tools/wmc/utils.c

   1 /*
   2  * Utility routines
   3  *
   4  * Copyright 1998,2000 Bertho A. Stultiens
   5  *
   6  * This library is free software; you can redistribute it and/or
   7  * modify it under the terms of the GNU Lesser General Public
   8  * License as published by the Free Software Foundation; either
   9  * version 2.1 of the License, or (at your option) any later version.
  10  *
  11  * This library is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14  * Lesser General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU Lesser General Public
  17  * License along with this library; if not, write to the Free Software
  18  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
  19  */
  20
  21 #include "config.h"
  22
  23 #include <stdio.h>
  24 #include <stdlib.h>
  25 #include <stdarg.h>
  26 #include <string.h>
  27 #include <assert.h>
  28 #include <ctype.h>
  29
  30 #include "wmc.h"
  31 #include "winternl.h"
  32 #include "winnls.h"
  33 #include "utils.h"
  34
  35 #define SUPPRESS_YACC_ERROR_MESSAGE
  36
  37 static void generic_msg(const char *s, const char *t, va_list ap)
  38 {
  39         fprintf(stderr, "%s:%d:%d: %s: ", input_name ? input_name : "stdin", line_number, char_number, t);
  40         vfprintf(stderr, s, ap);
  41 }
  42
  43 /*
  44  * The yyerror routine should not exit because we use the error-token
  45  * to determine the syntactic error in the source. However, YACC
  46  * uses the same routine to print an error just before the error
  47  * token is reduced.
  48  * The extra routine 'xyyerror' is used to exit after giving a real
  49  * message.
  50  */
  51 int mcy_error(const char *s, ...)
  52 {
  53 #ifndef SUPPRESS_YACC_ERROR_MESSAGE
  54         va_list ap;
  55         va_start(ap, s);
  56         generic_msg(s, "Yacc error", ap);
  57         va_end(ap);
  58 #endif
  59         return 1;
  60 }
  61
  62 int xyyerror(const char *s, ...)
  63 {
  64         va_list ap;
  65         va_start(ap, s);
  66         generic_msg(s, "Error", ap);
  67         va_end(ap);
  68         exit(1);
  69         return 1;
  70 }
  71
  72 int mcy_warning(const char *s, ...)
  73 {
  74         va_list ap;
  75         va_start(ap, s);
  76         generic_msg(s, "Warning", ap);
  77         va_end(ap);
  78         return 0;
  79 }
  80
  81 void internal_error(const char *file, int line, const char *s, ...)
  82 {
  83         va_list ap;
  84         va_start(ap, s);
  85         fprintf(stderr, "Internal error (please report) %s %d: ", file, line);
  86         vfprintf(stderr, s, ap);
  87         va_end(ap);
  88         exit(3);
  89 }
  90
  91 void fatal_perror( const char *msg, ... )
  92 {
  93         va_list valist;
  94         va_start( valist, msg );
  95         fprintf(stderr, "Error: ");
  96         vfprintf( stderr, msg, valist );
  97         perror( " " );
  98         va_end( valist );
  99         exit(2);
 100 }
 101
 102 void error(const char *s, ...)
 103 {
 104         va_list ap;
 105         va_start(ap, s);
 106         fprintf(stderr, "Error: ");
 107         vfprintf(stderr, s, ap);
 108         va_end(ap);
 109         exit(2);
 110 }
 111
 112 void warning(const char *s, ...)
