xref: /netbsd-src/sys/fs/udf/udf_osta.c (revision 466a16a118933bd295a8a104f095714fadf9cf68)
1 /* $NetBSD: udf_osta.c,v 1.7 2008/05/19 20:12:36 reinoud Exp $ */
2 
3 #include <sys/cdefs.h>
4 #ifndef lint
5 __KERNEL_RCSID(0, "$NetBSD: udf_osta.c,v 1.7 2008/05/19 20:12:36 reinoud Exp $");
6 #endif /* not lint */
7 
8 /*
9  * Various routines from the OSTA 2.01 specs.  Copyrights are included with
10  * each code segment.  Slight whitespace modifications have been made for
11  * formatting purposes.  Typos/bugs have been fixed.
12  *
13  */
14 
15 #include "udf_osta.h"
16 
17 #ifndef _KERNEL
18 #include <ctype.h>
19 #endif
20 
21 /*****************************************************************************/
22 /***********************************************************************
23  * OSTA compliant Unicode compression, uncompression routines.
24  * Copyright 1995 Micro Design International, Inc.
25  * Written by Jason M. Rinn.
26  * Micro Design International gives permission for the free use of the
27  * following source code.
28  */
29 
30 /***********************************************************************
31  * Takes an OSTA CS0 compressed unicode name, and converts
32  * it to Unicode.
33  * The Unicode output will be in the byte order
34  * that the local compiler uses for 16-bit values.
35  * NOTE: This routine only performs error checking on the compID.
36  * It is up to the user to ensure that the unicode buffer is large
37  * enough, and that the compressed unicode name is correct.
38  *
39  * RETURN VALUE
40  *
41  * The number of unicode characters which were uncompressed.
42  * A -1 is returned if the compression ID is invalid.
43  */
44 int
45 udf_UncompressUnicode(
46 	int numberOfBytes,	/* (Input) number of bytes read from media. */
47 	byte *UDFCompressed,	/* (Input) bytes read from media. */
48 	unicode_t *unicode)	/* (Output) uncompressed unicode characters. */
49 {
50 	unsigned int compID;
51 	int returnValue, unicodeIndex, byteIndex;
52 
53 	/* Use UDFCompressed to store current byte being read. */
54 	compID = UDFCompressed[0];
55 
56 	/* First check for valid compID. */
57 	if (compID != 8 && compID != 16) {
58 		returnValue = -1;
59 	} else {
60 		unicodeIndex = 0;
61 		byteIndex = 1;
62 
63 		/* Loop through all the bytes. */
64 		while (byteIndex < numberOfBytes) {
65 			if (compID == 16) {
66 				/* Move the first byte to the high bits of the
67 				 * unicode char.
68 				 */
69 				unicode[unicodeIndex] =
70 				    UDFCompressed[byteIndex++] << 8;
71 			} else {
72 				unicode[unicodeIndex] = 0;
73 			}
74 			if (byteIndex < numberOfBytes) {
75 				/*Then the next byte to the low bits. */
76 				unicode[unicodeIndex] |=
77 				    UDFCompressed[byteIndex++];
78 			}
79 			unicodeIndex++;
80 		}
81 		returnValue = unicodeIndex;
82 	}
83 	return(returnValue);
84 }
85 
86 /***********************************************************************
87  * DESCRIPTION:
88  * Takes a string of unicode wide characters and returns an OSTA CS0
89  * compressed unicode string. The unicode MUST be in the byte order of
90  * the compiler in order to obtain correct results. Returns an error
91  * if the compression ID is invalid.
92  *
93  * NOTE: This routine assumes the implementation already knows, by
94  * the local environment, how many bits are appropriate and
95  * therefore does no checking to test if the input characters fit
96  * into that number of bits or not.
97  *
98  * RETURN VALUE
99  *
100  * The total number of bytes in the compressed OSTA CS0 string,
101  * including the compression ID.
102  * A -1 is returned if the compression ID is invalid.
