char *) (pstr->raw_mbs + rawbuf_idx); |
497
|
0
|
|
|
|
|
|
mbclen = 1; |
498
|
0
|
|
|
|
|
|
pstr->cur_state = prev_st; |
499
|
|
|
|
|
|
|
} |
500
|
|
|
|
|
|
|
else |
501
|
36
|
|
|
|
|
|
wc = wc2; |
502
|
|
|
|
|
|
|
/* Then proceed the next character. */ |
503
|
36
|
|
|
|
|
|
rawbuf_idx += mbclen; |
504
|
|
|
|
|
|
|
} |
505
|
10
|
|
|
|
|
|
*last_wc = wc; |
506
|
10
|
|
|
|
|
|
return rawbuf_idx; |
507
|
|
|
|
|
|
|
} |
508
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
509
|
|
|
|
|
|
|
|
510
|
|
|
|
|
|
|
/* Build the buffer PSTR->MBS, and apply the translation if we need. |
511
|
|
|
|
|
|
|
This function is used in case of REG_ICASE. */ |
512
|
|
|
|
|
|
|
|
513
|
|
|
|
|
|
|
static void |
514
|
|
|
|
|
|
|
internal_function |
515
|
0
|
|
|
|
|
|
build_upper_buffer (pTHX_ re_string_t *pstr) |
516
|
|
|
|
|
|
|
{ |
517
|
|
|
|
|
|
|
Idx char_idx, end_idx; |
518
|
0
|
|
|
|
|
|
end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len; |
519
|
|
|
|
|
|
|
|
520
|
0
|
0
|
|
|
|
|
for (char_idx = pstr->valid_len; char_idx < end_idx; ++char_idx) |
521
|
|
|
|
|
|
|
{ |
522
|
0
|
|
|
|
|
|
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + char_idx]; |
523
|
0
|
0
|
|
|
|
|
if (BE (pstr->trans != NULL, 0)) |
524
|
0
|
|
|
|
|
|
ch = pstr->trans[ch]; |
525
|
0
|
0
|
|
|
|
|
if (rpl__islower (ch)) |
526
|
0
|
|
|
|
|
|
pstr->mbs[char_idx] = rpl__toupper (ch); |
527
|
|
|
|
|
|
|
else |
528
|
0
|
|
|
|
|
|
pstr->mbs[char_idx] = ch; |
529
|
|
|
|
|
|
|
} |
530
|
0
|
|
|
|
|
|
pstr->valid_len = char_idx; |
531
|
0
|
|
|
|
|
|
pstr->valid_raw_len = char_idx; |
532
|
0
|
|
|
|
|
|
} |
533
|
|
|
|
|
|
|
|
534
|
|
|
|
|
|
|
/* Apply TRANS to the buffer in PSTR. */ |
535
|
|
|
|
|
|
|
|
536
|
|
|
|
|
|
|
static void |
537
|
|
|
|
|
|
|
internal_function |
538
|
0
|
|
|
|
|
|
re_string_translate_buffer (pTHX_ re_string_t *pstr) |
539
|
|
|
|
|
|
|
{ |
540
|
|
|
|
|
|
|
Idx buf_idx, end_idx; |
541
|
0
|
|
|
|
|
|
end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len; |
542
|
|
|
|
|
|
|
|
543
|
0
|
0
|
|
|
|
|
for (buf_idx = pstr->valid_len; buf_idx < end_idx; ++buf_idx) |
544
|
|
|
|
|
|
|
{ |
545
|
0
|
|
|
|
|
|
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + buf_idx]; |
546
|
0
|
|
|
|
|
|
pstr->mbs[buf_idx] = pstr->trans[ch]; |
547
|
|
|
|
|
|
|
} |
548
|
|
|
|
|
|
|
|
549
|
0
|
|
|
|
|
|
pstr->valid_len = buf_idx; |
550
|
0
|
|
|
|
|
|
pstr->valid_raw_len = buf_idx; |
551
|
0
|
|
|
|
|
|
} |
552
|
|
|
|
|
|
|
|
553
|
|
|
|
|
|
|
/* This function re-construct the buffers. |
554
|
|
|
|
|
|
|
Concretely, convert to wide character in case of pstr->mb_cur_max > 1, |
555
|
|
|
|
|
|
|
convert to upper case in case of REG_ICASE, apply translation. */ |
556
|
|
|
|
|
|
|
|
557
|
|
|
|
|
|
|
static reg_errcode_t |
558
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
559
|
51
|
|
|
|
|
|
re_string_reconstruct (pTHX_ re_string_t *pstr, Idx idx, int eflags) |
560
|
|
|
|
|
|
|
{ |
561
|
|
|
|
|
|
|
Idx offset; |
562
|
|
|
|
|
|
|
|
563
|
51
|
50
|
|
|
|
|
if (BE (pstr->raw_mbs_idx <= idx, 0)) |
564
|
51
|
|
|
|
|
|
offset = idx - pstr->raw_mbs_idx; |
565
|
|
|
|
|
|
|
else |
566
|
|
|
|
|
|
|
{ |
567
|
|
|
|
|
|
|
/* Reset buffer. */ |
568
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
569
|
0
|
0
|
|
|
|
|
if (pstr->mb_cur_max > 1) |
570
|
0
|
|
|
|
|
|
memset (&pstr->cur_state, '\0', sizeof (rpl__mbstate_t)); |
571
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
572
|
0
|
|
|
|
|
|
pstr->len = pstr->raw_len; |
573
|
0
|
|
|
|
|
|
pstr->stop = pstr->raw_stop; |
574
|
0
|
|
|
|
|
|
pstr->valid_len = 0; |
575
|
0
|
|
|
|
|
|
pstr->raw_mbs_idx = 0; |
576
|
0
|
|
|
|
|
|
pstr->valid_raw_len = 0; |
577
|
0
|
|
|
|
|
|
pstr->offsets_needed = 0; |
578
|
0
|
0
|
|
|
|
|
pstr->tip_context = ((eflags & REG_NOTBOL) ? CONTEXT_BEGBUF |
579
|
|
|
|
|
|
|
: CONTEXT_NEWLINE | CONTEXT_BEGBUF); |
580
|
0
|
0
|
|
|
|
|
if (!pstr->mbs_allocated) |
581
|
0
|
|
|
|
|
|
pstr->mbs = (unsigned char *) pstr->raw_mbs; |
582
|
0
|
|
|
|
|
|
offset = idx; |
583
|
|
|
|
|
|
|
} |
584
|
|
|
|
|
|
|
|
585
|
51
|
100
|
|
|
|
|
if (BE (offset != 0, 1)) |
586
|
|
|
|
|
|
|
{ |
587
|
|
|
|
|
|
|
/* Should the already checked characters be kept? */ |
588
|
45
|
100
|
|
|
|
|
if (BE (offset < pstr->valid_raw_len, 1)) |
589
|
|
|
|
|
|
|
{ |
590
|
|
|
|
|
|
|
/* Yes, move them to the front of the buffer. */ |
591
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
592
|
35
|
100
|
|
|
|
|
if (BE (pstr->offsets_needed, 0)) |
593
|
|
|
|
|
|
|
{ |
594
|
11
|
|
|
|
|
|
Idx low = 0, high = pstr->valid_len, mid; |
595
|
|
|
|
|
|
|
do |
596
|
|
|
|
|
|
|
{ |
597
|
31
|
|
|
|
|
|
mid = (high + low) / 2; |
598
|
31
|
100
|
|
|
|
|
if (pstr->offsets[mid] > offset) |
599
|
20
|
|
|
|
|
|
high = mid; |
600
|
11
|
100
|
|
|
|
|
else if (pstr->offsets[mid] < offset) |
601
|
2
|
|
|
|
|
|
low = mid + 1; |
602
|
|
|
|
|
|
|
else |
603
|
9
|
|
|
|
|
|
break; |
604
|
|
|
|
|
|
|
} |
605
|
22
|
100
|
|
|
|
|
while (low < high); |
606
|
11
|
100
|
|
|
|
|
if (pstr->offsets[mid] < offset) |
607
|
2
|
|
|
|
|
|
++mid; |
608
|
11
|
|
|
|
|
|
pstr->tip_context = re_string_context_at (aTHX_ pstr, mid - 1, |
609
|
|
|
|
|
|
|
eflags); |
610
|
|
|
|
|
|
|
/* This can be quite complicated, so handle specially |
611
|
|
|
|
|
|
|
only the common and easy case where the character with |
612
|
|
|
|
|
|
|
different length representation of lower and upper |
613
|
|
|
|
|
|
|
case is present at or after offset. */ |
614
|
20
|
50
|
|
|
|
|
if (pstr->valid_len > offset |
615
|
11
|
50
|
|
|
|
|
&& mid == offset && pstr->offsets[mid] == offset) |
|
|
100
|
|
|
|
|
|
616
|
|
|
|
|
|
|
{ |
617
|
9
|
50
|
|
|
|
|
Move (pstr->wcs + offset, pstr->wcs, pstr->valid_len - offset, rpl__wint_t); |
618
|
9
|
|
|
|
|
|
Move (pstr->mbs + offset, pstr->mbs, pstr->valid_len - offset, char); |
619
|
9
|
|
|
|
|
|
pstr->valid_len -= offset; |
620
|
9
|
|
|
|
|
|
pstr->valid_raw_len -= offset; |
621
|
81
|
100
|
|
|
|
|
for (low = 0; low < pstr->valid_len; low++) |
622
|
72
|
|
|
|
|
|
pstr->offsets[low] = pstr->offsets[low + offset] - offset; |
623
|
|
|
|
|
|
|
} |
624
|
|
|
|
|
|
|
else |
625
|
|
|
|
|
|
|
{ |
626
|
|
|
|
|
|
|
/* Otherwise, just find out how long the partial multibyte |
627
|
|
|
|
|
|
|
character at offset is and fill it with rpl__WEOF/255. */ |
628
|
2
|
|
|
|
|
|
pstr->len = pstr->raw_len - idx + offset; |
629
|
2
|
|
|
|
|
|
pstr->stop = pstr->raw_stop - idx + offset; |
630
|
2
|
|
|
|
|
|
pstr->offsets_needed = 0; |
631
|
2
|
50
|
|
|
|
|
while (mid > 0 && pstr->offsets[mid - 1] == offset) |
|
|
50
|
|
|
|
|
|
632
|
0
|
|
|
|
|
|
--mid; |
633
|
2
|
50
|
|
|
|
|
while (mid < pstr->valid_len) |
634
|
2
|
50
|
|
|
|
|
if (pstr->wcs[mid] != rpl__WEOF) |
635
|
2
|
|
|
|
|
|
break; |
636
|
|
|
|
|
|
|
else |
637
|
0
|
|
|
|
|
|
++mid; |
638
|
2
|
50
|
|
|
|
|
if (mid == pstr->valid_len) |
639
|
0
|
|
|
|
|
|
pstr->valid_len = 0; |
640
|
|
|
|
|
|
|
else |
641
|
|
|
|
|
|
|
{ |
642
|
2
|
