1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
|
To: vim_dev@googlegroups.com
Subject: Patch 7.4.036
Fcc: outbox
From: Bram Moolenaar <Bram@moolenaar.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
------------
Patch 7.4.036
Problem: NFA engine does not capture group correctly when using \@>. (ZyX)
Solution: Copy submatches before doing the recursive match.
Files: src/regexp_nfa.c, src/testdir/test64.in, src/testdir/test64.ok
*** ../vim-7.4.035/src/regexp_nfa.c 2013-09-22 13:57:19.000000000 +0200
--- src/regexp_nfa.c 2013-09-25 16:35:54.000000000 +0200
***************
*** 36,42 ****
{
NFA_SPLIT = -1024,
NFA_MATCH,
! NFA_SKIP_CHAR, /* matches a 0-length char */
NFA_START_COLL, /* [abc] start */
NFA_END_COLL, /* [abc] end */
--- 36,42 ----
{
NFA_SPLIT = -1024,
NFA_MATCH,
! NFA_EMPTY, /* matches 0-length */
NFA_START_COLL, /* [abc] start */
NFA_END_COLL, /* [abc] end */
***************
*** 2005,2012 ****
{
/* Ignore result of previous call to nfa_regatom() */
post_ptr = post_start + my_post_start;
! /* NFA_SKIP_CHAR has 0-length and works everywhere */
! EMIT(NFA_SKIP_CHAR);
return OK;
}
--- 2005,2012 ----
{
/* Ignore result of previous call to nfa_regatom() */
post_ptr = post_start + my_post_start;
! /* NFA_EMPTY is 0-length and works everywhere */
! EMIT(NFA_EMPTY);
return OK;
}
***************
*** 2170,2185 ****
old_post_pos = (int)(post_ptr - post_start);
if (nfa_regconcat() == FAIL)
return FAIL;
! /* if concat is empty, skip a input char. But do emit a node */
if (old_post_pos == (int)(post_ptr - post_start))
! EMIT(NFA_SKIP_CHAR);
EMIT(NFA_CONCAT);
ch = peekchr();
}
! /* Even if a branch is empty, emit one node for it */
if (old_post_pos == (int)(post_ptr - post_start))
! EMIT(NFA_SKIP_CHAR);
return OK;
}
--- 2170,2185 ----
old_post_pos = (int)(post_ptr - post_start);
if (nfa_regconcat() == FAIL)
return FAIL;
! /* if concat is empty do emit a node */
if (old_post_pos == (int)(post_ptr - post_start))
! EMIT(NFA_EMPTY);
EMIT(NFA_CONCAT);
ch = peekchr();
}
! /* if a branch is empty, emit one node for it */
if (old_post_pos == (int)(post_ptr - post_start))
! EMIT(NFA_EMPTY);
return OK;
}
***************
*** 2423,2429 ****
case NFA_STAR_NONGREEDY: STRCPY(code, "NFA_STAR_NONGREEDY "); break;
case NFA_QUEST: STRCPY(code, "NFA_QUEST"); break;
case NFA_QUEST_NONGREEDY: STRCPY(code, "NFA_QUEST_NON_GREEDY"); break;
! case NFA_SKIP_CHAR: STRCPY(code, "NFA_SKIP_CHAR"); break;
case NFA_OR: STRCPY(code, "NFA_OR"); break;
case NFA_START_COLL: STRCPY(code, "NFA_START_COLL"); break;
--- 2423,2429 ----
case NFA_STAR_NONGREEDY: STRCPY(code, "NFA_STAR_NONGREEDY "); break;
case NFA_QUEST: STRCPY(code, "NFA_QUEST"); break;
case NFA_QUEST_NONGREEDY: STRCPY(code, "NFA_QUEST_NON_GREEDY"); break;
! case NFA_EMPTY: STRCPY(code, "NFA_EMPTY"); break;
case NFA_OR: STRCPY(code, "NFA_OR"); break;
case NFA_START_COLL: STRCPY(code, "NFA_START_COLL"); break;
***************
*** 3067,3073 ****
case NFA_ZSTART:
case NFA_ZEND:
case NFA_OPT_CHARS:
! case NFA_SKIP_CHAR:
case NFA_START_PATTERN:
case NFA_END_PATTERN:
case NFA_COMPOSING:
--- 3067,3073 ----
case NFA_ZSTART:
case NFA_ZEND:
case NFA_OPT_CHARS:
! case NFA_EMPTY:
case NFA_START_PATTERN:
case NFA_END_PATTERN:
case NFA_COMPOSING:
***************
*** 3265,3279 ****
PUSH(frag(e1.start, e2.out));
break;
! case NFA_SKIP_CHAR:
! /* Symbol of 0-length, Used in a repetition
! * with max/min count of 0 */
if (nfa_calc_size == TRUE)
{
nstate++;
break;
}
! s = alloc_state(NFA_SKIP_CHAR, NULL, NULL);
if (s == NULL)
goto theend;
PUSH(frag(s, list1(&s->out)));
