hfs_encodings/hfs_encodings.c

   1 /*
   2  * Copyright (c) 2000-2015 Apple Inc. All rights reserved.
   3  *
   4  * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
   5  *
   6  * This file contains Original Code and/or Modifications of Original Code
   7  * as defined in and that are subject to the Apple Public Source License
   8  * Version 2.0 (the 'License'). You may not use this file except in
   9  * compliance with the License. The rights granted to you under the License
  10  * may not be used to create, or enable the creation or redistribution of,
  11  * unlawful or unlicensed copies of an Apple operating system, or to
  12  * circumvent, violate, or enable the circumvention or violation of, any
  13  * terms of an Apple operating system software license agreement.
  14  *
  15  * Please obtain a copy of the License at
  16  * http://www.opensource.apple.com/apsl/ and read it before using this file.
  17  *
  18  * The Original Code and all software distributed under the License are
  19  * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  20  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  21  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
  22  * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
  23  * Please see the License for the specific language governing rights and
  24  * limitations under the License.
  25  *
  26  * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
  27  */
  28
  29 #include <IOKit/IOLib.h>
  30
  31 #include <sys/types.h>
  32 #include <sys/errno.h>
  33
  34 #include <sys/param.h>
  35 #include <sys/systm.h>
  36 #include <sys/kernel.h>
  37 #include <sys/malloc.h>
  38 #include <sys/queue.h>
  39 #include <sys/utfconv.h>
  40 #include <kern/host.h>
  41 #include <mach/host_priv.h>
  42 #include <libkern/OSKextLib.h>
  43 #include <libkern/OSKextLibPrivate.h>
  44
  45 #include "hfs_encodings.h"
  46 #include "../core/hfs_macos_defs.h"
  47
  48 lck_grp_t * encodinglst_lck_grp;
  49 lck_grp_attr_t * encodinglst_lck_grp_attr;
  50 lck_attr_t * encodinglst_lck_attr;
  51
  52
  53 /* hfs encoding converter list */
  54 SLIST_HEAD(encodinglst, hfs_encoding) hfs_encoding_list = {0};
  55
  56 lck_mtx_t  encodinglst_mutex;
  57
  58 /* hfs encoding converter entry */
  59 struct  hfs_encoding {
  60         SLIST_ENTRY(hfs_encoding)  link;
  61         int                     refcount;
  62         int                     kmod_id;
  63         u_int32_t       encoding;
  64         hfs_to_unicode_func_t   get_unicode_func;
  65         unicode_to_hfs_func_t   get_hfsname_func;
  66 };
  67
  68 void
  69 hfs_converterinit(void)
  70 {
  71         SLIST_INIT(&hfs_encoding_list);
  72
  73         encodinglst_lck_grp_attr= lck_grp_attr_alloc_init();
  74         encodinglst_lck_grp  = lck_grp_alloc_init("cnode_hash", encodinglst_lck_grp_attr);
  75         encodinglst_lck_attr = lck_attr_alloc_init();
  76
  77         lck_mtx_init(&encodinglst_mutex, encodinglst_lck_grp, encodinglst_lck_attr);
  78
  79         /*
  80          * add resident MacRoman converter and take a reference
  81          * since its always "loaded". MacRoman is the default converter
  82          * for HFS standard volumes.
  83          *
  84          * Only do this if we are actually supporting HFS standard
  85          * volumes. The converter is not used on configurations
  86          * that do not support HFS standard.
  87          */
  88         hfs_addconverter(0, kTextEncodingMacRoman, mac_roman_to_unicode, unicode_to_mac_roman);
  89         SLIST_FIRST(&hfs_encoding_list)->refcount++;
  90 }
  91
  92 void hfs_converterdone(void)
  93 {
  94         if (encodinglst_lck_grp_attr)
  95                 lck_grp_attr_free(encodinglst_lck_grp_attr);
  96         if (encodinglst_lck_grp)
  97                 lck_grp_free(encodinglst_lck_grp);
  98         if (encodinglst_lck_attr)
  99                 lck_attr_free(encodinglst_lck_attr);
 100 }
 101
 102 /*
 103  * For configurations that do support HFS standard, we need all of these..