|
|
|
|
|
pstr->valid_len = pstr->offsets[mid] - offset; |
643
|
2
|
50
|
|
|
|
|
if (pstr->valid_len) |
644
|
|
|
|
|
|
|
{ |
645
|
4
|
100
|
|
|
|
|
for (low = 0; low < pstr->valid_len; ++low) |
646
|
2
|
|
|
|
|
|
pstr->wcs[low] = rpl__WEOF; |
647
|
2
|
|
|
|
|
|
memset (pstr->mbs, 255, pstr->valid_len); |
648
|
|
|
|
|
|
|
} |
649
|
|
|
|
|
|
|
} |
650
|
11
|
|
|
|
|
|
pstr->valid_raw_len = pstr->valid_len; |
651
|
|
|
|
|
|
|
} |
652
|
|
|
|
|
|
|
} |
653
|
|
|
|
|
|
|
else |
654
|
|
|
|
|
|
|
#endif |
655
|
|
|
|
|
|
|
{ |
656
|
24
|
|
|
|
|
|
pstr->tip_context = re_string_context_at (aTHX_ pstr, offset - 1, |
657
|
|
|
|
|
|
|
eflags); |
658
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
659
|
24
|
50
|
|
|
|
|
if (pstr->mb_cur_max > 1) |
660
|
24
|
50
|
|
|
|
|
Move (pstr->wcs + offset, pstr->wcs, pstr->valid_len - offset, rpl__wint_t); |
661
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
662
|
24
|
100
|
|
|
|
|
if (BE (pstr->mbs_allocated, 0)) |
663
|
3
|
|
|
|
|
|
Move (pstr->mbs + offset, pstr->mbs, pstr->valid_len - offset, char); |
664
|
24
|
|
|
|
|
|
pstr->valid_len -= offset; |
665
|
35
|
|
|
|
|
|
pstr->valid_raw_len -= offset; |
666
|
|
|
|
|
|
|
#if DEBUG |
667
|
|
|
|
|
|
|
assert (pstr->valid_len > 0); |
668
|
|
|
|
|
|
|
#endif |
669
|
|
|
|
|
|
|
} |
670
|
|
|
|
|
|
|
} |
671
|
|
|
|
|
|
|
else |
672
|
|
|
|
|
|
|
{ |
673
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
674
|
|
|
|
|
|
|
/* No, skip all characters until IDX. */ |
675
|
10
|
|
|
|
|
|
Idx prev_valid_len = pstr->valid_len; |
676
|
|
|
|
|
|
|
|
677
|
10
|
50
|
|
|
|
|
if (BE (pstr->offsets_needed, 0)) |
678
|
|
|
|
|
|
|
{ |
679
|
0
|
|
|
|
|
|
pstr->len = pstr->raw_len - idx + offset; |
680
|
0
|
|
|
|
|
|
pstr->stop = pstr->raw_stop - idx + offset; |
681
|
0
|
|
|
|
|
|
pstr->offsets_needed = 0; |
682
|
|
|
|
|
|
|
} |
683
|
|
|
|
|
|
|
#endif |
684
|
10
|
|
|
|
|
|
pstr->valid_len = 0; |
685
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
686
|
10
|
50
|
|
|
|
|
if (pstr->mb_cur_max > 1) |
687
|
|
|
|
|
|
|
{ |
688
|
|
|
|
|
|
|
Idx wcs_idx; |
689
|
10
|
|
|
|
|
|
rpl__wint_t wc = rpl__WEOF; |
690
|
|
|
|
|
|
|
|
691
|
10
|
50
|
|
|
|
|
if (pstr->is_utf8) |
692
|
|
|
|
|
|
|
{ |
693
|
|
|
|
|
|
|
const unsigned char *raw, *p, *end; |
694
|
|
|
|
|
|
|
|
695
|
|
|
|
|
|
|
/* Special case UTF-8. Multi-byte chars start with any |
696
|
|
|
|
|
|
|
byte other than 0x80 - 0xbf. */ |
697
|
0
|
|
|
|
|
|
raw = pstr->raw_mbs + pstr->raw_mbs_idx; |
698
|
0
|
|
|
|
|
|
end = raw + (offset - pstr->mb_cur_max); |
699
|
0
|
0
|
|
|
|
|
if (end < pstr->raw_mbs) |
700
|
0
|
|
|
|
|
|
end = pstr->raw_mbs; |
701
|
0
|
|
|
|
|
|
p = raw + offset - 1; |
702
|
|
|
|
|
|
|
#if (defined(_LIBC) || defined(_PERL_I18N)) |
703
|
|
|
|
|
|
|
/* For the simple case, ASCII characters, skip the conversion step. */ |
704
|
0
|
0
|
|
|
|
|
if (rpl__isascii (*p) && BE (pstr->trans == NULL, 1)) |
|
|
0
|
|
|
|
|
|
705
|
|
|
|
|
|
|
{ |
706
|
0
|
|
|
|
|
|
memset (&pstr->cur_state, '\0', sizeof (rpl__mbstate_t)); |
707
|
|
|
|
|
|
|
/* pstr->valid_len = 0; */ |
708
|
0
|
|
|
|
|
|
wc = (rpl__wchar_t) *p; |
709
|
|
|
|
|
|
|
} |
710
|
|
|
|
|
|
|
else |
711
|
|
|
|
|
|
|
#endif |
712
|
0
|
0
|
|
|
|
|
for (; p >= end; --p) |
713
|
0
|
0
|
|
|
|
|
if ((*p & 0xc0) != 0x80) |
714
|
|
|
|
|
|
|
{ |
715
|
|
|
|
|
|
|
rpl__mbstate_t cur_state; |
716
|
|
|
|
|
|
|
rpl__wchar_t wc2; |
717
|
0
|
|
|
|
|
|
Idx mlen = raw + pstr->len - p; |
718
|
|
|
|
|
|
|
unsigned char buf[6]; |
719
|
|
|
|
|
|
|
size_t mbclen; |
720
|
|
|
|
|
|
|
|
721
|
0
|
|
|
|
|
|
const unsigned char *pp = p; |
722
|
0
|
0
|
|
|
|
|
if (BE (pstr->trans != NULL, 0)) |
723
|
|
|
|
|
|
|
{ |
724
|
0
|
|
|
|
|
|
int i = (int)mlen < 6 ? (int)mlen : 6; |
725
|
0
|
0
|
|
|
|
|
while (--i >= 0) |
726
|
0
|
|
|
|
|
|
buf[i] = pstr->trans[p[i]]; |
727
|
0
|
|
|
|
|
|
pp = buf; |
728
|
|
|
|
|
|
|
} |
729
|
|
|
|
|
|
|
/* XXX Don't use mbrtowc, we know which conversion |
730
|
|
|
|
|
|
|
to use (UTF-8 -> UCS4). */ |
731
|
0
|
|
|
|
|
|
memset (&cur_state, 0, sizeof (cur_state)); |
732
|
0
|
|
|
|
|
|
mbclen = rpl__mbrtowc (&wc2, (const char *) pp, mlen, |
733
|
|
|
|
|
|
|
&cur_state); |
734
|
0
|
0
|
|
|
|
|
if (raw + offset - p <= mbclen |
735
|
0
|
0
|
|
|
|
|
&& mbclen < (size_t) -2) |
736
|
|
|
|
|
|
|
{ |
737
|
0
|
|
|
|
|
|
memset (&pstr->cur_state, '\0', |
738
|
|
|
|
|
|
|
sizeof (rpl__mbstate_t)); |
739
|
0
|
|
|
|
|
|
pstr->valid_len = mbclen - (raw + offset - p); |
740
|
0
|
|
|
|
|
|
wc = wc2; |
741
|
|
|
|
|
|
|
} |
742
|
0
|
|
|
|
|
|
break; |
743
|
|
|
|
|
|
|
} |
744
|
|
|
|
|
|
|
} |
745
|
|
|
|
|
|
|
|
746
|
10
|
50
|
|
|
|
|
if (wc == rpl__WEOF) |
747
|
10
|
|
|
|
|
|
pstr->valid_len = re_string_skip_chars (aTHX_ pstr, idx, &wc) - idx; |
748
|
10
|
100
|
|
|
|
|
if (wc == rpl__WEOF) |
749
|
|
|
|
|
|
|
pstr->tip_context |
750
|
2
|
|
|
|
|
|
= re_string_context_at (aTHX_ pstr, prev_valid_len - 1, eflags); |
751
|
|
|
|
|
|
|
else |
752
|
8
|
50
|
|
|
|
|
pstr->tip_context = ((BE (pstr->word_ops_used != 0, 0) |
|
|
50
|
|
|
|
|
|
753
|
0
|
0
|
|
|
|
|
&& IS_WIDE_WORD_CHAR (wc)) |
|
|
0
|
|
|
|
|
|
754
|
|
|
|
|
|
|
? CONTEXT_WORD |
755
|
8
|
|
|
|
|
|
: ((IS_WIDE_NEWLINE (wc) |
756
|
0
|
0
|
|
|
|
|
&& pstr->newline_anchor) |
757
|
|
|
|
|
|
|
? CONTEXT_NEWLINE : 0)); |
758
|
10
|
50
|
|
|
|
|
if (BE (pstr->valid_len, 0)) |
759
|
|
|
|
|
|
|
{ |
760
|
0
|
0
|
|
|
|
|
for (wcs_idx = 0; wcs_idx < pstr->valid_len; ++wcs_idx) |
761
|
0
|
|
|
|
|
|
pstr->wcs[wcs_idx] = rpl__WEOF; |
762
|
0
|
0
|
|
|
|
|
if (pstr->mbs_allocated) |
763
|
0
|
|
|
|
|
|
memset (pstr->mbs, 255, pstr->valid_len); |
764
|
|
|
|
|
|
|
} |
765
|
10
|
|
|
|
|
|
pstr->valid_raw_len = pstr->valid_len; |
766
|
|
|
|
|
|
|
} |
767
|
|
|
|
|
|
|
else |
768
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
769
|
|
|
|
|
|
|
{ |
770
|
0
|
|
|
|
|
|
int c = pstr->raw_mbs[pstr->raw_mbs_idx + offset - 1]; |
771
|
0
|
|
|
|
|
|
pstr->valid_raw_len = 0; |
772
|
0
|
0
|
|
|
|
|
if (pstr->trans) |
773
|
0
|
|
|
|
|
|
c = pstr->trans[c]; |
774
|
0
|
0
|
|
|
|
|
pstr->tip_context = (bitset_contain (aTHX_ pstr->word_char, c) |
|
|
0
|
|
|
|
|
|
775
|
|
|
|
|
|
|
? CONTEXT_WORD |
776
|
0
|
0
|
|
|
|
|
: ((IS_NEWLINE (c) && pstr->newline_anchor) |
777
|
|
|
|
|
|
|
? CONTEXT_NEWLINE : 0)); |
778
|
|
|
|
|
|
|
} |
779
|
|
|
|
|
|
|
} |
780
|
45
|
100
|
|
|
|
|
if (!BE (pstr->mbs_allocated, 0)) |
781
|
30
|
|
|
|
|
|
pstr->mbs += offset; |
782
|
|
|
|
|
|
|
} |
783
|
51
|
|
|
|
|
|
pstr->raw_mbs_idx = idx; |
784
|
51
|
|
|
|
|
|
pstr->len -= offset; |
785
|
51
|
|
|
|
|
|
pstr->stop -= offset; |
786
|
|
|
|
|
|
|
|
787
|
|
|
|
|
|
|
/* Then build the buffers. */ |
788
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
789
|
51
|
50
|
|
|
|
|
if (pstr->mb_cur_max > 1) |
790
|
|
|
|
|
|
|
{ |
791
|
51
|
100
|
|
|
|
|
if (pstr->icase) |
792
|
|
|
|
|
|
|
{ |
793
|
16
|
|
|
|
|
|
reg_errcode_t ret = build_wcs_upper_buffer (aTHX_ pstr); |
794
|
16
|
