xref: /minix/sys/fs/udf/udf_osta.c (revision ebfedea0) 
1 /* $NetBSD: udf_osta.c,v 1.10 2013/08/05 17:02:54 joerg Exp $ */
2 #if HAVE_NBTOOL_CONFIG_H
3 #include "nbtool_config.h"
4 #endif
5 
6 #include <sys/cdefs.h>
7 __KERNEL_RCSID(0, "$NetBSD: udf_osta.c,v 1.10 2013/08/05 17:02:54 joerg Exp $");
8 
9 /*
10  * Various routines from the OSTA 2.01 specs.  Copyrights are included with
11  * each code segment.  Slight whitespace modifications have been made for
12  * formatting purposes.  Typos/bugs have been fixed.
13  *
14  */
15 
16 #include "udf_osta.h"
17 
18 #ifndef _KERNEL
19 #include <ctype.h>
20 #endif
21 
22 /*****************************************************************************/
23 /***********************************************************************
24  * OSTA compliant Unicode compression, uncompression routines.
25  * Copyright 1995 Micro Design International, Inc.
26  * Written by Jason M. Rinn.
27  * Micro Design International gives permission for the free use of the
28  * following source code.
29  */
30 
31 /***********************************************************************
32  * Takes an OSTA CS0 compressed unicode name, and converts
33  * it to Unicode.
34  * The Unicode output will be in the byte order
35  * that the local compiler uses for 16-bit values.
36  * NOTE: This routine only performs error checking on the compID.
37  * It is up to the user to ensure that the unicode buffer is large
38  * enough, and that the compressed unicode name is correct.
39  *
40  * RETURN VALUE
41  *
42  * The number of unicode characters which were uncompressed.
43  * A -1 is returned if the compression ID is invalid.
44  */
45 int
46 udf_UncompressUnicode(
47 	int numberOfBytes,	/* (Input) number of bytes read from media. */
48 	byte *UDFCompressed,	/* (Input) bytes read from media. */
49 	unicode_t *unicode)	/* (Output) uncompressed unicode characters. */
50 {
51 	unsigned int compID;
52 	int returnValue, unicodeIndex, byteIndex;
53 
54 	/* Use UDFCompressed to store current byte being read. */
55 	compID = UDFCompressed[0];
56 
57 	/* First check for valid compID. */
58 	if (compID != 8 && compID != 16) {
59 		returnValue = -1;
60 	} else {
61 		unicodeIndex = 0;
62 		byteIndex = 1;
63 
64 		/* Loop through all the bytes. */
65 		while (byteIndex < numberOfBytes) {
66 			if (compID == 16) {
67 				/* Move the first byte to the high bits of the
68 				 * unicode char.
69 				 */
70 				unicode[unicodeIndex] =
71 				    UDFCompressed[byteIndex++] << 8;
72 			} else {
73 				unicode[unicodeIndex] = 0;
74 			}
75 			if (byteIndex < numberOfBytes) {
76 				/*Then the next byte to the low bits. */
77 				unicode[unicodeIndex] |=
78 				    UDFCompressed[byteIndex++];
79 			}
80 			unicodeIndex++;
81 		}
82 		returnValue = unicodeIndex;
83 	}
84 	return(returnValue);
85 }
86 
87 /***********************************************************************
88  * DESCRIPTION:
89  * Takes a string of unicode wide characters and returns an OSTA CS0
90  * compressed unicode string. The unicode MUST be in the byte order of
91  * the compiler in order to obtain correct results. Returns an error
92  * if the compression ID is invalid.
93  *
94  * NOTE: This routine assumes the implementation already knows, by
95  * the local environment, how many bits are appropriate and
96  * therefore does no checking to test if the input characters fit
97  * into that number of bits or not.
98  *
99  * RETURN VALUE
100  *
101  * The total number of bytes in the compressed OSTA CS0 string,
102  * including the compression ID.
103  * A -1 is returned if the compression ID is invalid.
