]> git.saurik.com Git - apple/xnu.git/blob - bsd/hfs/hfs_encodings.c
fdf03bc701619538fb374d0ac28db89b7e95f9bd
[apple/xnu.git] / bsd / hfs / hfs_encodings.c
1 /*
2 * Copyright (c) 2000-2002 Apple Computer, Inc. All rights reserved.
3 *
4 * @APPLE_LICENSE_HEADER_START@
5 *
6 * Copyright (c) 1999-2003 Apple Computer, Inc. All Rights Reserved.
7 *
8 * This file contains Original Code and/or Modifications of Original Code
9 * as defined in and that are subject to the Apple Public Source License
10 * Version 2.0 (the 'License'). You may not use this file except in
11 * compliance with the License. Please obtain a copy of the License at
12 * http://www.opensource.apple.com/apsl/ and read it before using this
13 * file.
14 *
15 * The Original Code and all software distributed under the License are
16 * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
17 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
18 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
19 * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
20 * Please see the License for the specific language governing rights and
21 * limitations under the License.
22 *
23 * @APPLE_LICENSE_HEADER_END@
24 */
25
26 #include <sys/param.h>
27 #include <sys/systm.h>
28 #include <sys/kernel.h>
29 #include <sys/lock.h>
30 #include <sys/malloc.h>
31 #include <sys/queue.h>
32 #include <sys/utfconv.h>
33
34 #include "hfs.h"
35
36
37 /* hfs encoding converter list */
38 SLIST_HEAD(encodinglst, hfs_encoding) hfs_encoding_list = {0};
39 decl_simple_lock_data(,hfs_encoding_list_slock);
40
41
42 /* hfs encoding converter entry */
43 struct hfs_encoding {
44 SLIST_ENTRY(hfs_encoding) link;
45 int refcount;
46 int kmod_id;
47 UInt32 encoding;
48 hfs_to_unicode_func_t get_unicode_func;
49 unicode_to_hfs_func_t get_hfsname_func;
50 };
51
52 /* XXX We should use an "official" interface! */
53 extern kern_return_t kmod_destroy(host_priv_t host_priv, kmod_t id);
54 extern struct host realhost;
55
56 #define MAX_HFS_UNICODE_CHARS (15*5)
57
58 int mac_roman_to_unicode(Str31 hfs_str, UniChar *uni_str, UInt32 maxCharLen, UInt32 *usedCharLen);
59
60 static int unicode_to_mac_roman(UniChar *uni_str, UInt32 unicodeChars, Str31 hfs_str);
61
62
63 void
64 hfs_converterinit(void)
65 {
66 SLIST_INIT(&hfs_encoding_list);
67 simple_lock_init(&hfs_encoding_list_slock);
68
69 /*
70 * add resident MacRoman converter and take a reference
71 * since its always "loaded".
72 */
73 hfs_addconverter(0, kTextEncodingMacRoman, mac_roman_to_unicode, unicode_to_mac_roman);
74 SLIST_FIRST(&hfs_encoding_list)->refcount++;
75 }
76
77
78 /*
79 * hfs_addconverter - add an HFS encoding converter
80 *
81 * This is called exclusivly by kernel loadable modules
82 * (like HFS_Japanese.kmod) to register hfs encoding
83 * conversion routines.
84 *
85 */
86 int
87 hfs_addconverter(int id, UInt32 encoding, hfs_to_unicode_func_t get_unicode, unicode_to_hfs_func_t get_hfsname)
88 {
89 struct hfs_encoding *encp;
90
91 MALLOC(encp, struct hfs_encoding *, sizeof(struct hfs_encoding), M_TEMP, M_WAITOK);
92
93 simple_lock(&hfs_encoding_list_slock);
94
95 encp->link.sle_next = NULL;
96 encp->refcount = 0;
97 encp->encoding = encoding;
98 encp->get_unicode_func = get_unicode;
99 encp->get_hfsname_func = get_hfsname;
100 encp->kmod_id = id;
101 SLIST_INSERT_HEAD(&hfs_encoding_list, encp, link);
102
103 simple_unlock(&hfs_encoding_list_slock);
104 return (0);
105 }
106
107
108 /*
109 * hfs_remconverter - remove an HFS encoding converter
110 *
111 * Can be called by a kernel loadable module's finalize
112 * routine to remove an encoding converter so that the
113 * module (i.e. the code) can be unloaded.