50
|
|
|
|
|
if (BE (ret != REG_NOERROR, 0)) |
795
|
16
|
|
|
|
|
|
return ret; |
796
|
|
|
|
|
|
|
} |
797
|
|
|
|
|
|
|
else |
798
|
51
|
|
|
|
|
|
build_wcs_buffer (aTHX_ pstr); |
799
|
|
|
|
|
|
|
} |
800
|
|
|
|
|
|
|
else |
801
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
802
|
0
|
0
|
|
|
|
|
if (BE (pstr->mbs_allocated, 0)) |
803
|
|
|
|
|
|
|
{ |
804
|
0
|
0
|
|
|
|
|
if (pstr->icase) |
805
|
0
|
|
|
|
|
|
build_upper_buffer (aTHX_ pstr); |
806
|
0
|
0
|
|
|
|
|
else if (pstr->trans != NULL) |
807
|
0
|
|
|
|
|
|
re_string_translate_buffer (aTHX_ pstr); |
808
|
|
|
|
|
|
|
} |
809
|
|
|
|
|
|
|
else |
810
|
0
|
|
|
|
|
|
pstr->valid_len = pstr->len; |
811
|
|
|
|
|
|
|
|
812
|
51
|
|
|
|
|
|
pstr->cur_idx = 0; |
813
|
51
|
|
|
|
|
|
return REG_NOERROR; |
814
|
|
|
|
|
|
|
} |
815
|
|
|
|
|
|
|
|
816
|
|
|
|
|
|
|
static unsigned char |
817
|
|
|
|
|
|
|
internal_function __attribute__ ((pure)) |
818
|
12
|
|
|
|
|
|
re_string_peek_byte_case (pTHX_ const re_string_t *pstr, Idx idx) |
819
|
|
|
|
|
|
|
{ |
820
|
|
|
|
|
|
|
int ch; |
821
|
|
|
|
|
|
|
Idx off; |
822
|
|
|
|
|
|
|
|
823
|
|
|
|
|
|
|
/* Handle the common (easiest) cases first. */ |
824
|
12
|
50
|
|
|
|
|
if (BE (!pstr->mbs_allocated, 1)) |
825
|
12
|
|
|
|
|
|
return re_string_peek_byte (aTHX_ pstr, idx); |
826
|
|
|
|
|
|
|
|
827
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
828
|
0
|
0
|
|
|
|
|
if (pstr->mb_cur_max > 1 |
829
|
0
|
0
|
|
|
|
|
&& ! re_string_is_single_byte_char (aTHX_ pstr, pstr->cur_idx + idx)) |
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
830
|
0
|
|
|
|
|
|
return re_string_peek_byte (aTHX_ pstr, idx); |
831
|
|
|
|
|
|
|
#endif |
832
|
|
|
|
|
|
|
|
833
|
0
|
|
|
|
|
|
off = pstr->cur_idx + idx; |
834
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
835
|
0
|
0
|
|
|
|
|
if (pstr->offsets_needed) |
836
|
0
|
|
|
|
|
|
off = pstr->offsets[off]; |
837
|
|
|
|
|
|
|
#endif |
838
|
|
|
|
|
|
|
|
839
|
0
|
|
|
|
|
|
ch = pstr->raw_mbs[pstr->raw_mbs_idx + off]; |
840
|
|
|
|
|
|
|
|
841
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
842
|
|
|
|
|
|
|
/* Ensure that e.g. for tr_TR.UTF-8 BACKSLASH DOTLESS SMALL LETTER I |
843
|
|
|
|
|
|
|
this function returns CAPITAL LETTER I instead of first byte of |
844
|
|
|
|
|
|
|
DOTLESS SMALL LETTER I. The latter would confuse the parser, |
845
|
|
|
|
|
|
|
since peek_byte_case doesn't advance cur_idx in any way. */ |
846
|
0
|
0
|
|
|
|
|
if (pstr->offsets_needed && !rpl__isascii (ch)) |
|
|
0
|
|
|
|
|
|
847
|
0
|
|
|
|
|
|
return re_string_peek_byte (aTHX_ pstr, idx); |
848
|
|
|
|
|
|
|
#endif |
849
|
|
|
|
|
|
|
|
850
|
0
|
|
|
|
|
|
return ch; |
851
|
|
|
|
|
|
|
} |
852
|
|
|
|
|
|
|
|
853
|
|
|
|
|
|
|
static unsigned char |
854
|
|
|
|
|
|
|
internal_function |
855
|
0
|
|
|
|
|
|
re_string_fetch_byte_case (pTHX_ re_string_t *pstr) |
856
|
|
|
|
|
|
|
{ |
857
|
0
|
0
|
|
|
|
|
if (BE (!pstr->mbs_allocated, 1)) |
858
|
0
|
|
|
|
|
|
return re_string_fetch_byte (aTHX_ pstr); |
859
|
|
|
|
|
|
|
|
860
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
861
|
0
|
0
|
|
|
|
|
if (pstr->offsets_needed) |
862
|
|
|
|
|
|
|
{ |
863
|
|
|
|
|
|
|
Idx off; |
864
|
|
|
|
|
|
|
int ch; |
865
|
|
|
|
|
|
|
|
866
|
|
|
|
|
|
|
/* For tr_TR.UTF-8 [[:islower:]] there is |
867
|
|
|
|
|
|
|
[[: CAPITAL LETTER I WITH DOT lower:]] in mbs. Skip |
868
|
|
|
|
|
|
|
in that case the whole multi-byte character and return |
869
|
|
|
|
|
|
|
the original letter. On the other side, with |
870
|
|
|
|
|
|
|
[[: DOTLESS SMALL LETTER I return [[:I, as doing |
871
|
|
|
|
|
|
|
anything else would complicate things too much. */ |
872
|
|
|
|
|
|
|
|
873
|
0
|
0
|
|
|
|
|
if (!re_string_first_byte (aTHX_ pstr, pstr->cur_idx)) |
|
|
0
|
|
|
|
|
|
874
|
0
|
|
|
|
|
|
return re_string_fetch_byte (aTHX_ pstr); |
875
|
|
|
|
|
|
|
|
876
|
0
|
|
|
|
|
|
off = pstr->offsets[pstr->cur_idx]; |
877
|
0
|
|
|
|
|
|
ch = pstr->raw_mbs[pstr->raw_mbs_idx + off]; |
878
|
|
|
|
|
|
|
|
879
|
0
|
0
|
|
|
|
|
if (! rpl__isascii (ch)) |
880
|
0
|
|
|
|
|
|
return re_string_fetch_byte (aTHX_ pstr); |
881
|
|
|
|
|
|
|
|
882
|
0
|
|
|
|
|
|
re_string_skip_bytes (aTHX_ pstr, |
883
|
|
|
|
|
|
|
re_string_char_size_at (aTHX_ pstr, pstr->cur_idx)); |
884
|
0
|
|
|
|
|
|
return ch; |
885
|
|
|
|
|
|
|
} |
886
|
|
|
|
|
|
|
#endif |
887
|
|
|
|
|
|
|
|
888
|
0
|
|
|
|
|
|
return pstr->raw_mbs[pstr->raw_mbs_idx + pstr->cur_idx++]; |
889
|
|
|
|
|
|
|
} |
890
|
|
|
|
|
|
|
|
891
|
|
|
|
|
|
|
static void |
892
|
|
|
|
|
|
|
internal_function |
893
|
20
|
|
|
|
|
|
re_string_destruct (pTHX_ re_string_t *pstr) |
894
|
|
|
|
|
|
|
{ |
895
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
896
|
20
|
50
|
|
|
|
|
re_free (pstr->wcs); |
897
|
20
|
100
|
|
|
|
|
re_free (pstr->offsets); |
898
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
899
|
20
|
100
|
|
|
|
|
if (pstr->mbs_allocated) |
900
|
2
|
50
|
|
|
|
|
re_free (pstr->mbs); |
901
|
20
|
|
|
|
|
|
} |
902
|
|
|
|
|
|
|
|
903
|
|
|
|
|
|
|
/* Return the context at IDX in INPUT. */ |
904
|
|
|
|
|
|
|
|
905
|
|
|
|
|
|
|
static unsigned int |
906
|
|
|
|
|
|
|
internal_function |
907
|
81
|
|
|
|
|
|
re_string_context_at (pTHX_ const re_string_t *input, Idx idx, int eflags) |
908
|
|
|
|
|
|
|
{ |
909
|
|
|
|
|
|
|
int c; |
910
|
81
|
50
|
|
|
|
|
if (BE (! REG_VALID_INDEX (idx), 0)) |
911
|
|
|
|
|
|
|
/* In this case, we use the value stored in input->tip_context, |
912
|
|
|
|
|
|
|
since we can't know the character in input->mbs[-1] here. */ |
913
|
0
|
|
|
|
|
|
return input->tip_context; |
914
|
81
|
100
|
|
|
|
|
if (BE (idx == input->len, 0)) |
915
|
3
|
50
|
|
|
|
|
return ((eflags & REG_NOTEOL) ? CONTEXT_ENDBUF |
916
|
|
|
|
|
|
|
: CONTEXT_NEWLINE | CONTEXT_ENDBUF); |
917
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
918
|
78
|
50
|
|
|
|
|
if (input->mb_cur_max > 1) |
919
|
|
|
|
|
|
|
{ |
920
|
|
|
|
|
|
|
rpl__wint_t wc; |
921
|
78
|
|
|
|
|
|
Idx wc_idx = idx; |
922
|
142
|
100
|
|
|
|
|
while(input->wcs[wc_idx] == rpl__WEOF) |
923
|
|
|
|
|
|
|
{ |
924
|
|
|
|
|
|
|
#ifdef DEBUG |
925
|
|
|
|
|
|
|
/* It must not happen. */ |
926
|
|
|
|
|
|
|
assert (REG_VALID_INDEX (wc_idx)); |
927
|
|
|
|
|
|
|
#endif |
928
|
80
|
|
|
|
|
|
--wc_idx; |
929
|
80
|
100
|
|
|
|
|
if (! REG_VALID_INDEX (wc_idx)) |
930
|
16
|
|
|
|
|
|
return input->tip_context; |
931
|
|
|
|
|
|
|
} |
932
|
62
|
|
|
|
|
|
wc = input->wcs[wc_idx]; |
933
|
62
|
50
|
|
|
|
|
if (BE (input->word_ops_used != 0, 0) && IS_WIDE_WORD_CHAR (wc)) |
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
934
|
0
|
|
|
|
|
|
return CONTEXT_WORD; |
935
|
62
|
50
|
|
|
|
|
return (IS_WIDE_NEWLINE (wc) && input->newline_anchor |
|
|
0
|
|
|
|
|
|
936
|
|
|
|
|
|
|
? CONTEXT_NEWLINE : 0); |
937
|
|
|
|
|
|
|
} |
938
|
|
|
|
|
|
|
else |
939
|
|
|
|
|
|
|
#endif |
940
|
|
|
|
|
|
|
{ |
941
|
0
|
|
|
|
|
|
c = re_string_byte_at (aTHX_ input, idx); |
942
|
0
|
0
|
|
|
|
|
if (bitset_contain (aTHX_ input->word_char, c)) |