104  */
105 int
106 udf_CompressUnicode(
107 	int numberOfChars,	/* (Input) number of unicode characters. */
108 	int compID,		/* (Input) compression ID to be used. */
109 	unicode_t *unicode,	/* (Input) unicode characters to compress. */
110 	byte *UDFCompressed)	/* (Output) compressed string, as bytes. */
111 {
112 	int byteIndex, unicodeIndex;
113 
114 	if (compID != 8 && compID != 16) {
115 		byteIndex = -1; /* Unsupported compression ID ! */
116 	} else {
117 		/* Place compression code in first byte. */
118 		UDFCompressed[0] = compID;
119 
120 		byteIndex = 1;
121 		unicodeIndex = 0;
122 		while (unicodeIndex < numberOfChars) {
123 			if (compID == 16) {
124 				/* First, place the high bits of the char
125 				 * into the byte stream.
126 				 */
127 				UDFCompressed[byteIndex++] =
128 				    (unicode[unicodeIndex] & 0xFF00) >> 8;
129 			}
130 			/*Then place the low bits into the stream. */
131 			UDFCompressed[byteIndex++] =
132 			    unicode[unicodeIndex] & 0x00FF;
133 			unicodeIndex++;
134 		}
135 	}
136 	return(byteIndex);
137 }
138 
139 /*****************************************************************************/
140 /*
141  * CRC 010041
142  */
143 static unsigned short crc_table[256] = {
144 	0x0000, 0x1021, 0x2042, 0x3063, 0x4084, 0x50A5, 0x60C6, 0x70E7,
145 	0x8108, 0x9129, 0xA14A, 0xB16B, 0xC18C, 0xD1AD, 0xE1CE, 0xF1EF,
146 	0x1231, 0x0210, 0x3273, 0x2252, 0x52B5, 0x4294, 0x72F7, 0x62D6,
147 	0x9339, 0x8318, 0xB37B, 0xA35A, 0xD3BD, 0xC39C, 0xF3FF, 0xE3DE,
148 	0x2462, 0x3443, 0x0420, 0x1401, 0x64E6, 0x74C7, 0x44A4, 0x5485,
149 	0xA56A, 0xB54B, 0x8528, 0x9509, 0xE5EE, 0xF5CF, 0xC5AC, 0xD58D,
150 	0x3653, 0x2672, 0x1611, 0x0630, 0x76D7, 0x66F6, 0x5695, 0x46B4,
151 	0xB75B, 0xA77A, 0x9719, 0x8738, 0xF7DF, 0xE7FE, 0xD79D, 0xC7BC,
152 	0x48C4, 0x58E5, 0x6886, 0x78A7, 0x0840, 0x1861, 0x2802, 0x3823,
153 	0xC9CC, 0xD9ED, 0xE98E, 0xF9AF, 0x8948, 0x9969, 0xA90A, 0xB92B,
154 	0x5AF5, 0x4AD4, 0x7AB7, 0x6A96, 0x1A71, 0x0A50, 0x3A33, 0x2A12,
155 	0xDBFD, 0xCBDC, 0xFBBF, 0xEB9E, 0x9B79, 0x8B58, 0xBB3B, 0xAB1A,
156 	0x6CA6, 0x7C87, 0x4CE4, 0x5CC5, 0x2C22, 0x3C03, 0x0C60, 0x1C41,
157 	0xEDAE, 0xFD8F, 0xCDEC, 0xDDCD, 0xAD2A, 0xBD0B, 0x8D68, 0x9D49,
158 	0x7E97, 0x6EB6, 0x5ED5, 0x4EF4, 0x3E13, 0x2E32, 0x1E51, 0x0E70,
159 	0xFF9F, 0xEFBE, 0xDFDD, 0xCFFC, 0xBF1B, 0xAF3A, 0x9F59, 0x8F78,
160 	0x9188, 0x81A9, 0xB1CA, 0xA1EB, 0xD10C, 0xC12D, 0xF14E, 0xE16F,
161 	0x1080, 0x00A1, 0x30C2, 0x20E3, 0x5004, 0x4025, 0x7046, 0x6067,
162 	0x83B9, 0x9398, 0xA3FB, 0xB3DA, 0xC33D, 0xD31C, 0xE37F, 0xF35E,
163 	0x02B1, 0x1290, 0x22F3, 0x32D2, 0x4235, 0x5214, 0x6277, 0x7256,
164 	0xB5EA, 0xA5CB, 0x95A8, 0x8589, 0xF56E, 0xE54F, 0xD52C, 0xC50D,
165 	0x34E2, 0x24C3, 0x14A0, 0x0481, 0x7466, 0x6447, 0x5424, 0x4405,
166 	0xA7DB, 0xB7FA, 0x8799, 0x97B8, 0xE75F, 0xF77E, 0xC71D, 0xD73C,
167 	0x26D3, 0x36F2, 0x0691, 0x16B0, 0x6657, 0x7676, 0x4615, 0x5634,
168 	0xD94C, 0xC96D, 0xF90E, 0xE92F, 0x99C8, 0x89E9, 0xB98A, 0xA9AB,