114 *
115 * However, in the normal case, the removing and unloading
116 * of these converters is done in hfs_relconverter.
117 * The call is initiated from within the kernel during the unmounting of an hfs voulume.
118 */
119 int
120 hfs_remconverter(int id, UInt32 encoding)
121 {
122 struct hfs_encoding *encp;
123 int busy = 0;
124
125 simple_lock(&hfs_encoding_list_slock);
126 SLIST_FOREACH(encp, &hfs_encoding_list, link) {
127 if (encp->encoding == encoding && encp->kmod_id == id) {
128 encp->refcount--;
129
130 /* if converter is no longer in use, release it */
131 if (encp->refcount <= 0 && encp->kmod_id != 0) {
132 SLIST_REMOVE(&hfs_encoding_list, encp, hfs_encoding, link);
133 FREE(encp, M_TEMP);
134 } else {
135 busy = 1;
136 }
137 break;
138 }
139 }
140 simple_unlock(&hfs_encoding_list_slock);
141
142 return (busy);
143 }
144
145
146 /*
147 * hfs_getconverter - get HFS encoding converters
148 *
149 * Normally called during the mounting of an hfs voulume.
150 */
151 int
152 hfs_getconverter(UInt32 encoding, hfs_to_unicode_func_t *get_unicode, unicode_to_hfs_func_t *get_hfsname)
153 {
154 struct hfs_encoding *encp;
155 int found = 0;
156
157 simple_lock(&hfs_encoding_list_slock);
158 SLIST_FOREACH(encp, &hfs_encoding_list, link) {
159 if (encp->encoding == encoding) {
160 found = 1;
161 *get_unicode = encp->get_unicode_func;
162 *get_hfsname = encp->get_hfsname_func;
163 ++encp->refcount;
164 break;
165 }
166 }
167 simple_unlock(&hfs_encoding_list_slock);
168
169 if (!found) {
170 *get_unicode = NULL;
171 *get_hfsname = NULL;
172 return (EINVAL);
173 }
174
175 return (0);
176 }
177
178
179 /*
180 * hfs_relconverter - release interest in an HFS encoding converter
181 *
182 * Normally called during the unmounting of an hfs voulume.
183 */
184 int
185 hfs_relconverter(UInt32 encoding)
186 {
187 struct hfs_encoding *encp;
188 int found = 0;
189
190 simple_lock(&hfs_encoding_list_slock);
191 SLIST_FOREACH(encp, &hfs_encoding_list, link) {
192 if (encp->encoding == encoding) {
193 found = 1;
194 encp->refcount--;
195
196 /* if converter is no longer in use, release it */
197 if (encp->refcount <= 0 && encp->kmod_id != 0) {
198 int id = encp->kmod_id;
199
200 SLIST_REMOVE(&hfs_encoding_list, encp, hfs_encoding, link);
201 FREE(encp, M_TEMP);
202 encp = NULL;
203
204 simple_unlock(&hfs_encoding_list_slock);
205 kmod_destroy(host_priv_self(), id);
206 simple_lock(&hfs_encoding_list_slock);
207 }
208 break;
209 }
210 }
211 simple_unlock(&hfs_encoding_list_slock);
212
213 return (found ? 0 : EINVAL);
214 }
215
216
217 /*
218 * Convert HFS encoded string into UTF-8
219 *
220 * Unicode output is fully decomposed
221 * '/' chars are converted to ':'
222 */
223 int
224 hfs_to_utf8(ExtendedVCB *vcb, Str31 hfs_str, ByteCount maxDstLen, ByteCount *actualDstLen, unsigned char* dstStr)
225 {
226 int error;
227 UniChar uniStr[MAX_HFS_UNICODE_CHARS];
228 ItemCount uniCount;
229 size_t utf8len;
230 hfs_to_unicode_func_t hfs_get_unicode = VCBTOHFS(vcb)->hfs_get_unicode;
231
232 error = hfs_get_unicode(hfs_str, uniStr, MAX_HFS_UNICODE_CHARS, &uniCount);
233
234 if (uniCount == 0)
235 error = EINVAL;
236
237 if (error == 0) {
238 error = utf8_encodestr(uniStr, uniCount * sizeof(UniChar), dstStr, &utf8len, maxDstLen , ':', 0);
239 if (error == ENAMETOOLONG)
240 *actualDstLen = utf8_encodelen(uniStr, uniCount * sizeof(UniChar), ':', 0);
241 else
242 *actualDstLen = utf8len;
243 }
244
245 return error;
246 }
247
248
249 /*
250 * When an HFS name cannot be encoded with the current
251 * volume encoding then MacRoman is used as a fallback.