943
|
0
|
|
|
|
|
|
return CONTEXT_WORD; |
944
|
0
|
0
|
|
|
|
|
return IS_NEWLINE (c) && input->newline_anchor ? CONTEXT_NEWLINE : 0; |
|
|
0
|
|
|
|
|
|
945
|
|
|
|
|
|
|
} |
946
|
|
|
|
|
|
|
} |
947
|
|
|
|
|
|
|
|
948
|
|
|
|
|
|
|
/* Functions for set operation. */ |
949
|
|
|
|
|
|
|
|
950
|
|
|
|
|
|
|
static reg_errcode_t |
951
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
952
|
100
|
|
|
|
|
|
re_node_set_alloc (pTHX_ re_node_set *set, Idx size) |
953
|
|
|
|
|
|
|
{ |
954
|
100
|
|
|
|
|
|
set->alloc = size; |
955
|
100
|
|
|
|
|
|
set->nelem = 0; |
956
|
100
|
50
|
|
|
|
|
re_malloc (set->elems, Idx, size); |
957
|
100
|
|
|
|
|
|
return REG_NOERROR; |
958
|
|
|
|
|
|
|
} |
959
|
|
|
|
|
|
|
|
960
|
|
|
|
|
|
|
static reg_errcode_t |
961
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
962
|
50
|
|
|
|
|
|
re_node_set_init_1 (pTHX_ re_node_set *set, Idx elem) |
963
|
|
|
|
|
|
|
{ |
964
|
50
|
|
|
|
|
|
set->alloc = 1; |
965
|
50
|
|
|
|
|
|
set->nelem = 1; |
966
|
50
|
|
|
|
|
|
re_malloc (set->elems, Idx, 1); |
967
|
50
|
|
|
|
|
|
set->elems[0] = elem; |
968
|
50
|
|
|
|
|
|
return REG_NOERROR; |
969
|
|
|
|
|
|
|
} |
970
|
|
|
|
|
|
|
|
971
|
|
|
|
|
|
|
static reg_errcode_t |
972
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
973
|
2
|
|
|
|
|
|
re_node_set_init_2 (pTHX_ re_node_set *set, Idx elem1, Idx elem2) |
974
|
|
|
|
|
|
|
{ |
975
|
2
|
|
|
|
|
|
set->alloc = 2; |
976
|
2
|
|
|
|
|
|
re_malloc (set->elems, Idx, 2); |
977
|
2
|
50
|
|
|
|
|
if (elem1 == elem2) |
978
|
|
|
|
|
|
|
{ |
979
|
0
|
|
|
|
|
|
set->nelem = 1; |
980
|
0
|
|
|
|
|
|
set->elems[0] = elem1; |
981
|
|
|
|
|
|
|
} |
982
|
|
|
|
|
|
|
else |
983
|
|
|
|
|
|
|
{ |
984
|
2
|
|
|
|
|
|
set->nelem = 2; |
985
|
2
|
50
|
|
|
|
|
if (elem1 < elem2) |
986
|
|
|
|
|
|
|
{ |
987
|
2
|
|
|
|
|
|
set->elems[0] = elem1; |
988
|
2
|
|
|
|
|
|
set->elems[1] = elem2; |
989
|
|
|
|
|
|
|
} |
990
|
|
|
|
|
|
|
else |
991
|
|
|
|
|
|
|
{ |
992
|
0
|
|
|
|
|
|
set->elems[0] = elem2; |
993
|
0
|
|
|
|
|
|
set->elems[1] = elem1; |
994
|
|
|
|
|
|
|
} |
995
|
|
|
|
|
|
|
} |
996
|
2
|
|
|
|
|
|
return REG_NOERROR; |
997
|
|
|
|
|
|
|
} |
998
|
|
|
|
|
|
|
|
999
|
|
|
|
|
|
|
static reg_errcode_t |
1000
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1001
|
36
|
|
|
|
|
|
re_node_set_init_copy (pTHX_ re_node_set *dest, const re_node_set *src) |
1002
|
|
|
|
|
|
|
{ |
1003
|
36
|
|
|
|
|
|
dest->nelem = src->nelem; |
1004
|
36
|
50
|
|
|
|
|
if (src->nelem > 0) |
1005
|
|
|
|
|
|
|
{ |
1006
|
36
|
|
|
|
|
|
dest->alloc = dest->nelem; |
1007
|
36
|
50
|
|
|
|
|
re_malloc (dest->elems, Idx, dest->alloc); |
1008
|
36
|
50
|
|
|
|
|
Copy (src->elems, dest->elems, src->nelem, Idx); |
1009
|
|
|
|
|
|
|
} |
1010
|
|
|
|
|
|
|
else |
1011
|
0
|
|
|
|
|
|
re_node_set_init_empty (dest); |
1012
|
36
|
|
|
|
|
|
return REG_NOERROR; |
1013
|
|
|
|
|
|
|
} |
1014
|
|
|
|
|
|
|
|
1015
|
|
|
|
|
|
|
/* Calculate the intersection of the sets SRC1 and SRC2. And merge it to |
1016
|
|
|
|
|
|
|
DEST. Return value indicate the error code or REG_NOERROR if succeeded. |
1017
|
|
|
|
|
|
|
Note: We assume dest->elems is NULL, when dest->alloc is 0. */ |
1018
|
|
|
|
|
|
|
|
1019
|
|
|
|
|
|
|
static reg_errcode_t |
1020
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1021
|
19
|
|
|
|
|
|
re_node_set_add_intersect (pTHX_ re_node_set *dest, const re_node_set *src1, |
1022
|
|
|
|
|
|
|
const re_node_set *src2) |
1023
|
|
|
|
|
|
|
{ |
1024
|
|
|
|
|
|
|
Idx i1, i2, is, id, delta, sbase; |
1025
|
19
|
50
|
|
|
|
|
if (src1->nelem == 0 || src2->nelem == 0) |
|
|
50
|
|
|
|
|
|
1026
|
0
|
|
|
|
|
|
return REG_NOERROR; |
1027
|
|
|
|
|
|
|
|
1028
|
|
|
|
|
|
|
/* We need dest->nelem + 2 * elems_in_intersection; this is a |
1029
|
|
|
|
|
|
|
conservative estimate. */ |
1030
|
19
|
100
|
|
|
|
|
if (src1->nelem + src2->nelem + dest->nelem > dest->alloc) |
1031
|
|
|
|
|
|
|
{ |
1032
|
18
|
|
|
|
|
|
Idx new_alloc = src1->nelem + src2->nelem + dest->alloc; |
1033
|
18
|
50
|
|
|
|
|
re_realloc (dest->elems, Idx, new_alloc); |
1034
|
18
|
|
|
|
|
|
dest->alloc = new_alloc; |
1035
|
|
|
|
|
|
|
} |
1036
|
|
|
|
|
|
|
|
1037
|
|
|
|
|
|
|
/* Find the items in the intersection of SRC1 and SRC2, and copy |
1038
|
|
|
|
|
|
|
into the top of DEST those that are not already in DEST itself. */ |
1039
|
19
|
|
|
|
|
|
sbase = dest->nelem + src1->nelem + src2->nelem; |
1040
|
19
|
|
|
|
|
|
i1 = src1->nelem - 1; |
1041
|
19
|
|
|
|
|
|
i2 = src2->nelem - 1; |
1042
|
19
|
|
|
|
|
|
id = dest->nelem - 1; |
1043
|
|
|
|
|
|
|
for (;;) |
1044
|
|
|
|
|
|
|
{ |
1045
|
57
|
100
|
|
|
|
|
if (src1->elems[i1] == src2->elems[i2]) |
1046
|
|
|
|
|
|
|
{ |
1047
|
|
|
|
|
|
|
/* Try to find the item in DEST. Maybe we could binary search? */ |
1048
|
67
|
100
|
|
|
|
|
while (REG_VALID_INDEX (id) && dest->elems[id] > src1->elems[i1]) |
|
|
100
|
|
|
|
|
|
1049
|
19
|
|
|
|
|
|
--id; |
1050
|
|
|
|
|
|
|
|
1051
|
48
|
100
|
|
|
|
|
if (! REG_VALID_INDEX (id) || dest->elems[id] != src1->elems[i1]) |
|
|
100
|
|
|
|
|
|
1052
|
29
|
|
|
|
|
|
dest->elems[--sbase] = src1->elems[i1]; |
1053
|
|
|
|
|
|
|
|
1054
|
48
|
100
|
|
|
|
|
if (! REG_VALID_INDEX (--i1) || ! REG_VALID_INDEX (--i2)) |
|
|
50
|
|
|
|
|
|
1055
|
|
|
|
|
|
|
break; |
1056
|
|
|
|
|
|
|
} |
1057
|
|
|
|
|
|
|
|
1058
|
|
|
|
|
|
|
/* Lower the highest of the two items. */ |
1059
|
9
|
50
|
|
|
|
|
else if (src1->elems[i1] < src2->elems[i2]) |
1060
|
|
|
|
|
|
|
{ |
1061
|
0
|
0
|
|
|
|
|
if (! REG_VALID_INDEX (--i2)) |
1062
|
0
|
|
|
|
|
|
break; |
1063
|
|
|
|
|
|
|
} |
1064
|
|
|
|
|
|
|
else |
1065
|
|
|
|
|
|
|
{ |
1066
|
9
|
50
|
|
|
|
|
if (! REG_VALID_INDEX (--i1)) |
1067
|
0
|
|
|
|
|
|
break; |
1068
|
|
|
|
|
|
|
} |
1069
|
38
|
|
|
|
|
|
} |
1070
|
|
|
|
|
|
|
|
1071
|
19
|
|
|
|
|
|
id = dest->nelem - 1; |
1072
|
19
|
|
|
|
|
|
is = dest->nelem + src1->nelem + src2->nelem - 1; |
1073
|
19
|
|
|
|
|
|
delta = is - sbase + 1; |
1074
|
|
|
|
|
|
|
|
1075
|
|
|
|
|
|
|
/* Now copy. When DELTA becomes zero, the remaining |
1076
|
|
|
|
|
|
|
DEST elements are already in place; this is more or |
1077
|
|
|
|
|
|
|
less the same loop that is in re_node_set_merge. */ |
1078
|
19
|
|
|
|
|
|
dest->nelem += delta; |
1079
|
19
|
50
|
|
|
|
|
if (delta > 0 && REG_VALID_INDEX (id)) |
|
|
50
|
|
|
|
|
|
1080
|
|
|
|
|
|
|
for (;;) |
1081
|
|
|
|
|
|
|
{ |
1082
|
28
|
100
|
|
|
|
|
if (dest->elems[is] > dest->elems[id]) |
1083
|
|
|
|
|
|
|
{ |
1084
|
|
|
|
|
|
|
/* Copy from the top. */ |
1085
|
9
|
|
|
|
|
|
dest->elems[id + delta--] = dest->elems[is--]; |
1086
|
9
|
50
|
|
|
|
|
if (delta == 0) |
1087
|
0
|
|
|
|
|
|
break; |
1088
|
|
|
|
|
|
|
} |
1089
|
|
|
|
|
|
|
else |
1090
|
|
|
|
|
|
|
{ |
1091
|
|
|
|
|
|
|
/* Slide from the bottom. */ |
1092
|
19
|
|
|
|
|
|
dest->elems[id + delta] = dest->elems[id]; |
1093
|
19
|
50
|
|
|
|
|
if (! REG_VALID_INDEX (--id)) |
1094
|
19
|
|
|
|
|
|
break; |
1095
|
|
|
|
|
|
|
} |
1096
|