169 	0x5844, 0x4865, 0x7806, 0x6827, 0x18C0, 0x08E1, 0x3882, 0x28A3,
170 	0xCB7D, 0xDB5C, 0xEB3F, 0xFB1E, 0x8BF9, 0x9BD8, 0xABBB, 0xBB9A,
171 	0x4A75, 0x5A54, 0x6A37, 0x7A16, 0x0AF1, 0x1AD0, 0x2AB3, 0x3A92,
172 	0xFD2E, 0xED0F, 0xDD6C, 0xCD4D, 0xBDAA, 0xAD8B, 0x9DE8, 0x8DC9,
173 	0x7C26, 0x6C07, 0x5C64, 0x4C45, 0x3CA2, 0x2C83, 0x1CE0, 0x0CC1,
174 	0xEF1F, 0xFF3E, 0xCF5D, 0xDF7C, 0xAF9B, 0xBFBA, 0x8FD9, 0x9FF8,
175 	0x6E17, 0x7E36, 0x4E55, 0x5E74, 0x2E93, 0x3EB2, 0x0ED1, 0x1EF0
176 };
177 
178 unsigned short
179 udf_cksum(unsigned char *s, int n)
180 {
181 	unsigned short crc=0;
182 
183 	while (n-- > 0)
184 		crc = crc_table[(crc>>8 ^ *s++) & 0xff] ^ (crc<<8);
185 	return crc;
186 }
187 
188 /* UNICODE Checksum */
189 unsigned short
190 udf_unicode_cksum(unsigned short *s, int n)
191 {
192 	unsigned short crc=0;
193 
194 	while (n-- > 0) {
195 		/* Take high order byte first--corresponds to a big endian
196 		 * byte stream.
197 		 */
198 		crc = crc_table[(crc>>8 ^ (*s>>8)) & 0xff] ^ (crc<<8);
199 		crc = crc_table[(crc>>8 ^ (*s++ & 0xff)) & 0xff] ^ (crc<<8);
200 	}
201 	return crc;
202 }
203 
204 
205 /*
206   * Calculates a 16-bit checksum of the Implementation Use
207   * Extended Attribute header or Application Use Extended Attribute
208   * header. The fields AttributeType through ImplementationIdentifier
209   * (or ApplicationIdentifier) inclusively represent the
210   * data covered by the checksum (48 bytes).
211   *
212   */
213 uint16_t udf_ea_cksum(uint8_t *data) {
214         uint16_t checksum = 0;
215         int      count;
216 
217         for (count = 0; count < 48; count++) {
218                checksum += *data++;
219         }
220 
221         return checksum;
222 }
223 
224 
225 #ifdef MAIN
226 unsigned char bytes[] = { 0x70, 0x6A, 0x77 };
227 
228 main(void)
229 {
230 	unsigned short x;
231 	x = cksum(bytes, sizeof bytes);
232 	printf("checksum: calculated=%4.4x, correct=%4.4x\en", x, 0x3299);
233 	exit(0);
234 }
235 #endif
236 
237 /*****************************************************************************/
238 /* #ifdef NEEDS_ISPRINT */
239 /***********************************************************************
240  * OSTA UDF compliant file name translation routine for OS/2,
241  * Windows 95, Windows NT, Macintosh and UNIX.
242  * Copyright 1995 Micro Design International, Inc.
243  * Written by Jason M. Rinn.
244  * Micro Design International gives permission for the free use of the
245  * following source code.
246  */
247 
248 /***********************************************************************
249  * To use these routines with different operating systems.
250  *
251  * OS/2
252  * Define OS2
253  * Define MAXLEN = 254
254  *
255  * Windows 95
256  * Define WIN_95
257  * Define MAXLEN = 255
258  *
259  * Windows NT
260  * Define WIN_NT
261  * Define MAXLEN = 255
262  *
263  * Macintosh:
264  * Define MAC.
