| 1 | /* winduni.c -- unicode support for the windres program. |
| 2 | Copyright 1997, 1998, 2000, 2001, 2003, 2007 |
| 3 | Free Software Foundation, Inc. |
| 4 | Written by Ian Lance Taylor, Cygnus Support. |
| 5 | Rewritten by Kai Tietz, Onevision. |
| 6 | |
| 7 | This file is part of GNU Binutils. |
| 8 | |
| 9 | This program is free software; you can redistribute it and/or modify |
| 10 | it under the terms of the GNU General Public License as published by |
| 11 | the Free Software Foundation; either version 2 of the License, or |
| 12 | (at your option) any later version. |
| 13 | |
| 14 | This program is distributed in the hope that it will be useful, |
| 15 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 16 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 17 | GNU General Public License for more details. |
| 18 | |
| 19 | You should have received a copy of the GNU General Public License |
| 20 | along with this program; if not, write to the Free Software |
| 21 | Foundation, Inc., 51 Franklin Street - Fifth Floor, Boston, MA |
| 22 | 02110-1301, USA. */ |
| 23 | |
| 24 | /* This file contains unicode support routines for the windres |
| 25 | program. Ideally, we would have generic unicode support which |
| 26 | would work on all systems. However, we don't. Instead, on a |
| 27 | Windows host, we are prepared to call some Windows routines. This |
| 28 | means that we will generate different output on Windows and Unix |
| 29 | hosts, but that seems better than not really supporting unicode at |
| 30 | all. */ |
| 31 | |
| 32 | #include "sysdep.h" |
| 33 | #include "bfd.h" |
| 34 | #include "bucomm.h" |
| 35 | #include "winduni.h" |
| 36 | #include "safe-ctype.h" |
| 37 | |
| 38 | #ifdef _WIN32 |
| 39 | #include <windows.h> |
| 40 | #endif |
| 41 | |
| 42 | /* Prototypes. */ |
| 43 | static int unichar_isascii (const unichar *, rc_uint_type); |
| 44 | |
| 45 | /* Convert an ASCII string to a unicode string. We just copy it, |
| 46 | expanding chars to shorts, rather than doing something intelligent. */ |
| 47 | |
| 48 | void |
| 49 | unicode_from_ascii (rc_uint_type *length, unichar **unicode, const char *ascii) |
| 50 | { |
| 51 | rc_uint_type len; |
| 52 | #ifndef _WIN32 |
| 53 | const char *s; |
| 54 | unsigned short *w; |
| 55 | |
| 56 | len = strlen (ascii); |
| 57 | *unicode = ((unichar *) res_alloc ((len + 1) * sizeof (unichar))); |
| 58 | for (s = ascii, w = *unicode; *s != '\0'; s++, w++) |
| 59 | *w = *s & 0xff; |
| 60 | *w = 0; |
| 61 | #else |
| 62 | /* We use MultiByteToWideChar rather than strlen to get the unicode |
| 63 | string length to allow multibyte "ascii" chars. The value returned |
| 64 | by this function includes the trailing '\0'. */ |
| 65 | len = (rc_uint_type) MultiByteToWideChar (CP_ACP, 0, ascii, -1, NULL, 0); |
| 66 | if (len) |
| 67 | { |
| 68 | *unicode = ((unichar *) res_alloc (len * sizeof (unichar))); |
| 69 | MultiByteToWideChar (CP_ACP, 0, ascii, -1, *unicode, (int) len); |
| 70 | } |
| 71 | /* Discount the trailing '/0'. If MultiByteToWideChar failed, |
| 72 | this will set *length to -1. */ |
| 73 | len--; |
| 74 | #endif |
| 75 | |
| 76 | if (length != NULL) |
| 77 | *length = len; |
| 78 | } |
| 79 | |
| 80 | /* Convert an unicode string to an ASCII string. We just copy it, |
| 81 | shrink shorts to chars, rather than doing something intelligent. |