9
|
|
|
|
|
|
} |
1097
|
|
|
|
|
|
|
|
1098
|
|
|
|
|
|
|
/* Copy remaining SRC elements. */ |
1099
|
19
|
50
|
|
|
|
|
Copy (dest->elems + sbase, dest->elems, delta, Idx); |
1100
|
|
|
|
|
|
|
|
1101
|
19
|
|
|
|
|
|
return REG_NOERROR; |
1102
|
|
|
|
|
|
|
} |
1103
|
|
|
|
|
|
|
|
1104
|
|
|
|
|
|
|
/* Calculate the union set of the sets SRC1 and SRC2. And store it to |
1105
|
|
|
|
|
|
|
DEST. Return value indicate the error code or REG_NOERROR if succeeded. */ |
1106
|
|
|
|
|
|
|
|
1107
|
|
|
|
|
|
|
static reg_errcode_t |
1108
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1109
|
0
|
|
|
|
|
|
re_node_set_init_union (pTHX_ re_node_set *dest, const re_node_set *src1, |
1110
|
|
|
|
|
|
|
const re_node_set *src2) |
1111
|
|
|
|
|
|
|
{ |
1112
|
|
|
|
|
|
|
Idx i1, i2, id; |
1113
|
0
|
0
|
|
|
|
|
if (src1 != NULL && src1->nelem > 0 && src2 != NULL && src2->nelem > 0) |
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
1114
|
|
|
|
|
|
|
{ |
1115
|
0
|
|
|
|
|
|
dest->alloc = src1->nelem + src2->nelem; |
1116
|
0
|
0
|
|
|
|
|
re_malloc (dest->elems, Idx, dest->alloc); |
1117
|
|
|
|
|
|
|
} |
1118
|
|
|
|
|
|
|
else |
1119
|
|
|
|
|
|
|
{ |
1120
|
0
|
0
|
|
|
|
|
if (src1 != NULL && src1->nelem > 0) |
|
|
0
|
|
|
|
|
|
1121
|
0
|
|
|
|
|
|
return re_node_set_init_copy (aTHX_ dest, src1); |
1122
|
0
|
0
|
|
|
|
|
else if (src2 != NULL && src2->nelem > 0) |
|
|
0
|
|
|
|
|
|
1123
|
0
|
|
|
|
|
|
return re_node_set_init_copy (aTHX_ dest, src2); |
1124
|
|
|
|
|
|
|
else |
1125
|
0
|
|
|
|
|
|
re_node_set_init_empty (dest); |
1126
|
0
|
|
|
|
|
|
return REG_NOERROR; |
1127
|
|
|
|
|
|
|
} |
1128
|
0
|
0
|
|
|
|
|
for (i1 = i2 = id = 0 ; i1 < src1->nelem && i2 < src2->nelem ;) |
|
|
0
|
|
|
|
|
|
1129
|
|
|
|
|
|
|
{ |
1130
|
0
|
0
|
|
|
|
|
if (src1->elems[i1] > src2->elems[i2]) |
1131
|
|
|
|
|
|
|
{ |
1132
|
0
|
|
|
|
|
|
dest->elems[id++] = src2->elems[i2++]; |
1133
|
0
|
|
|
|
|
|
continue; |
1134
|
|
|
|
|
|
|
} |
1135
|
0
|
0
|
|
|
|
|
if (src1->elems[i1] == src2->elems[i2]) |
1136
|
0
|
|
|
|
|
|
++i2; |
1137
|
0
|
|
|
|
|
|
dest->elems[id++] = src1->elems[i1++]; |
1138
|
|
|
|
|
|
|
} |
1139
|
0
|
0
|
|
|
|
|
if (i1 < src1->nelem) |
1140
|
|
|
|
|
|
|
{ |
1141
|
0
|
0
|
|
|
|
|
Copy (src1->elems + i1, dest->elems + id, src1->nelem - i1, Idx); |
1142
|
0
|
|
|
|
|
|
id += src1->nelem - i1; |
1143
|
|
|
|
|
|
|
} |
1144
|
0
|
0
|
|
|
|
|
else if (i2 < src2->nelem) |
1145
|
|
|
|
|
|
|
{ |
1146
|
0
|
0
|
|
|
|
|
Copy (src2->elems + i2, dest->elems + id, src2->nelem - i2, Idx); |
1147
|
0
|
|
|
|
|
|
id += src2->nelem - i2; |
1148
|
|
|
|
|
|
|
} |
1149
|
0
|
|
|
|
|
|
dest->nelem = id; |
1150
|
0
|
|
|
|
|
|
return REG_NOERROR; |
1151
|
|
|
|
|
|
|
} |
1152
|
|
|
|
|
|
|
|
1153
|
|
|
|
|
|
|
/* Calculate the union set of the sets DEST and SRC. And store it to |
1154
|
|
|
|
|
|
|
DEST. Return value indicate the error code or REG_NOERROR if succeeded. */ |
1155
|
|
|
|
|
|
|
|
1156
|
|
|
|
|
|
|
static reg_errcode_t |
1157
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1158
|
38
|
|
|
|
|
|
re_node_set_merge (pTHX_ re_node_set *dest, const re_node_set *src) |
1159
|
|
|
|
|
|
|
{ |
1160
|
|
|
|
|
|
|
Idx is, id, sbase, delta; |
1161
|
38
|
50
|
|
|
|
|
if (src == NULL || src->nelem == 0) |
|
|
50
|
|
|
|
|
|
1162
|
0
|
|
|
|
|
|
return REG_NOERROR; |
1163
|
38
|
100
|
|
|
|
|
if (dest->alloc < 2 * src->nelem + dest->nelem) |
1164
|
|
|
|
|
|
|
{ |
1165
|
14
|
|
|
|
|
|
Idx new_alloc = 2 * (src->nelem + dest->alloc); |
1166
|
14
|
50
|
|
|
|
|
re_realloc (dest->elems, Idx, new_alloc); |
1167
|
14
|
|
|
|
|
|
dest->alloc = new_alloc; |
1168
|
|
|
|
|
|
|
} |
1169
|
|
|
|
|
|
|
|
1170
|
38
|
100
|
|
|
|
|
if (BE (dest->nelem == 0, 0)) |
1171
|
|
|
|
|
|
|
{ |
1172
|
36
|
|
|
|
|
|
dest->nelem = src->nelem; |
1173
|
36
|
50
|
|
|
|
|
Copy (src->elems, dest->elems, src->nelem, Idx); |
1174
|
36
|
|
|
|
|
|
return REG_NOERROR; |
1175
|
|
|
|
|
|
|
} |
1176
|
|
|
|
|
|
|
|
1177
|
|
|
|
|
|
|
/* Copy into the top of DEST the items of SRC that are not |
1178
|
|
|
|
|
|
|
found in DEST. Maybe we could binary search in DEST? */ |
1179
|
4
|
100
|
|
|
|
|
for (sbase = dest->nelem + 2 * src->nelem, |
1180
|
2
|
|
|
|
|
|
is = src->nelem - 1, id = dest->nelem - 1; |
1181
|
2
|
50
|
|
|
|
|
REG_VALID_INDEX (is) && REG_VALID_INDEX (id); ) |
1182
|
|
|
|
|
|
|
{ |
1183
|
2
|
50
|
|
|
|
|
if (dest->elems[id] == src->elems[is]) |
1184
|
0
|
|
|
|
|
|
is--, id--; |
1185
|
2
|
50
|
|
|
|
|
else if (dest->elems[id] < src->elems[is]) |
1186
|
2
|
|
|
|
|
|
dest->elems[--sbase] = src->elems[is--]; |
1187
|
|
|
|
|
|
|
else /* if (dest->elems[id] > src->elems[is]) */ |
1188
|
0
|
|
|
|
|
|
--id; |
1189
|
|
|
|
|
|
|
} |
1190
|
|
|
|
|
|
|
|
1191
|
2
|
50
|
|
|
|
|
if (REG_VALID_INDEX (is)) |
1192
|
|
|
|
|
|
|
{ |
1193
|
|
|
|
|
|
|
/* If DEST is exhausted, the remaining items of SRC must be unique. */ |
1194
|
0
|
|
|
|
|
|
sbase -= is + 1; |
1195
|
0
|
0
|
|
|
|
|
Copy (src->elems, dest->elems + sbase, is + 1, Idx); |
1196
|
|
|
|
|
|
|
} |
1197
|
|
|
|
|
|
|
|
1198
|
2
|
|
|
|
|
|
id = dest->nelem - 1; |
1199
|
2
|
|
|
|
|
|
is = dest->nelem + 2 * src->nelem - 1; |
1200
|
2
|
|
|
|
|
|
delta = is - sbase + 1; |
1201
|
2
|
50
|
|
|
|
|
if (delta == 0) |
1202
|
0
|
|
|
|
|
|
return REG_NOERROR; |
1203
|
|
|
|
|
|
|
|
1204
|
|
|
|
|
|
|
/* Now copy. When DELTA becomes zero, the remaining |
1205
|
|
|
|
|
|
|
DEST elements are already in place. */ |
1206
|
2
|
|
|
|
|
|
dest->nelem += delta; |
1207
|
|
|
|
|
|
|
for (;;) |
1208
|
|
|
|
|
|
|
{ |
1209
|
2
|
50
|
|
|
|
|
if (dest->elems[is] > dest->elems[id]) |
1210
|
|
|
|
|
|
|
{ |
1211
|
|
|
|
|
|
|
/* Copy from the top. */ |
1212
|
2
|
|
|
|
|
|
dest->elems[id + delta--] = dest->elems[is--]; |
1213
|
2
|
50
|
|
|
|
|
if (delta == 0) |
1214
|
2
|
|
|
|
|
|
break; |
1215
|
|
|
|
|
|
|
} |
1216
|
|
|
|
|
|
|
else |
1217
|
|
|
|
|
|
|
{ |
1218
|
|
|
|
|
|
|
/* Slide from the bottom. */ |
1219
|
0
|
|
|
|
|
|
dest->elems[id + delta] = dest->elems[id]; |
1220
|
0
|
0
|
|
|
|
|
if (! REG_VALID_INDEX (--id)) |
1221
|
|
|
|
|
|
|
{ |
1222
|
|
|
|
|
|
|
/* Copy remaining SRC elements. */ |
1223
|
0
|
0
|
|
|
|
|
Copy (dest->elems + sbase, dest->elems, delta, Idx); |
1224
|
0
|
|
|
|
|
|
break; |
1225
|
|
|
|
|
|
|
} |
1226
|
|
|
|
|
|
|
} |
1227
|
0
|
|
|
|
|
|
} |
1228
|
|
|
|
|
|
|
|
1229
|
2
|
|
|
|
|
|
return REG_NOERROR; |
1230
|
|
|
|
|
|
|
} |
1231
|
|
|
|
|
|
|
|
1232
|
|
|
|
|
|
|
/* Insert the new element ELEM to the re_node_set* SET. |
1233
|
|
|
|
|
|
|
SET should not already have ELEM. |
1234
|
|
|
|
|
|
|
Return true if successful. */ |
1235
|
|
|
|
|
|
|
|
1236
|
|
|
|
|
|
|
static bool |
1237
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1238
|
93
|
|
|
|
|
|
re_node_set_insert (pTHX_ re_node_set *set, Idx elem) |
1239
|
|
|
|
|
|
|
{ |
1240
|
|
|
|
|
|
|
Idx idx; |
1241
|
|
|
|
|
|
|
/* In case the set is empty. */ |
1242
|
93
|
100
|
|
|
|
|
if (set->alloc == 0) |
1243
|
12
|
|
|
|
|
|