252 */
253 int
254 mac_roman_to_utf8(Str31 hfs_str, ByteCount maxDstLen, ByteCount *actualDstLen, unsigned char* dstStr)
255 {
256 int error;
257 UniChar uniStr[MAX_HFS_UNICODE_CHARS];
258 ItemCount uniCount;
259 size_t utf8len;
260
261 error = mac_roman_to_unicode(hfs_str, uniStr, MAX_HFS_UNICODE_CHARS, &uniCount);
262
263 if (uniCount == 0)
264 error = EINVAL;
265
266 if (error == 0) {
267 error = utf8_encodestr(uniStr, uniCount * sizeof(UniChar), dstStr, &utf8len, maxDstLen , ':', 0);
268 if (error == ENAMETOOLONG)
269 *actualDstLen = utf8_encodelen(uniStr, uniCount * sizeof(UniChar), ':', 0);
270 else
271 *actualDstLen = utf8len;
272 }
273
274 return error;
275 }
276
277
278 /*
279 * Convert Unicode string into HFS encoding
280 *
281 * ':' chars are converted to '/'
282 * Assumes input represents fully decomposed Unicode
283 */
284 int
285 unicode_to_hfs(ExtendedVCB *vcb, ByteCount srcLen, u_int16_t* srcStr, Str31 dstStr, int retry)
286 {
287 int error;
288 unicode_to_hfs_func_t hfs_get_hfsname = VCBTOHFS(vcb)->hfs_get_hfsname;
289
290 error = hfs_get_hfsname(srcStr, srcLen/sizeof(UniChar), dstStr);
291 if (error && retry) {
292 error = unicode_to_mac_roman(srcStr, srcLen/sizeof(UniChar), dstStr);
293 }
294 return error;
295 }
296
297 /*
298 * Convert UTF-8 string into HFS encoding
299 *
300 * ':' chars are converted to '/'
301 * Assumes input represents fully decomposed Unicode
302 */
303 int
304 utf8_to_hfs(ExtendedVCB *vcb, ByteCount srcLen, const unsigned char* srcStr, Str31 dstStr/*, int retry*/)
305 {
306 int error;
307 UniChar uniStr[MAX_HFS_UNICODE_CHARS];
308 size_t ucslen;
309
310 error = utf8_decodestr(srcStr, srcLen, uniStr, &ucslen, sizeof(uniStr), ':', 0);
311 if (error == 0)
312 error = unicode_to_hfs(vcb, ucslen, uniStr, dstStr, 1);
313
314 return error;
315 }
316
317 int
318 utf8_to_mac_roman(ByteCount srcLen, const unsigned char* srcStr, Str31 dstStr)
319 {
320 int error;
321 UniChar uniStr[MAX_HFS_UNICODE_CHARS];
322 size_t ucslen;
323
324 error = utf8_decodestr(srcStr, srcLen, uniStr, &ucslen, sizeof(uniStr), ':', 0);
325 if (error == 0)
326 error = unicode_to_mac_roman(uniStr, ucslen/sizeof(UniChar), dstStr);
327
328 return error;
329 }
330
331 /*
332 * HFS MacRoman to/from Unicode conversions are built into the kernel
333 * All others hfs encodings are loadable.