265  * Define MAXLEN = 31.
266  *
267  * UNIX
268  * Define UNIX.
269  * Define MAXLEN as specified by unix version.
270  */
271 
272 #define	ILLEGAL_CHAR_MARK	0x005F
273 #define	CRC_MARK	0x0023
274 #define	EXT_SIZE	5
275 #define	PERIOD	0x002E
276 #define	SPACE	0x0020
277 
278 /*** PROTOTYPES ***/
279 int IsIllegal(unicode_t ch);
280 
281 /* Define a function or macro which determines if a Unicode character is
282  * printable under your implementation.
283  */
284 
285 
286 /* #include <stdio.h> */
287 static int UnicodeIsPrint(unicode_t ch) {
288 	return (ch >=' ') && (ch != 127);
289 }
290 
291 
292 int UnicodeLength(unicode_t *string) {
293 	int length;
294 	length = 0;
295 	while (*string++) length++;
296 
297 	return length;
298 }
299 
300 
301 #ifdef _KERNEL
302 static int isprint(int c) {
303 	return (c >= ' ') && (c != 127);
304 }
305 #endif
306 
307 
308 /***********************************************************************
309  * Translates a long file name to one using a MAXLEN and an illegal
310  * char set in accord with the OSTA requirements. Assumes the name has
311  * already been translated to Unicode.
312  *
313  * RETURN VALUE
314  *
315  * Number of unicode characters in translated name.
316  */
317 int UDFTransName(
318 	unicode_t *newName,	/* (Output)Translated name. Must be of length
319 				 * MAXLEN */
320 	unicode_t *udfName,	/* (Input) Name from UDF volume.*/
321 	int udfLen)		/* (Input) Length of UDF Name. */
322 {
323 	int Index, newIndex = 0, needsCRC = false;	/* index is shadowed */
324 	int extIndex = 0, newExtIndex = 0, hasExt = false;
325 #if defined OS2 || defined WIN_95 || defined WIN_NT
326 	int trailIndex = 0;
327 #endif
328 	unsigned short valueCRC;
329 	unicode_t current;
330 	const char hexChar[] = "0123456789ABCDEF";
331 
332 	for (Index = 0; Index < udfLen; Index++) {
333 		current = udfName[Index];
334 
335 		if (IsIllegal(current) || !UnicodeIsPrint(current)) {
336 			needsCRC = true;
337 			/* Replace Illegal and non-displayable chars with
338 			 * underscore.
339 			 */
340 			current = ILLEGAL_CHAR_MARK;
341 			/* Skip any other illegal or non-displayable
342 			 * characters.
343 			 */
344 			while(Index+1 < udfLen && (IsIllegal(udfName[Index+1])
345 			    || !UnicodeIsPrint(udfName[Index+1]))) {
346 				Index++;
347 			}
348 		}
349 
350 		/* Record position of extension, if one is found. */
351 		if (current == PERIOD && (udfLen - Index -1) <= EXT_SIZE) {
352 			if (udfLen == Index + 1) {
353 				/* A trailing period is NOT an extension. */
354 				hasExt = false;
355 			} else {
356 				hasExt = true;
357 				extIndex = Index;
358 				newExtIndex = newIndex;
359 			}
360 		}
361 
362 #if defined OS2 || defined WIN_95 || defined WIN_NT
363 		/* Record position of last char which is NOT period or space. */
364 		else if (current != PERIOD && current != SPACE) {
365 			trailIndex = newIndex;
366 		}
367 #endif
368 
369 		if (newIndex < MAXLEN) {
370 			newName[newIndex++] = current;
371 		} else {
372 			needsCRC = true;
373 		}
374 	}
375 
376 #if defined OS2 || defined WIN_95 || defined WIN_NT
377 	/* For OS2, 95 & NT, truncate any trailing periods and\or spaces. */
378 	if (trailIndex != newIndex - 1) {
379 		newIndex = trailIndex + 1;
380 		needsCRC = true;
381 		hasExt = false; /* Trailing period does not make an
382 				 * extension. */
383 	}
384 #endif
385 
386 	if (needsCRC) {
387 		unicode_t ext[EXT_SIZE];
388 		int localExtIndex = 0;
389 		if (hasExt) {
390 			int maxFilenameLen;
391 			/* Translate extension, and store it in ext. */
392 			for(Index = 0; Index<EXT_SIZE &&
393 			    extIndex + Index +1 < udfLen; Index++ ) {
394 				current = udfName[extIndex + Index + 1];
395 				if (IsIllegal(current) ||
396 				    !UnicodeIsPrint(current)) {
397 					needsCRC = 1;
398 					/* Replace Illegal and non-displayable
399 					 * chars with underscore.
