DRC | 2ff39b8 | 2011-07-28 08:38:59 +0000 | [diff] [blame] | 1 | // |
| 2 | // "$Id: fl_utf8.cxx 7975 2010-12-08 12:15:48Z AlbrechtS $" |
| 3 | // |
| 4 | // Unicode to UTF-8 conversion functions. |
| 5 | // |
| 6 | // Author: Jean-Marc Lienher ( http://oksid.ch ) |
| 7 | // Copyright 2000-2010 by O'ksi'D. |
| 8 | // |
| 9 | // This library is free software; you can redistribute it and/or |
| 10 | // modify it under the terms of the GNU Library General Public |
| 11 | // License as published by the Free Software Foundation; either |
| 12 | // version 2 of the License, or (at your option) any later version. |
| 13 | // |
| 14 | // This library 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 GNU |
| 17 | // Library General Public License for more details. |
| 18 | // |
| 19 | // You should have received a copy of the GNU Library General Public |
| 20 | // License along with this library; if not, write to the Free Software |
| 21 | // Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 |
| 22 | // USA. |
| 23 | // |
| 24 | // Please report all bugs and problems on the following page: |
| 25 | // |
| 26 | // http://www.fltk.org/str.php |
| 27 | |
| 28 | |
| 29 | #include <config.h> |
| 30 | #include <FL/filename.H> |
| 31 | #include <stdarg.h> |
| 32 | |
| 33 | #if defined(WIN32) && !defined(__CYGWIN__) |
| 34 | # include <ctype.h> |
| 35 | # include <io.h> |
| 36 | # include <windows.h> |
| 37 | # include <winbase.h> |
| 38 | # include <process.h> |
| 39 | #ifdef __CYGWIN__ |
| 40 | #include <wchar.h> |
| 41 | #else |
| 42 | #include <direct.h> |
| 43 | #endif |
| 44 | extern "C" { |
| 45 | int XUtf8Tolower(int ucs); |
| 46 | unsigned short XUtf8IsNonSpacing(unsigned int ucs); |
| 47 | }; |
| 48 | |
| 49 | #elif defined(__APPLE__) |
| 50 | # include <stdio.h> |
| 51 | # include <time.h> |
| 52 | //# include <unix.h> |
| 53 | # include <fcntl.h> |
| 54 | # include <unistd.h> |
| 55 | # include <wchar.h> |
| 56 | # include <stdlib.h> |
| 57 | # include <sys/types.h> |
| 58 | # include <sys/stat.h> |
| 59 | |
| 60 | extern "C" { |
| 61 | int XUtf8Tolower(int ucs); |
| 62 | unsigned short XUtf8IsNonSpacing(unsigned int ucs); |
| 63 | } |
| 64 | |
| 65 | #else // X-windows platform |
| 66 | |
| 67 | # include <FL/Xutf8.h> |
| 68 | # include <sys/types.h> |
| 69 | # include <sys/stat.h> |
| 70 | # include <fcntl.h> |
| 71 | # include <unistd.h> |
| 72 | #endif // WIN32 |
| 73 | |
| 74 | #include <FL/fl_utf8.h> |
| 75 | #include <string.h> |
| 76 | #include <stdlib.h> |
| 77 | |
| 78 | #undef fl_open |
| 79 | |
| 80 | /** \addtogroup fl_unicode |
| 81 | @{ |
| 82 | */ |
| 83 | |
| 84 | /*** NOTE : all functions are LIMITED to 24 bits Unicode values !!! ***/ |
| 85 | /*** but only 16 bits are really used under Linux and win32 ***/ |
| 86 | |
| 87 | |
| 88 | #define NBC 0xFFFF + 1 |
| 89 | |
| 90 | static int |
| 91 | Toupper( |
| 92 | int ucs) |
| 93 | { |
| 94 | long i; |
| 95 | static unsigned short *table = NULL; |
| 96 | |
| 97 | if (!table) { |
| 98 | table = (unsigned short*) malloc( |
| 99 | sizeof(unsigned short) * (NBC)); |
| 100 | for (i = 0; i < NBC; i++) { |
| 101 | table[i] = (unsigned short) i; |
| 102 | } |
| 103 | for (i = 0; i < NBC; i++) { |
| 104 | int l; |
| 105 | l = XUtf8Tolower(i); |
| 106 | if (l != i) table[l] = (unsigned short) i; |
| 107 | } |
| 108 | |
| 109 | } |
| 110 | if (ucs >= NBC || ucs < 0) return ucs; |
| 111 | return table[ucs]; |