 113 {
 114         va_list ap;
 115         va_start(ap, s);
 116         fprintf(stderr, "Warning: ");
 117         vfprintf(stderr, s, ap);
 118         va_end(ap);
 119 }
 120
 121 int unistrlen(const WCHAR *s)
 122 {
 123         int n;
 124         for(n = 0; *s; n++, s++)
 125                 ;
 126         return n;
 127 }
 128
 129 WCHAR *unistrcpy(WCHAR *dst, const WCHAR *src)
 130 {
 131         WCHAR *t = dst;
 132         while(*src)
 133                 *t++ = *src++;
 134         *t = 0;
 135         return dst;
 136 }
 137
 138 WCHAR *xunistrdup(const WCHAR * str)
 139 {
 140         WCHAR *s;
 141
 142         assert(str != NULL);
 143         s = xmalloc((unistrlen(str)+1) * sizeof(WCHAR));
 144         return unistrcpy(s, str);
 145 }
 146
 147 int unistricmp(const WCHAR *s1, const WCHAR *s2)
 148 {
 149         int i;
 150         int once = 0;
 151         static const char warn[] = "Don't know the uppercase equivalent of non ascii characters;"
 152                              "comparison might yield wrong results";
 153         while(*s1 && *s2)
 154         {
 155                 if((*s1 & 0xffff) > 0x7f || (*s2 & 0xffff) > 0x7f)
 156                 {
 157                         if(!once)
 158                         {
 159                                 once++;
 160                                 mcy_warning(warn);
 161                         }
 162                         i = *s1++ - *s2++;
 163                 }
 164                 else
 165                         i = toupper(*s1++) - toupper(*s2++);
 166                 if(i)
 167                         return i;
 168         }
 169
 170         if((*s1 & 0xffff) > 0x7f || (*s2 & 0xffff) > 0x7f)
 171         {
 172                 if(!once)
 173                         mcy_warning(warn);
 174                 return *s1 - *s2;
 175         }
 176         else
 177                 return  toupper(*s1) - toupper(*s2);
 178 }
 179
 180 int unistrcmp(const WCHAR *s1, const WCHAR *s2)
 181 {
 182         int i;
 183         while(*s1 && *s2)
 184         {
 185                 i = *s1++ - *s2++;
 186                 if(i)
 187                         return i;
 188         }
 189
 190         return *s1 - *s2;
 191 }
 192
 193 WCHAR *utf8_to_unicode( const char *src, int srclen, int *dstlen )
 194 {
 195     static const char utf8_length[128] =
 196     {
 197         0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x80-0x8f */
 198         0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x90-0x9f */
 199         0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xa0-0xaf */
 200         0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xb0-0xbf */
 201         0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xc0-0xcf */
 202         1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xd0-0xdf */
 203         2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* 0xe0-0xef */
 204         3,3,3,3,3,0,0,0,0,0,0,0,0,0,0,0  /* 0xf0-0xff */
 205     };
 206     static const unsigned char utf8_mask[4] = { 0x7f, 0x1f, 0x0f, 0x07 };
 207
 208     const char *srcend = src + srclen;
 209     int len, res;
 210     WCHAR *ret, *dst;
 211
 212     dst = ret = xmalloc( (srclen + 1) * sizeof(WCHAR) );
 213     while (src < srcend)
 214     {
 215         unsigned char ch = *src++;
 216         if (ch < 0x80)  /* special fast case for 7-bit ASCII */
 217         {
 218             *dst++ = ch;
 219             continue;
 220         }
 221         len = utf8_length[ch - 0x80];
 222         if (len && src + len <= srcend)
 223         {
 224             res = ch & utf8_mask[len];
 225             switch (len)
 226             {
 227             case 3:
 228                 if ((ch = *src ^ 0x80) >= 0x40) break;
 229                 res = (res << 6) | ch;
 230                 src++;
 231                 if (res < 0x10) break;
 232             case 2:
 233                 if ((ch = *src ^ 0x80) >= 0x40) break;
 234                 res = (res << 6) | ch;
 235                 if (res >= 0x110000 >> 6) break;
 236                 src++;
 237                 if (res < 0x20) break;
 238                 if (res >= 0xd800 >> 6 && res <= 0xdfff >> 6) break;
 239             case 1:
 240                 if ((ch = *src ^ 0x80) >= 0x40) break;
 241                 res = (res << 6) | ch;
 242                 src++;
 243                 if (res < 0x80) break;
 244                 if (res <= 0xffff) *dst++ = res;