| 82 | Shorts with not within the char range are replaced by '_'. */ |
| 83 | |
| 84 | void |
| 85 | ascii_from_unicode (rc_uint_type *length, const unichar *unicode, char **ascii) |
| 86 | { |
| 87 | rc_uint_type len; |
| 88 | #ifndef _WIN32 |
| 89 | char *s; |
| 90 | const unsigned short *w; |
| 91 | |
| 92 | len = 0; |
| 93 | while (unicode[len] != 0) |
| 94 | ++len; |
| 95 | *ascii = ((char *) res_alloc (len + 1)); |
| 96 | for (s = *ascii, w = unicode; *w != '\0'; w++, s++) |
| 97 | { |
| 98 | if(w[0]==(w[0]&0xff)) |
| 99 | *s = (char) w[0]; |
| 100 | else |
| 101 | *s = '_'; |
| 102 | } |
| 103 | *s = 0; |
| 104 | #else |
| 105 | WINBOOL used_def = FALSE; |
| 106 | /* We use MultiByteToWideChar rather than strlen to get the unicode |
| 107 | string length to allow multibyte "ascii" chars. The value returned |
| 108 | by this function includes the trailing '\0'. */ |
| 109 | len = (rc_uint_type) WideCharToMultiByte (CP_ACP, WC_DEFAULTCHAR, unicode, -1, NULL, |
| 110 | 0, "_", &used_def); |
| 111 | if (len) |
| 112 | { |
| 113 | *ascii = (char *) res_alloc (len * sizeof (char)); |
| 114 | WideCharToMultiByte (CP_ACP, WC_DEFAULTCHAR, unicode, -1, *ascii, (int) len, |
| 115 | "_", &used_def); |
| 116 | } |
| 117 | /* Discount the trailing '/0'. If MultiByteToWideChar failed, |
| 118 | this will set *length to -1. */ |
| 119 | len--; |
| 120 | #endif |
| 121 | |
| 122 | if (length != NULL) |
| 123 | *length = len; |
| 124 | } |
| 125 | |
| 126 | /* Print the unicode string UNICODE to the file E. LENGTH is the |
| 127 | number of characters to print, or -1 if we should print until the |
| 128 | end of the string. FIXME: On a Windows host, we should be calling |
| 129 | some Windows function, probably WideCharToMultiByte. */ |
| 130 | |
| 131 | void |
| 132 | unicode_print (FILE *e, const unichar *unicode, rc_uint_type length) |
| 133 | { |
| 134 | while (1) |
| 135 | { |
| 136 | unichar ch; |
| 137 | |
| 138 | if (length == 0) |
| 139 | return; |
| 140 | if ((bfd_signed_vma) length > 0) |
| 141 | --length; |
| 142 | |
| 143 | ch = *unicode; |
| 144 | |
| 145 | if (ch == 0 && (bfd_signed_vma) length < 0) |
| 146 | return; |
| 147 | |
| 148 | ++unicode; |
| 149 | |
| 150 | if ((ch & 0x7f) == ch) |
| 151 | { |
| 152 | if (ch == '\\') |
| 153 | fputs ("\\\\", e); |
| 154 | else if (ch == '"') |
| 155 | fputs ("\"\"", e); |
| 156 | else if (ISPRINT (ch)) |
| 157 | putc (ch, e); |
| 158 | else |
| 159 | { |
| 160 | switch (ch) |
| 161 | { |
| 162 | case ESCAPE_A: |
| 163 | fputs ("\\a", e); |
| 164 | break; |
| 165 | |
| 166 | case ESCAPE_B: |
| 167 | fputs ("\\b", e); |
| 168 | break; |
| 169 | |
| 170 | case ESCAPE_F: |
| 171 | fputs ("\\f", e); |
| 172 | break; |
| 173 | |
| 174 | case ESCAPE_N: |
| 175 | fputs ("\\n", e); |
| 176 | break; |
| 177 | |
| 178 | case ESCAPE_R: |
| 179 | fputs ("\\r", e); |
| 180 | break; |
| 181 | |
| 182 | case ESCAPE_T: |
| 183 | fputs ("\\t", e); |
| 184 | break; |
| 185 | |
| 186 | case ESCAPE_V: |
| 187 | fputs ("\\v", e); |
| 188 | break; |
| 189 | |
| 190 | default: |
| 191 | fprintf (e, "\\%03o", (unsigned int) ch); |
| 192 | break; |
| 193 | } |
| 194 | } |
| 195 | } |
| 196 | else if ((ch & 0xff) == ch) |
| 197 | fprintf (e, "\\%03o", (unsigned int) ch); |
| 198 | else |
| 199 | fprintf (e, "\\x%x", (unsigned int) ch); |
