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 "winnls.h"
  32 #include "utils.h"
  33
  34 #define SUPPRESS_YACC_ERROR_MESSAGE
  35
  36 static void generic_msg(const char *s, const char *t, va_list ap)
  37 {
  38         fprintf(stderr, "%s:%d:%d: %s: ", input_name ? input_name : "stdin", line_number, char_number, t);
  39         vfprintf(stderr, s, ap);
  40 }
  41
  42 /*
  43  * The yyerror routine should not exit because we use the error-token
  44  * to determine the syntactic error in the source. However, YACC
  45  * uses the same routine to print an error just before the error
  46  * token is reduced.
  47  * The extra routine 'xyyerror' is used to exit after giving a real
  48  * message.
  49  */
  50 int mcy_error(const char *s, ...)
  51 {
  52 #ifndef SUPPRESS_YACC_ERROR_MESSAGE
  53         va_list ap;
  54         va_start(ap, s);
  55         generic_msg(s, "Yacc error", ap);
  56         va_end(ap);
  57 #endif
  58         return 1;
  59 }
  60
  61 int xyyerror(const char *s, ...)
  62 {
  63         va_list ap;
  64         va_start(ap, s);
  65         generic_msg(s, "Error", ap);
  66         va_end(ap);
  67         exit(1);
  68         return 1;
  69 }
  70
  71 int mcy_warning(const char *s, ...)
  72 {
  73         va_list ap;
  74         va_start(ap, s);
  75         generic_msg(s, "Warning", ap);
  76         va_end(ap);
  77         return 0;
  78 }
  79
  80 void internal_error(const char *file, int line, const char *s, ...)
  81 {
  82         va_list ap;
  83         va_start(ap, s);
  84         fprintf(stderr, "Internal error (please report) %s %d: ", file, line);
  85         vfprintf(stderr, s, ap);
  86         va_end(ap);
  87         exit(3);
  88 }
  89
  90 void fatal_perror( const char *msg, ... )
  91 {
  92         va_list valist;
  93         va_start( valist, msg );
  94         fprintf(stderr, "Error: ");
  95         vfprintf( stderr, msg, valist );
  96         perror( " " );
  97         va_end( valist );
  98         exit(2);
  99 }
 100
 101 void error(const char *s, ...)
 102 {
 103         va_list ap;
 104         va_start(ap, s);
 105         fprintf(stderr, "Error: ");
 106         vfprintf(stderr, s, ap);
 107         va_end(ap);
 108         exit(2);
 109 }
 110
 111 void warning(const char *s, ...)
 112 {
 113         va_list ap;
 114         va_start(ap, s);
 115         fprintf(stderr, "Warning: ");
 116         vfprintf(stderr, s, ap);
 117         va_end(ap);
 118 }
 119
 120 int unistrlen(const WCHAR *s)
 121 {
 122         int n;
 123         for(n = 0; *s; n++, s++)
 124                 ;
 125         return n;
 126 }
 127
 128 WCHAR *unistrcpy(WCHAR *dst, const WCHAR *src)
 129 {
 130         WCHAR *t = dst;
 131         while(*src)
 132                 *t++ = *src++;
 133         *t = 0;
 134         return dst;
 135 }
 136
 137 WCHAR *xunistrdup(const WCHAR * str)
 138 {
 139         WCHAR *s;
 140
 141         assert(str != NULL);
 142         s = xmalloc((unistrlen(str)+1) * sizeof(WCHAR));
 143         return unistrcpy(s, str);
 144 }
 145
 146 int unistricmp(const WCHAR *s1, const WCHAR *s2)
 147 {
 148         int i;
 149         int once = 0;
 150         static const char warn[] = "Don't know the uppercase equivalent of non ascii characters;"
 151                              "comparison might yield wrong results";
 152         while(*s1 && *s2)
 153         {
 154                 if((*s1 & 0xffff) > 0x7f || (*s2 & 0xffff) > 0x7f)
 155                 {
 156                         if(!once)
 157                         {
 158                                 once++;
 159                                 mcy_warning(warn);
 160                         }
 161                         i = *s1++ - *s2++;
 162                 }
 163                 else
 164                         i = toupper(*s1++) - toupper(*s2++);
 165                 if(i)
 166                         return i;
 167         }
 168
 169         if((*s1 & 0xffff) > 0x7f || (*s2 & 0xffff) > 0x7f)
 170         {
 171                 if(!once)
 172                         mcy_warning(warn);
 173                 return *s1 - *s2;
 174         }
 175         else
