xref: /netbsd/sys/fs/udf/udf_osta.c (revision b3404b22)
1 /* $NetBSD: udf_osta.c,v 1.11 2022/02/11 16:33:18 reinoud 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 1.11 2022/02/11 16:33:18 reinoud 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
udf_UncompressUnicode(int numberOfBytes,byte * UDFCompressed,unicode_t * unicode)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 	/* Translate 254/255 compID values used for deleted entries */
58 	if (compID == 254)
59 		compID = 8;
60 	if (compID == 255)
61 		compID = 16;
62 
63 	/* First check for valid compID. */
64 	if (compID != 8 && compID != 16) {
65 		returnValue = -1;
66 	} else {
67 		unicodeIndex = 0;
68 		byteIndex = 1;
69 
70 		/* Loop through all the bytes. */
71 		while (byteIndex < numberOfBytes) {
72 			if (compID == 16) {
73 				/* Move the first byte to the high bits of the
74 				 * unicode char.
75 				 */
76 				unicode[unicodeIndex] =
77 				    UDFCompressed[byteIndex++] << 8;
78 			} else {
79 				unicode[unicodeIndex] = 0;
80 			}
81 			if (byteIndex < numberOfBytes) {
82 				/*Then the next byte to the low bits. */
83 				unicode[unicodeIndex] |=
84 				    UDFCompressed[byteIndex++];
85 			}
86 			unicodeIndex++;
87 		}
88 		returnValue = unicodeIndex;
89 	}
90 	return(returnValue);
91 }
92 
93 /***********************************************************************
94  * DESCRIPTION:
95  * Takes a string of unicode wide characters and returns an OSTA CS0
96  * compressed unicode string. The unicode MUST be in the byte order of
97  * the compiler in order to obtain correct results. Returns an error
98  * if the compression ID is invalid.
99  *
100  * NOTE: This routine assumes the implementation already knows, by
101  * the local environment, how many bits are appropriate and
102  * therefore does no checking to test if the input characters fit
103  * into that number of bits or not.
104  *
105  * RETURN VALUE
106  *
107  * The total number of bytes in the compressed OSTA CS0 string,
108  * including the compression ID.
109  * A -1 is returned if the compression ID is invalid.
110  */
111 int
udf_CompressUnicode(int numberOfChars,int compID,unicode_t * unicode,byte * UDFCompressed)112 udf_CompressUnicode(
113 	int numberOfChars,	/* (Input) number of unicode characters. */
114 	int compID,		/* (Input) compression ID to be used. */
115 	unicode_t *unicode,	/* (Input) unicode characters to compress. */
116 	byte *UDFCompressed)	/* (Output) compressed string, as bytes. */
117 {
118 	int byteIndex, unicodeIndex;
119 
120 	if (compID != 8 && compID != 16) {
121 		byteIndex = -1; /* Unsupported compression ID ! */
122 	} else {
123 		/* Place compression code in first byte. */
124 		UDFCompressed[0] = compID;
125 
126 		byteIndex = 1;
127 		unicodeIndex = 0;
128 		while (unicodeIndex < numberOfChars) {
129 			if (compID == 16) {
130 				/* First, place the high bits of the char
131 				 * into the byte stream.