| 200 | } |
| 201 | } |
| 202 | |
| 203 | /* Print a unicode string to a file. */ |
| 204 | void |
| 205 | ascii_print (FILE *e, const char *s, rc_uint_type length) |
| 206 | { |
| 207 | while (1) |
| 208 | { |
| 209 | char ch; |
| 210 | |
| 211 | if (length == 0) |
| 212 | return; |
| 213 | if ((bfd_signed_vma) length > 0) |
| 214 | --length; |
| 215 | |
| 216 | ch = *s; |
| 217 | |
| 218 | if (ch == 0 && (bfd_signed_vma) length < 0) |
| 219 | return; |
| 220 | |
| 221 | ++s; |
| 222 | |
| 223 | if ((ch & 0x7f) == ch) |
| 224 | { |
| 225 | if (ch == '\\') |
| 226 | fputs ("\\\\", e); |
| 227 | else if (ch == '"') |
| 228 | fputs ("\"\"", e); |
| 229 | else if (ISPRINT (ch)) |
| 230 | putc (ch, e); |
| 231 | else |
| 232 | { |
| 233 | switch (ch) |
| 234 | { |
| 235 | case ESCAPE_A: |
| 236 | fputs ("\\a", e); |
| 237 | break; |
| 238 | |
| 239 | case ESCAPE_B: |
| 240 | fputs ("\\b", e); |
| 241 | break; |
| 242 | |
| 243 | case ESCAPE_F: |
| 244 | fputs ("\\f", e); |
| 245 | break; |
| 246 | |
| 247 | case ESCAPE_N: |
| 248 | fputs ("\\n", e); |
| 249 | break; |
| 250 | |
| 251 | case ESCAPE_R: |
| 252 | fputs ("\\r", e); |
| 253 | break; |
| 254 | |
| 255 | case ESCAPE_T: |
| 256 | fputs ("\\t", e); |
| 257 | break; |
| 258 | |
| 259 | case ESCAPE_V: |
| 260 | fputs ("\\v", e); |
| 261 | break; |
| 262 | |
| 263 | default: |
| 264 | fprintf (e, "\\%03o", (unsigned int) ch); |
| 265 | break; |
| 266 | } |
| 267 | } |
| 268 | } |
| 269 | else |
| 270 | fprintf (e, "\\%03o", (unsigned int) ch & 0xff); |
| 271 | } |
| 272 | } |
| 273 | |
| 274 | rc_uint_type |
| 275 | unichar_len (const unichar *unicode) |
| 276 | { |
| 277 | rc_uint_type r = 0; |
| 278 | if (unicode) |
| 279 | while (unicode[r] != 0) |
| 280 | r++; |
| 281 | else |
| 282 | --r; |
| 283 | return r; |
| 284 | } |
| 285 | |
| 286 | unichar * |
| 287 | unichar_dup (const unichar *unicode) |
| 288 | { |
| 289 | unichar *r; |
| 290 | int len; |
| 291 | |
| 292 | if (! unicode) |
| 293 | return NULL; |
| 294 | for (len = 0; unicode[len] != 0; ++len) |
| 295 | ; |
| 296 | ++len; |
| 297 | r = ((unichar *) res_alloc (len * sizeof (unichar))); |
| 298 | memcpy (r, unicode, len * sizeof (unichar)); |
| 299 | return r; |
| 300 | } |
| 301 | |
| 302 | unichar * |
| 303 | unichar_dup_uppercase (const unichar *u) |
| 304 | { |
| 305 | unichar *r = unichar_dup (u); |
| 306 | int i; |
| 307 | |
| 308 | if (! r) |
| 309 | return NULL; |
| 310 | |
| 311 | for (i = 0; r[i] != 0; ++i) |
| 312 | { |
| 313 | if (r[i] >= 'a' && r[i] <= 'z') |
| 314 | r[i] &= 0xdf; |
| 315 | } |
| 316 | return r; |
| 317 | } |
| 318 | |
| 319 | static int |
| 320 | unichar_isascii (const unichar *u, rc_uint_type len) |
| 321 | { |
| 322 | rc_uint_type i; |
| 323 | if ((bfd_signed_vma) len < 0) |
| 324 | { |
| 325 | if (u) |
| 326 | len = (rc_uint_type) unichar_len (u); |
| 327 | else |
| 328 | len = 0; |
| 329 | } |
| 330 | |
| 331 | for (i = 0; i < len; i++) |
| 332 | if ((u[i] & 0xff80) != 0) |
| 333 | return 0; |
| 334 | return 1; |
| 335 | } |
| 336 | |
| 337 | void |
| 338 | unicode_print_quoted (FILE *e, const unichar *u, rc_uint_type len) |
| 339 | { |
| 340 | if (! unichar_isascii (u, len)) |
| 341 | fputc ('L', e); |
| 342 | fputc ('"', e); |
| 343 | unicode_print (e, u, len); |
| 344 | fputc ('"', e); |
| 345 | } |