--- 3265,3278 ----
PUSH(frag(e1.start, e2.out));
break;
! case NFA_EMPTY:
! /* 0-length, used in a repetition with max/min count of 0 */
if (nfa_calc_size == TRUE)
{
nstate++;
break;
}
! s = alloc_state(NFA_EMPTY, NULL, NULL);
if (s == NULL)
goto theend;
PUSH(frag(s, list1(&s->out)));
***************
*** 4209,4215 ****
case NFA_MOPEN:
case NFA_ZEND:
case NFA_SPLIT:
! case NFA_SKIP_CHAR:
/* These nodes are not added themselves but their "out" and/or
* "out1" may be added below. */
break;
--- 4208,4214 ----
case NFA_MOPEN:
case NFA_ZEND:
case NFA_SPLIT:
! case NFA_EMPTY:
/* These nodes are not added themselves but their "out" and/or
* "out1" may be added below. */
break;
***************
*** 4337,4343 ****
subs = addstate(l, state->out1, subs, pim, off);
break;
! case NFA_SKIP_CHAR:
case NFA_NOPEN:
case NFA_NCLOSE:
subs = addstate(l, state->out, subs, pim, off);
--- 4336,4342 ----
subs = addstate(l, state->out1, subs, pim, off);
break;
! case NFA_EMPTY:
case NFA_NOPEN:
case NFA_NCLOSE:
subs = addstate(l, state->out, subs, pim, off);
***************
*** 5604,5612 ****
{
int in_use = m->norm.in_use;
! /* Copy submatch info for the recursive call, so that
! * \1 can be matched. */
copy_sub_off(&m->norm, &t->subs.norm);
/*
* First try matching the invisible match, then what
--- 5603,5615 ----
{
int in_use = m->norm.in_use;
! /* Copy submatch info for the recursive call, opposite
! * of what happens on success below. */
copy_sub_off(&m->norm, &t->subs.norm);
+ #ifdef FEAT_SYN_HL
+ if (nfa_has_zsubexpr)
+ copy_sub_off(&m->synt, &t->subs.synt);
+ #endif
/*
* First try matching the invisible match, then what
***************
*** 5713,5718 ****
--- 5716,5728 ----
#endif
break;
}
+ /* Copy submatch info to the recursive call, opposite of what
+ * happens afterwards. */
+ copy_sub_off(&m->norm, &t->subs.norm);
+ #ifdef FEAT_SYN_HL
+ if (nfa_has_zsubexpr)
+ copy_sub_off(&m->synt, &t->subs.synt);
+ #endif
/* First try matching the pattern. */
result = recursive_regmatch(t->state, NULL, prog,
*** ../vim-7.4.035/src/testdir/test64.in 2013-09-22 13:57:19.000000000 +0200
--- src/testdir/test64.in 2013-09-25 15:51:12.000000000 +0200
***************
*** 430,435 ****
--- 430,436 ----
:call add(tl, [2, '\(a*\)\@>a', 'aaaa'])
:call add(tl, [2, '\(a*\)\@>b', 'aaab', 'aaab', 'aaa'])
:call add(tl, [2, '^\(.\{-}b\)\@>.', ' abcbd', ' abc', ' ab'])
+ :call add(tl, [2, '\(.\{-}\)\(\)\@>$', 'abc', 'abc', 'abc', ''])
:" TODO: BT engine does not restore submatch after failure
:call add(tl, [1, '\(a*\)\@>a\|a\+', 'aaaa', 'aaaa'])
:"
*** ../vim-7.4.035/src/testdir/test64.ok 2013-09-22 13:57:19.000000000 +0200
--- src/testdir/test64.ok 2013-09-25 16:39:31.000000000 +0200
***************
*** 992,997 ****
--- 992,1000 ----
OK 0 - ^\(.\{-}b\)\@>.
OK 1 - ^\(.\{-}b\)\@>.
OK 2 - ^\(.\{-}b\)\@>.
+ OK 0 - \(.\{-}\)\(\)\@>$
+ OK 1 - \(.\{-}\)\(\)\@>$
+ OK 2 - \(.\{-}\)\(\)\@>$
OK 0 - \(a*\)\@>a\|a\+
OK 2 - \(a*\)\@>a\|a\+
OK 0 - \_[^8-9]\+
*** ../vim-7.4.035/src/version.c 2013-09-22 15:43:34.000000000 +0200
--- src/version.c 2013-09-25 16:40:01.000000000 +0200
***************
*** 740,741 ****
--- 740,743 ----
{ /* Add new patch number below this line */
+ /**/
+ 36,
/**/
--
There is a fine line between courage and foolishness.
Unfortunately, it's not a fence.
/// Bram Moolenaar -- Bram@Moolenaar.net -- http://www.Moolenaar.net \\\
/// sponsor Vim, vote for features -- http://www.Vim.org/sponsor/ \\\
\\\ an exciting new programming language -- http://www.Zimbu.org ///
\\\ help me help AIDS victims -- http://ICCF-Holland.org ///
|