| 112 | } |
| 113 | |
| 114 | /** |
| 115 | return the byte length of the UTF-8 sequence with first byte \p c, |
| 116 | or -1 if \p c is not valid. |
| 117 | This function is helpful for finding faulty UTF8 sequences. |
| 118 | \see fl_utf8len1 |
| 119 | */ |
| 120 | int fl_utf8len(char c) |
| 121 | { |
| 122 | if (!(c & 0x80)) return 1; |
| 123 | if (c & 0x40) { |
| 124 | if (c & 0x20) { |
| 125 | if (c & 0x10) { |
| 126 | if (c & 0x08) { |
| 127 | if (c & 0x04) { |
| 128 | return 6; |
| 129 | } |
| 130 | return 5; |
| 131 | } |
| 132 | return 4; |
| 133 | } |
| 134 | return 3; |
| 135 | } |
| 136 | return 2; |
| 137 | } |
| 138 | return -1; |
| 139 | } // fl_utf8len |
| 140 | |
| 141 | |
| 142 | /** |
| 143 | Return the byte length of the UTF-8 sequence with first byte \p c, |
| 144 | or 1 if \p c is not valid. |
| 145 | This function can be used to scan faulty UTF8 sequence, albeit ignoring invalid |
| 146 | codes. |
| 147 | \see fl_utf8len |
| 148 | */ |
| 149 | int fl_utf8len1(char c) |
| 150 | { |
| 151 | if (!(c & 0x80)) return 1; |
| 152 | if (c & 0x40) { |
| 153 | if (c & 0x20) { |
| 154 | if (c & 0x10) { |
| 155 | if (c & 0x08) { |
| 156 | if (c & 0x04) { |
| 157 | return 6; |
| 158 | } |
| 159 | return 5; |
| 160 | } |
| 161 | return 4; |
| 162 | } |
| 163 | return 3; |
| 164 | } |
| 165 | return 2; |
| 166 | } |
| 167 | return 1; |
| 168 | } // fl_utf8len1 |
| 169 | |
| 170 | |
| 171 | /** |
| 172 | returns the number of Unicode chars in the UTF-8 string |
| 173 | */ |
| 174 | int |
| 175 | fl_utf_nb_char( |
| 176 | const unsigned char *buf, |
| 177 | int len) |
| 178 | { |
| 179 | int i = 0; |
| 180 | int nbc = 0; |
| 181 | while (i < len) { |
| 182 | int cl = fl_utf8len((buf+i)[0]);//fl_utflen(buf + i, len - i); |
| 183 | if (cl < 1) cl = 1; |
| 184 | nbc++; |
| 185 | i += cl; |
| 186 | } |
| 187 | return nbc; |
| 188 | } |
| 189 | |
| 190 | /* |
| 191 | * compare only the first n bytes |
| 192 | * return 0 if the strings are equal; |
| 193 | * return 1 if s1 is greater than s2 |
| 194 | * return -1 if s1 is less than s2 |
| 195 | */ |
| 196 | /** |
| 197 | UTF-8 aware strncasecmp - converts to lower case Unicode and tests. |
| 198 | |
| 199 | \todo Correct the incorrect logic where length of strings tested |
| 200 | \todo Clarify whether n means number of bytes, or characters. |
| 201 | */ |
| 202 | int fl_utf_strncasecmp(const char *s1, const char *s2, int n) |
| 203 | { |
| 204 | int i; |
| 205 | int s1_l; |
| 206 | int s2_l; |
| 207 | char *e1, *e2; // string end pointers |
| 208 | |
| 209 | s1_l = 0; |
| 210 | while (s1_l < n && s1[s1_l]) s1_l++; |
| 211 | s2_l = 0; |
| 212 | while (s2_l < n && s2[s2_l]) s2_l++; |
| 213 | |
| 214 | if (s1_l < s2_l) { |
| 215 | return -1; |
| 216 | } else if (s1_l > s2_l) { |
| 217 | return 1; |
| 218 | } |
| 219 | e1 = (char *)&s1[s1_l]; // last char to test |
| 220 | e2 = (char *)&s2[s2_l]; |
| 221 | for (i = 0; i < n;) { |
| 222 | int l1, l2; |
| 223 | unsigned int u1, u2; |
| 224 | int res; |
| 225 | |
| 226 | // l1 = fl_utf2ucs((unsigned char*)s1 + i, n - i, &u1); |
| 227 | u1 = fl_utf8decode(s1 + i, e1, &l1); |
| 228 | // l2 = fl_utf2ucs((unsigned char*)s2 + i, n - i, &u2); |
| 229 | u2 = fl_utf8decode(s2 + i, e2, &l2); |
| 230 | if (l1 - l2 != 0) return l1 - l2; |
| 231 | res = XUtf8Tolower(u1) - XUtf8Tolower(u2); |
| 232 | if (res != 0) return res; |
| 233 | if (l1 < 1) { |
| 234 | i += 1; |