 176                 return  toupper(*s1) - toupper(*s2);
 177 }
 178
 179 int unistrcmp(const WCHAR *s1, const WCHAR *s2)
 180 {
 181         int i;
 182         while(*s1 && *s2)
 183         {
 184                 i = *s1++ - *s2++;
 185                 if(i)
 186                         return i;
 187         }
 188
 189         return *s1 - *s2;
 190 }
 191
 192 WCHAR *utf8_to_unicode( const char *src, int srclen, int *dstlen )
 193 {
 194     static const char utf8_length[128] =
 195     {
 196         0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x80-0x8f */
 197         0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x90-0x9f */
 198         0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xa0-0xaf */
 199         0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xb0-0xbf */
 200         0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xc0-0xcf */
 201         1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xd0-0xdf */
 202         2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* 0xe0-0xef */
 203         3,3,3,3,3,0,0,0,0,0,0,0,0,0,0,0  /* 0xf0-0xff */
 204     };
 205     static const unsigned char utf8_mask[4] = { 0x7f, 0x1f, 0x0f, 0x07 };
 206
 207     const char *srcend = src + srclen;
 208     int len, res;
 209     WCHAR *ret, *dst;
 210
 211     dst = ret = xmalloc( (srclen + 1) * sizeof(WCHAR) );
 212     while (src < srcend)
 213     {
 214         unsigned char ch = *src++;
 215         if (ch < 0x80)  /* special fast case for 7-bit ASCII */
 216         {
 217             *dst++ = ch;
 218             continue;
 219         }
 220         len = utf8_length[ch - 0x80];
 221         if (len && src + len <= srcend)
 222         {
 223             res = ch & utf8_mask[len];
 224             switch (len)
 225             {
 226             case 3:
 227                 if ((ch = *src ^ 0x80) >= 0x40) break;
 228                 res = (res << 6) | ch;
 229                 src++;
 230                 if (res < 0x10) break;
 231             case 2:
 232                 if ((ch = *src ^ 0x80) >= 0x40) break;
 233                 res = (res << 6) | ch;
 234                 if (res >= 0x110000 >> 6) break;
 235                 src++;
 236                 if (res < 0x20) break;
 237                 if (res >= 0xd800 >> 6 && res <= 0xdfff >> 6) break;
 238             case 1:
 239                 if ((ch = *src ^ 0x80) >= 0x40) break;
 240                 res = (res << 6) | ch;
 241                 src++;
 242                 if (res < 0x80) break;
 243                 if (res <= 0xffff) *dst++ = res;
 244                 else
 245                 {
 246                     res -= 0x10000;
 247                     *dst++ = 0xd800 | (res >> 10);
 248                     *dst++ = 0xdc00 | (res & 0x3ff);
 249                 }
 250                 continue;
 251             }
 252         }
 253         *dst++ = 0xfffd;
 254     }
 255     *dst = 0;
 256     *dstlen = dst - ret;
 257     return ret;
 258 }
 259
 260 char *unicode_to_utf8( const WCHAR *src, int srclen, int *dstlen )
 261 {
 262     char *ret, *dst;
 263
 264     dst = ret = xmalloc( srclen * 3 + 1 );
 265     for ( ; srclen; srclen--, src++)
 266     {
 267         unsigned int ch = *src;
 268
 269         if (ch < 0x80)  /* 0x00-0x7f: 1 byte */
 270         {
 271             *dst++ = ch;
 272             continue;
 273         }
 274         if (ch < 0x800)  /* 0x80-0x7ff: 2 bytes */
 275         {
 276             dst[1] = 0x80 | (ch & 0x3f);
 277             ch >>= 6;
 278             dst[0] = 0xc0 | ch;
 279             dst += 2;
 280             continue;
 281         }
 282         if (ch >= 0xd800 && ch <= 0xdbff && srclen > 1 && src[1] >= 0xdc00 && src[1] <= 0xdfff)
 283         {
 284             /* 0x10000-0x10ffff: 4 bytes */
 285             ch = 0x10000 + ((ch & 0x3ff) << 10) + (src[1] & 0x3ff);
 286             dst[3] = 0x80 | (ch & 0x3f);
 287             ch >>= 6;
 288             dst[2] = 0x80 | (ch & 0x3f);
 289             ch >>= 6;
 290             dst[1] = 0x80 | (ch & 0x3f);
 291             ch >>= 6;
 292             dst[0] = 0xf0 | ch;
 293             dst += 4;
 294             src++;
 295             srclen--;