 245                 else
 246                 {
 247                     res -= 0x10000;
 248                     *dst++ = 0xd800 | (res >> 10);
 249                     *dst++ = 0xdc00 | (res & 0x3ff);
 250                 }
 251                 continue;
 252             }
 253         }
 254         *dst++ = 0xfffd;
 255     }
 256     *dst = 0;
 257     *dstlen = dst - ret;
 258     return ret;
 259 }
 260
 261 char *unicode_to_utf8( const WCHAR *src, int srclen, int *dstlen )
 262 {
 263     char *ret, *dst;
 264
 265     dst = ret = xmalloc( srclen * 3 + 1 );
 266     for ( ; srclen; srclen--, src++)
 267     {
 268         unsigned int ch = *src;
 269
 270         if (ch < 0x80)  /* 0x00-0x7f: 1 byte */
 271         {
 272             *dst++ = ch;
 273             continue;
 274         }
 275         if (ch < 0x800)  /* 0x80-0x7ff: 2 bytes */
 276         {
 277             dst[1] = 0x80 | (ch & 0x3f);
 278             ch >>= 6;
 279             dst[0] = 0xc0 | ch;
 280             dst += 2;
 281             continue;
 282         }
 283         if (ch >= 0xd800 && ch <= 0xdbff && srclen > 1 && src[1] >= 0xdc00 && src[1] <= 0xdfff)
 284         {
 285             /* 0x10000-0x10ffff: 4 bytes */
 286             ch = 0x10000 + ((ch & 0x3ff) << 10) + (src[1] & 0x3ff);
 287             dst[3] = 0x80 | (ch & 0x3f);
 288             ch >>= 6;
 289             dst[2] = 0x80 | (ch & 0x3f);
 290             ch >>= 6;
 291             dst[1] = 0x80 | (ch & 0x3f);
 292             ch >>= 6;
 293             dst[0] = 0xf0 | ch;
 294             dst += 4;
 295             src++;
 296             srclen--;
 297             continue;
 298         }
 299         if (ch >= 0xd800 && ch <= 0xdfff) ch = 0xfffd;  /* invalid surrogate pair */
 300
 301         /* 0x800-0xffff: 3 bytes */
 302         dst[2] = 0x80 | (ch & 0x3f);
 303         ch >>= 6;
 304         dst[1] = 0x80 | (ch & 0x3f);
 305         ch >>= 6;
 306         dst[0] = 0xe0 | ch;
 307         dst += 3;
 308     }
 309     *dst = 0;
 310     *dstlen = dst - ret;
 311     return ret;
 312 }
 313
 314 #ifdef _WIN32
 315
 316 int is_valid_codepage(int id)
 317 {
 318     return IsValidCodePage( id );
 319 }
 320
 321 WCHAR *codepage_to_unicode( int codepage, const char *src, int srclen, int *dstlen )
 322 {
 323     WCHAR *dst = xmalloc( (srclen + 1) * sizeof(WCHAR) );
 324     DWORD ret = MultiByteToWideChar( codepage, MB_ERR_INVALID_CHARS, src, srclen, dst, srclen );
 325     if (!ret) return NULL;
 326     dst[ret] = 0;
 327     *dstlen = ret;
 328     return dst;
 329 }
 330
 331 unsigned int get_language_from_name( const char *name )
 332 {
 333     WCHAR nameW[LOCALE_NAME_MAX_LENGTH];
 334
 335     MultiByteToWideChar( 1252, 0, name, -1, nameW, ARRAY_SIZE(nameW) );
 336     return LocaleNameToLCID( nameW, LOCALE_ALLOW_NEUTRAL_NAMES );
 337 }
 338
 339 #else  /* _WIN32 */
 340
 341 struct nls_info
 342 {
 343     unsigned short  codepage;
 344     unsigned short  unidef;
 345     unsigned short  trans_unidef;
 346     unsigned short *cp2uni;
 347     unsigned short *dbcs_offsets;
 348 };
 349
 350 static struct nls_info nlsinfo[128];
 351
 352 static void init_nls_info( struct nls_info *info, unsigned short *ptr )
 353 {
 354     unsigned short hdr_size = ptr[0];
 355
 356     info->codepage      = ptr[1];
 357     info->unidef        = ptr[4];
 358     info->trans_unidef  = ptr[6];
 359     ptr += hdr_size;
 360     info->cp2uni = ++ptr;
 361     ptr += 256;
 362     if (*ptr++) ptr += 256;  /* glyph table */
 363     info->dbcs_offsets  = *ptr ? ptr + 1 : NULL;
 364 }
 365
 366 static void *load_nls_file( const char *name )
 367 {
 368     unsigned int i;
 369     void *data;
 370     size_t size;
 371
 372     for (i = 0; nlsdirs[i]; i++)
 373     {
 374         char *path = strmake( "%s/%s", nlsdirs[i], name );
 375         if ((data = read_file( path, &size )))
 376         {
 377             free( path );
 378             return data;
 379         }
 380         free( path );
 381     }
 382     return NULL;
 383 }
 384
 385 static const struct nls_info *get_nls_info( unsigned int codepage )
 386 {
 387     unsigned short *data;
 388     unsigned int i;
 389
 390     for (i = 0; i < ARRAY_SIZE(nlsinfo) && nlsinfo[i].codepage; i++)
 391         if (nlsinfo[i].codepage == codepage) return &nlsinfo[i];
 392
 393     assert( i < ARRAY_SIZE(nlsinfo) );
 394
 395     if ((data = load_nls_file( strmake( "c_%03u.nls", codepage ))))
 396     {