return BE (re_node_set_init_1 (aTHX_ set, elem) == REG_NOERROR, 1); |
1244
|
|
|
|
|
|
|
|
1245
|
81
|
100
|
|
|
|
|
if (BE (set->nelem, 0) == 0) |
1246
|
|
|
|
|
|
|
{ |
1247
|
|
|
|
|
|
|
/* We already guaranteed above that set->alloc != 0. */ |
1248
|
57
|
|
|
|
|
|
set->elems[0] = elem; |
1249
|
57
|
|
|
|
|
|
++set->nelem; |
1250
|
57
|
|
|
|
|
|
return true; |
1251
|
|
|
|
|
|
|
} |
1252
|
|
|
|
|
|
|
|
1253
|
|
|
|
|
|
|
/* Realloc if we need. */ |
1254
|
24
|
100
|
|
|
|
|
if (set->alloc == set->nelem) |
1255
|
|
|
|
|
|
|
{ |
1256
|
9
|
|
|
|
|
|
set->alloc = set->alloc * 2; |
1257
|
9
|
50
|
|
|
|
|
re_realloc (set->elems, Idx, set->alloc); |
1258
|
|
|
|
|
|
|
} |
1259
|
|
|
|
|
|
|
|
1260
|
|
|
|
|
|
|
/* Move the elements which follows the new element. Test the |
1261
|
|
|
|
|
|
|
first element separately to skip a check in the inner loop. */ |
1262
|
24
|
100
|
|
|
|
|
if (elem < set->elems[0]) |
1263
|
|
|
|
|
|
|
{ |
1264
|
12
|
|
|
|
|
|
idx = 0; |
1265
|
29
|
100
|
|
|
|
|
for (idx = set->nelem; idx > 0; idx--) |
1266
|
17
|
|
|
|
|
|
set->elems[idx] = set->elems[idx - 1]; |
1267
|
|
|
|
|
|
|
} |
1268
|
|
|
|
|
|
|
else |
1269
|
|
|
|
|
|
|
{ |
1270
|
12
|
50
|
|
|
|
|
for (idx = set->nelem; set->elems[idx - 1] > elem; idx--) |
1271
|
0
|
|
|
|
|
|
set->elems[idx] = set->elems[idx - 1]; |
1272
|
|
|
|
|
|
|
} |
1273
|
|
|
|
|
|
|
|
1274
|
|
|
|
|
|
|
/* Insert the new element. */ |
1275
|
24
|
|
|
|
|
|
set->elems[idx] = elem; |
1276
|
24
|
|
|
|
|
|
++set->nelem; |
1277
|
24
|
|
|
|
|
|
return true; |
1278
|
|
|
|
|
|
|
} |
1279
|
|
|
|
|
|
|
|
1280
|
|
|
|
|
|
|
/* Insert the new element ELEM to the re_node_set* SET. |
1281
|
|
|
|
|
|
|
SET should not already have any element greater than or equal to ELEM. |
1282
|
|
|
|
|
|
|
Return true if successful. */ |
1283
|
|
|
|
|
|
|
|
1284
|
|
|
|
|
|
|
static bool |
1285
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1286
|
62
|
|
|
|
|
|
re_node_set_insert_last (pTHX_ re_node_set *set, Idx elem) |
1287
|
|
|
|
|
|
|
{ |
1288
|
|
|
|
|
|
|
/* Realloc if we need. */ |
1289
|
62
|
100
|
|
|
|
|
if (set->alloc == set->nelem) |
1290
|
|
|
|
|
|
|
{ |
1291
|
20
|
|
|
|
|
|
set->alloc = (set->alloc + 1) * 2; |
1292
|
20
|
50
|
|
|
|
|
re_realloc (set->elems, Idx, set->alloc); |
1293
|
|
|
|
|
|
|
} |
1294
|
|
|
|
|
|
|
|
1295
|
|
|
|
|
|
|
/* Insert the new element. */ |
1296
|
62
|
|
|
|
|
|
set->elems[set->nelem++] = elem; |
1297
|
62
|
|
|
|
|
|
return true; |
1298
|
|
|
|
|
|
|
} |
1299
|
|
|
|
|
|
|
|
1300
|
|
|
|
|
|
|
/* Compare two node sets SET1 and SET2. |
1301
|
|
|
|
|
|
|
Return true if SET1 and SET2 are equivalent. */ |
1302
|
|
|
|
|
|
|
|
1303
|
|
|
|
|
|
|
static bool |
1304
|
|
|
|
|
|
|
internal_function __attribute__ ((pure)) |
1305
|
66
|
|
|
|
|
|
re_node_set_compare (pTHX_ const re_node_set *set1, const re_node_set *set2) |
1306
|
|
|
|
|
|
|
{ |
1307
|
|
|
|
|
|
|
Idx i; |
1308
|
66
|
50
|
|
|
|
|
if (set1 == NULL || set2 == NULL || set1->nelem != set2->nelem) |
|
|
50
|
|
|
|
|
|
|
|
100
|
|
|
|
|
|
1309
|
3
|
|
|
|
|
|
return false; |
1310
|
199
|
100
|
|
|
|
|
for (i = set1->nelem ; REG_VALID_INDEX (--i) ; ) |
1311
|
136
|
50
|
|
|
|
|
if (set1->elems[i] != set2->elems[i]) |
1312
|
0
|
|
|
|
|
|
return false; |
1313
|
63
|
|
|
|
|
|
return true; |
1314
|
|
|
|
|
|
|
} |
1315
|
|
|
|
|
|
|
|
1316
|
|
|
|
|
|
|
/* Return (idx + 1) if SET contains the element ELEM, return 0 otherwise. */ |
1317
|
|
|
|
|
|
|
|
1318
|
|
|
|
|
|
|
static Idx |
1319
|
|
|
|
|
|
|
internal_function __attribute__ ((pure)) |
1320
|
54
|
|
|
|
|
|
re_node_set_contains (pTHX_ const re_node_set *set, Idx elem) |
1321
|
|
|
|
|
|
|
{ |
1322
|
|
|
|
|
|
|
__re_size_t idx, right, mid; |
1323
|
54
|
50
|
|
|
|
|
if (! REG_VALID_NONZERO_INDEX (set->nelem)) |
1324
|
0
|
|
|
|
|
|
return 0; |
1325
|
|
|
|
|
|
|
|
1326
|
|
|
|
|
|
|
/* Binary search the element. */ |
1327
|
54
|
|
|
|
|
|
idx = 0; |
1328
|
54
|
|
|
|
|
|
right = set->nelem - 1; |
1329
|
128
|
100
|
|
|
|
|
while (idx < right) |
1330
|
|
|
|
|
|
|
{ |
1331
|
74
|
|
|
|
|
|
mid = (idx + right) / 2; |
1332
|
74
|
100
|
|
|
|
|
if (set->elems[mid] < elem) |
1333
|
44
|
|
|
|
|
|
idx = mid + 1; |
1334
|
|
|
|
|
|
|
else |
1335
|
30
|
|
|
|
|
|
right = mid; |
1336
|
|
|
|
|
|
|
} |
1337
|
54
|
100
|
|
|
|
|
return set->elems[idx] == elem ? idx + 1 : 0; |
1338
|
|
|
|
|
|
|
} |
1339
|
|
|
|
|
|
|
|
1340
|
|
|
|
|
|
|
static void |
1341
|
|
|
|
|
|
|
internal_function |
1342
|
0
|
|
|
|
|
|
re_node_set_remove_at (pTHX_ re_node_set *set, Idx idx) |
1343
|
|
|
|
|
|
|
{ |
1344
|
0
|
0
|
|
|
|
|
if (idx < 0 || idx >= set->nelem) |
1345
|
0
|
|
|
|
|
|
return; |
1346
|
0
|
|
|
|
|
|
--set->nelem; |
1347
|
0
|
0
|
|
|
|
|
for (; idx < set->nelem; idx++) |
1348
|
0
|
|
|
|
|
|
set->elems[idx] = set->elems[idx + 1]; |
1349
|
|
|
|
|
|
|
} |
1350
|
|
|
|
|
|
|
|
1351
|
|
|
|
|
|
|
|
1352
|
|
|
|
|
|
|
/* Add the token TOKEN to dfa->nodes, and return the index of the token. |
1353
|
|
|
|
|
|
|
Or return REG_MISSING if an error occurred. */ |
1354
|
|
|
|
|
|
|
|
1355
|
|
|
|
|
|
|
static Idx |
1356
|
|
|
|
|
|
|
internal_function |
1357
|
48
|
|
|
|
|
|
re_dfa_add_node (pTHX_ re_dfa_t *dfa, re_token_t token) |
1358
|
|
|
|
|
|
|
{ |
1359
|
48
|
50
|
|
|
|
|
if (BE (dfa->nodes_len >= dfa->nodes_alloc, 0)) |
1360
|
|
|
|
|
|
|
{ |
1361
|
0
|
|
|
|
|
|
size_t new_nodes_alloc = dfa->nodes_alloc * 2; |
1362
|
|
|
|
|
|
|
|
1363
|
|
|
|
|
|
|
/* Avoid overflows in realloc. */ |
1364
|
0
|
|
|
|
|
|
const size_t max_object_size = MAX (sizeof (re_token_t), |
1365
|
|
|
|
|
|
|
MAX (sizeof (re_node_set), |
1366
|
|
|
|
|
|
|
sizeof (Idx))); |
1367
|
0
|
0
|
|
|
|
|
if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) < new_nodes_alloc, 0)) |
1368
|
0
|
|
|
|
|
|
return REG_MISSING; |
1369
|
|
|
|
|
|
|
|
1370
|
0
|
0
|
|
|
|
|
re_realloc (dfa->nodes, re_token_t, new_nodes_alloc); |
1371
|
0
|
0
|
|
|
|
|
re_realloc (dfa->nexts, Idx, new_nodes_alloc); |
1372
|
0
|
0
|
|
|
|
|
re_realloc (dfa->org_indices, Idx, new_nodes_alloc); |
1373
|
0
|
0
|
|
|
|
|
re_realloc (dfa->edests, re_node_set, new_nodes_alloc); |
1374
|
0
|
0
|
|
|
|
|
re_realloc (dfa->eclosures, re_node_set, new_nodes_alloc); |
1375
|
0
|
|
|
|
|
|
dfa->nodes_alloc = new_nodes_alloc; |
1376
|
|
|
|
|
|
|
} |
1377
|
48
|
|
|
|
|
|
dfa->nodes[dfa->nodes_len] = token; |
1378
|
48
|
|
|
|
|
|
dfa->nodes[dfa->nodes_len].constraint = 0; |
1379
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
1380
|
96
|
|
|
|
|
|
dfa->nodes[dfa->nodes_len].accept_mb = |
1381
|
0
|
0
|
|
|
|
|
((token.type == OP_PERIOD && dfa->mb_cur_max > 1) |
1382
|
48
|
50
|
|
|
|
|
|| token.type == COMPLEX_BRACKET); |
|
|
100
|
|
|
|
|
|
1383
|
|
|
|
|
|
|
#endif |
1384
|
48
|
|
|
|
|
|
dfa->nexts[dfa->nodes_len] = REG_MISSING; |
1385
|
48
|
|
|
|
|
|
re_node_set_init_empty (dfa->edests + dfa->nodes_len); |
1386
|
48
|
|
|
|
|
|
re_node_set_init_empty (dfa->eclosures + dfa->nodes_len); |
1387
|
48
|
|
|
|
|
|
return dfa->nodes_len++; |