| 235 | } else { |
| 236 | i += l1; |
| 237 | } |
| 238 | } |
| 239 | return 0; |
| 240 | } |
| 241 | |
| 242 | /* |
| 243 | * return 0 if the strings are equal; |
| 244 | * return 1 if s1 is greater than s2 |
| 245 | * return -1 if s1 is less than s2 |
| 246 | */ |
| 247 | /** |
| 248 | UTF-8 aware strcasecmp - converts to Unicode and tests. |
| 249 | |
| 250 | \todo Correct the incorrect logic where length of strings tested |
| 251 | */ |
| 252 | int fl_utf_strcasecmp(const char *s1, const char *s2) |
| 253 | { |
| 254 | int s1_l = strlen(s1); |
| 255 | int s2_l = strlen(s2); |
| 256 | |
| 257 | if (s1_l < s2_l) { |
| 258 | return -1; |
| 259 | } else if (s1_l > s2_l) { |
| 260 | return 1; |
| 261 | } |
| 262 | return fl_utf_strncasecmp(s1, s2, s1_l); |
| 263 | } |
| 264 | |
| 265 | /** |
| 266 | return the Unicode lower case value of \p ucs |
| 267 | */ |
| 268 | int fl_tolower(unsigned int ucs) |
| 269 | { |
| 270 | return XUtf8Tolower(ucs); |
| 271 | } |
| 272 | |
| 273 | /** |
| 274 | return the Unicode upper case value of \p ucs |
| 275 | */ |
| 276 | int fl_toupper(unsigned int ucs) |
| 277 | { |
| 278 | return Toupper(ucs); |
| 279 | } |
| 280 | |
| 281 | /** |
| 282 | converts the str string to the lower case equivalent into buf. |
| 283 | Warning: to be safe buf length must be at least 3 * len [for 16-bit Unicode] |
| 284 | */ |
| 285 | int fl_utf_tolower(const unsigned char *str, int len, char *buf) |
| 286 | { |
| 287 | int i; |
| 288 | int l = 0; |
| 289 | char *end = (char *)&str[len]; |
| 290 | for (i = 0; i < len;) { |
| 291 | int l1, l2; |
| 292 | unsigned int u1; |
| 293 | |
| 294 | // l1 = fl_utf2ucs((unsigned char*)str + i, len - i, &u1); |
| 295 | u1 = fl_utf8decode((const char*)(str + i), end, &l1); |
| 296 | l2 = fl_utf8encode((unsigned int) XUtf8Tolower(u1), buf + l); |
| 297 | if (l1 < 1) { |
| 298 | i += 1; |
| 299 | } else { |
| 300 | i += l1; |
| 301 | } |
| 302 | if (l2 < 1) { |
| 303 | l += 1; |
| 304 | } else { |
| 305 | l += l2; |
| 306 | } |
| 307 | |
| 308 | } |
| 309 | return l; |
| 310 | } |
| 311 | |
| 312 | /** |
| 313 | converts the str string to the upper case equivalent into buf. |
| 314 | Warning: to be safe buf length must be at least 3 * len [for 16-bit Unicode] |
| 315 | */ |
| 316 | int fl_utf_toupper(const unsigned char *str, int len, char *buf) |
| 317 | { |
| 318 | int i; |
| 319 | int l = 0; |
| 320 | char *end = (char *)&str[len]; |
| 321 | for (i = 0; i < len;) { |
| 322 | int l1, l2; |
| 323 | unsigned int u1; |
| 324 | |
| 325 | // l1 = fl_utf2ucs((unsigned char*)str + i, len - i, &u1); |
| 326 | u1 = fl_utf8decode((const char*)(str + i), end, &l1); |
| 327 | l2 = fl_utf8encode((unsigned int) Toupper(u1), buf + l); |
| 328 | if (l1 < 1) { |
| 329 | i += 1; |
| 330 | } else { |
| 331 | i += l1; |
| 332 | } |
| 333 | if (l2 < 1) { |
| 334 | l += 1; |
| 335 | } else { |
| 336 | l += l2; |
| 337 | } |
| 338 | } |
| 339 | return l; |
| 340 | } |
| 341 | |
| 342 | #if 0 // deprecated in favour of FLTK2's fl_utf8toa |
| 343 | /* |
| 344 | * convert UTF-8 str to latin1 |
| 345 | * Warning: buf must be at least len long |
| 346 | */ |
| 347 | int fl_utf2latin1(const unsigned char *str, int len, char *buf) |
| 348 | { |
| 349 | int i; |
| 350 | int l = 0; |
| 351 | char *end = (char *)&str[len]; |
| 352 | for (i = 0; i < len;) { |
| 353 | unsigned int u1; |
| 354 | int l1; |
| 355 | |
| 356 | // l1 = fl_utf2ucs((unsigned char*)str + i, len - i, &u1); |
