1 |
/* |
2 |
* Copyright (C) 2006 Anders Gavare. All rights reserved. |
3 |
* |
4 |
* Redistribution and use in source and binary forms, with or without |
5 |
* modification, are permitted provided that the following conditions are met: |
6 |
* |
7 |
* 1. Redistributions of source code must retain the above copyright |
8 |
* notice, this list of conditions and the following disclaimer. |
9 |
* 2. Redistributions in binary form must reproduce the above copyright |
10 |
* notice, this list of conditions and the following disclaimer in the |
11 |
* documentation and/or other materials provided with the distribution. |
12 |
* 3. The name of the author may not be used to endorse or promote products |
13 |
* derived from this software without specific prior written permission. |
14 |
* |
15 |
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND |
16 |
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
17 |
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
18 |
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE |
19 |
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL |
20 |
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS |
21 |
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) |
22 |
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
23 |
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY |
24 |
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF |
25 |
* SUCH DAMAGE. |
26 |
* |
27 |
* |
28 |
* $Id: symbol_demangle.c,v 1.5 2006/05/01 09:00:19 debug Exp $ |
29 |
* |
30 |
* C++ symbol name demangling. |
31 |
* |
32 |
* For obvious performance reasons, we cannot call the external c++filt |
33 |
* utility. The host's version of this utility might be incompatible with the |
34 |
* binary being emulated. |
35 |
* |
36 |
* TODO: Constructors, destructors, and lots of other stuff. See |
37 |
* http://www.codesourcery.com/cxx-abi/abi.html#mangling for details. |
38 |
*/ |
39 |
|
40 |
#include <stdio.h> |
41 |
#include <stdlib.h> |
42 |
#include <string.h> |
43 |
#include <ctype.h> |
44 |
|
45 |
#include "symbol.h" |
46 |
|
47 |
|
48 |
#define MAXLEN 1000 |
49 |
|
50 |
|
51 |
static void add_string(char *p, size_t *curlenp, char *to_add) |
52 |
{ |
53 |
size_t curlen = *curlenp; |
54 |
while (curlen < MAXLEN && *to_add) |
55 |
p[curlen++] = *to_add++; |
56 |
*curlenp = curlen; |
57 |
} |
58 |
|
59 |
|
60 |
/* |
61 |
* symbol_demangle_cplusplus_nested(): |
62 |
* |
63 |
* Try to demangle a nested cplusplus name. name points to the first character |
64 |
* after "_ZN". |
65 |
*/ |
66 |
static char *symbol_demangle_cplusplus_nested(char *name) |
67 |
{ |
68 |
char *result = malloc(MAXLEN + 1); |
69 |
size_t result_len = 0, len; |
70 |
int first = 1, type_added = 0, pointercounter, reference; |
71 |
|
72 |
result[0] = '\0'; |
73 |
|
74 |
while (name[0] && name[0] != 'E' && result_len < MAXLEN) { |
75 |
/* Read length of the next part: */ |
76 |
len = 0; |
77 |
if (*name == '0') { |
78 |
name ++; |
79 |
} else { |
80 |
while (isdigit((int)*name)) { |
81 |
len *= 10; |
82 |
len += (*name - '0'); |
83 |
name ++; |
84 |
} |
85 |
} |
86 |
|
87 |
/* Add :: */ |
88 |
if (!first) |
89 |
add_string(result, &result_len, "::"); |
90 |
|
91 |
/* Read the part itself: */ |
92 |
while (len-- >= 1 && result_len < MAXLEN) |
93 |
result[result_len ++] = *name++; |
94 |
|
95 |
first = 0; |
96 |
} |
97 |
|
98 |
if (name[0] != 'E') |
99 |
goto fail; |
100 |
|
101 |
name ++; |
102 |
|
103 |
if (*name) |
104 |
add_string(result, &result_len, "{"); |
105 |
|
106 |
/* Type: */ |
107 |
pointercounter = reference = 0; |
108 |
while (*name) { |
109 |
int argument_done = 0; |
110 |
char t = *name++; |
111 |
switch (t) { |
112 |
case 'c': |
113 |
add_string(result, &result_len, "char"); |
114 |
argument_done = 1; |
115 |
break; |
116 |
case 'a': |
117 |
add_string(result, &result_len, "signed char"); |
118 |
argument_done = 1; |
119 |
break; |
120 |
case 'h': |
121 |
add_string(result, &result_len, "unsigned char"); |
122 |
argument_done = 1; |
123 |
break; |
124 |
case 'i': |
125 |
add_string(result, &result_len, "int"); |
126 |
argument_done = 1; |
127 |
break; |
128 |
case 'j': |
129 |
add_string(result, &result_len, "unsigned int"); |
130 |
argument_done = 1; |
131 |
break; |
132 |
case 'w': |