1388
|
|
|
|
|
|
|
} |
1389
|
|
|
|
|
|
|
|
1390
|
|
|
|
|
|
|
static re_hashval_t |
1391
|
|
|
|
|
|
|
internal_function |
1392
|
93
|
|
|
|
|
|
calc_state_hash (pTHX_ const re_node_set *nodes, unsigned int context) |
1393
|
|
|
|
|
|
|
{ |
1394
|
93
|
|
|
|
|
|
re_hashval_t hash = nodes->nelem + context; |
1395
|
|
|
|
|
|
|
Idx i; |
1396
|
279
|
100
|
|
|
|
|
for (i = 0 ; i < nodes->nelem ; i++) |
1397
|
186
|
|
|
|
|
|
hash += nodes->elems[i]; |
1398
|
93
|
|
|
|
|
|
return hash; |
1399
|
|
|
|
|
|
|
} |
1400
|
|
|
|
|
|
|
|
1401
|
|
|
|
|
|
|
/* Search for the state whose node_set is equivalent to NODES. |
1402
|
|
|
|
|
|
|
Return the pointer to the state, if we found it in the DFA. |
1403
|
|
|
|
|
|
|
Otherwise create the new one and return it. In case of an error |
1404
|
|
|
|
|
|
|
return NULL and set the error code in ERR. |
1405
|
|
|
|
|
|
|
Note: - We assume NULL as the invalid state, then it is possible that |
1406
|
|
|
|
|
|
|
return value is NULL and ERR is REG_NOERROR. |
1407
|
|
|
|
|
|
|
- We never return non-NULL value in case of any errors, it is for |
1408
|
|
|
|
|
|
|
optimization. */ |
1409
|
|
|
|
|
|
|
|
1410
|
|
|
|
|
|
|
static re_dfastate_t * |
1411
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1412
|
38
|
|
|
|
|
|
re_acquire_state (pTHX_ reg_errcode_t *err, const re_dfa_t *dfa, |
1413
|
|
|
|
|
|
|
const re_node_set *nodes) |
1414
|
|
|
|
|
|
|
{ |
1415
|
|
|
|
|
|
|
re_hashval_t hash; |
1416
|
|
|
|
|
|
|
re_dfastate_t *new_state; |
1417
|
|
|
|
|
|
|
struct re_state_table_entry *spot; |
1418
|
|
|
|
|
|
|
Idx i; |
1419
|
|
|
|
|
|
|
#ifdef lint |
1420
|
|
|
|
|
|
|
/* Suppress bogus uninitialized-variable warnings. */ |
1421
|
|
|
|
|
|
|
*err = REG_NOERROR; |
1422
|
|
|
|
|
|
|
#endif |
1423
|
38
|
50
|
|
|
|
|
if (BE (nodes->nelem == 0, 0)) |
1424
|
|
|
|
|
|
|
{ |
1425
|
0
|
|
|
|
|
|
*err = REG_NOERROR; |
1426
|
0
|
|
|
|
|
|
return NULL; |
1427
|
|
|
|
|
|
|
} |
1428
|
38
|
|
|
|
|
|
hash = calc_state_hash (aTHX_ nodes, 0); |
1429
|
38
|
|
|
|
|
|
spot = dfa->state_table + (hash & dfa->state_hash_mask); |
1430
|
|
|
|
|
|
|
|
1431
|
38
|
100
|
|
|
|
|
for (i = 0 ; i < spot->num ; i++) |
1432
|
|
|
|
|
|
|
{ |
1433
|
31
|
|
|
|
|
|
re_dfastate_t *state = spot->array[i]; |
1434
|
31
|
50
|
|
|
|
|
if (hash != state->hash) |
1435
|
0
|
|
|
|
|
|
continue; |
1436
|
31
|
50
|
|
|
|
|
if (re_node_set_compare (aTHX_ &state->nodes, nodes)) |
1437
|
31
|
|
|
|
|
|
return state; |
1438
|
|
|
|
|
|
|
} |
1439
|
|
|
|
|
|
|
|
1440
|
|
|
|
|
|
|
/* There are no appropriate state in the dfa, create the new one. */ |
1441
|
7
|
|
|
|
|
|
new_state = create_ci_newstate (aTHX_ dfa, nodes, hash); |
1442
|
7
|
50
|
|
|
|
|
if (BE (new_state == NULL, 0)) |
1443
|
0
|
|
|
|
|
|
*err = REG_ESPACE; |
1444
|
|
|
|
|
|
|
|
1445
|
7
|
|
|
|
|
|
return new_state; |
1446
|
|
|
|
|
|
|
} |
1447
|
|
|
|
|
|
|
|
1448
|
|
|
|
|
|
|
/* Search for the state whose node_set is equivalent to NODES and |
1449
|
|
|
|
|
|
|
whose context is equivalent to CONTEXT. |
1450
|
|
|
|
|
|
|
Return the pointer to the state, if we found it in the DFA. |
1451
|
|
|
|
|
|
|
Otherwise create the new one and return it. In case of an error |
1452
|
|
|
|
|
|
|
return NULL and set the error code in ERR. |
1453
|
|
|
|
|
|
|
Note: - We assume NULL as the invalid state, then it is possible that |
1454
|
|
|
|
|
|
|
return value is NULL and ERR is REG_NOERROR. |
1455
|
|
|
|
|
|
|
- We never return non-NULL value in case of any errors, it is for |
1456
|
|
|
|
|
|
|
optimization. */ |
1457
|
|
|
|
|
|
|
|
1458
|
|
|
|
|
|
|
static re_dfastate_t * |
1459
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1460
|
55
|
|
|
|
|
|
re_acquire_state_context (pTHX_ reg_errcode_t *err, const re_dfa_t *dfa, |
1461
|
|
|
|
|
|
|
const re_node_set *nodes, unsigned int context) |
1462
|
|
|
|
|
|
|
{ |
1463
|
|
|
|
|
|
|
re_hashval_t hash; |
1464
|
|
|
|
|
|
|
re_dfastate_t *new_state; |
1465
|
|
|
|
|
|
|
struct re_state_table_entry *spot; |
1466
|
|
|
|
|
|
|
Idx i; |
1467
|
|
|
|
|
|
|
#ifdef lint |
1468
|
|
|
|
|
|
|
/* Suppress bogus uninitialized-variable warnings. */ |
1469
|
|
|
|
|
|
|
*err = REG_NOERROR; |
1470
|
|
|
|
|
|
|
#endif |
1471
|
55
|
50
|
|
|
|
|
if (nodes->nelem == 0) |
1472
|
|
|
|
|
|
|
{ |
1473
|
0
|
|
|
|
|
|
*err = REG_NOERROR; |
1474
|
0
|
|
|
|
|
|
return NULL; |
1475
|
|
|
|
|
|
|
} |
1476
|
55
|
|
|
|
|
|
hash = calc_state_hash (aTHX_ nodes, context); |
1477
|
55
|
|
|
|
|
|
spot = dfa->state_table + (hash & dfa->state_hash_mask); |
1478
|
|
|
|
|
|
|
|
1479
|
58
|
100
|
|
|
|
|
for (i = 0 ; i < spot->num ; i++) |
1480
|
|
|
|
|
|
|
{ |
1481
|
35
|
|
|
|
|
|
re_dfastate_t *state = spot->array[i]; |
1482
|
35
|
50
|
|
|
|
|
if (state->hash == hash |
1483
|
35
|
50
|
|
|
|
|
&& state->context == context |
1484
|
35
|
100
|
|
|
|
|
&& re_node_set_compare (aTHX_ state->entrance_nodes, nodes)) |
1485
|
32
|
|
|
|
|
|
return state; |
1486
|
|
|
|
|
|
|
} |
1487
|
|
|
|
|
|
|
/* There are no appropriate state in 'dfa', create the new one. */ |
1488
|
23
|
|
|
|
|
|
new_state = create_cd_newstate (aTHX_ dfa, nodes, context, hash); |
1489
|
23
|
50
|
|
|
|
|
if (BE (new_state == NULL, 0)) |
1490
|
0
|
|
|
|
|
|
*err = REG_ESPACE; |
1491
|
|
|
|
|
|
|
|
1492
|
23
|
|
|
|
|
|
return new_state; |
1493
|
|
|
|
|
|
|
} |
1494
|
|
|
|
|
|
|
|
1495
|
|
|
|
|
|
|
/* Finish initialization of the new state NEWSTATE, and using its hash value |
1496
|
|
|
|
|
|
|
HASH put in the appropriate bucket of DFA's state table. Return value |
1497
|
|
|
|
|
|
|
indicates the error code if failed. */ |
1498
|
|
|
|
|
|
|
|
1499
|
|
|
|
|
|
|
static reg_errcode_t |
1500
|
|
|
|
|
|
|
__attribute_warn_unused_result__ |
1501
|
30
|
|
|
|
|
|
register_state (pTHX_ const re_dfa_t *dfa, re_dfastate_t *newstate, |
1502
|
|
|
|
|
|
|
re_hashval_t hash) |
1503
|
|
|
|
|
|
|
{ |
1504
|
|
|
|
|
|
|
struct re_state_table_entry *spot; |
1505
|
|
|
|
|
|
|
reg_errcode_t err; |
1506
|
|
|
|
|
|
|
Idx i; |
1507
|
|
|
|
|
|
|
|
1508
|
30
|
|
|
|
|
|
newstate->hash = hash; |
1509
|
30
|
|
|
|
|
|
err = re_node_set_alloc (aTHX_ &newstate->non_eps_nodes, newstate->nodes.nelem); |
1510
|
30
|
50
|
|
|
|
|
if (BE (err != REG_NOERROR, 0)) |
1511
|
0
|
|
|
|
|
|
return REG_ESPACE; |
1512
|
80
|
100
|
|
|
|
|
for (i = 0; i < newstate->nodes.nelem; i++) |
1513
|
|
|
|
|
|
|
{ |
1514
|
50
|
|
|
|
|
|
Idx elem = newstate->nodes.elems[i]; |
1515
|
50
|
100
|
|
|
|
|
if (!IS_EPSILON_NODE (dfa->nodes[elem].type)) |
1516
|
32
|
50
|
|
|
|
|
if (! re_node_set_insert_last (aTHX_ &newstate->non_eps_nodes, elem)) |
1517
|
0
|
|
|
|
|
|
return REG_ESPACE; |
1518
|
|
|
|
|
|
|
} |
1519
|
|
|
|
|
|
|
|
1520
|
30
|
|
|
|
|
|
spot = dfa->state_table + (hash & dfa->state_hash_mask); |
1521
|
30
|
100
|
|
|
|
|
if (BE (spot->alloc <= spot->num, 0)) |
1522
|
|
|
|
|
|
|
{ |
1523
|
27
|
|
|
|
|
|
Idx new_alloc = 2 * spot->num + 2; |
1524
|
27
|
50
|
|
|
|
|
re_realloc (spot->array, re_dfastate_t *, new_alloc); |