334 */
335
336 /* 0x00A0 - 0x00FF = Latin 1 Supplement (30 total) */
337 static UInt8 gLatin1Table[] = {
338 /* 0 1 2 3 4 5 6 7 8 9 A B C D E F */
339 /* 0x00A0 */ 0xCA, 0xC1, 0xA2, 0xA3, 0xDB, 0xB4, '?', 0xA4, 0xAC, 0xA9, 0xBB, 0xC7, 0xC2, '?', 0xA8, 0xF8,
340 /* 0x00B0 */ 0xA1, 0XB1, '?', '?', 0xAB, 0xB5, 0xA6, 0xe1, 0xFC, '?', 0xBC, 0xC8, '?', '?', '?', 0xC0,
341 /* 0x00C0 */ '?', '?', '?', '?', '?', '?', 0xAE, '?', '?', '?', '?', '?', '?', '?', '?', '?',
342 /* 0x00D0 */ '?', '?', '?', '?', '?', '?', '?', '?', 0xAF, '?', '?', '?', '?', '?', '?', 0xA7,
343 /* 0x00E0 */ '?', '?', '?', '?', '?', '?', 0xBE, '?', '?', '?', '?', '?', '?', '?', '?', '?',
344 /* 0x00F0 */ '?', '?', '?', '?', '?', '?', '?', 0xD6, 0xBF, '?', '?', '?', '?', '?', '?', '?'
345 };
346
347 /* 0x02C0 - 0x02DF = Spacing Modifiers (8 total) */
348 static UInt8 gSpaceModsTable[] = {
349 /* 0 1 2 3 4 5 6 7 8 9 A B C D E F */
350 /* 0x02C0 */ '?', '?', '?', '?', '?', '?', 0xF6, 0xFF, '?', '?', '?', '?', '?', '?', '?', '?',
351 /* 0x02D0 */ '?', '?', '?', '?', '?', '?', '?', '?', 0xF9, 0xFA, 0xFB, 0xFE, 0xF7, 0xFD, '?', '?'
352 };
353
354 /* 0x2010 - 0x20AF = General Punctuation (17 total) */
355 static UInt8 gPunctTable[] = {
356 /* 0 1 2 3 4 5 6 7 8 9 A B C D E F */
357 /* 0x2010 */ '?', '?', '?', 0xd0, 0xd1, '?', '?', '?', 0xd4, 0xd5, 0xe2, '?', 0xd2, 0xd3, 0xe3, '?',
358 /* 0x2020 */ 0xa0, 0xe0, 0xa5, '?', '?', '?', 0xc9, '?', '?', '?', '?', '?', '?', '?', '?', '?',
359 /* 0x2030 */ 0xe4, '?', '?', '?', '?', '?', '?', '?', '?', 0xdc, 0xdd, '?', '?', '?', '?', '?',
360 /* 0x2040 */ '?', '?', '?', '?', 0xda, '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?',
361 /* 0x2050 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?',
362 /* 0x2060 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?',
363 /* 0x2070 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?',
364 /* 0x2080 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?',
365 /* 0x2090 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?',
366 /* 0x20A0 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', 0xdb, '?', '?', '?'
367 };
368
369 /* 0x22xx = Mathematical Operators (11 total) */
370 static UInt8 gMathTable[] = {
371 /* 0 1 2 3 4 5 6 7 8 9 A B C D E F */
372 /* 0x2200 */ '?', '?', 0xb6, '?', '?', '?', 0xc6, '?', '?', '?', '?', '?', '?', '?', '?', 0xb8,
373 /* 0x2210 */ '?', 0xb7, '?', '?', '?', '?', '?', '?', '?', '?', 0xc3, '?', '?', '?', 0xb0, '?',
374 /* 0x2220 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', 0xba, '?', '?', '?', '?',
375 /* 0x2230 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?',
376 /* 0x2240 */ '?', '?', '?', '?', '?', '?', '?', '?', 0xc5, '?', '?', '?', '?', '?', '?', '?',
377 /* 0x2250 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?',
378 /* 0x2260 */ 0xad, '?', '?', '?', 0xb2, 0xb3, '?', '?'