400 					 */
401 					current = ILLEGAL_CHAR_MARK;
402 					/* Skip any other illegal or
403 					 * non-displayable characters.
404 					 */
405 					while(Index + 1 < EXT_SIZE
406 					    && (IsIllegal(udfName[extIndex +
407 					    Index + 2]) ||
408 					    !isprint(udfName[extIndex +
409 					    Index + 2]))) {
410 						Index++;
411 					}
412 				}
413 				ext[localExtIndex++] = current;
414 			}
415 
416 			/* Truncate filename to leave room for extension and
417 			 * CRC.
418 			 */
419 			maxFilenameLen = ((MAXLEN - 5) - localExtIndex - 1);
420 			if (newIndex > maxFilenameLen) {
421 				newIndex = maxFilenameLen;
422 			} else {
423 				newIndex = newExtIndex;
424 			}
425 		} else if (newIndex > MAXLEN - 5) {
426 			/*If no extension, make sure to leave room for CRC. */
427 			newIndex = MAXLEN - 5;
428 		}
429 		newName[newIndex++] = CRC_MARK; /* Add mark for CRC. */
430 
431 		/*Calculate CRC from original filename from FileIdentifier. */
432 		valueCRC = udf_unicode_cksum(udfName, udfLen);
433 		/* Convert 16-bits of CRC to hex characters. */
434 		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
435 		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
436 		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
437 		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
438 
439 		/* Place a translated extension at end, if found. */
440 		if (hasExt) {
441 			newName[newIndex++] = PERIOD;
442 			for (Index = 0;Index < localExtIndex ;Index++ ) {
443 				newName[newIndex++] = ext[Index];
444 			}
445 		}
446 	}
447 	return(newIndex);
448 }
449 
450 #if defined OS2 || defined WIN_95 || defined WIN_NT
451 /***********************************************************************
452  * Decides if a Unicode character matches one of a list
453  * of ASCII characters.
454  * Used by OS2 version of IsIllegal for readability, since all of the
455  * illegal characters above 0x0020 are in the ASCII subset of Unicode.
456  * Works very similarly to the standard C function strchr().
457  *
458  * RETURN VALUE
459  *
460  * Non-zero if the Unicode character is in the given ASCII string.
461  */
462 int UnicodeInString(
463 	unsigned char *string,	/* (Input) String to search through. */
464 	unicode_t ch)		/* (Input) Unicode char to search for. */
465 {
466 	int found = false;
467 	while (*string != '\0' && found == false) {
468 		/* These types should compare, since both are unsigned
469 		 * numbers. */
470 		if (*string == ch) {
471 			found = true;
472 		}
473 		string++;
474 	}
475 	return(found);
476 }
477 #endif /* OS2 */
478 
479 /***********************************************************************
480  * Decides whether the given character is illegal for a given OS.
481  *
482  * RETURN VALUE
483  *
484  * Non-zero if char is illegal.
485  */
486 int IsIllegal(unicode_t ch)
487 {
488 #ifdef MAC
489 	/* Only illegal character on the MAC is the colon. */
490 	if (ch == 0x003A) {
491 		return(1);
492 	} else {
493 		return(0);
494 	}
495 
496 #elif defined UNIX
497 	/* Illegal UNIX characters are NULL and slash. */
498 	if (ch == 0x0000 || ch == 0x002F) {
499 		return(1);
500 	} else {
501 		return(0);
502 	}
503 
504 #elif defined OS2 || defined WIN_95 || defined WIN_NT
505 	/* Illegal char's for OS/2 according to WARP toolkit. */
506 	if (ch < 0x0020 || UnicodeInString("\\/:*?\"<>|", ch)) {
507 		return(1);
508 	} else {
509 		return(0);
510 	}
511 #endif
512 }
513 /* #endif*/	/* NEEDS_ISPRINT */
514 
515