103  */
104 int
105 udf_CompressUnicode(
106 	int numberOfChars,	/* (Input) number of unicode characters. */
107 	int compID,		/* (Input) compression ID to be used. */
108 	unicode_t *unicode,	/* (Input) unicode characters to compress. */
109 	byte *UDFCompressed)	/* (Output) compressed string, as bytes. */
110 {
111 	int byteIndex, unicodeIndex;
112 
113 	if (compID != 8 && compID != 16) {
114 		byteIndex = -1; /* Unsupported compression ID ! */
115 	} else {
116 		/* Place compression code in first byte. */
117 		UDFCompressed[0] = compID;
118 
119 		byteIndex = 1;
120 		unicodeIndex = 0;
121 		while (unicodeIndex < numberOfChars) {
122 			if (compID == 16) {
123 				/* First, place the high bits of the char
124 				 * into the byte stream.
125 				 */
126 				UDFCompressed[byteIndex++] =
127 				    (unicode[unicodeIndex] & 0xFF00) >> 8;
128 			}
129 			/*Then place the low bits into the stream. */
130 			UDFCompressed[byteIndex++] =
131 			    unicode[unicodeIndex] & 0x00FF;
132 			unicodeIndex++;
133 		}
134 	}
135 	return(byteIndex);
136 }
137 
138 /*****************************************************************************/
139 /*
140  * CRC 010041
141  */
142 static unsigned short crc_table[256] = {
143 	0x0000, 0x1021, 0x2042, 0x3063, 0x4084, 0x50A5, 0x60C6, 0x70E7,
144 	0x8108, 0x9129, 0xA14A, 0xB16B, 0xC18C, 0xD1AD, 0xE1CE, 0xF1EF,
145 	0x1231, 0x0210, 0x3273, 0x2252, 0x52B5, 0x4294, 0x72F7, 0x62D6,
146 	0x9339, 0x8318, 0xB37B, 0xA35A, 0xD3BD, 0xC39C, 0xF3FF, 0xE3DE,
147 	0x2462, 0x3443, 0x0420, 0x1401, 0x64E6, 0x74C7, 0x44A4, 0x5485,
148 	0xA56A, 0xB54B, 0x8528, 0x9509, 0xE5EE, 0xF5CF, 0xC5AC, 0xD58D,
149 	0x3653, 0x2672, 0x1611, 0x0630, 0x76D7, 0x66F6, 0x5695, 0x46B4,
150 	0xB75B, 0xA77A, 0x9719, 0x8738, 0xF7DF, 0xE7FE, 0xD79D, 0xC7BC,
151 	0x48C4, 0x58E5, 0x6886, 0x78A7, 0x0840, 0x1861, 0x2802, 0x3823,
152 	0xC9CC, 0xD9ED, 0xE98E, 0xF9AF, 0x8948, 0x9969, 0xA90A, 0xB92B,
153 	0x5AF5, 0x4AD4, 0x7AB7, 0x6A96, 0x1A71, 0x0A50, 0x3A33, 0x2A12,
154 	0xDBFD, 0xCBDC, 0xFBBF, 0xEB9E, 0x9B79, 0x8B58, 0xBB3B, 0xAB1A,
155 	0x6CA6, 0x7C87, 0x4CE4, 0x5CC5, 0x2C22, 0x3C03, 0x0C60, 0x1C41,
156 	0xEDAE, 0xFD8F, 0xCDEC, 0xDDCD, 0xAD2A, 0xBD0B, 0x8D68, 0x9D49,
157 	0x7E97, 0x6EB6, 0x5ED5, 0x4EF4, 0x3E13, 0x2E32, 0x1E51, 0x0E70,
158 	0xFF9F, 0xEFBE, 0xDFDD, 0xCFFC, 0xBF1B, 0xAF3A, 0x9F59, 0x8F78,
159 	0x9188, 0x81A9, 0xB1CA, 0xA1EB, 0xD10C, 0xC12D, 0xF14E, 0xE16F,
160 	0x1080, 0x00A1, 0x30C2, 0x20E3, 0x5004, 0x4025, 0x7046, 0x6067,
161 	0x83B9, 0x9398, 0xA3FB, 0xB3DA, 0xC33D, 0xD31C, 0xE37F, 0xF35E,
162 	0x02B1, 0x1290, 0x22F3, 0x32D2, 0x4235, 0x5214, 0x6277, 0x7256,
163 	0xB5EA, 0xA5CB, 0x95A8, 0x8589, 0xF56E, 0xE54F, 0xD52C, 0xC50D,
164 	0x34E2, 0x24C3, 0x14A0, 0x0481, 0x7466, 0x6447, 0x5424, 0x4405,
165 	0xA7DB, 0xB7FA, 0x8799, 0x97B8, 0xE75F, 0xF77E, 0xC71D, 0xD73C,
166 	0x26D3, 0x36F2, 0x0691, 0x16B0, 0x6657, 0x7676, 0x4615, 0x5634,
167 	0xD94C, 0xC96D, 0xF90E, 0xE92F, 0x99C8, 0x89E9, 0xB98A, 0xA9AB,