379 };
380
381 /* */
382 static UInt8 gReverseCombTable[] = {
383 /* 0 1 2 3 4 5 6 7 8 9 A B C D E F */
384 /* 0x40 */ 0xDA, 0x40, 0xDA, 0xDA, 0xDA, 0x56, 0xDA, 0xDA, 0xDA, 0x6C, 0xDA, 0xDA, 0xDA, 0xDA, 0x82, 0x98,
385 /* 0x50 */ 0xDA, 0xDA, 0xDA, 0xDA, 0xDA, 0xAE, 0xDA, 0xDA, 0xDA, 0xC4, 0xDA, 0xDA, 0xDA, 0xDA, 0xDA, 0xDA,
386 /* 0x60 */ 0xDA, 0x4B, 0xDA, 0xDA, 0xDA, 0x61, 0xDA, 0xDA, 0xDA, 0x77, 0xDA, 0xDA, 0xDA, 0xDA, 0x8D, 0xA3,
387 /* 0x70 */ 0xDA, 0xDA, 0xDA, 0xDA, 0xDA, 0xB9, 0xDA, 0xDA, 0xDA, 0xCF, 0xDA, 0xDA, 0xDA, 0xDA, 0xDA, 0xDA,
388
389 /* Combining Diacritical Marks (0x0300 - 0x030A) */
390 /* 0 1 2 3 4 5 6 7 8 9 A */
391 /* 'A' */
392 /* 0x0300 */ 0xCB, 0xE7, 0xE5, 0xCC, '?', '?', '?', '?', 0x80, '?', 0x81,
393
394 /* 'a' */
395 /* 0x0300 */ 0x88, 0x87, 0x89, 0x8B, '?', '?', '?', '?', 0x8A, '?', 0x8C,
396
397 /* 'E' */
398 /* 0x0300 */ 0xE9, 0x83, 0xE6, '?', '?', '?', '?', '?', 0xE8, '?', '?',
399
400 /* 'e' */
401 /* 0x0300 */ 0x8F, 0x8E, 0x90, '?', '?', '?', '?', '?', 0x91, '?', '?',
402
403 /* 'I' */
404 /* 0x0300 */ 0xED, 0xEA, 0xEB, '?', '?', '?', '?', '?', 0xEC, '?', '?',
405
406 /* 'i' */
407 /* 0x0300 */ 0x93, 0x92, 0x94, '?', '?', '?', '?', '?', 0x95, '?', '?',
408
409 /* 'N' */
410 /* 0x0300 */ '?', '?', '?', 0x84, '?', '?', '?', '?', '?', '?', '?',
411
412 /* 'n' */
413 /* 0x0300 */ '?', '?', '?', 0x96, '?', '?', '?', '?', '?', '?', '?',
414
415 /* 'O' */
416 /* 0x0300 */ 0xF1, 0xEE, 0xEF, 0xCD, '?', '?', '?', '?', 0x85, '?', '?',
417
418 /* 'o' */
419 /* 0x0300 */ 0x98, 0x97, 0x99, 0x9B, '?', '?', '?', '?', 0x9A, '?', '?',
420
421 /* 'U' */
422 /* 0x0300 */ 0xF4, 0xF2, 0xF3, '?', '?', '?', '?', '?', 0x86, '?', '?',
423
424 /* 'u' */
425 /* 0x0300 */ 0x9D, 0x9C, 0x9E, '?', '?', '?', '?', '?', 0x9F, '?', '?',
426
427 /* 'Y' */
428 /* 0x0300 */ '?', '?', '?', '?', '?', '?', '?', '?', 0xD9, '?', '?',
429
430 /* 'y' */
431 /* 0x0300 */ '?', '?', '?', '?', '?', '?', '?', '?', 0xD8, '?', '?',
432
433 /* else */
434 /* 0x0300 */ '?', '?', '?', '?', '?', '?', '?', '?', '?', '?', '?'