 104  */
 105
 106 /*
 107  * hfs_addconverter - add an HFS encoding converter
 108  *
 109  * This is called exclusivly by kernel loadable modules
 110  * (like HFS_Japanese.kmod) to register hfs encoding
 111  * conversion routines.
 112  *
 113  */
 114 int
 115 hfs_addconverter(int id, u_int32_t encoding, hfs_to_unicode_func_t get_unicode, unicode_to_hfs_func_t get_hfsname)
 116 {
 117         struct hfs_encoding *encp = (struct hfs_encoding *)IOMalloc(sizeof(*encp));
 118
 119         lck_mtx_lock(&encodinglst_mutex);
 120
 121         encp->link.sle_next = NULL;
 122         encp->refcount = 0;
 123         encp->encoding = encoding;
 124         encp->get_unicode_func = get_unicode;
 125         encp->get_hfsname_func = get_hfsname;
 126         encp->kmod_id = id;
 127         SLIST_INSERT_HEAD(&hfs_encoding_list, encp, link);
 128
 129         lck_mtx_unlock(&encodinglst_mutex);
 130
 131         if (id)
 132                 OSKextRetainKextWithLoadTag(id);
 133
 134         return (0);
 135 }
 136
 137
 138 /*
 139  * hfs_remconverter - remove an HFS encoding converter
 140  *
 141  * Can be called by a kernel loadable module's finalize
 142  * routine to remove an encoding converter so that the
 143  * module (i.e. the code) can be unloaded.
 144  *
 145  * However, in the normal case, the removing and unloading
 146  * of these converters is done in hfs_relconverter.
 147  * The call is initiated from within the kernel during the unmounting of an hfs voulume.
 148  */
 149 int
 150 hfs_remconverter(int id, u_int32_t encoding)
 151 {
 152         struct hfs_encoding *encp;
 153
 154         lck_mtx_lock(&encodinglst_mutex);
 155         SLIST_FOREACH(encp, &hfs_encoding_list, link) {
 156                 if (encp->encoding == encoding && encp->kmod_id == id) {
 157                         encp->refcount--;
 158
 159                         /* if converter is no longer in use, release it */
 160                         if (encp->refcount <= 0 && encp->kmod_id != 0) {
 161                                 SLIST_REMOVE(&hfs_encoding_list, encp, hfs_encoding, link);
 162                                 lck_mtx_unlock(&encodinglst_mutex);
 163                                 OSKextReleaseKextWithLoadTag(encp->kmod_id);
 164                                 IOFree(encp, sizeof(*encp));
 165                                 return (0);
 166                         } else {
 167                                 lck_mtx_unlock(&encodinglst_mutex);
 168                                 return (1);   /* busy */
 169                         }
 170                         break;
 171                 }
 172         }
 173         lck_mtx_unlock(&encodinglst_mutex);
 174
 175         return (0);
 176 }
 177
 178 /*
 179  * hfs_getconverter - get HFS encoding converters
 180  *
 181  * Normally called during the mounting of an hfs voulume.
 182  */
 183 int
 184 hfs_getconverter(u_int32_t encoding, hfs_to_unicode_func_t *get_unicode, unicode_to_hfs_func_t *get_hfsname)
 185 {
 186         struct hfs_encoding *encp;
 187         int found = 0;
 188
 189         lck_mtx_lock(&encodinglst_mutex);
 190         SLIST_FOREACH(encp, &hfs_encoding_list, link) {
 191                 if (encp->encoding == encoding) {
 192                         found = 1;
 193                         *get_unicode = encp->get_unicode_func;
 194                         *get_hfsname = encp->get_hfsname_func;
 195                         ++encp->refcount;
 196                         break;
 197                 }
 198         }
 199         lck_mtx_unlock(&encodinglst_mutex);
 200
 201         if (!found) {
 202                 *get_unicode = NULL;
 203                 *get_hfsname = NULL;
 204                 return (EINVAL);
 205         }
 206
 207         return (0);
 208 }
 209
 210
 211 /*
 212  * hfs_relconverter - release interest in an HFS encoding converter
 213  *
 214  * Normally called during the unmounting of an hfs voulume.