 397         init_nls_info( &nlsinfo[i], data );
 398         return &nlsinfo[i];
 399     }
 400     return NULL;
 401 }
 402
 403 int is_valid_codepage(int cp)
 404 {
 405     return cp == CP_UTF8 || get_nls_info( cp );
 406 }
 407
 408 WCHAR *codepage_to_unicode( int codepage, const char *src, int srclen, int *dstlen )
 409 {
 410     const struct nls_info *info = get_nls_info( codepage );
 411     unsigned int i;
 412     WCHAR dbch, *dst = xmalloc( (srclen + 1) * sizeof(WCHAR) );
 413
 414     if (!info) error( "codepage %u not supported\n", codepage );
 415
 416     if (info->dbcs_offsets)
 417     {
 418         for (i = 0; srclen; i++, srclen--, src++)
 419         {
 420             unsigned short off = info->dbcs_offsets[(unsigned char)*src];
 421             if (off)
 422             {
 423                 if (srclen == 1) return NULL;
 424                 dbch = (src[0] << 8) | (unsigned char)src[1];
 425                 src++;
 426                 srclen--;
 427                 dst[i] = info->dbcs_offsets[off + (unsigned char)*src];
 428                 if (dst[i] == info->unidef && dbch != info->trans_unidef) return NULL;
 429             }
 430             else
 431             {
 432                 dst[i] = info->cp2uni[(unsigned char)*src];
 433                 if (dst[i] == info->unidef && *src != info->trans_unidef) return NULL;
 434             }
 435         }
 436     }
 437     else
 438     {
 439         for (i = 0; i < srclen; i++)
 440         {
 441             dst[i] = info->cp2uni[(unsigned char)src[i]];
 442             if (dst[i] == info->unidef && src[i] != info->trans_unidef) return NULL;
 443         }
 444     }
 445     dst[i] = 0;
 446     *dstlen = i;
 447     return dst;
 448 }
 449
 450 static const NLS_LOCALE_LCID_INDEX *lcids_index;
 451 static const NLS_LOCALE_HEADER *locale_table;
 452 static const NLS_LOCALE_LCNAME_INDEX *lcnames_index;
 453 static const WCHAR *locale_strings;
 454
 455 static void load_locale_nls(void)
 456 {
 457     struct
 458     {
 459         unsigned int ctypes;
 460         unsigned int unknown1;
 461         unsigned int unknown2;
 462         unsigned int unknown3;
 463         unsigned int locales;
 464         unsigned int charmaps;
 465         unsigned int geoids;
 466         unsigned int scripts;
 467     } *header;
 468
 469     if (!(header = load_nls_file( "locale.nls" ))) error( "unable to load locale.nls\n" );
 470     locale_table = (const NLS_LOCALE_HEADER *)((char *)header + header->locales);
 471     lcids_index = (const NLS_LOCALE_LCID_INDEX *)((char *)locale_table + locale_table->lcids_offset);
 472     lcnames_index = (const NLS_LOCALE_LCNAME_INDEX *)((char *)locale_table + locale_table->lcnames_offset);
 473     locale_strings = (const WCHAR *)((char *)locale_table + locale_table->strings_offset);
 474 }
 475
 476 static int compare_locale_names( const char *n1, const WCHAR *n2 )
 477 {
 478     for (;;)
 479     {
 480         WCHAR ch1 = (unsigned char)*n1++;
 481         WCHAR ch2 = *n2++;
 482         if (ch1 >= 'a' && ch1 <= 'z') ch1 -= 'a' - 'A';
 483         if (ch2 >= 'a' && ch2 <= 'z') ch2 -= 'a' - 'A';
 484         if (!ch1 || ch1 != ch2) return ch1 - ch2;
 485     }
 486 }
 487
 488 static const NLS_LOCALE_LCNAME_INDEX *find_lcname_entry( const char *name )
 489 {
 490     int min = 0, max = locale_table->nb_lcnames - 1;
 491
 492     if (!name) return NULL;
 493     while (min <= max)
 494     {
 495         int res, pos = (min + max) / 2;
 496         const WCHAR *str = locale_strings + lcnames_index[pos].name;
 497         res = compare_locale_names( name, str + 1 );
 498         if (res < 0) max = pos - 1;
 499         else if (res > 0) min = pos + 1;
 500         else return &lcnames_index[pos];
 501     }
 502     return NULL;
 503 }
 504
 505 static const NLS_LOCALE_DATA *get_locale_data( UINT idx )
 506 {
 507     ULONG offset = locale_table->locales_offset + idx * locale_table->locale_size;
 508     return (const NLS_LOCALE_DATA *)((const char *)locale_table + offset);
 509 }
 510
 511 unsigned int get_language_from_name( const char *name )
 512 {
 513     const NLS_LOCALE_LCNAME_INDEX *entry;
 514
 515     if (!locale_table) load_locale_nls();
 516     if (!(entry = find_lcname_entry( name ))) return 0;
 517     return get_locale_data( entry->idx )->unique_lcid;
 518 }
 519
 520 #endif  /* _WIN32 */
 521
 522 unsigned char *output_buffer;
 523 size_t output_buffer_pos;
 524 size_t output_buffer_size;