1525
|
27
|
|
|
|
|
|
spot->alloc = new_alloc; |
1526
|
|
|
|
|
|
|
} |
1527
|
30
|
|
|
|
|
|
spot->array[spot->num++] = newstate; |
1528
|
30
|
|
|
|
|
|
return REG_NOERROR; |
1529
|
|
|
|
|
|
|
} |
1530
|
|
|
|
|
|
|
|
1531
|
|
|
|
|
|
|
static void |
1532
|
30
|
|
|
|
|
|
free_state (pTHX_ re_dfastate_t *state) |
1533
|
|
|
|
|
|
|
{ |
1534
|
30
|
50
|
|
|
|
|
re_node_set_free (&state->non_eps_nodes); |
1535
|
30
|
100
|
|
|
|
|
re_node_set_free (&state->inveclosure); |
1536
|
30
|
50
|
|
|
|
|
if (state->entrance_nodes != &state->nodes) |
1537
|
|
|
|
|
|
|
{ |
1538
|
0
|
0
|
|
|
|
|
re_node_set_free (state->entrance_nodes); |
1539
|
0
|
0
|
|
|
|
|
re_free (state->entrance_nodes); |
1540
|
|
|
|
|
|
|
} |
1541
|
30
|
50
|
|
|
|
|
re_node_set_free (&state->nodes); |
1542
|
30
|
50
|
|
|
|
|
re_free (state->word_trtable); |
1543
|
30
|
100
|
|
|
|
|
re_free (state->trtable); |
1544
|
30
|
50
|
|
|
|
|
re_free (state); |
1545
|
30
|
|
|
|
|
|
} |
1546
|
|
|
|
|
|
|
|
1547
|
|
|
|
|
|
|
/* Create the new state which is independent of contexts. |
1548
|
|
|
|
|
|
|
Return the new state if succeeded, otherwise return NULL. */ |
1549
|
|
|
|
|
|
|
|
1550
|
|
|
|
|
|
|
static re_dfastate_t * |
1551
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1552
|
7
|
|
|
|
|
|
create_ci_newstate (pTHX_ const re_dfa_t *dfa, const re_node_set *nodes, |
1553
|
|
|
|
|
|
|
re_hashval_t hash) |
1554
|
|
|
|
|
|
|
{ |
1555
|
|
|
|
|
|
|
Idx i; |
1556
|
|
|
|
|
|
|
reg_errcode_t err; |
1557
|
|
|
|
|
|
|
re_dfastate_t *newstate; |
1558
|
|
|
|
|
|
|
|
1559
|
7
|
|
|
|
|
|
re_calloc(newstate, re_dfastate_t, 1); |
1560
|
|
|
|
|
|
|
/* newstate = (re_dfastate_t *) calloc (sizeof (re_dfastate_t), 1); */ |
1561
|
7
|
50
|
|
|
|
|
if (BE (newstate == NULL, 0)) |
1562
|
0
|
|
|
|
|
|
return NULL; |
1563
|
7
|
|
|
|
|
|
err = re_node_set_init_copy (aTHX_ &newstate->nodes, nodes); |
1564
|
7
|
50
|
|
|
|
|
if (BE (err != REG_NOERROR, 0)) |
1565
|
|
|
|
|
|
|
{ |
1566
|
0
|
0
|
|
|
|
|
re_free (newstate); |
1567
|
0
|
|
|
|
|
|
return NULL; |
1568
|
|
|
|
|
|
|
} |
1569
|
|
|
|
|
|
|
|
1570
|
7
|
|
|
|
|
|
newstate->entrance_nodes = &newstate->nodes; |
1571
|
18
|
100
|
|
|
|
|
for (i = 0 ; i < nodes->nelem ; i++) |
1572
|
|
|
|
|
|
|
{ |
1573
|
11
|
|
|
|
|
|
re_token_t *node = dfa->nodes + nodes->elems[i]; |
1574
|
11
|
|
|
|
|
|
re_token_type_t type = node->type; |
1575
|
11
|
100
|
|
|
|
|
if (type == CHARACTER && !node->constraint) |
|
|
50
|
|
|
|
|
|
1576
|
1
|
|
|
|
|
|
continue; |
1577
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
1578
|
10
|
|
|
|
|
|
newstate->accept_mb |= node->accept_mb; |
1579
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
1580
|
|
|
|
|
|
|
|
1581
|
|
|
|
|
|
|
/* If the state has the halt node, the state is a halt state. */ |
1582
|
10
|
100
|
|
|
|
|
if (type == END_OF_RE) |
1583
|
2
|
|
|
|
|
|
newstate->halt = 1; |
1584
|
8
|
50
|
|
|
|
|
else if (type == OP_BACK_REF) |
1585
|
0
|
|
|
|
|
|
newstate->has_backref = 1; |
1586
|
8
|
50
|
|
|
|
|
else if (type == ANCHOR || node->constraint) |
|
|
50
|
|
|
|
|
|
1587
|
0
|
|
|
|
|
|
newstate->has_constraint = 1; |
1588
|
|
|
|
|
|
|
} |
1589
|
7
|
|
|
|
|
|
err = register_state (aTHX_ dfa, newstate, hash); |
1590
|
7
|
50
|
|
|
|
|
if (BE (err != REG_NOERROR, 0)) |
1591
|
|
|
|
|
|
|
{ |
1592
|
0
|
|
|
|
|
|
free_state (aTHX_ newstate); |
1593
|
0
|
|
|
|
|
|
newstate = NULL; |
1594
|
|
|
|
|
|
|
} |
1595
|
7
|
|
|
|
|
|
return newstate; |
1596
|
|
|
|
|
|
|
} |
1597
|
|
|
|
|
|
|
|
1598
|
|
|
|
|
|
|
/* Create the new state which is depend on the context CONTEXT. |
1599
|
|
|
|
|
|
|
Return the new state if succeeded, otherwise return NULL. */ |
1600
|
|
|
|
|
|
|
|
1601
|
|
|
|
|
|
|
static re_dfastate_t * |
1602
|
|
|
|
|
|
|
internal_function __attribute_warn_unused_result__ |
1603
|
23
|
|
|
|
|
|
create_cd_newstate (pTHX_ const re_dfa_t *dfa, const re_node_set *nodes, |
1604
|
|
|
|
|
|
|
unsigned int context, re_hashval_t hash) |
1605
|
|
|
|
|
|
|
{ |
1606
|
23
|
|
|
|
|
|
Idx i, nctx_nodes = 0; |
1607
|
|
|
|
|
|
|
reg_errcode_t err; |
1608
|
|
|
|
|
|
|
re_dfastate_t *newstate; |
1609
|
|
|
|
|
|
|
|
1610
|
23
|
|
|
|
|
|
re_calloc(newstate, re_dfastate_t, 1); |
1611
|
|
|
|
|
|
|
/* newstate = (re_dfastate_t *) calloc (sizeof (re_dfastate_t), 1); */ |
1612
|
23
|
50
|
|
|
|
|
if (BE (newstate == NULL, 0)) |
1613
|
0
|
|
|
|
|
|
return NULL; |
1614
|
23
|
|
|
|
|
|
err = re_node_set_init_copy (aTHX_ &newstate->nodes, nodes); |
1615
|
23
|
50
|
|
|
|
|
if (BE (err != REG_NOERROR, 0)) |
1616
|
|
|
|
|
|
|
{ |
1617
|
0
|
0
|
|
|
|
|
re_free (newstate); |
1618
|
0
|
|
|
|
|
|
return NULL; |
1619
|
|
|
|
|
|
|
} |
1620
|
|
|
|
|
|
|
|
1621
|
23
|
|
|
|
|
|
newstate->context = context; |
1622
|
23
|
|
|
|
|
|
newstate->entrance_nodes = &newstate->nodes; |
1623
|
|
|
|
|
|
|
|
1624
|
62
|
100
|
|
|
|
|
for (i = 0 ; i < nodes->nelem ; i++) |
1625
|
|
|
|
|
|
|
{ |
1626
|
39
|
|
|
|
|
|
re_token_t *node = dfa->nodes + nodes->elems[i]; |
1627
|
39
|
|
|
|
|
|
re_token_type_t type = node->type; |
1628
|
39
|
|
|
|
|
|
unsigned int constraint = node->constraint; |
1629
|
|
|
|
|
|
|
|
1630
|
39
|
100
|
|
|
|
|
if (type == CHARACTER && !constraint) |
|
|
50
|
|
|
|
|
|
1631
|
16
|
|
|
|
|
|
continue; |
1632
|
|
|
|
|
|
|
#ifdef RE_ENABLE_I18N |
1633
|
23
|
|
|
|
|
|
newstate->accept_mb |= node->accept_mb; |
1634
|
|
|
|
|
|
|
#endif /* RE_ENABLE_I18N */ |
1635
|
|
|
|
|
|
|
|
1636
|
|
|
|
|
|
|
/* If the state has the halt node, the state is a halt state. */ |
1637
|
23
|
100
|
|
|
|
|
if (type == END_OF_RE) |
1638
|
5
|
|
|
|
|
|
newstate->halt = 1; |
1639
|
18
|
50
|
|
|
|
|
else if (type == OP_BACK_REF) |
1640
|
0
|
|
|
|
|
|
newstate->has_backref = 1; |
1641
|
|
|
|
|
|
|
|
1642
|
23
|
50
|
|
|
|
|
if (constraint) |
1643
|
|
|
|
|
|
|
{ |
1644
|
0
|
0
|
|
|
|
|
if (newstate->entrance_nodes == &newstate->nodes) |
1645
|
|
|
|
|
|
|
{ |
1646
|
0
|
|
|
|
|
|
re_malloc (newstate->entrance_nodes, re_node_set, 1); |
1647
|
0
|
0
|
|
|
|
|
if (re_node_set_init_copy (aTHX_ newstate->entrance_nodes, nodes) |
1648
|
|
|
|
|
|
|
!= REG_NOERROR) |
1649
|
0
|
|
|
|
|
|
return NULL; |
1650
|
0
|
|
|
|
|
|
nctx_nodes = 0; |
1651
|
0
|
|
|
|
|
|
newstate->has_constraint = 1; |
1652
|
|
|
|
|
|
|
} |
1653
|
|
|
|
|
|
|
|
1654
|
0
|
0
|
|
|
|
|
if (NOT_SATISFY_PREV_CONSTRAINT (constraint,context)) |
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
1655
|
|
|
|
|
|
|
{ |
1656
|
0
|
|
|
|
|
|
re_node_set_remove_at (aTHX_ &newstate->nodes, i - nctx_nodes); |
1657
|
0
|
|
|
|
|
|
++nctx_nodes; |
1658
|
|
|
|
|
|
|
} |
1659
|
|
|
|
|
|
|
} |
1660
|
|
|
|
|
|
|
} |
1661
|
23
|
|
|
|
|
|
err = register_state (aTHX_ dfa, newstate, hash); |
1662
|
23
|
50
|
|
|
|
|
if (BE (err != REG_NOERROR, 0)) |
1663
|
|
|
|
|
|
|
{ |
1664
|
0
|
|
|
|
|
|
free_state (aTHX_ newstate); |
1665
|
0
|
|
|
|
|
|
newstate = NULL; |
1666
|
|
|
|
|
|
|
} |
1667
|
23
|
|
|
|
|
|
return newstate; |
1668
|
|
|
|
|
|
|
} |