435 };
436
437
438 /*
439 * Convert Unicode string into HFS MacRoman encoding
440 *
441 * Assumes Unicode input is fully decomposed
442 */
443 static int unicode_to_mac_roman(UniChar *uni_str, UInt32 unicodeChars, Str31 hfs_str)
444 {
445 UInt8 *p;
446 const UniChar *u;
447 UniChar c;
448 UniChar mask;
449 UInt16 inputChars;
450 UInt16 pascalChars;
451 OSErr result = noErr;
452 UInt8 lsb;
453 UInt8 prevChar;
454 UInt8 mc;
455
456 mask = (UniChar) 0xFF80;
457 p = &hfs_str[1];
458 u = uni_str;
459 inputChars = unicodeChars;
460 pascalChars = prevChar = 0;
461
462 while (inputChars) {
463 c = *(u++);
464 lsb = (UInt8) c;
465
466 /*
467 * If its not 7-bit ascii, then we need to map it
468 */
469 if ( c & mask ) {
470 mc = '?';
471 switch (c & 0xFF00) {
472 case 0x0000:
473 if (lsb >= 0xA0)
474 mc = gLatin1Table[lsb - 0xA0];
475 break;
476
477 case 0x0200:
478 if (lsb >= 0xC0 && lsb <= 0xDF)
479 mc = gSpaceModsTable[lsb - 0xC0];
480 break;
481
482 case 0x2000:
483 if (lsb >= 0x10 && lsb <= 0xAF)
484 mc = gPunctTable[lsb- 0x10];
485 break;
486
487 case 0x2200:
488 if (lsb <= 0x68)
489 mc = gMathTable[lsb];
490 break;
491
492 case 0x0300:
493 if (c <= 0x030A) {
494 if (prevChar >= 'A' && prevChar < 'z') {
495 mc = gReverseCombTable[gReverseCombTable[prevChar - 0x40] + lsb];
496 --p; /* backup over base char */
497 --pascalChars;
498 }
499 } else {
500 switch (c) {
501 case 0x0327: /* combining cedilla */
502 if (prevChar == 'C')
503 mc = 0x82;
504 else if (prevChar == 'c')
505 mc = 0x8D;
506 else
507 break;
508 --p; /* backup over base char */
509 --pascalChars;
510 break;
511
512 case 0x03A9: mc = 0xBD; break; /* omega */
513
514 case 0x03C0: mc = 0xB9; break; /* pi */
515 }
516 }
517 break;
518
519 default:
520 switch (c) {
521 case 0x0131: mc = 0xf5; break; /* dotless i */
522
523 case 0x0152: mc = 0xce; break; /* OE */
524
525 case 0x0153: mc = 0xcf; break; /* oe */
526
527 case 0x0192: mc = 0xc4; break; /* Ä */
528
529 case 0x2122: mc = 0xaa; break; /* TM */
530
531 case 0x25ca: mc = 0xd7; break; /* diamond */
532
533 case 0xf8ff: mc = 0xf0; break; /* apple logo */
534
535 case 0xfb01: mc = 0xde; break; /* fi */
536
537 case 0xfb02: mc = 0xdf; break; /* fl */
538 }
539 } /* end switch (c & 0xFF00) */
540
541 /*
542 * If we have an unmapped character then we need to mangle the name...
543 */
544 if (mc == '?')
545 result = kTECUsedFallbacksStatus;
546
547 prevChar = 0;
548 lsb = mc;
549
550 } else {
551 prevChar = lsb;
552 }
553
554 if (pascalChars >= 31)
555 break;
556
557 *(p++) = lsb;
558 ++pascalChars;
559 --inputChars;
560
561 } /* end while */
562
563 hfs_str[0] = pascalChars;
564
565 if (inputChars > 0)
566 result = ENAMETOOLONG; /* ran out of room! */
567
568 return result;
569 }
570
571
572 static UniChar gHiBitBaseUnicode[128] = {
573 /* 0x80 */ 0x0041, 0x0041, 0x0043, 0x0045, 0x004e, 0x004f, 0x0055, 0x0061,
574 /* 0x88 */ 0x0061, 0x0061, 0x0061, 0x0061, 0x0061, 0x0063, 0x0065, 0x0065,
575 /* 0x90 */ 0x0065, 0x0065, 0x0069, 0x0069, 0x0069, 0x0069, 0x006e, 0x006f,
576 /* 0x98 */ 0x006f, 0x006f, 0x006f, 0x006f, 0x0075, 0x0075, 0x0075, 0x0075,
577 /* 0xa0 */ 0x2020, 0x00b0, 0x00a2, 0x00a3, 0x00a7, 0x2022, 0x00b6, 0x00df,
