1 |
/* |
2 |
* Copyright (C) 2006-2007 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 1.1 2007/06/20 14:25:14 debug Exp $ |
29 |
* |
30 |
* C++ symbol name demangling. |
31 |
* |
32 |
* For obvious performance reasons, the external c++filt utility cannot be |
33 |
* used. Also, the host's version of this utility might be incompatible with |
34 |
* the 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; |
69 |
size_t result_len = 0, len; |
70 |
int first = 1, type_added = 0, pointercounter, reference; |
71 |
|
72 |
CHECK_ALLOCATION(result = malloc(MAXLEN + 1)); |
73 |
result[0] = '\0'; |
74 |
|
75 |
while (name[0] && name[0] != 'E' && result_len < MAXLEN) { |
76 |
/* Read length of the next part: */ |
77 |
len = 0; |
78 |
if (*name == '0') { |
79 |
name ++; |
80 |
} else { |
81 |
while (isdigit((int)*name)) { |
82 |
len *= 10; |
83 |
len += (*name - '0'); |
84 |
name ++; |
85 |
} |
86 |
} |
87 |
|
88 |
/* Add :: */ |
89 |
if (!first) |
90 |
add_string(result, &result_len, "::"); |
91 |
|
92 |
/* Read the part itself: */ |
93 |
while (len-- >= 1 && result_len < MAXLEN) |
94 |
result[result_len ++] = *name++; |
95 |
|
96 |
first = 0; |
97 |
} |
98 |
|
99 |
if (name[0] != 'E') |
100 |
goto fail; |
101 |
|
102 |
name ++; |
103 |
|
104 |
if (*name) |
105 |
add_string(result, &result_len, "{"); |
106 |
|
107 |
/* Type: */ |
108 |
pointercounter = reference = 0; |
109 |
while (*name) { |
110 |
int argument_done = 0; |
111 |
char t = *name++; |
112 |
switch (t) { |
113 |
case 'c': |
114 |
add_string(result, &result_len, "char"); |
115 |
argument_done = 1; |
116 |
break; |
117 |
case 'a': |
118 |
add_string(result, &result_len, "signed char"); |
119 |
argument_done = 1; |
120 |
break; |
121 |
case 'h': |
122 |
add_string(result, &result_len, "unsigned char"); |
123 |
argument_done = 1; |
124 |
break; |
125 |
case 'i': |
126 |
add_string(result, &result_len, "int"); |
127 |
argument_done = 1; |
128 |
break; |
129 |
case 'j': |
130 |
add_string(result, &result_len, "unsigned int"); |
131 |
argument_done = 1; |
132 |
break; |
133 |
case 'w': |
134 |
add_string(result, &result_len, "wchar_t"); |
135 |
argument_done = 1; |
136 |
break; |
137 |
case 'b': |
138 |
add_string(result, &result_len, "bool"); |
139 |
argument_done = 1; |
140 |
break; |
141 |
case 's': |
142 |
add_string(result, &result_len, "short"); |
143 |
argument_done = 1; |
144 |
break; |
145 |
case 't': |
146 |
add_string(result, &result_len, "unsigned short"); |
147 |
argument_done = 1; |
148 |
break; |
149 |
case 'l': |
150 |
add_string(result, &result_len, "long"); |
151 |
argument_done = 1; |
152 |
break; |
153 |
case 'm': |
154 |
add_string(result, &result_len, "unsigned long"); |
155 |
argument_done = 1; |
156 |
break; |
157 |
case 'x': |
158 |
add_string(result, &result_len, "long long"); |
159 |
argument_done = 1; |
160 |
break; |
161 |
case 'y': |
162 |
add_string(result, &result_len, "unsigned long long"); |
163 |
argument_done = 1; |
164 |
break; |
165 |
case 'n': |
166 |
add_string(result, &result_len, "__int128"); |
167 |
argument_done = 1; |
168 |
break; |
169 |
case 'o': |
170 |
add_string(result, &result_len, "unsigned __int128"); |
171 |