132 				 */
133 				UDFCompressed[byteIndex++] =
134 				    (unicode[unicodeIndex] & 0xFF00) >> 8;
135 			}
136 			/*Then place the low bits into the stream. */
137 			UDFCompressed[byteIndex++] =
138 			    unicode[unicodeIndex] & 0x00FF;
139 			unicodeIndex++;
140 		}
141 	}
142 	return(byteIndex);
143 }
144 
145 /*****************************************************************************/
146 /*
147  * CRC 010041
148  */
149 static unsigned short crc_table[256] = {
150 	0x0000, 0x1021, 0x2042, 0x3063, 0x4084, 0x50A5, 0x60C6, 0x70E7,
151 	0x8108, 0x9129, 0xA14A, 0xB16B, 0xC18C, 0xD1AD, 0xE1CE, 0xF1EF,
152 	0x1231, 0x0210, 0x3273, 0x2252, 0x52B5, 0x4294, 0x72F7, 0x62D6,
153 	0x9339, 0x8318, 0xB37B, 0xA35A, 0xD3BD, 0xC39C, 0xF3FF, 0xE3DE,
154 	0x2462, 0x3443, 0x0420, 0x1401, 0x64E6, 0x74C7, 0x44A4, 0x5485,
155 	0xA56A, 0xB54B, 0x8528, 0x9509, 0xE5EE, 0xF5CF, 0xC5AC, 0xD58D,
156 	0x3653, 0x2672, 0x1611, 0x0630, 0x76D7, 0x66F6, 0x5695, 0x46B4,
157 	0xB75B, 0xA77A, 0x9719, 0x8738, 0xF7DF, 0xE7FE, 0xD79D, 0xC7BC,
158 	0x48C4, 0x58E5, 0x6886, 0x78A7, 0x0840, 0x1861, 0x2802, 0x3823,
159 	0xC9CC, 0xD9ED, 0xE98E, 0xF9AF, 0x8948, 0x9969, 0xA90A, 0xB92B,
160 	0x5AF5, 0x4AD4, 0x7AB7, 0x6A96, 0x1A71, 0x0A50, 0x3A33, 0x2A12,
161 	0xDBFD, 0xCBDC, 0xFBBF, 0xEB9E, 0x9B79, 0x8B58, 0xBB3B, 0xAB1A,
162 	0x6CA6, 0x7C87, 0x4CE4, 0x5CC5, 0x2C22, 0x3C03, 0x0C60, 0x1C41,
163 	0xEDAE, 0xFD8F, 0xCDEC, 0xDDCD, 0xAD2A, 0xBD0B, 0x8D68, 0x9D49,
164 	0x7E97, 0x6EB6, 0x5ED5, 0x4EF4, 0x3E13, 0x2E32, 0x1E51, 0x0E70,
165 	0xFF9F, 0xEFBE, 0xDFDD, 0xCFFC, 0xBF1B, 0xAF3A, 0x9F59, 0x8F78,
166 	0x9188, 0x81A9, 0xB1CA, 0xA1EB, 0xD10C, 0xC12D, 0xF14E, 0xE16F,
167 	0x1080, 0x00A1, 0x30C2, 0x20E3, 0x5004, 0x4025, 0x7046, 0x6067,
168 	0x83B9, 0x9398, 0xA3FB, 0xB3DA, 0xC33D, 0xD31C, 0xE37F, 0xF35E,
169 	0x02B1, 0x1290, 0x22F3, 0x32D2, 0x4235, 0x5214, 0x6277, 0x7256,
170 	0xB5EA, 0xA5CB, 0x95A8, 0x8589, 0xF56E, 0xE54F, 0xD52C, 0xC50D,
171 	0x34E2, 0x24C3, 0x14A0, 0x0481, 0x7466, 0x6447, 0x5424, 0x4405,
172 	0xA7DB, 0xB7FA, 0x8799, 0x97B8, 0xE75F, 0xF77E, 0xC71D, 0xD73C,
173 	0x26D3, 0x36F2, 0x0691, 0x16B0, 0x6657, 0x7676, 0x4615, 0x5634,
174 	0xD94C, 0xC96D, 0xF90E, 0xE92F, 0x99C8, 0x89E9, 0xB98A, 0xA9AB,
175 	0x5844, 0x4865, 0x7806, 0x6827, 0x18C0, 0x08E1, 0x3882, 0x28A3,
176 	0xCB7D, 0xDB5C, 0xEB3F, 0xFB1E, 0x8BF9, 0x9BD8, 0xABBB, 0xBB9A,
177 	0x4A75, 0x5A54, 0x6A37, 0x7A16, 0x0AF1, 0x1AD0, 0x2AB3, 0x3A92,
178 	0xFD2E, 0xED0F, 0xDD6C, 0xCD4D, 0xBDAA, 0xAD8B, 0x9DE8, 0x8DC9,
179 	0x7C26, 0x6C07, 0x5C64, 0x4C45, 0x3CA2, 0x2C83, 0x1CE0, 0x0CC1,
180 	0xEF1F, 0xFF3E, 0xCF5D, 0xDF7C, 0xAF9B, 0xBFBA, 0x8FD9, 0x9FF8,
181 	0x6E17, 0x7E36, 0x4E55, 0x5E74, 0x2E93, 0x3EB2, 0x0ED1, 0x1EF0
182 };
183 
184 unsigned short
udf_cksum(unsigned char * s,int n)185 udf_cksum(unsigned char *s, int n)
186 {
187 	unsigned short crc=0;
188 
189 	while (n-- > 0)
190 		crc = crc_table[(crc>>8 ^ *s++) & 0xff] ^ (crc<<8);
191 	return crc;
192 }
193 
194 /* UNICODE Checksum */
195 unsigned short
udf_unicode_cksum(unsigned short * s,int n)196 udf_unicode_cksum(unsigned short *s, int n)
197 {
198 	unsigned short crc=0;
199 
200 	while (n-- > 0) {
201 		/* Take high order byte first--corresponds to a big endian
202 		 * byte stream.