 215  */
 216 int
 217 hfs_relconverter(u_int32_t encoding)
 218 {
 219         struct hfs_encoding *encp;
 220
 221         lck_mtx_lock(&encodinglst_mutex);
 222         SLIST_FOREACH(encp, &hfs_encoding_list, link) {
 223                 if (encp->encoding == encoding) {
 224                         encp->refcount--;
 225
 226                         /* if converter is no longer in use, release it */
 227                         if (encp->refcount <= 0 && encp->kmod_id != 0) {
 228                                 SLIST_REMOVE(&hfs_encoding_list, encp, hfs_encoding, link);
 229                                 lck_mtx_unlock(&encodinglst_mutex);
 230
 231                                 OSKextReleaseKextWithLoadTag(encp->kmod_id);
 232                                 IOFree(encp, sizeof(*encp));
 233                                 return (0);
 234                         }
 235                         lck_mtx_unlock(&encodinglst_mutex);
 236                         return (0);
 237                 }
 238         }
 239         lck_mtx_unlock(&encodinglst_mutex);
 240
 241         return (EINVAL);
 242 }
 243
 244 /*
 245  * When an HFS name cannot be encoded with the current
 246  * volume encoding then MacRoman is used as a fallback.
 247  */
 248 int
 249 mac_roman_to_utf8(const Str31 hfs_str, ByteCount maxDstLen, ByteCount *actualDstLen, unsigned char* dstStr)
 250 {
 251         int error;
 252         UniChar uniStr[MAX_HFS_UNICODE_CHARS];
 253         ItemCount uniCount;
 254         size_t utf8len;
 255         u_int8_t pascal_length = 0;
 256
 257         /*
 258          * Validate the length of the Pascal-style string before passing it
 259          * down to the decoding engine.
 260          */
 261         pascal_length = *((const u_int8_t*)(hfs_str));
 262         if (pascal_length > 31) {
 263                 /* invalid string; longer than 31 bytes */
 264                 error = EINVAL;
 265                 return error;
 266         }
 267
 268         error = mac_roman_to_unicode(hfs_str, uniStr, MAX_HFS_UNICODE_CHARS, &uniCount);
 269
 270         if (uniCount == 0)
 271                 error = EINVAL;
 272
 273         if (error == 0) {
 274                 error = utf8_encodestr(uniStr, uniCount * sizeof(UniChar), dstStr, &utf8len, maxDstLen , ':', 0);
 275                 if (error == ENAMETOOLONG)
 276                         *actualDstLen = utf8_encodelen(uniStr, uniCount * sizeof(UniChar), ':', 0);
 277                 else
 278                         *actualDstLen = utf8len;
 279         }
 280
 281         return error;
 282 }
 283
 284 int
 285 utf8_to_mac_roman(ByteCount srcLen, const unsigned char* srcStr, Str31 dstStr)
 286 {
 287         int error;
 288         UniChar uniStr[MAX_HFS_UNICODE_CHARS];
 289         size_t ucslen;
 290
 291         error = utf8_decodestr(srcStr, srcLen, uniStr, &ucslen, sizeof(uniStr), ':', 0);
 292         if (error == 0)
 293                 error = unicode_to_mac_roman(uniStr, ucslen/sizeof(UniChar), dstStr);
 294
 295         return error;
 296 }
 297
 298 /*
 299  * HFS MacRoman to/from Unicode conversions are built into the kernel
 300  * All others hfs encodings are loadable.
 301  */
 302
 303 /* 0x00A0 - 0x00FF = Latin 1 Supplement (30 total) */
 304 static u_int8_t gLatin1Table[] = {
 305   /*              0     1     2     3     4     5     6     7     8     9     A     B     C     D     E     F  */
 306   /* 0x00A0 */  0xCA, 0xC1, 0xA2, 0xA3, 0xDB, 0xB4,  '?', 0xA4, 0xAC, 0xA9, 0xBB, 0xC7, 0xC2,  '?', 0xA8, 0xF8,
 307   /* 0x00B0 */  0xA1, 0XB1,  '?',  '?', 0xAB, 0xB5, 0xA6, 0xe1, 0xFC,  '?', 0xBC, 0xC8,  '?',  '?',  '?', 0xC0,
 308   /* 0x00C0 */   '?',  '?',  '?',  '?',  '?',  '?', 0xAE,  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 309   /* 0x00D0 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xAF,  '?',  '?',  '?',  '?',  '?',  '?', 0xA7,
 310   /* 0x00E0 */   '?',  '?',  '?',  '?',  '?',  '?', 0xBE,  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 311   /* 0x00F0 */   '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xD6, 0xBF,  '?',  '?',  '?',  '?',  '?',  '?',  '?'