168 	0x5844, 0x4865, 0x7806, 0x6827, 0x18C0, 0x08E1, 0x3882, 0x28A3,
169 	0xCB7D, 0xDB5C, 0xEB3F, 0xFB1E, 0x8BF9, 0x9BD8, 0xABBB, 0xBB9A,
170 	0x4A75, 0x5A54, 0x6A37, 0x7A16, 0x0AF1, 0x1AD0, 0x2AB3, 0x3A92,
171 	0xFD2E, 0xED0F, 0xDD6C, 0xCD4D, 0xBDAA, 0xAD8B, 0x9DE8, 0x8DC9,
172 	0x7C26, 0x6C07, 0x5C64, 0x4C45, 0x3CA2, 0x2C83, 0x1CE0, 0x0CC1,
173 	0xEF1F, 0xFF3E, 0xCF5D, 0xDF7C, 0xAF9B, 0xBFBA, 0x8FD9, 0x9FF8,
174 	0x6E17, 0x7E36, 0x4E55, 0x5E74, 0x2E93, 0x3EB2, 0x0ED1, 0x1EF0
175 };
176 
177 unsigned short
178 udf_cksum(s, n)
179 	unsigned char *s;
180 	int n;
181 {
182 	unsigned short crc=0;
183 
184 	while (n-- > 0)
185 		crc = crc_table[(crc>>8 ^ *s++) & 0xff] ^ (crc<<8);
186 	return crc;
187 }
188 
189 /* UNICODE Checksum */
190 unsigned short
191 udf_unicode_cksum(s, n)
192 	unsigned short *s;
193 	int n;
194 {
195 	unsigned short crc=0;
196 
197 	while (n-- > 0) {
198 		/* Take high order byte first--corresponds to a big endian
199 		 * byte stream.
200 		 */
201 		crc = crc_table[(crc>>8 ^ (*s>>8)) & 0xff] ^ (crc<<8);
202 		crc = crc_table[(crc>>8 ^ (*s++ & 0xff)) & 0xff] ^ (crc<<8);
203 	}
204 	return crc;
205 }
206 
207 
208 /*
209   * Calculates a 16-bit checksum of the Implementation Use
210   * Extended Attribute header or Application Use Extended Attribute
211   * header. The fields AttributeType through ImplementationIdentifier
212   * (or ApplicationIdentifier) inclusively represent the
213   * data covered by the checksum (48 bytes).
214   *
215   */
216 uint16_t udf_ea_cksum(uint8_t *data) {
217         uint16_t checksum = 0;
218         int      count;
219 
220         for (count = 0; count < 48; count++) {
221                checksum += *data++;
222         }
223 
224         return checksum;
225 }
226 
227 
228 #ifdef MAIN
229 unsigned char bytes[] = { 0x70, 0x6A, 0x77 };
230 
231 main()
232 {
233 	unsigned short x;
234 	x = cksum(bytes, sizeof bytes);
235 	printf("checksum: calculated=%4.4x, correct=%4.4x\en", x, 0x3299);
236 	exit(0);
237 }
238 #endif
239 
240 /*****************************************************************************/
241 /* #ifdef NEEDS_ISPRINT */
242 /***********************************************************************
243  * OSTA UDF compliant file name translation routine for OS/2,
244  * Windows 95, Windows NT, Macintosh and UNIX.
245  * Copyright 1995 Micro Design International, Inc.
246  * Written by Jason M. Rinn.
247  * Micro Design International gives permission for the free use of the
248  * following source code.
249  */
250 
251 /***********************************************************************
252  * To use these routines with different operating systems.
253  *
254  * OS/2
255  * Define OS2
256  * Define MAXLEN = 254
257  *
258  * Windows 95
259  * Define WIN_95
260  * Define MAXLEN = 255
261  *
262  * Windows NT
263  * Define WIN_NT
264  * Define MAXLEN = 255
265  *
266  * Macintosh:
267  * Define MAC.