203 		 */
204 		crc = crc_table[(crc>>8 ^ (*s>>8)) & 0xff] ^ (crc<<8);
205 		crc = crc_table[(crc>>8 ^ (*s++ & 0xff)) & 0xff] ^ (crc<<8);
206 	}
207 	return crc;
208 }
209 
210 
211 /*
212   * Calculates a 16-bit checksum of the Implementation Use
213   * Extended Attribute header or Application Use Extended Attribute
214   * header. The fields AttributeType through ImplementationIdentifier
215   * (or ApplicationIdentifier) inclusively represent the
216   * data covered by the checksum (48 bytes).
217   *
218   */
udf_ea_cksum(uint8_t * data)219 uint16_t udf_ea_cksum(uint8_t *data) {
220         uint16_t checksum = 0;
221         int      count;
222 
223         for (count = 0; count < 48; count++) {
224                checksum += *data++;
225         }
226 
227         return checksum;
228 }
229 
230 
231 #ifdef MAIN
232 unsigned char bytes[] = { 0x70, 0x6A, 0x77 };
233 
main(void)234 main(void)
235 {
236 	unsigned short x;
237 	x = cksum(bytes, sizeof bytes);
238 	printf("checksum: calculated=%4.4x, correct=%4.4x\en", x, 0x3299);
239 	exit(0);
240 }
241 #endif
242 
243 /*****************************************************************************/
244 /* #ifdef NEEDS_ISPRINT */
245 /***********************************************************************
246  * OSTA UDF compliant file name translation routine for OS/2,
247  * Windows 95, Windows NT, Macintosh and UNIX.
248  * Copyright 1995 Micro Design International, Inc.
249  * Written by Jason M. Rinn.
250  * Micro Design International gives permission for the free use of the
251  * following source code.
252  */
253 
254 /***********************************************************************
255  * To use these routines with different operating systems.
256  *
257  * OS/2
258  * Define OS2
259  * Define MAXLEN = 254
260  *
261  * Windows 95
262  * Define WIN_95
263  * Define MAXLEN = 255
264  *
265  * Windows NT
266  * Define WIN_NT
267  * Define MAXLEN = 255
268  *
269  * Macintosh:
270  * Define MAC.
271  * Define MAXLEN = 31.
272  *
273  * UNIX
274  * Define UNIX.
275  * Define MAXLEN as specified by unix version.
276  */
277 
278 #define	ILLEGAL_CHAR_MARK	0x005F
279 #define	CRC_MARK	0x0023
280 #define	EXT_SIZE	5
281 #define	PERIOD	0x002E
282 #define	SPACE	0x0020
283 
284 /*** PROTOTYPES ***/
285 int IsIllegal(unicode_t ch);
286 
287 /* Define a function or macro which determines if a Unicode character is
288  * printable under your implementation.
