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
|
// Copyright 2014 Paul Sokolovsky.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
#include "re1.5.h"
// Matches: DSWdsw
#define MATCH_NAMED_CLASS_CHAR(c) (((c) | 0x20) == 'd' || ((c) | 0x24) == 'w')
#define INSERT_CODE(at, num, pc) \
((code ? memmove(code + at + num, code + at, pc - at) : 0), pc += num)
#define REL(at, to) (to - at - 2)
#define EMIT(at, byte) (code ? (code[at] = byte) : (at))
#define EMIT_CHECKED(at, byte) (_emit_checked(at, code, byte, &err))
#define PC (prog->bytelen)
static void _emit_checked(int at, char *code, int val, bool *err) {
*err |= val != (int8_t)val;
if (code) {
code[at] = val;
}
}
static const char *_compilecode(const char *re, ByteProg *prog, int sizecode)
{
char *code = sizecode ? NULL : prog->insts;
bool err = false;
int start = PC;
int term = PC;
int alt_label = 0;
for (; *re && *re != ')'; re++) {
switch (*re) {
case '\\':
re++;
if (!*re) return NULL; // Trailing backslash
if (MATCH_NAMED_CLASS_CHAR(*re)) {
term = PC;
EMIT(PC++, NamedClass);
EMIT(PC++, *re);
prog->len++;
break;
}
MP_FALLTHROUGH
default:
term = PC;
EMIT(PC++, Char);
EMIT(PC++, *re);
prog->len++;
break;
case '.':
term = PC;
EMIT(PC++, Any);
prog->len++;
break;
case '[': {
int cnt;
term = PC;
re++;
if (*re == '^') {
EMIT(PC++, ClassNot);
re++;
} else {
EMIT(PC++, Class);
}
PC++; // Skip # of pair byte
prog->len++;
for (cnt = 0; *re != ']'; re++, cnt++) {
char c = *re;
if (c == '\\') {
++re;
c = *re;
if (MATCH_NAMED_CLASS_CHAR(c)) {
c = RE15_CLASS_NAMED_CLASS_INDICATOR;
goto emit_char_pair;
}
}
if (!c) return NULL;
if (re[1] == '-' && re[2] != ']') {
re += 2;
}
emit_char_pair:
EMIT(PC++, c);
EMIT(PC++, *re);
}
EMIT_CHECKED(term + 1, cnt);
break;
}
case '(': {
term = PC;
int sub = 0;
int capture = re[1] != '?' || re[2] != ':';
if (capture) {
sub = ++prog->sub;
EMIT(PC++, Save);
EMIT_CHECKED(PC++, 2 * sub);
prog->len++;
} else {
re += 2;
}
re = _compilecode(re + 1, prog, sizecode);
if (re == NULL || *re != ')') return NULL; // error, or no matching paren
if (capture) {
EMIT(PC++, Save);
EMIT_CHECKED(PC++, 2 * sub + 1);
prog->len++;
}
break;
}
case '?':
if (PC == term) return NULL; // nothing to repeat
INSERT_CODE(term, 2, PC);
if (re[1] == '?') {
EMIT(term, RSplit);
re++;
} else {
EMIT(term, Split);
}
EMIT_CHECKED(term + 1, REL(term, PC));
prog->len++;
term = PC;
break;
case '*':
if (PC == term) return NULL; // nothing to repeat
INSERT_CODE(term, 2, PC);
EMIT(PC, Jmp);
EMIT_CHECKED(PC + 1, REL(PC, term));
PC += 2;
if (re[1] == '?') {
EMIT(term, RSplit);
re++;
} else {
EMIT(term, Split);
}
EMIT_CHECKED(term + 1, REL(term, PC));
prog->len += 2;
term = PC;
break;
case '+':
if (PC == term) return NULL; // nothing to repeat
if (re[1] == '?') {
EMIT(PC, Split);
re++;
} else {
EMIT(PC, RSplit);
}
EMIT_CHECKED(PC + 1, REL(PC, term));
PC += 2;
prog->len++;
term = PC;
break;
case '|':
if (alt_label) {
EMIT_CHECKED(alt_label, REL(alt_label, PC) + 1);
}
INSERT_CODE(start, 2, PC);
EMIT(PC++, Jmp);
alt_label = PC++;
EMIT(start, Split);
EMIT_CHECKED(start + 1, REL(start, PC));
prog->len += 2;
term = PC;
break;
case '^':
EMIT(PC++, Bol);
prog->len++;
term = PC;
break;
case '$':
EMIT(PC++, Eol);
prog->len++;
term = PC;
break;
}
}
if (alt_label) {
EMIT_CHECKED(alt_label, REL(alt_label, PC) + 1);
}
return err ? NULL : re;
}
int re1_5_sizecode(const char *re)
{
ByteProg dummyprog = {
// Save 0, Save 1, Match; more bytes for "search" (vs "match") prefix code
.bytelen = 5 + NON_ANCHORED_PREFIX
};
if (_compilecode(re, &dummyprog, /*sizecode*/1) == NULL) return -1;
return dummyprog.bytelen;
}
int re1_5_compilecode(ByteProg *prog, const char *re)
{
prog->len = 0;
prog->bytelen = 0;
prog->sub = 0;
// Add code to implement non-anchored operation ("search"),
// for anchored operation ("match"), this code will be just skipped.
// TODO: Implement search in much more efficient manner
prog->insts[prog->bytelen++] = RSplit;
prog->insts[prog->bytelen++] = 3;
prog->insts[prog->bytelen++] = Any;
prog->insts[prog->bytelen++] = Jmp;
prog->insts[prog->bytelen++] = -5;
prog->len += 3;
prog->insts[prog->bytelen++] = Save;
prog->insts[prog->bytelen++] = 0;
prog->len++;
re = _compilecode(re, prog, /*sizecode*/0);
if (re == NULL || *re) return 1;
prog->insts[prog->bytelen++] = Save;
prog->insts[prog->bytelen++] = 1;
prog->len++;
prog->insts[prog->bytelen++] = Match;
prog->len++;
return 0;
}
#if 0
int main(int argc, char *argv[])
{
int pc = 0;
ByteProg *code = re1_5_compilecode(argv[1]);
re1_5_dumpcode(code);
}
#endif
|