268  * Define MAXLEN = 31.
269  *
270  * UNIX
271  * Define UNIX.
272  * Define MAXLEN as specified by unix version.
273  */
274 
275 #define	ILLEGAL_CHAR_MARK	0x005F
276 #define	CRC_MARK	0x0023
277 #define	EXT_SIZE	5
278 #define	PERIOD	0x002E
279 #define	SPACE	0x0020
280 
281 /*** PROTOTYPES ***/
282 int IsIllegal(unicode_t ch);
283 
284 /* Define a function or macro which determines if a Unicode character is
285  * printable under your implementation.
286  */
287 
288 
289 /* #include <stdio.h> */
290 static int UnicodeIsPrint(unicode_t ch) {
291 	return (ch >=' ') && (ch != 127);
292 }
293 
294 
295 int UnicodeLength(unicode_t *string) {
296 	int length;
297 	length = 0;
298 	while (*string++) length++;
299 
300 	return length;
301 }
302 
303 
304 #ifdef _KERNEL
305 static int isprint(int c) {
306 	return (c >= ' ') && (c != 127);
307 }
308 #endif
309 
310 
311 /***********************************************************************
312  * Translates a long file name to one using a MAXLEN and an illegal
313  * char set in accord with the OSTA requirements. Assumes the name has
314  * already been translated to Unicode.
315  *
316  * RETURN VALUE
317  *
318  * Number of unicode characters in translated name.
319  */
320 int UDFTransName(
321 	unicode_t *newName,	/* (Output)Translated name. Must be of length
322 				 * MAXLEN */
323 	unicode_t *udfName,	/* (Input) Name from UDF volume.*/
324 	int udfLen)		/* (Input) Length of UDF Name. */
325 {
326 	int Index, newIndex = 0, needsCRC = false;	/* index is shadowed */
327 	int extIndex = 0, newExtIndex = 0, hasExt = false;
328 #if defined OS2 || defined WIN_95 || defined WIN_NT
329 	int trailIndex = 0;
330 #endif
331 	unsigned short valueCRC;
332 	unicode_t current;
333 	const char hexChar[] = "0123456789ABCDEF";
334 
335 	for (Index = 0; Index < udfLen; Index++) {
336 		current = udfName[Index];
337 
338 		if (IsIllegal(current) || !UnicodeIsPrint(current)) {
339 			needsCRC = true;
340 			/* Replace Illegal and non-displayable chars with
341 			 * underscore.
342 			 */
343 			current = ILLEGAL_CHAR_MARK;
344 			/* Skip any other illegal or non-displayable
345 			 * characters.
346 			 */
347 			while(Index+1 < udfLen && (IsIllegal(udfName[Index+1])
348 			    || !UnicodeIsPrint(udfName[Index+1]))) {
349 				Index++;
350 			}
351 		}
352 
353 		/* Record position of extension, if one is found. */
354 		if (current == PERIOD && (udfLen - Index -1) <= EXT_SIZE) {
355 			if (udfLen == Index + 1) {
356 				/* A trailing period is NOT an extension. */
357 				hasExt = false;
358 			} else {
359 				hasExt = true;
360 				extIndex = Index;
361 				newExtIndex = newIndex;
362 			}
363 		}
364 
365 #if defined OS2 || defined WIN_95 || defined WIN_NT
366 		/* Record position of last char which is NOT period or space. */
367 		else if (current != PERIOD && current != SPACE) {
368 			trailIndex = newIndex;
369 		}
370 #endif
371 
372 		if (newIndex < MAXLEN) {
373 			newName[newIndex++] = current;
374 		} else {
375 			needsCRC = true;
376 		}
377 	}
378 
379 #if defined OS2 || defined WIN_95 || defined WIN_NT
380 	/* For OS2, 95 & NT, truncate any trailing periods and\or spaces. */
381 	if (trailIndex != newIndex - 1) {
382 		newIndex = trailIndex + 1;
383 		needsCRC = true;
384 		hasExt = false; /* Trailing period does not make an
385 				 * extension. */
386 	}
387 #endif
388 
389 	if (needsCRC) {
390 		unicode_t ext[EXT_SIZE];
391 		int localExtIndex = 0;
392 		if (hasExt) {
393 			int maxFilenameLen;
394 			/* Translate extension, and store it in ext. */
395 			for(Index = 0; Index<EXT_SIZE &&
396 			    extIndex + Index +1 < udfLen; Index++ ) {
397 				current = udfName[extIndex + Index + 1];
398 				if (IsIllegal(current) ||
399 				    !UnicodeIsPrint(current)) {
400 					needsCRC = 1;
401 					/* Replace Illegal and non-displayable
402 					 * chars with underscore.