289  */
290 
291 
292 /* #include <stdio.h> */
UnicodeIsPrint(unicode_t ch)293 static int UnicodeIsPrint(unicode_t ch) {
294 	return (ch >=' ') && (ch != 127);
295 }
296 
297 
UnicodeLength(unicode_t * string)298 int UnicodeLength(unicode_t *string) {
299 	int length;
300 	length = 0;
301 	while (*string++) length++;
302 
303 	return length;
304 }
305 
306 
307 #ifdef _KERNEL
isprint(int c)308 static int isprint(int c) {
309 	return (c >= ' ') && (c != 127);
310 }
311 #endif
312 
313 
314 /***********************************************************************
315  * Translates a long file name to one using a MAXLEN and an illegal
316  * char set in accord with the OSTA requirements. Assumes the name has
317  * already been translated to Unicode.
318  *
319  * RETURN VALUE
320  *
321  * Number of unicode characters in translated name.
322  */
UDFTransName(unicode_t * newName,unicode_t * udfName,int udfLen)323 int UDFTransName(
324 	unicode_t *newName,	/* (Output)Translated name. Must be of length
325 				 * MAXLEN */
326 	unicode_t *udfName,	/* (Input) Name from UDF volume.*/
327 	int udfLen)		/* (Input) Length of UDF Name. */
328 {
329 	int Index, newIndex = 0, needsCRC = false;	/* index is shadowed */
330 	int extIndex = 0, newExtIndex = 0, hasExt = false;
331 #if defined OS2 || defined WIN_95 || defined WIN_NT
332 	int trailIndex = 0;
333 #endif
334 	unsigned short valueCRC;
335 	unicode_t current;
336 	const char hexChar[] = "0123456789ABCDEF";
337 
338 	for (Index = 0; Index < udfLen; Index++) {
339 		current = udfName[Index];
340 
341 		if (IsIllegal(current) || !UnicodeIsPrint(current)) {
342 			needsCRC = true;
343 			/* Replace Illegal and non-displayable chars with
344 			 * underscore.
345 			 */
346 			current = ILLEGAL_CHAR_MARK;
347 			/* Skip any other illegal or non-displayable
348 			 * characters.
349 			 */
350 			while(Index+1 < udfLen && (IsIllegal(udfName[Index+1])
351 			    || !UnicodeIsPrint(udfName[Index+1]))) {
352 				Index++;
353 			}
354 		}
355 
356 		/* Record position of extension, if one is found. */
357 		if (current == PERIOD && (udfLen - Index -1) <= EXT_SIZE) {
358 			if (udfLen == Index + 1) {
359 				/* A trailing period is NOT an extension. */
360 				hasExt = false;
361 			} else {
362 				hasExt = true;
363 				extIndex = Index;
364 				newExtIndex = newIndex;
365 			}
366 		}
367 
368 #if defined OS2 || defined WIN_95 || defined WIN_NT
369 		/* Record position of last char which is NOT period or space. */
370 		else if (current != PERIOD && current != SPACE) {
371 			trailIndex = newIndex;
372 		}
373 #endif
374 
375 		if (newIndex < MAXLEN) {
376 			newName[newIndex++] = current;
377 		} else {
378 			needsCRC = true;
379 		}
380 	}
381 
382 #if defined OS2 || defined WIN_95 || defined WIN_NT
383 	/* For OS2, 95 & NT, truncate any trailing periods and\or spaces. */
384 	if (trailIndex != newIndex - 1) {
385 		newIndex = trailIndex + 1;
386 		needsCRC = true;
387 		hasExt = false; /* Trailing period does not make an
388 				 * extension. */
389 	}
390 #endif
391 
392 	if (needsCRC) {
393 		unicode_t ext[EXT_SIZE];
394 		int localExtIndex = 0;
395 		if (hasExt) {
396 			int maxFilenameLen;
397 			/* Translate extension, and store it in ext. */
398 			for(Index = 0; Index<EXT_SIZE &&
399 			    extIndex + Index +1 < udfLen; Index++ ) {
400 				current = udfName[extIndex + Index + 1];
401 				if (IsIllegal(current) ||
402 				    !UnicodeIsPrint(current)) {
403 					needsCRC = 1;
404 					/* Replace Illegal and non-displayable
405 					 * chars with underscore.