argument_done = 1; |
172 |
break; |
173 |
case 'f': |
174 |
add_string(result, &result_len, "float"); |
175 |
argument_done = 1; |
176 |
break; |
177 |
case 'd': |
178 |
add_string(result, &result_len, "double"); |
179 |
argument_done = 1; |
180 |
break; |
181 |
case 'e': |
182 |
add_string(result, &result_len, "__float80"); |
183 |
argument_done = 1; |
184 |
break; |
185 |
case 'g': |
186 |
add_string(result, &result_len, "__float128"); |
187 |
argument_done = 1; |
188 |
break; |
189 |
case 'z': |
190 |
add_string(result, &result_len, "..."); |
191 |
argument_done = 1; |
192 |
break; |
193 |
case 'P': |
194 |
pointercounter ++; |
195 |
break; |
196 |
case 'R': |
197 |
reference ++; |
198 |
break; |
199 |
case 'v': /* void */ |
200 |
break; |
201 |
default:/* Unknown */ |
202 |
goto fail; |
203 |
} |
204 |
if (argument_done) { |
205 |
while (pointercounter-- > 0) |
206 |
add_string(result, &result_len, "*"); |
207 |
while (reference-- > 0) |
208 |
add_string(result, &result_len, "&"); |
209 |
if (*name) |
210 |
add_string(result, &result_len, ","); |
211 |
} |
212 |
type_added = 1; |
213 |
} |
214 |
|
215 |
if (type_added) |
216 |
add_string(result, &result_len, "}"); |
217 |
|
218 |
if (result_len == MAXLEN) |
219 |
goto fail; |
220 |
|
221 |
result[result_len] = '\0'; |
222 |
|
223 |
return result; |
224 |
|
225 |
fail: |
226 |
free(result); |
227 |
return NULL; |
228 |
} |
229 |
|
230 |
|
231 |
/* |
232 |
* symbol_demangle_cplusplus(): |
233 |
* |
234 |
* Try to demangle name. If name was not a valid/known C++ symbol, then NULL |
235 |
* is returned. Otherwise, a newly allocated string is returned, containing |
236 |
* the demangled name. |
237 |
*/ |
238 |
char *symbol_demangle_cplusplus(char *name) |
239 |
{ |
240 |
/* Only support _Z-style mangled names, for now: */ |
241 |
if (strlen(name) < 2 || name[0] != '_' || name[1] != 'Z') |
242 |
return NULL; |
243 |
|
244 |
name += 2; |
245 |
|
246 |
switch (name[0]) { |
247 |
case 'N': |
248 |
return symbol_demangle_cplusplus_nested(name + 1); |
249 |
break; |
250 |
} |
251 |
|
252 |
return NULL; |
253 |
} |
254 |
|
255 |
|
256 |
|
257 |
#ifdef TEST |
258 |
|
259 |
void test(char *mangled, char *result) |
260 |
{ |
261 |
char *p = symbol_demangle_cplusplus(mangled); |
262 |
if (p == NULL) { |
263 |
if (result == NULL) { |
264 |
return; |
265 |
} else { |
266 |
printf("FAILURE for %s!\n", mangled); |
267 |
exit(1); |
268 |
} |
269 |
} |
270 |
if (strcmp(p, result) == 0) |
271 |
return; |
272 |
printf("FAILURE for %s! (result = %s)\n", mangled, p); |
273 |
exit(1); |
274 |
} |
275 |
|
276 |
int main(int argc, char *argv[]) |
277 |
{ |
278 |
test("monkey", NULL); |
279 |
test("_monkey", NULL); |
280 |
test("_zmonkey", NULL); |
281 |
test("_Zmonkey", NULL); |
282 |
test("_ZQ5abcde", NULL); |
283 |
test("_ZN3abc5defghE", "abc::defgh"); |
284 |
test("_ZN05defghEv", "::defgh{}"); |
285 |
test("_ZN5defghEv", "defgh{}"); |
286 |
test("_ZN3abc5defghEv", "abc::defgh{}"); |
287 |
test("_ZN3abc5defghEc", "abc::defgh{char}"); |
288 |
test("_ZN1a2bcEjij", "a::bc{unsigned int,int,unsigned int}"); |
289 |
printf("OK\n"); |
290 |
return 0; |
291 |
} |
292 |
|
293 |
#endif |
294 |
|