 312 };
 313
 314 /* 0x02C0 - 0x02DF = Spacing Modifiers (8 total) */
 315 static u_int8_t gSpaceModsTable[] = {
 316   /*              0     1     2     3     4     5     6     7     8     9     A     B     C     D     E     F  */
 317   /* 0x02C0 */   '?',  '?',  '?',  '?',  '?',  '?', 0xF6, 0xFF,  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 318   /* 0x02D0 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xF9, 0xFA, 0xFB, 0xFE, 0xF7, 0xFD,  '?',  '?'
 319 };
 320
 321 /* 0x2010 - 0x20AF = General Punctuation (17 total) */
 322 static u_int8_t gPunctTable[] = {
 323   /*              0     1     2     3     4     5     6     7     8     9     A     B     C     D     E     F  */
 324   /* 0x2010 */   '?',  '?',  '?', 0xd0, 0xd1,  '?',  '?',  '?', 0xd4, 0xd5, 0xe2,  '?', 0xd2, 0xd3, 0xe3,  '?',
 325   /* 0x2020 */  0xa0, 0xe0, 0xa5,  '?',  '?',  '?', 0xc9,  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 326   /* 0x2030 */  0xe4,  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xdc, 0xdd,  '?',  '?',  '?',  '?',  '?',
 327   /* 0x2040 */   '?',  '?',  '?',  '?', 0xda,  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 328   /* 0x2050 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 329   /* 0x2060 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 330   /* 0x2070 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 331   /* 0x2080 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 332   /* 0x2090 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 333   /* 0x20A0 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xdb,  '?',  '?',  '?'
 334 };
 335
 336 /* 0x22xx = Mathematical Operators (11 total) */
 337 static u_int8_t gMathTable[] = {
 338   /*              0     1     2     3     4     5     6     7     8     9     A     B     C     D     E     F  */
 339   /* 0x2200 */   '?',  '?', 0xb6,  '?',  '?',  '?', 0xc6,  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xb8,
 340   /* 0x2210 */   '?', 0xb7,  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xc3,  '?',  '?',  '?', 0xb0,  '?',
 341   /* 0x2220 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xba,  '?',  '?',  '?',  '?',
 342   /* 0x2230 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 343   /* 0x2240 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xc5,  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 344   /* 0x2250 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 345   /* 0x2260 */  0xad,  '?',  '?',  '?', 0xb2, 0xb3,  '?',  '?'
 346 };
 347
 348 /* */
 349 static u_int8_t gReverseCombTable[] = {
 350   /*              0     1     2     3     4     5     6     7     8     9     A     B     C     D     E     F  */
 351   /* 0x40 */    0xDA, 0x40, 0xDA, 0xDA, 0xDA, 0x56, 0xDA, 0xDA, 0xDA, 0x6C, 0xDA, 0xDA, 0xDA, 0xDA, 0x82, 0x98,
 352   /* 0x50 */    0xDA, 0xDA, 0xDA, 0xDA, 0xDA, 0xAE, 0xDA, 0xDA, 0xDA, 0xC4, 0xDA, 0xDA, 0xDA, 0xDA, 0xDA, 0xDA,
 353   /* 0x60 */    0xDA, 0x4B, 0xDA, 0xDA, 0xDA, 0x61, 0xDA, 0xDA, 0xDA, 0x77, 0xDA, 0xDA, 0xDA, 0xDA, 0x8D, 0xA3,
 354   /* 0x70 */    0xDA, 0xDA, 0xDA, 0xDA, 0xDA, 0xB9, 0xDA, 0xDA, 0xDA, 0xCF, 0xDA, 0xDA, 0xDA, 0xDA, 0xDA, 0xDA,
 355
 356   /* Combining Diacritical Marks (0x0300 - 0x030A) */
 357   /*              0     1     2     3     4     5     6     7     8     9     A  */
 358   /*  'A'   */
 359   /* 0x0300 */  0xCB, 0xE7, 0xE5, 0xCC,  '?',  '?',  '?',  '?', 0x80,  '?', 0x81,
 360
 361   /*  'a'   */
 362   /* 0x0300 */  0x88, 0x87, 0x89, 0x8B,  '?',  '?',  '?',  '?', 0x8A,  '?', 0x8C,
 363
 364   /*  'E'   */
 365   /* 0x0300 */  0xE9, 0x83, 0xE6,  '?',  '?',  '?',  '?',  '?', 0xE8,  '?',  '?',
 366
 367   /*  'e'   */
 368   /* 0x0300 */  0x8F, 0x8E, 0x90,  '?',  '?',  '?',  '?',  '?', 0x91,  '?',  '?',