406 					 */
407 					current = ILLEGAL_CHAR_MARK;
408 					/* Skip any other illegal or
409 					 * non-displayable characters.
410 					 */
411 					while(Index + 1 < EXT_SIZE
412 					    && (IsIllegal(udfName[extIndex +
413 					    Index + 2]) ||
414 					    !isprint(udfName[extIndex +
415 					    Index + 2]))) {
416 						Index++;
417 					}
418 				}
419 				ext[localExtIndex++] = current;
420 			}
421 
422 			/* Truncate filename to leave room for extension and
423 			 * CRC.
424 			 */
425 			maxFilenameLen = ((MAXLEN - 5) - localExtIndex - 1);
426 			if (newIndex > maxFilenameLen) {
427 				newIndex = maxFilenameLen;
428 			} else {
429 				newIndex = newExtIndex;
430 			}
431 		} else if (newIndex > MAXLEN - 5) {
432 			/*If no extension, make sure to leave room for CRC. */
433 			newIndex = MAXLEN - 5;
434 		}
435 		newName[newIndex++] = CRC_MARK; /* Add mark for CRC. */
436 
437 		/*Calculate CRC from original filename from FileIdentifier. */
438 		valueCRC = udf_unicode_cksum(udfName, udfLen);
439 		/* Convert 16-bits of CRC to hex characters. */
440 		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
441 		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
442 		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
443 		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
444 
445 		/* Place a translated extension at end, if found. */
446 		if (hasExt) {
447 			newName[newIndex++] = PERIOD;
448 			for (Index = 0;Index < localExtIndex ;Index++ ) {
449 				newName[newIndex++] = ext[Index];
450 			}
451 		}
452 	}
453 	return(newIndex);
454 }
455 
456 #if defined OS2 || defined WIN_95 || defined WIN_NT
457 /***********************************************************************
458  * Decides if a Unicode character matches one of a list
459  * of ASCII characters.
460  * Used by OS2 version of IsIllegal for readability, since all of the
461  * illegal characters above 0x0020 are in the ASCII subset of Unicode.
462  * Works very similarly to the standard C function strchr().
463  *
464  * RETURN VALUE
465  *
466  * Non-zero if the Unicode character is in the given ASCII string.
467  */
UnicodeInString(unsigned char * string,unicode_t ch)468 int UnicodeInString(
469 	unsigned char *string,	/* (Input) String to search through. */
470 	unicode_t ch)		/* (Input) Unicode char to search for. */
471 {
472 	int found = false;
473 	while (*string != '\0' && found == false) {
474 		/* These types should compare, since both are unsigned
475 		 * numbers. */
476 		if (*string == ch) {
477 			found = true;
478 		}
479 		string++;
480 	}
481 	return(found);
482 }
483 #endif /* OS2 */
484 
485 /***********************************************************************
486  * Decides whether the given character is illegal for a given OS.
487  *
488  * RETURN VALUE
489  *
490  * Non-zero if char is illegal.
491  */
IsIllegal(unicode_t ch)492 int IsIllegal(unicode_t ch)
493 {
494 #ifdef MAC
495 	/* Only illegal character on the MAC is the colon. */
496 	if (ch == 0x003A) {
497 		return(1);
498 	} else {
499 		return(0);
500 	}
501 
502 #elif defined UNIX
503 	/* Illegal UNIX characters are NULL and slash. */
504 	if (ch == 0x0000 || ch == 0x002F) {
505 		return(1);
506 	} else {
507 		return(0);
508 	}
509 
510 #elif defined OS2 || defined WIN_95 || defined WIN_NT
511 	/* Illegal char's for OS/2 according to WARP toolkit. */
512 	if (ch < 0x0020 || UnicodeInString("\\/:*?\"<>|", ch)) {
513 		return(1);
514 	} else {
515 		return(0);
516 	}
517 #endif
518 }
519 /* #endif*/	/* NEEDS_ISPRINT */
520 
521