403 					 */
404 					current = ILLEGAL_CHAR_MARK;
405 					/* Skip any other illegal or
406 					 * non-displayable characters.
407 					 */
408 					while(Index + 1 < EXT_SIZE
409 					    && (IsIllegal(udfName[extIndex +
410 					    Index + 2]) ||
411 					    !isprint(udfName[extIndex +
412 					    Index + 2]))) {
413 						Index++;
414 					}
415 				}
416 				ext[localExtIndex++] = current;
417 			}
418 
419 			/* Truncate filename to leave room for extension and
420 			 * CRC.
421 			 */
422 			maxFilenameLen = ((MAXLEN - 5) - localExtIndex - 1);
423 			if (newIndex > maxFilenameLen) {
424 				newIndex = maxFilenameLen;
425 			} else {
426 				newIndex = newExtIndex;
427 			}
428 		} else if (newIndex > MAXLEN - 5) {
429 			/*If no extension, make sure to leave room for CRC. */
430 			newIndex = MAXLEN - 5;
431 		}
432 		newName[newIndex++] = CRC_MARK; /* Add mark for CRC. */
433 
434 		/*Calculate CRC from original filename from FileIdentifier. */
435 		valueCRC = udf_unicode_cksum(udfName, udfLen);
436 		/* Convert 16-bits of CRC to hex characters. */
437 		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
438 		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
439 		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
440 		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
441 
442 		/* Place a translated extension at end, if found. */
443 		if (hasExt) {
444 			newName[newIndex++] = PERIOD;
445 			for (Index = 0;Index < localExtIndex ;Index++ ) {
446 				newName[newIndex++] = ext[Index];
447 			}
448 		}
449 	}
450 	return(newIndex);
451 }
452 
453 #if defined OS2 || defined WIN_95 || defined WIN_NT
454 /***********************************************************************
455  * Decides if a Unicode character matches one of a list
456  * of ASCII characters.
457  * Used by OS2 version of IsIllegal for readability, since all of the
458  * illegal characters above 0x0020 are in the ASCII subset of Unicode.
459  * Works very similarly to the standard C function strchr().
460  *
461  * RETURN VALUE
462  *
463  * Non-zero if the Unicode character is in the given ASCII string.
464  */
465 int UnicodeInString(
466 	unsigned char *string,	/* (Input) String to search through. */
467 	unicode_t ch)		/* (Input) Unicode char to search for. */
468 {
469 	int found = false;
470 	while (*string != '\0' && found == false) {
471 		/* These types should compare, since both are unsigned
472 		 * numbers. */
473 		if (*string == ch) {
474 			found = true;
475 		}
476 		string++;
477 	}
478 	return(found);
479 }
480 #endif /* OS2 */
481 
482 /***********************************************************************
483  * Decides whether the given character is illegal for a given OS.
484  *
485  * RETURN VALUE
486  *
487  * Non-zero if char is illegal.
488  */
489 int IsIllegal(unicode_t ch)
490 {
491 #ifdef MAC
492 	/* Only illegal character on the MAC is the colon. */
493 	if (ch == 0x003A) {
494 		return(1);
495 	} else {
496 		return(0);
497 	}
498 
499 #elif defined UNIX
500 	/* Illegal UNIX characters are NULL and slash. */
501 	if (ch == 0x0000 || ch == 0x002F) {
502 		return(1);
503 	} else {
504 		return(0);
505 	}
506 
507 #elif defined OS2 || defined WIN_95 || defined WIN_NT
508 	/* Illegal char's for OS/2 according to WARP toolkit. */
509 	if (ch < 0x0020 || UnicodeInString("\\/:*?\"<>|", ch)) {
510 		return(1);
511 	} else {
512 		return(0);
513 	}
514 #endif
515 }
516 /* #endif*/	/* NEEDS_ISPRINT */
517 
518