| 357 | u1 = fl_utf8decode((const char*)(str + i), end, &l1); |
| 358 | if (u1 > 0xFF) u1 = '?'; |
| 359 | buf[l] = (char) u1; |
| 360 | if (l1 < 1) { |
| 361 | i += 1; |
| 362 | } else { |
| 363 | i += l1; |
| 364 | } |
| 365 | l++; |
| 366 | |
| 367 | } |
| 368 | return l; |
| 369 | } |
| 370 | #endif |
| 371 | |
| 372 | #if 0 // deprecated in favour of FLTK2's fl_utf8froma |
| 373 | /* |
| 374 | * convert latin1 str to UTF-8 |
| 375 | * Warning: buf must be at least 2 * len long |
| 376 | */ |
| 377 | int fl_latin12utf(const unsigned char *str, int len, char *buf) |
| 378 | { |
| 379 | int i; |
| 380 | int l = 0; |
| 381 | int l1 = 0; |
| 382 | for (i = 0; i < len; i++) { |
| 383 | unsigned int n = (unsigned int) str[i]; |
| 384 | l1 = fl_utf8encode(n, buf + l); |
| 385 | if (l1 < 1) { |
| 386 | l = l + 1; |
| 387 | } else { |
| 388 | l = l + l1; |
| 389 | } |
| 390 | |
| 391 | } |
| 392 | return l; |
| 393 | } |
| 394 | #endif |
| 395 | |
| 396 | /** |
| 397 | returns true if the character is non-spacing. |
| 398 | \todo explain what non-spacing means. |
| 399 | */ |
| 400 | unsigned int fl_nonspacing(unsigned int ucs) |
| 401 | { |
| 402 | #ifdef __APPLE__ |
| 403 | return (ucs==0x20); // FIXME: what does this really do? |
| 404 | #else |
| 405 | return (unsigned int) XUtf8IsNonSpacing(ucs); |
| 406 | #endif |
| 407 | } |
| 408 | |
| 409 | #if defined(WIN32) && !defined(__CYGWIN__) |
| 410 | static xchar *mbwbuf = NULL; |
| 411 | #endif |
| 412 | |
| 413 | #ifdef WIN32 |
| 414 | unsigned int fl_codepage = 0; |
| 415 | #endif |
| 416 | |
| 417 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 418 | |
| 419 | static char *buf = NULL; |
| 420 | static int buf_len = 0; |
| 421 | static unsigned short *wbufa = NULL; |
| 422 | |
| 423 | // FIXME: This should *maybe* return 'const char *' instead of 'char *' |
| 424 | char *fl_utf8_to_locale(const char *s, int len, UINT codepage) |
| 425 | { |
| 426 | if (!s) return (char *)""; |
| 427 | int l = 0; |
| 428 | // if (buf_len < len * 2 + 1) { |
| 429 | // buf_len = len * 2 + 1; |
| 430 | // buf = (char*) realloc(buf, buf_len); |
| 431 | // wbufa = (unsigned short*) realloc(wbufa, buf_len * sizeof(short)); |
| 432 | // } |
| 433 | unsigned wn = fl_utf8toUtf16(s, len, NULL, 0); // Query length |
| 434 | wn = wn * 2 + 1; |
| 435 | if (wn >= (unsigned)buf_len) { |
| 436 | buf_len = wn; |
| 437 | buf = (char*) realloc(buf, buf_len); |
| 438 | wbufa = (unsigned short*) realloc(wbufa, buf_len * sizeof(short)); |
| 439 | } |
| 440 | if (codepage < 1) codepage = fl_codepage; |
| 441 | // l = fl_utf2unicode((const unsigned char *)s, len, (xchar*) wbufa); |
| 442 | l = fl_utf8toUtf16(s, len, wbufa, wn); // Convert string |
| 443 | wbufa[l] = 0; |
| 444 | buf[l] = 0; |
| 445 | l = WideCharToMultiByte(codepage, 0, (WCHAR*)wbufa, l, buf, buf_len, NULL, NULL); |
| 446 | if (l < 0) l = 0; |
| 447 | buf[l] = 0; |
| 448 | return buf; |
| 449 | } |
| 450 | |
| 451 | // FIXME: This should maybe return 'const char *' instead of 'char *' |
| 452 | char *fl_locale_to_utf8(const char *s, int len, UINT codepage) |
| 453 | { |
| 454 | if (!s) return (char *)""; |
| 455 | int l = 0; |
| 456 | if (buf_len < len * 5 + 1) { |
| 457 | buf_len = len * 5 + 1; |
| 458 | buf = (char*) realloc(buf, buf_len); |
| 459 | wbufa = (unsigned short*) realloc(wbufa, buf_len * sizeof(short)); |
| 460 | } |