578 /* 0xa8 */ 0x00ae, 0x00a9, 0x2122, 0x00b4, 0x00a8, 0x2260, 0x00c6, 0x00d8,
579 /* 0xb0 */ 0x221e, 0x00b1, 0x2264, 0x2265, 0x00a5, 0x00b5, 0x2202, 0x2211,
580 /* 0xb8 */ 0x220f, 0x03c0, 0x222b, 0x00aa, 0x00ba, 0x03a9, 0x00e6, 0x00f8,
581 /* 0xc0 */ 0x00bf, 0x00a1, 0x00ac, 0x221a, 0x0192, 0x2248, 0x2206, 0x00ab,
582 /* 0xc8 */ 0x00bb, 0x2026, 0x00a0, 0x0041, 0x0041, 0x004f, 0x0152, 0x0153,
583 /* 0xd0 */ 0x2013, 0x2014, 0x201c, 0x201d, 0x2018, 0x2019, 0x00f7, 0x25ca,
584 /* 0xd8 */ 0x0079, 0x0059, 0x2044, 0x20ac, 0x2039, 0x203a, 0xfb01, 0xfb02,
585 /* 0xe0 */ 0x2021, 0x00b7, 0x201a, 0x201e, 0x2030, 0x0041, 0x0045, 0x0041,
586 /* 0xe8 */ 0x0045, 0x0045, 0x0049, 0x0049, 0x0049, 0x0049, 0x004f, 0x004f,
587 /* 0xf0 */ 0xf8ff, 0x004f, 0x0055, 0x0055, 0x0055, 0x0131, 0x02c6, 0x02dc,
588 /* 0xf8 */ 0x00af, 0x02d8, 0x02d9, 0x02da, 0x00b8, 0x02dd, 0x02db, 0x02c7
589 };
590
591 static UniChar gHiBitCombUnicode[128] = {
592 /* 0x80 */ 0x0308, 0x030a, 0x0327, 0x0301, 0x0303, 0x0308, 0x0308, 0x0301,
593 /* 0x88 */ 0x0300, 0x0302, 0x0308, 0x0303, 0x030a, 0x0327, 0x0301, 0x0300,
594 /* 0x90 */ 0x0302, 0x0308, 0x0301, 0x0300, 0x0302, 0x0308, 0x0303, 0x0301,
595 /* 0x98 */ 0x0300, 0x0302, 0x0308, 0x0303, 0x0301, 0x0300, 0x0302, 0x0308,
596 /* 0xa0 */ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
597 /* 0xa8 */ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
598 /* 0xb0 */ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
599 /* 0xb8 */ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
600 /* 0xc0 */ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
601 /* 0xc8 */ 0x0000, 0x0000, 0x0000, 0x0300, 0x0303, 0x0303, 0x0000, 0x0000,
602 /* 0xd0 */ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
603 /* 0xd8 */ 0x0308, 0x0308, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
604 /* 0xe0 */ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0302, 0x0302, 0x0301,
605 /* 0xe8 */ 0x0308, 0x0300, 0x0301, 0x0302, 0x0308, 0x0300, 0x0301, 0x0302,
606 /* 0xf0 */ 0x0000, 0x0300, 0x0301, 0x0302, 0x0300, 0x0000, 0x0000, 0x0000,
607 /* 0xf8 */ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
608 };
609
610
611 /*
612 * Convert HFS MacRoman encoded string into Unicode
613 *
614 * Unicode output is fully decomposed
615 */
616 int
617 mac_roman_to_unicode(Str31 hfs_str, UniChar *uni_str,
618 UInt32 maxCharLen, UInt32 *unicodeChars)
619 {
620 const UInt8 *p;
621 UniChar *u;
622 UInt16 pascalChars;
623 UInt8 c;
624
625 p = hfs_str;
626 u = uni_str;
627
628 *unicodeChars = pascalChars = *(p++); /* pick up length byte */
629
630 while (pascalChars--) {
631 c = *(p++);
632
633 if ( (SInt8) c >= 0 ) { /* check if seven bit ascii */
634 *(u++) = (UniChar) c; /* just pad high byte with zero */
635 } else { /* its a hi bit character */
636 UniChar uc;
637
638 c &= 0x7F;
639 *(u++) = uc = gHiBitBaseUnicode[c];
640
641 /*
642 * if the unicode character we get back is an alpha char
643 * then we must have an additional combining character
644 */
645 if ((uc <= (UniChar) 'z') && (uc >= (UniChar) 'A')) {
646 *(u++) = gHiBitCombUnicode[c];
647 ++(*unicodeChars);
648 }
649 }
650 }
651
652 return noErr;
653 }
654