 369
 370   /*  'I'   */
 371   /* 0x0300 */  0xED, 0xEA, 0xEB,  '?',  '?',  '?',  '?',  '?', 0xEC,  '?',  '?',
 372
 373   /*  'i'   */
 374   /* 0x0300 */  0x93, 0x92, 0x94,  '?',  '?',  '?',  '?',  '?', 0x95,  '?',  '?',
 375
 376   /*  'N'   */
 377   /* 0x0300 */   '?',  '?',  '?', 0x84,  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 378
 379   /*  'n'   */
 380   /* 0x0300 */   '?',  '?',  '?', 0x96,  '?',  '?',  '?',  '?',  '?',  '?',  '?',
 381
 382   /*  'O'   */
 383   /* 0x0300 */  0xF1, 0xEE, 0xEF, 0xCD,  '?',  '?',  '?',  '?', 0x85,  '?',  '?',
 384
 385   /*  'o'   */
 386   /* 0x0300 */  0x98, 0x97, 0x99, 0x9B,  '?',  '?',  '?',  '?', 0x9A,  '?',  '?',
 387
 388   /*  'U'   */
 389   /* 0x0300 */  0xF4, 0xF2, 0xF3,  '?',  '?',  '?',  '?',  '?', 0x86,  '?',  '?',
 390
 391   /*  'u'   */
 392   /* 0x0300 */  0x9D, 0x9C, 0x9E,  '?',  '?',  '?',  '?',  '?', 0x9F,  '?',  '?',
 393
 394   /*  'Y'   */
 395   /* 0x0300 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xD9,  '?',  '?',
 396
 397   /*  'y'   */
 398   /* 0x0300 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?', 0xD8,  '?',  '?',
 399
 400   /*  else  */
 401   /* 0x0300 */   '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?',  '?'
 402 };
 403
 404
 405 /*
 406  * Convert Unicode string into HFS MacRoman encoding
 407  *
 408  * Assumes Unicode input is fully decomposed
 409  */
 410 int unicode_to_mac_roman(UniChar *uni_str, u_int32_t unicodeChars, Str31 hfs_str)
 411 {
 412         u_int8_t        *p;
 413         const UniChar   *u;
 414         UniChar         c;
 415         UniChar         mask;
 416         u_int16_t       inputChars;
 417         u_int16_t       pascalChars;
 418         OSErr           result = noErr;
 419         u_int8_t        lsb;
 420         u_int8_t        prevChar;
 421         u_int8_t        mc;
 422
 423         mask = (UniChar) 0xFF80;
 424         p = &hfs_str[1];
 425         u = uni_str;
 426         inputChars = unicodeChars;
 427         pascalChars = prevChar = 0;
 428
 429         while (inputChars) {
 430                 c = *(u++);
 431                 lsb = (u_int8_t) c;
 432
 433                 /*
 434                  * If its not 7-bit ascii, then we need to map it
 435                  */
 436                 if ( c & mask ) {
 437                         mc = '?';
 438                         switch (c & 0xFF00) {
 439                         case 0x0000:
 440                                 if (lsb >= 0xA0)
 441                                         mc = gLatin1Table[lsb - 0xA0];
 442                                 break;
 443
 444                         case 0x0200:
 445                                 if (lsb >= 0xC0 && lsb <= 0xDF)
 446                                         mc = gSpaceModsTable[lsb - 0xC0];
 447                                 break;
 448
 449                         case 0x2000:
 450                                 if (lsb >= 0x10 && lsb <= 0xAF)
 451                                         mc = gPunctTable[lsb- 0x10];
 452                                 break;
 453
 454                         case 0x2200:
 455                                 if (lsb < 0x68)
 456                                         mc = gMathTable[lsb];
 457                                 break;
 458
 459                         case 0x0300:
 460                                 if (c <= 0x030A) {
 461                                         if (prevChar >= 'A' && prevChar < 'z') {
 462                                                 mc = gReverseCombTable[gReverseCombTable[prevChar - 0x40] + lsb];
 463                                                 --p;    /* backup over base char */