| 461 | if (codepage < 1) codepage = fl_codepage; |
| 462 | buf[l] = 0; |
| 463 | |
| 464 | l = MultiByteToWideChar(codepage, 0, s, len, (WCHAR*)wbufa, buf_len); |
| 465 | if (l < 0) l = 0; |
| 466 | wbufa[l] = 0; |
| 467 | // l = fl_unicode2utf((xchar*)wbufa, l, buf); |
| 468 | l = fl_utf8fromwc(buf, buf_len, (xchar*)wbufa, l); |
| 469 | buf[l] = 0; |
| 470 | return buf; |
| 471 | } |
| 472 | #endif |
| 473 | |
| 474 | /** |
| 475 | converts UTF8 to a local multi-byte character string. |
| 476 | */ |
| 477 | char * fl_utf2mbcs(const char *s) |
| 478 | { |
| 479 | if (!s) return NULL; |
| 480 | #if defined(WIN32) && !defined(__CYGWIN__) |
| 481 | int l = strlen(s); |
| 482 | static char *buf = NULL; |
| 483 | |
| 484 | // mbwbuf = (xchar*)realloc(mbwbuf, (l+6) * sizeof(xchar)); |
| 485 | // l = fl_utf2unicode((unsigned char*)s, l, mbwbuf); |
| 486 | // mbwbuf[l] = 0; |
| 487 | unsigned wn = fl_utf8toUtf16(s, l, NULL, 0) + 7; // Query length |
| 488 | mbwbuf = (xchar*)realloc(mbwbuf, sizeof(xchar)*wn); |
| 489 | l = fl_utf8toUtf16(s, l, (unsigned short *)mbwbuf, wn); // Convert string |
| 490 | mbwbuf[l] = 0; |
| 491 | |
| 492 | buf = (char*)realloc(buf, l * 6 + 1); |
| 493 | l = wcstombs(buf, mbwbuf, l * 6); |
| 494 | buf[l] = 0; |
| 495 | return buf; |
| 496 | #else |
| 497 | return (char*) s; |
| 498 | #endif |
| 499 | } |
| 500 | |
| 501 | |
| 502 | #if 0 // deprecated in favour of FLTK2's fl_utf8from_mb |
| 503 | char * fl_mbcs2utf(const char *s) |
| 504 | { |
| 505 | if (!s) return NULL; |
| 506 | #if defined(WIN32) |
| 507 | int l = strlen(s); |
| 508 | unsigned dstlen; |
| 509 | static char *buf = NULL; |
| 510 | |
| 511 | mbwbuf = (xchar*)realloc(mbwbuf,(l * 6 + 6) * sizeof(xchar)); |
| 512 | l = mbstowcs(mbwbuf, s, l); |
| 513 | dstlen = l * 6 + 1; |
| 514 | buf = (char*)realloc(buf, dstlen); |
| 515 | // l = fl_unicode2utf(mbwbuf, l, buf); |
| 516 | l = fl_utf8fromwc(buf, dstlen, mbwbuf, l); |
| 517 | buf[l] = 0; |
| 518 | return buf; |
| 519 | #else |
| 520 | return (char*) s; |
| 521 | #endif |
| 522 | } // fl_mbcs2utf |
| 523 | #endif |
| 524 | |
| 525 | #if defined(WIN32) && !defined(__CYGWIN__) |
| 526 | static xchar *wbuf = NULL; |
| 527 | static xchar *wbuf1 = NULL; |
| 528 | #endif |
| 529 | |
| 530 | |
| 531 | char *fl_getenv(const char* v) |
| 532 | { |
| 533 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 534 | int l = strlen(v); |
| 535 | // static xchar* wbuf = NULL; |
| 536 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 537 | // wbuf[fl_utf2unicode((const unsigned char*)v, l, wbuf)] = 0; |
| 538 | unsigned wn = fl_utf8toUtf16(v, l, NULL, 0) + 1; // Query length |
| 539 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 540 | wn = fl_utf8toUtf16(v, l, (unsigned short *)wbuf, wn); // Convert string |
| 541 | wbuf[wn] = 0; |
| 542 | xchar *ret = _wgetenv(wbuf); |
| 543 | static char *buf = NULL; |
| 544 | if (ret) { |
| 545 | l = wcslen(ret); |
| 546 | wn = fl_utf8fromwc(NULL, 0, ret, l) + 1; // query length |
| 547 | buf = (char*) realloc(buf, wn); |
| 548 | // buf[fl_unicode2utf(ret, l, buf)] = 0; |
| 549 | wn = fl_utf8fromwc(buf, wn, ret, l); // convert string |
| 550 | buf[wn] = 0; |
| 551 | return buf; |
| 552 | } else { |
| 553 | return NULL; |
| 554 | } |
| 555 | #else |
| 556 | return getenv(v); |
| 557 | #endif |
| 558 | } |
| 559 | |
| 560 | int fl_open(const char* f, int oflags, ...) |
| 561 | { |
| 562 | int pmode; |
| 563 | va_list ap; |
