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
|
/*
* Copyright (C) 2002 Andrew Tridgell
*
* This program is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License as published by the Free
* Software Foundation; either version 3 of the License, or (at your option)
* any later version.
*
* This program is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*
* You should have received a copy of the GNU General Public License along with
* this program; if not, write to the Free Software Foundation, Inc., 51
* Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
/*
* C/C++ unifier
*
* The idea is that changes that don't affect the resulting C code should not
* change the hash. This is achieved by folding white-space and other
* non-semantic fluff in the input into a single unified format.
*
* This unifier was design to match the output of the unifier in compilercache,
* which is flex based. The major difference is that this unifier is much
* faster (about 2x) and more forgiving of syntactic errors. Continuing on
* syntactic errors is important to cope with C/C++ extensions in the local
* compiler (for example, inline assembly systems).
*/
#include "ccache.h"
static const char *const s_tokens[] = {
"...", ">>=", "<<=", "+=", "-=", "*=", "/=", "%=", "&=", "^=",
"|=", ">>", "<<", "++", "--", "->", "&&", "||", "<=", ">=",
"==", "!=", ";", "{", "<%", "}", "%>", ",", ":", "=",
"(", ")", "[", "<:", "]", ":>", ".", "&", "!", "~",
"-", "+", "*", "/", "%", "<", ">", "^", "|", "?",
0
};
#define C_ALPHA 1
#define C_SPACE 2
#define C_TOKEN 4
#define C_QUOTE 8
#define C_DIGIT 16
#define C_HEX 32
#define C_FLOAT 64
#define C_SIGN 128
static struct {
unsigned char type;
unsigned char num_toks;
const char *toks[7];
} tokens[256];
/* build up the table used by the unifier */
static void
build_table(void)
{
unsigned char c;
int i;
static bool done;
if (done) return;
done = true;
memset(tokens, 0, sizeof(tokens));
for (c = 0; c < 128; c++) {
if (isalpha(c) || c == '_') tokens[c].type |= C_ALPHA;
if (isdigit(c)) tokens[c].type |= C_DIGIT;
if (isspace(c)) tokens[c].type |= C_SPACE;
if (isxdigit(c)) tokens[c].type |= C_HEX;
}
tokens['\''].type |= C_QUOTE;
tokens['"'].type |= C_QUOTE;
tokens['l'].type |= C_FLOAT;
tokens['L'].type |= C_FLOAT;
tokens['f'].type |= C_FLOAT;
tokens['F'].type |= C_FLOAT;
tokens['U'].type |= C_FLOAT;
tokens['u'].type |= C_FLOAT;
tokens['-'].type |= C_SIGN;
tokens['+'].type |= C_SIGN;
for (i = 0; s_tokens[i]; i++) {
c = s_tokens[i][0];
tokens[c].type |= C_TOKEN;
tokens[c].toks[tokens[c].num_toks] = s_tokens[i];
tokens[c].num_toks++;
}
}
/* buffer up characters before hashing them */
static void
pushchar(struct mdfour *hash, unsigned char c)
{
static unsigned char buf[64];
static size_t len;
if (c == 0) {
if (len > 0) {
hash_buffer(hash, (char *)buf, len);
len = 0;
}
hash_buffer(hash, NULL, 0);
return;
}
buf[len++] = c;
if (len == 64) {
hash_buffer(hash, (char *)buf, len);
len = 0;
}
}
/* hash some C/C++ code after unifying */
static void
unify(struct mdfour *hash, unsigned char *p, size_t size)
{
size_t ofs;
unsigned char q;
int i;
build_table();
for (ofs = 0; ofs < size;) {
if (p[ofs] == '#') {
if ((size-ofs) > 2 && p[ofs+1] == ' ' && isdigit(p[ofs+2])) {
do {
ofs++;
} while (ofs < size && p[ofs] != '\n');
ofs++;
} else {
do {
pushchar(hash, p[ofs]);
ofs++;
} while (ofs < size && p[ofs] != '\n');
pushchar(hash, '\n');
ofs++;
}
continue;
}
if (tokens[p[ofs]].type & C_ALPHA) {
do {
pushchar(hash, p[ofs]);
ofs++;
} while (ofs < size && (tokens[p[ofs]].type & (C_ALPHA|C_DIGIT)));
pushchar(hash, '\n');
continue;
}
if (tokens[p[ofs]].type & C_DIGIT) {
do {
pushchar(hash, p[ofs]);
ofs++;
} while (ofs < size &&
((tokens[p[ofs]].type & C_DIGIT) || p[ofs] == '.'));
if (ofs < size && (p[ofs] == 'x' || p[ofs] == 'X')) {
do {
pushchar(hash, p[ofs]);
ofs++;
} while (ofs < size && (tokens[p[ofs]].type & C_HEX));
}
if (ofs < size && (p[ofs] == 'E' || p[ofs] == 'e')) {
pushchar(hash, p[ofs]);
ofs++;
while (ofs < size && (tokens[p[ofs]].type & (C_DIGIT|C_SIGN))) {
pushchar(hash, p[ofs]);
ofs++;
}
}
while (ofs < size && (tokens[p[ofs]].type & C_FLOAT)) {
pushchar(hash, p[ofs]);
ofs++;
}
pushchar(hash, '\n');
continue;
}
if (tokens[p[ofs]].type & C_SPACE) {
do {
ofs++;
} while (ofs < size && (tokens[p[ofs]].type & C_SPACE));
continue;
}
if (tokens[p[ofs]].type & C_QUOTE) {
q = p[ofs];
pushchar(hash, p[ofs]);
do {
ofs++;
while (ofs < size-1 && p[ofs] == '\\') {
pushchar(hash, p[ofs]);
pushchar(hash, p[ofs+1]);
ofs += 2;
}
pushchar(hash, p[ofs]);
} while (ofs < size && p[ofs] != q);
pushchar(hash, '\n');
ofs++;
continue;
}
if (tokens[p[ofs]].type & C_TOKEN) {
q = p[ofs];
for (i = 0; i < tokens[q].num_toks; i++) {
unsigned char *s = (unsigned char *)tokens[q].toks[i];
int len = strlen((char *)s);
if (size >= ofs+len && memcmp(&p[ofs], s, len) == 0) {
int j;
for (j = 0; s[j]; j++) {
pushchar(hash, s[j]);
ofs++;
}
pushchar(hash, '\n');
break;
}
}
if (i < tokens[q].num_toks) {
continue;
}
}
pushchar(hash, p[ofs]);
pushchar(hash, '\n');
ofs++;
}
pushchar(hash, 0);
}
/* hash a file that consists of preprocessor output, but remove any line
number information from the hash
*/
int
unify_hash(struct mdfour *hash, const char *fname)
{
char *data;
size_t size;
if (!read_file(fname, 0, &data, &size)) {
stats_update(STATS_PREPROCESSOR);
return -1;
}
unify(hash, (unsigned char *)data, size);
free(data);
return 0;
}
|