133 |
add_string(result, &result_len, "wchar_t"); |
134 |
argument_done = 1; |
135 |
break; |
136 |
case 'b': |
137 |
add_string(result, &result_len, "bool"); |
138 |
argument_done = 1; |
139 |
break; |
140 |
case 's': |
141 |
add_string(result, &result_len, "short"); |
142 |
argument_done = 1; |
143 |
break; |
144 |
case 't': |
145 |
add_string(result, &result_len, "unsigned short"); |
146 |
argument_done = 1; |
147 |
break; |
148 |
case 'l': |
149 |
add_string(result, &result_len, "long"); |
150 |
argument_done = 1; |
151 |
break; |
152 |
case 'm': |
153 |
add_string(result, &result_len, "unsigned long"); |
154 |
argument_done = 1; |
155 |
break; |
156 |
case 'x': |
157 |
add_string(result, &result_len, "long long"); |
158 |
argument_done = 1; |
159 |
break; |
160 |
case 'y': |
161 |
add_string(result, &result_len, "unsigned long long"); |
162 |
argument_done = 1; |
163 |
break; |
164 |
case 'n': |
165 |
add_string(result, &result_len, "__int128"); |
166 |
argument_done = 1; |
167 |
break; |
168 |
case 'o': |
169 |
add_string(result, &result_len, "unsigned __int128"); |
170 |
argument_done = 1; |
171 |
break; |
172 |
case 'f': |
173 |
add_string(result, &result_len, "float"); |
174 |
argument_done = 1; |
175 |
break; |
176 |
case 'd': |
177 |
add_string(result, &result_len, "double"); |
178 |
argument_done = 1; |
179 |
break; |
180 |
case 'e': |
181 |
add_string(result, &result_len, "__float80"); |
182 |
argument_done = 1; |
183 |
break; |
184 |
case 'g': |
185 |
add_string(result, &result_len, "__float128"); |
186 |
argument_done = 1; |
187 |
break; |
188 |
case 'z': |
189 |
add_string(result, &result_len, "..."); |
190 |
argument_done = 1; |
191 |
break; |
192 |
case 'P': |
193 |
pointercounter ++; |
194 |
break; |
195 |
case 'R': |
196 |
reference ++; |
197 |
break; |
198 |
case 'v': /* void */ |
199 |
break; |
200 |
default:/* Unknown */ |
201 |
goto fail; |
202 |
} |
203 |
if (argument_done) { |
204 |
while (pointercounter-- > 0) |
205 |
add_string(result, &result_len, "*"); |
206 |
while (reference-- > 0) |
207 |
add_string(result, &result_len, "&"); |
208 |
if (*name) |
209 |
add_string(result, &result_len, ","); |
210 |
} |
211 |
type_added = 1; |
212 |
} |
213 |
|
214 |
if (type_added) |
215 |
add_string(result, &result_len, "}"); |
216 |
|
217 |
if (result_len == MAXLEN) |
218 |
goto fail; |
219 |
|
220 |
result[result_len] = '\0'; |
221 |
|
222 |
return result; |
223 |
|
224 |
fail: |
225 |
free(result); |
226 |
return NULL; |
227 |
} |
228 |
|
229 |
|
230 |
/* |
231 |
* symbol_demangle_cplusplus(): |
232 |
* |
233 |
* Try to demangle name. If name was not a valid/known C++ symbol, then NULL |
234 |
* is returned. Otherwise, a newly allocated string is returned, containing |
235 |
* the demangled name. |
236 |
*/ |
237 |
char *symbol_demangle_cplusplus(char *name) |
238 |
{ |
239 |
/* Only support _Z-style mangled names, for now: */ |
240 |
if (strlen(name) < 2 || name[0] != '_' || name[1] != 'Z') |
241 |
return NULL; |
242 |
|
243 |
name += 2; |
244 |
|
245 |
switch (name[0]) { |
246 |
case 'N': |
247 |
return symbol_demangle_cplusplus_nested(name + 1); |
248 |
break; |
249 |
} |
250 |
|
251 |
return NULL; |
252 |
} |
253 |
|
254 |
|
255 |
|
256 |
#ifdef TEST |
257 |
|
258 |
void test(char *mangled, char *result) |
259 |
{ |
260 |
char *p = symbol_demangle_cplusplus(mangled); |
261 |
if (p == NULL) { |
262 |
if (result == NULL) { |
263 |
return; |
264 |
} else { |
265 |
printf("FAILURE for %s!\n", mangled); |
266 |
exit(1); |
267 |
} |
268 |
} |
269 |
if (strcmp(p, result) == 0) |
270 |
return; |
271 |
printf("FAILURE for %s! (result = %s)\n", mangled, p); |
272 |
exit(1); |
273 |
} |
274 |
|
275 |
int main(int argc, char *argv[]) |
276 |
{ |
277 |
test("monkey", NULL); |
278 |
test("_monkey", NULL); |
279 |
test("_zmonkey", NULL); |
280 |
test("_Zmonkey", NULL); |
281 |
test("_ZQ5abcde", NULL); |
282 |
test("_ZN3abc5defghE", "abc::defgh"); |
283 |
test("_ZN05defghEv", "::defgh{}"); |
284 |
test("_ZN5defghEv", "defgh{}"); |
285 |
test("_ZN3abc5defghEv", "abc::defgh{}"); |
286 |
test("_ZN3abc5defghEc", "abc::defgh{char}"); |
287 |
test("_ZN1a2bcEjij", "a::bc{unsigned int,int,unsigned int}"); |
288 |
printf("OK\n"); |
289 |
return 0; |
290 |
} |
291 |
|
292 |
#endif |
293 |
|