1 /* Shared functions related to mangling names for the GNU compiler
2 for the Java(TM) language.
3 Copyright (C) 2001, 2003 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 2, 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 COPYING. If not, write to
19 the Free Software Foundation, 59 Temple Place - Suite 330,
20 Boston, MA 02111-1307, USA.
22 Java and all Java-based marks are trademarks or registered trademarks
23 of Sun Microsystems, Inc. in the United States and other countries.
24 The Free Software Foundation is independent of Sun Microsystems, Inc. */
26 /* Written by Alexandre Petit-Bianco <apbianco@cygnus.com> */
30 #include "coretypes.h"
34 #include "java-tree.h"
38 static void append_unicode_mangled_name PARAMS ((const char *, int));
40 static int unicode_mangling_length PARAMS ((const char *, int));
43 extern struct obstack *mangle_obstack;
45 /* If the assembler doesn't support UTF8 in symbol names, some
46 characters might need to be escaped. */
50 /* Assuming (NAME, LEN) is a Utf8-encoding string, emit the string
51 appropriately mangled (with Unicode escapes if needed) to
52 MANGLE_OBSTACK. Note that `java', `lang' and `Object' are used so
53 frequently that they could be cached. */
56 append_gpp_mangled_name (name, len)
60 int encoded_len = unicode_mangling_length (name, len);
61 int needs_escapes = encoded_len > 0;
64 sprintf (buf, "%d", (needs_escapes ? encoded_len : len));
65 obstack_grow (mangle_obstack, buf, strlen (buf));
68 append_unicode_mangled_name (name, len);
70 obstack_grow (mangle_obstack, name, len);
73 /* Assuming (NAME, LEN) is a Utf8-encoded string, emit the string
74 appropriately mangled (with Unicode escapes) to MANGLE_OBSTACK.
75 Characters needing an escape are encoded `__UNN_' to `__UNNNN_', in
76 which case `__U' will be mangled `__U_'. */
79 append_unicode_mangled_name (name, len)
83 const unsigned char *ptr;
84 const unsigned char *limit = (const unsigned char *)name + len;
86 for (ptr = (const unsigned char *) name; ptr < limit; )
88 int ch = UTF8_GET(ptr, limit);
90 if ((ISALNUM (ch) && ch != 'U') || ch == '$')
91 obstack_1grow (mangle_obstack, ch);
92 /* Everything else needs encoding */
96 if (ch == '_' || ch == 'U')
98 /* Prepare to recognize __U */
99 if (ch == '_' && (uuU < 3))
102 obstack_1grow (mangle_obstack, ch);
104 /* We recognize __U that we wish to encode
105 __U_. Finish the encoding. */
106 else if (ch == 'U' && (uuU == 2))
109 obstack_grow (mangle_obstack, "U_", 2);
111 /* Otherwise, just reset uuU and emit the character we
116 obstack_1grow (mangle_obstack, ch);
120 sprintf (buf, "__U%x_", ch);
121 obstack_grow (mangle_obstack, buf, strlen (buf));
127 /* Assuming (NAME, LEN) is a Utf8-encoding string, calculate the
128 length of the string as mangled (a la g++) including Unicode
129 escapes. If no escapes are needed, return 0. */
132 unicode_mangling_length (name, len)
136 const unsigned char *ptr;
137 const unsigned char *limit = (const unsigned char *)name + len;
138 int need_escapes = 0; /* Whether we need an escape or not */
139 int num_chars = 0; /* Number of characters in the mangled name */
140 int uuU = 0; /* Help us to find __U. 0: '_', 1: '__' */
141 for (ptr = (const unsigned char *) name; ptr < limit; )
143 int ch = UTF8_GET(ptr, limit);
146 error ("internal error - invalid Utf8 name");
147 if ((ISALNUM (ch) && ch != 'U') || ch == '$')
149 /* Everything else needs encoding */
152 int encoding_length = 2;
154 if (ch == '_' || ch == 'U')
156 /* It's always at least one character. */
159 /* Prepare to recognize __U */
160 if (ch == '_' && (uuU < 3))
163 /* We recognize __U that we wish to encode __U_, we
164 count one more character. */
165 else if (ch == 'U' && (uuU == 2))
171 /* Otherwise, just reset uuU */
183 num_chars += (4 + encoding_length);
196 /* The assembler supports UTF8, we don't use escapes. Mangling is
197 simply <N>NAME. <N> is the number of UTF8 encoded characters that
198 are found in NAME. Note that `java', `lang' and `Object' are used
199 so frequently that they could be cached. */
202 append_gpp_mangled_name (name, len)
206 const unsigned char *ptr;
207 const unsigned char *limit = (const unsigned char *)name + len;
211 /* Compute the length of the string we wish to mangle. */
212 for (encoded_len = 0, ptr = (const unsigned char *) name;
213 ptr < limit; encoded_len++)
215 int ch = UTF8_GET(ptr, limit);
218 error ("internal error - invalid Utf8 name");
221 sprintf (buf, "%d", encoded_len);
222 obstack_grow (mangle_obstack, buf, strlen (buf));
223 obstack_grow (mangle_obstack, name, len);
226 #endif /* HAVE_AS_UTF8 */