| 564 | va_start(ap, oflags); |
| 565 | pmode = va_arg (ap, int); |
| 566 | va_end(ap); |
| 567 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 568 | int l = strlen(f); |
| 569 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 570 | // wbuf[fl_utf2unicode((const unsigned char*)f, l, wbuf)] = 0; |
| 571 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 572 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 573 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 574 | wbuf[wn] = 0; |
| 575 | if (pmode == -1) return _wopen(wbuf, oflags); |
| 576 | else return _wopen(wbuf, oflags, pmode); |
| 577 | #else |
| 578 | if (pmode == -1) return open(f, oflags); |
| 579 | else return open(f, oflags, pmode); |
| 580 | #endif |
| 581 | } |
| 582 | |
| 583 | FILE *fl_fopen(const char* f, const char *mode) |
| 584 | { |
| 585 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 586 | int l = strlen(f); |
| 587 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 588 | // wbuf[fl_utf2unicode((const unsigned char*)f, l, wbuf)] = 0; |
| 589 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 590 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 591 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 592 | wbuf[wn] = 0; |
| 593 | l = strlen(mode); |
| 594 | // wbuf1 = (xchar*)realloc(wbuf1, sizeof(xchar) * (l+1)); |
| 595 | // wbuf1[fl_utf2unicode((const unsigned char*)mode, l, wbuf1)] = 0; |
| 596 | wn = fl_utf8toUtf16(mode, l, NULL, 0) + 1; // Query length |
| 597 | wbuf1 = (xchar*)realloc(wbuf1, sizeof(xchar)*wn); |
| 598 | wn = fl_utf8toUtf16(mode, l, (unsigned short *)wbuf1, wn); // Convert string |
| 599 | wbuf1[wn] = 0; |
| 600 | return _wfopen(wbuf, wbuf1); |
| 601 | #else |
| 602 | return fopen(f, mode); |
| 603 | #endif |
| 604 | } |
| 605 | |
| 606 | int fl_system(const char* f) |
| 607 | { |
| 608 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 609 | # ifdef __MINGW32__ |
| 610 | return system(fl_utf2mbcs(f)); |
| 611 | # else |
| 612 | int l = strlen(f); |
| 613 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 614 | // wbuf[fl_utf2unicode((const unsigned char*)f, l, wbuf)] = 0; |
| 615 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 616 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 617 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 618 | wbuf[wn] = 0; |
| 619 | return _wsystem(wbuf); |
| 620 | # endif |
| 621 | #else |
| 622 | return system(f); |
| 623 | #endif |
| 624 | } |
| 625 | |
| 626 | int fl_execvp(const char *file, char *const *argv) |
| 627 | { |
| 628 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 629 | #ifdef __MINGW32__ |
| 630 | return _execvp(fl_utf2mbcs(file), argv); |
| 631 | #else |
| 632 | int l = strlen(file); |
| 633 | int i, n, ret; |
| 634 | xchar **ar; |
| 635 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 636 | // wbuf[fl_utf2unicode((const unsigned char*)file, l, wbuf)] = 0; |
| 637 | unsigned wn = fl_utf8toUtf16(file, l, NULL, 0) + 1; // Query length |
| 638 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 639 | wn = fl_utf8toUtf16(file, l, (unsigned short *)wbuf, wn); // Convert string |
| 640 | wbuf[wn] = 0; |
| 641 | |
| 642 | i = 0; n = 0; |
| 643 | while (argv[i]) {i++; n++;} |
| 644 | ar = (xchar**) malloc(sizeof(xchar*) * (n + 1)); |
| 645 | i = 0; |
| 646 | while (i <= n) { |
| 647 | unsigned wn; |
| 648 | l = strlen(argv[i]); |