 464                                                 --pascalChars;
 465                                         }
 466                                 } else {
 467                                         switch (c) {
 468                                         case 0x0327:    /* combining cedilla */
 469                                                 if (prevChar == 'C')
 470                                                         mc = 0x82;
 471                                                 else if (prevChar == 'c')
 472                                                         mc = 0x8D;
 473                                                 else
 474                                                         break;
 475                                                 --p;    /* backup over base char */
 476                                                 --pascalChars;
 477                                                 break;
 478
 479                                         case 0x03A9: mc = 0xBD; break;  /* omega */
 480
 481                                         case 0x03C0: mc = 0xB9; break;  /* pi */
 482                                         }
 483                                 }
 484                                 break;
 485
 486                         default:
 487                                 switch (c) {
 488                                 case 0x0131: mc = 0xf5; break;  /* dotless i */
 489
 490                                 case 0x0152: mc = 0xce; break;  /* OE */
 491
 492                                 case 0x0153: mc = 0xcf; break;  /* oe */
 493
 494                                 case 0x0192: mc = 0xc4; break;  /* Ä */
 495
 496                                 case 0x2122: mc = 0xaa; break;  /* TM */
 497
 498                                 case 0x25ca: mc = 0xd7; break;  /* diamond */
 499
 500                                 case 0xf8ff: mc = 0xf0; break;  /* apple logo */
 501
 502                                 case 0xfb01: mc = 0xde; break;  /* fi */
 503
 504                                 case 0xfb02: mc = 0xdf; break;  /* fl */
 505                                 }
 506                         } /* end switch (c & 0xFF00) */
 507
 508                         /*
 509                          * If we have an unmapped character then we need to mangle the name...
 510                          */
 511                         if (mc == '?')
 512                                 result = kTECUsedFallbacksStatus;
 513
 514                         prevChar = 0;
 515                         lsb = mc;
 516
 517                 } else {
 518                         prevChar = lsb;
 519                 }
 520
 521                 if (pascalChars >= 31)
 522                         break;
 523
 524                 *(p++) = lsb;
 525                 ++pascalChars;
 526                 --inputChars;
 527
 528         } /* end while */
 529
 530         hfs_str[0] = pascalChars;
 531
 532         if (inputChars > 0)
 533                 result = ENAMETOOLONG;  /* ran out of room! */
 534
 535         return result;
 536 }
 537
 538
 539 static UniChar gHiBitBaseUnicode[128] = {
 540   /* 0x80 */    0x0041, 0x0041, 0x0043, 0x0045, 0x004e, 0x004f, 0x0055, 0x0061,
 541   /* 0x88 */    0x0061, 0x0061, 0x0061, 0x0061, 0x0061, 0x0063, 0x0065, 0x0065,
 542   /* 0x90 */    0x0065, 0x0065, 0x0069, 0x0069, 0x0069, 0x0069, 0x006e, 0x006f,
 543   /* 0x98 */    0x006f, 0x006f, 0x006f, 0x006f, 0x0075, 0x0075, 0x0075, 0x0075,
 544   /* 0xa0 */    0x2020, 0x00b0, 0x00a2, 0x00a3, 0x00a7, 0x2022, 0x00b6, 0x00df,
 545   /* 0xa8 */    0x00ae, 0x00a9, 0x2122, 0x00b4, 0x00a8, 0x2260, 0x00c6, 0x00d8,
 546   /* 0xb0 */    0x221e, 0x00b1, 0x2264, 0x2265, 0x00a5, 0x00b5, 0x2202, 0x2211,
