1 /* Shared functions related to mangling names for the GNU compiler
2 for the Java(TM) language.
3 Copyright (C) 2001, 2002, 2003, 2007 Free Software Foundation, Inc.
5 This file is part of GCC.
7 GCC is free software; you can redistribute it and/or modify
8 it under the terms of the GNU General Public License as published by
9 the Free Software Foundation; either version 3, or (at your option)
12 GCC is distributed in the hope that it will be useful,
13 but WITHOUT ANY WARRANTY; without even the implied warranty of
14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 GNU General Public License for more details.
17 You should have received a copy of the GNU General Public License
18 along with GCC; see the file COPYING3. If not see
19 <http://www.gnu.org/licenses/>.
21 Java and all Java-based marks are trademarks or registered trademarks
22 of Sun Microsystems, Inc. in the United States and other countries.
23 The Free Software Foundation is independent of Sun Microsystems, Inc. */
25 /* Written by Alexandre Petit-Bianco <apbianco@cygnus.com> */
29 #include "coretypes.h"
32 #include "java-tree.h"
36 static void append_unicode_mangled_name (const char *, int);
38 static int unicode_mangling_length (const char *, int);
41 extern struct obstack *mangle_obstack;
44 utf8_cmp (const unsigned char *str, int length, const char *name)
46 const unsigned char *limit = str + length;
49 for (i = 0; name[i]; ++i)
51 int ch = UTF8_GET (str, limit);
56 return str == limit ? 0 : 1;
59 /* A sorted list of all C++ keywords. If you change this, be sure
60 also to change the list in
61 libjava/classpath/tools/gnu/classpath/tools/javah/Keywords.java. */
62 static const char *const cxx_keywords[] =
170 /* Return true if NAME is a C++ keyword. */
172 cxx_keyword_p (const char *name, int length)
174 int last = ARRAY_SIZE (cxx_keywords);
176 int mid = (last + first) / 2;
179 for (mid = (last + first) / 2;
181 old = mid, mid = (last + first) / 2)
183 int kwl = strlen (cxx_keywords[mid]);
184 int min_length = kwl > length ? length : kwl;
185 int r = utf8_cmp ((const unsigned char *) name, min_length, cxx_keywords[mid]);
190 /* We've found a match if all the remaining characters are `$'. */
191 for (i = min_length; i < length && name[i] == '$'; ++i)
206 /* If NAME happens to be a C++ keyword, add `$'. */
207 #define MANGLE_CXX_KEYWORDS(NAME, LEN) \
210 if (cxx_keyword_p ((NAME), (LEN))) \
212 char *tmp_buf = (char *)alloca ((LEN)+1); \
213 memcpy (tmp_buf, (NAME), (LEN)); \
222 /* If the assembler doesn't support UTF8 in symbol names, some
223 characters might need to be escaped. */
227 /* Assuming (NAME, LEN) is a Utf8-encoding string, emit the string
228 appropriately mangled (with Unicode escapes if needed) to
229 MANGLE_OBSTACK. Note that `java', `lang' and `Object' are used so
230 frequently that they could be cached. */
233 append_gpp_mangled_name (const char *name, int len)
235 int encoded_len, needs_escapes;
238 MANGLE_CXX_KEYWORDS (name, len);
240 encoded_len = unicode_mangling_length (name, len);
241 needs_escapes = encoded_len > 0;
243 sprintf (buf, "%d", (needs_escapes ? encoded_len : len));
244 obstack_grow (mangle_obstack, buf, strlen (buf));
247 append_unicode_mangled_name (name, len);
249 obstack_grow (mangle_obstack, name, len);
252 /* Assuming (NAME, LEN) is a Utf8-encoded string, emit the string
253 appropriately mangled (with Unicode escapes) to MANGLE_OBSTACK.
254 Characters needing an escape are encoded `__UNN_' to `__UNNNN_', in
255 which case `__U' will be mangled `__U_'. */
258 append_unicode_mangled_name (const char *name, int len)
260 const unsigned char *ptr;
261 const unsigned char *limit = (const unsigned char *)name + len;
263 for (ptr = (const unsigned char *) name; ptr < limit; )
265 int ch = UTF8_GET(ptr, limit);
267 if ((ISALNUM (ch) && ch != 'U') || ch == '$')
269 obstack_1grow (mangle_obstack, ch);
272 /* Everything else needs encoding */
276 if (ch == '_' || ch == 'U')
278 /* Prepare to recognize __U */
279 if (ch == '_' && (uuU < 3))
282 obstack_1grow (mangle_obstack, ch);
284 /* We recognize __U that we wish to encode
285 __U_. Finish the encoding. */
286 else if (ch == 'U' && (uuU == 2))
289 obstack_grow (mangle_obstack, "U_", 2);
291 /* Otherwise, just reset uuU and emit the character we
296 obstack_1grow (mangle_obstack, ch);
300 sprintf (buf, "__U%x_", ch);
301 obstack_grow (mangle_obstack, buf, strlen (buf));
307 /* Assuming (NAME, LEN) is a Utf8-encoding string, calculate the
308 length of the string as mangled (a la g++) including Unicode
309 escapes. If no escapes are needed, return 0. */
312 unicode_mangling_length (const char *name, int len)
314 const unsigned char *ptr;
315 const unsigned char *limit = (const unsigned char *)name + len;
316 int need_escapes = 0; /* Whether we need an escape or not */
317 int num_chars = 0; /* Number of characters in the mangled name */
318 int uuU = 0; /* Help us to find __U. 0: '_', 1: '__' */
319 for (ptr = (const unsigned char *) name; ptr < limit; )
321 int ch = UTF8_GET(ptr, limit);
324 error ("internal error - invalid Utf8 name");
325 if ((ISALNUM (ch) && ch != 'U') || ch == '$')
330 /* Everything else needs encoding */
333 int encoding_length = 2;
335 if (ch == '_' || ch == 'U')
337 /* It's always at least one character. */
340 /* Prepare to recognize __U */
341 if (ch == '_' && (uuU < 3))
344 /* We recognize __U that we wish to encode __U_, we
345 count one more character. */
346 else if (ch == 'U' && (uuU == 2))
352 /* Otherwise, just reset uuU */
364 num_chars += (4 + encoding_length);
377 /* The assembler supports UTF8, we don't use escapes. Mangling is
378 simply <N>NAME. <N> is the number of UTF8 encoded characters that
379 are found in NAME. Note that `java', `lang' and `Object' are used
380 so frequently that they could be cached. */
383 append_gpp_mangled_name (const char *name, int len)
385 const unsigned char *ptr;
386 const unsigned char *limit;
390 MANGLE_CXX_KEYWORDS (name, len);
392 limit = (const unsigned char *)name + len;
394 /* Compute the length of the string we wish to mangle. */
395 for (encoded_len = 0, ptr = (const unsigned char *) name;
396 ptr < limit; encoded_len++)
398 int ch = UTF8_GET(ptr, limit);
401 error ("internal error - invalid Utf8 name");
404 sprintf (buf, "%d", encoded_len);
405 obstack_grow (mangle_obstack, buf, strlen (buf));
406 obstack_grow (mangle_obstack, name, len);
409 #endif /* HAVE_AS_UTF8 */