| 649 | // ar[i] = (xchar *)malloc(sizeof(xchar) * (l+1)); |
| 650 | // ar[i][fl_utf2unicode((const unsigned char*)argv[i], l, ar[i])] = 0; |
| 651 | wn = fl_utf8toUtf16(argv[i], l, NULL, 0) + 1; // Query length |
| 652 | ar[i] = (xchar *)malloc(sizeof(xchar)*wn); |
| 653 | wn = fl_utf8toUtf16(argv[i], l, (unsigned short *)ar[i], wn); // Convert string |
| 654 | ar[i][wn] = 0; |
| 655 | i++; |
| 656 | } |
| 657 | ar[n] = NULL; |
| 658 | ret = _wexecvp(wbuf, ar); |
| 659 | i = 0; |
| 660 | while (i <= n) { |
| 661 | free(ar[i]); |
| 662 | i++; |
| 663 | } |
| 664 | free(ar); |
| 665 | return ret; |
| 666 | #endif |
| 667 | #else |
| 668 | return execvp(file, argv); |
| 669 | #endif |
| 670 | } |
| 671 | |
| 672 | |
| 673 | |
| 674 | int fl_chmod(const char* f, int mode) |
| 675 | { |
| 676 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 677 | int l = strlen(f); |
| 678 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 679 | // wbuf[fl_utf2unicode((const unsigned char*)f, l, wbuf)] = 0; |
| 680 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 681 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 682 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 683 | wbuf[wn] = 0; |
| 684 | return _wchmod(wbuf, mode); |
| 685 | #else |
| 686 | return chmod(f, mode); |
| 687 | #endif |
| 688 | } |
| 689 | |
| 690 | int fl_access(const char* f, int mode) |
| 691 | { |
| 692 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 693 | int l = strlen(f); |
| 694 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 695 | // wbuf[fl_utf2unicode((const unsigned char*)f, l, wbuf)] = 0; |
| 696 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 697 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 698 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 699 | wbuf[wn] = 0; |
| 700 | return _waccess(wbuf, mode); |
| 701 | #else |
| 702 | return access(f, mode); |
| 703 | #endif |
| 704 | } |
| 705 | |
| 706 | |
| 707 | int fl_stat(const char* f, struct stat *b) |
| 708 | { |
| 709 | #if defined(WIN32) && !defined(__CYGWIN__) |
| 710 | int l = strlen(f); |
| 711 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 712 | // wbuf[fl_utf2unicode((const unsigned char*)f, l, wbuf)] = 0; |
| 713 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 714 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 715 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 716 | wbuf[wn] = 0; |
| 717 | return _wstat(wbuf, (struct _stat*)b); |
| 718 | #else |
| 719 | return stat(f, b); |
| 720 | #endif |
| 721 | } |
| 722 | |
| 723 | char *fl_getcwd(char* b, int l) |
| 724 | { |
| 725 | if (b == NULL) { |
| 726 | b = (char*) malloc(l+1); |
| 727 | } |
| 728 | #if defined(WIN32) && !defined(__CYGWIN__) |
| 729 | static xchar *wbuf = NULL; |
| 730 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 731 | // xchar *ret = _wgetcwd(wbuf, l / 5); |
| 732 | xchar *ret = _wgetcwd(wbuf, l); |
| 733 | if (ret) { |
| 734 | unsigned dstlen = l; |
| 735 | l = wcslen(wbuf); |
| 736 | // b[fl_unicode2utf(wbuf, l, b)] = 0; |
| 737 | dstlen = fl_utf8fromwc(b, dstlen, wbuf, l); |
| 738 | b[dstlen] = 0; |
| 739 | return b; |
| 740 | } else { |
| 741 | return NULL; |
| 742 | } |
| 743 | #else |
| 744 | return getcwd(b, l); |
| 745 | #endif |
| 746 | } |
| 747 | |
| 748 | |
| 749 | int fl_unlink(const char* f) |
| 750 | { |
| 751 | #if defined(WIN32) && !defined(__CYGWIN__) |
| 752 | int l = strlen(f); |