 547   /* 0xb8 */    0x220f, 0x03c0, 0x222b, 0x00aa, 0x00ba, 0x03a9, 0x00e6, 0x00f8,
 548   /* 0xc0 */    0x00bf, 0x00a1, 0x00ac, 0x221a, 0x0192, 0x2248, 0x2206, 0x00ab,
 549   /* 0xc8 */    0x00bb, 0x2026, 0x00a0, 0x0041, 0x0041, 0x004f, 0x0152, 0x0153,
 550   /* 0xd0 */    0x2013, 0x2014, 0x201c, 0x201d, 0x2018, 0x2019, 0x00f7, 0x25ca,
 551   /* 0xd8 */    0x0079, 0x0059, 0x2044, 0x20ac, 0x2039, 0x203a, 0xfb01, 0xfb02,
 552   /* 0xe0 */    0x2021, 0x00b7, 0x201a, 0x201e, 0x2030, 0x0041, 0x0045, 0x0041,
 553   /* 0xe8 */    0x0045, 0x0045, 0x0049, 0x0049, 0x0049, 0x0049, 0x004f, 0x004f,
 554   /* 0xf0 */    0xf8ff, 0x004f, 0x0055, 0x0055, 0x0055, 0x0131, 0x02c6, 0x02dc,
 555   /* 0xf8 */    0x00af, 0x02d8, 0x02d9, 0x02da, 0x00b8, 0x02dd, 0x02db, 0x02c7
 556 };
 557
 558 static UniChar gHiBitCombUnicode[128] = {
 559   /* 0x80 */    0x0308, 0x030a, 0x0327, 0x0301, 0x0303, 0x0308, 0x0308, 0x0301,
 560   /* 0x88 */    0x0300, 0x0302, 0x0308, 0x0303, 0x030a, 0x0327, 0x0301, 0x0300,
 561   /* 0x90 */    0x0302, 0x0308, 0x0301, 0x0300, 0x0302, 0x0308, 0x0303, 0x0301,
 562   /* 0x98 */    0x0300, 0x0302, 0x0308, 0x0303, 0x0301, 0x0300, 0x0302, 0x0308,
 563   /* 0xa0 */    0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
 564   /* 0xa8 */    0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
 565   /* 0xb0 */    0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
 566   /* 0xb8 */    0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
 567   /* 0xc0 */    0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
 568   /* 0xc8 */    0x0000, 0x0000, 0x0000, 0x0300, 0x0303, 0x0303, 0x0000, 0x0000,
 569   /* 0xd0 */    0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
 570   /* 0xd8 */    0x0308, 0x0308, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
 571   /* 0xe0 */    0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0302, 0x0302, 0x0301,
 572   /* 0xe8 */    0x0308, 0x0300, 0x0301, 0x0302, 0x0308, 0x0300, 0x0301, 0x0302,
 573   /* 0xf0 */    0x0000, 0x0300, 0x0301, 0x0302, 0x0300, 0x0000, 0x0000, 0x0000,
 574   /* 0xf8 */    0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
 575 };
 576
 577
 578 /*
 579  * Convert HFS MacRoman encoded string into Unicode
 580  *
 581  * Unicode output is fully decomposed
 582  */
 583 int
 584 mac_roman_to_unicode(const Str31 hfs_str, UniChar *uni_str,
 585                                 __unused u_int32_t maxCharLen, u_int32_t *unicodeChars)
 586 {
 587         const u_int8_t  *p;
 588         UniChar  *u;
 589         u_int16_t  pascalChars;
 590         u_int8_t  c;
 591
 592         p = hfs_str;
 593         u = uni_str;
 594
 595         *unicodeChars = pascalChars = *(p++);   /* pick up length byte */
 596
 597         while (pascalChars--) {
 598                 c = *(p++);
 599
 600                 if ( (int8_t) c >= 0 ) {                /* check if seven bit ascii */
 601                         *(u++) = (UniChar) c;   /* just pad high byte with zero */
 602                 } else { /* its a hi bit character */
 603                         UniChar uc;
 604
 605                         c &= 0x7F;
 606                         *(u++) = uc = gHiBitBaseUnicode[c];
 607
 608                         /*
 609                          * if the unicode character we get back is an alpha char
 610                          * then we must have an additional combining character
 611                          */
 612                         if ((uc <= (UniChar) 'z') && (uc >= (UniChar) 'A')) {
 613                                 *(u++) = gHiBitCombUnicode[c];
 614                                 ++(*unicodeChars);
 615                         }
 616                 }
 617         }
 618
 619         return noErr;
 620 }