 296             continue;
 297         }
 298         if (ch >= 0xd800 && ch <= 0xdfff) ch = 0xfffd;  /* invalid surrogate pair */
 299
 300         /* 0x800-0xffff: 3 bytes */
 301         dst[2] = 0x80 | (ch & 0x3f);
 302         ch >>= 6;
 303         dst[1] = 0x80 | (ch & 0x3f);
 304         ch >>= 6;
 305         dst[0] = 0xe0 | ch;
 306         dst += 3;
 307     }
 308     *dst = 0;
 309     *dstlen = dst - ret;
 310     return ret;
 311 }
 312
 313 #ifdef _WIN32
 314
 315 int is_valid_codepage(int id)
 316 {
 317     return IsValidCodePage( id );
 318 }
 319
 320 WCHAR *codepage_to_unicode( int codepage, const char *src, int srclen, int *dstlen )
 321 {
 322     WCHAR *dst = xmalloc( (srclen + 1) * sizeof(WCHAR) );
 323     DWORD ret = MultiByteToWideChar( codepage, MB_ERR_INVALID_CHARS, src, srclen, dst, srclen );
 324     if (!ret) return NULL;
 325     dst[ret] = 0;
 326     *dstlen = ret;
 327     return dst;
 328 }
 329
 330 #else  /* _WIN32 */
 331
 332 struct nls_info
 333 {
 334     unsigned short  codepage;
 335     unsigned short  unidef;
 336     unsigned short  trans_unidef;
 337     unsigned short *cp2uni;
 338     unsigned short *dbcs_offsets;
 339 };
 340
 341 static struct nls_info nlsinfo[128];
 342
 343 static void init_nls_info( struct nls_info *info, unsigned short *ptr )
 344 {
 345     unsigned short hdr_size = ptr[0];
 346
 347     info->codepage      = ptr[1];
 348     info->unidef        = ptr[4];
 349     info->trans_unidef  = ptr[6];
 350     ptr += hdr_size;
 351     info->cp2uni = ++ptr;
 352     ptr += 256;
 353     if (*ptr++) ptr += 256;  /* glyph table */
 354     info->dbcs_offsets  = *ptr ? ptr + 1 : NULL;
 355 }
 356
 357 static const struct nls_info *get_nls_info( unsigned int codepage )
 358 {
 359     unsigned short *data;
 360     char *path;
 361     unsigned int i;
 362     size_t size;
 363
 364     for (i = 0; i < ARRAY_SIZE(nlsinfo) && nlsinfo[i].codepage; i++)
 365         if (nlsinfo[i].codepage == codepage) return &nlsinfo[i];
 366
 367     assert( i < ARRAY_SIZE(nlsinfo) );
 368
 369     for (i = 0; nlsdirs[i]; i++)
 370     {
 371         path = strmake( "%s/c_%03u.nls", nlsdirs[i], codepage );
 372         if ((data = read_file( path, &size )))
 373         {
 374             free( path );
 375             init_nls_info( &nlsinfo[i], data );
 376             return &nlsinfo[i];
 377         }
 378         free( path );
 379     }
 380     return NULL;
 381 }
 382
 383 int is_valid_codepage(int cp)
 384 {
 385     return cp == CP_UTF8 || get_nls_info( cp );
 386 }
 387
 388 WCHAR *codepage_to_unicode( int codepage, const char *src, int srclen, int *dstlen )
 389 {
 390     const struct nls_info *info = get_nls_info( codepage );
 391     unsigned int i;
 392     WCHAR dbch, *dst = xmalloc( (srclen + 1) * sizeof(WCHAR) );
 393
 394     if (!info) error( "codepage %u not supported\n", codepage );
 395
 396     if (info->dbcs_offsets)
 397     {
 398         for (i = 0; srclen; i++, srclen--, src++)
 399         {
 400             unsigned short off = info->dbcs_offsets[(unsigned char)*src];
 401             if (off)
 402             {
 403                 if (srclen == 1) return NULL;
 404                 dbch = (src[0] << 8) | (unsigned char)src[1];
 405                 src++;
 406                 srclen--;
 407                 dst[i] = info->dbcs_offsets[off + (unsigned char)*src];
 408                 if (dst[i] == info->unidef && dbch != info->trans_unidef) return NULL;
 409             }
 410             else
 411             {
 412                 dst[i] = info->cp2uni[(unsigned char)*src];
 413                 if (dst[i] == info->unidef && *src != info->trans_unidef) return NULL;
 414             }
 415         }
 416     }
 417     else
 418     {
 419         for (i = 0; i < srclen; i++)
 420         {
 421             dst[i] = info->cp2uni[(unsigned char)src[i]];
 422             if (dst[i] == info->unidef && src[i] != info->trans_unidef) return NULL;
 423         }
 424     }
 425     dst[i] = 0;
 426     *dstlen = i;
 427     return dst;
 428 }
 429
 430 #endif  /* _WIN32 */
 431
 432 unsigned char *output_buffer;
 433 size_t output_buffer_pos;
 434 size_t output_buffer_size;