| 753 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 754 | // wbuf[fl_utf2unicode((const unsigned char*)f, l, wbuf)] = 0; |
| 755 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 756 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 757 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 758 | wbuf[wn] = 0; |
| 759 | return _wunlink(wbuf); |
| 760 | #else |
| 761 | return unlink(f); |
| 762 | #endif |
| 763 | } |
| 764 | |
| 765 | int fl_mkdir(const char* f, int mode) |
| 766 | { |
| 767 | #if defined(WIN32) && !defined(__CYGWIN__) |
| 768 | int l = strlen(f); |
| 769 | // wbuf = (xchar*)realloc(wbuf, sizeof(short) * (l+1)); |
| 770 | // wbuf[fl_utf2unicode((const unsigned char*)f, l, wbuf)] = 0; |
| 771 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 772 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 773 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 774 | wbuf[wn] = 0; |
| 775 | return _wmkdir(wbuf); |
| 776 | #else |
| 777 | return mkdir(f, mode); |
| 778 | #endif |
| 779 | } |
| 780 | |
| 781 | |
| 782 | int fl_rmdir(const char* f) |
| 783 | { |
| 784 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 785 | int l = strlen(f); |
| 786 | // wbuf = (xchar*)realloc(wbuf, sizeof(xchar) * (l+1)); |
| 787 | // wbuf[fl_utf2unicode((const unsigned char*)f, l, wbuf)] = 0; |
| 788 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 789 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 790 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 791 | wbuf[wn] = 0; |
| 792 | return _wrmdir(wbuf); |
| 793 | #else |
| 794 | return rmdir(f); |
| 795 | #endif |
| 796 | } |
| 797 | |
| 798 | int fl_rename(const char* f, const char *n) |
| 799 | { |
| 800 | #if defined (WIN32) && !defined(__CYGWIN__) |
| 801 | int l = strlen(f); |
| 802 | unsigned wn = fl_utf8toUtf16(f, l, NULL, 0) + 1; // Query length |
| 803 | wbuf = (xchar*)realloc(wbuf, sizeof(xchar)*wn); |
| 804 | wn = fl_utf8toUtf16(f, l, (unsigned short *)wbuf, wn); // Convert string |
| 805 | wbuf[wn] = 0; |
| 806 | l = strlen(n); |
| 807 | wn = fl_utf8toUtf16(n, l, NULL, 0) + 1; // Query length |
| 808 | wbuf1 = (xchar*)realloc(wbuf1, sizeof(xchar)*wn); |
| 809 | wn = fl_utf8toUtf16(n, l, (unsigned short *)wbuf1, wn); // Convert string |
| 810 | wbuf1[wn] = 0; |
| 811 | return _wrename(wbuf, wbuf1); |
| 812 | #else |
| 813 | return rename(f, n); |
| 814 | #endif |
| 815 | } |
| 816 | |
| 817 | // recursively create a path in the file system |
| 818 | char fl_make_path( const char *path ) { |
| 819 | if (fl_access(path, 0)) { |
| 820 | const char *s = strrchr( path, '/' ); |
| 821 | if ( !s ) return 0; |
| 822 | int len = s-path; |
| 823 | char *p = (char*)malloc( len+1 ); |
| 824 | memcpy( p, path, len ); |
| 825 | p[len] = 0; |
| 826 | fl_make_path( p ); |
| 827 | free( p ); |
| 828 | fl_mkdir(path, 0700); |
| 829 | } |
| 830 | return 1; |
| 831 | } |
| 832 | |
| 833 | // strip the filename and create a path |
| 834 | void fl_make_path_for_file( const char *path ) |
| 835 | { |
| 836 | const char *s = strrchr( path, '/' ); |
| 837 | if ( !s ) return; |
| 838 | int len = s-path; |
| 839 | char *p = (char*)malloc( len+1 ); |
| 840 | memcpy( p, path, len ); |
| 841 | p[len] = 0; |
| 842 | fl_make_path( p ); |
| 843 | free( p ); |
| 844 | } |
| 845 | |
| 846 | /** @} */ |
| 847 | |
| 848 | // |
| 849 | // End of "$Id: fl_utf8.cxx 7975 2010-12-08 12:15:48Z AlbrechtS $". |
| 850 | // |