xref: /freebsd/sys/fs/udf/osta.c (revision d9a42747950146bf03cda7f6e25d219253f8a57a)
1 /*
2  * Various routines from the OSTA 2.01 specs.  Copyrights are included with
3  * each code segment.  Slight whitespace modifications have been made for
4  * formatting purposes.  Typos/bugs have been fixed.
5  *
6  * $FreeBSD$
7  */
8 
9 #include <fs/udf/osta.h>
10 
11 /*****************************************************************************/
12 /*-
13  **********************************************************************
14  * OSTA compliant Unicode compression, uncompression routines.
15  * Copyright 1995 Micro Design International, Inc.
16  * Written by Jason M. Rinn.
17  * Micro Design International gives permission for the free use of the
18  * following source code.
19  */
20 
21 /***********************************************************************
22  * Takes an OSTA CS0 compressed unicode name, and converts
23  * it to Unicode.
24  * The Unicode output will be in the byte order
25  * that the local compiler uses for 16-bit values.
26  * NOTE: This routine only performs error checking on the compID.
27  * It is up to the user to ensure that the unicode buffer is large
28  * enough, and that the compressed unicode name is correct.
29  *
30  * RETURN VALUE
31  *
32  * The number of unicode characters which were uncompressed.
33  * A -1 is returned if the compression ID is invalid.
34  */
35 int
36 udf_UncompressUnicode(
37 	int numberOfBytes,	/* (Input) number of bytes read from media. */
38 	byte *UDFCompressed,	/* (Input) bytes read from media. */
39 	unicode_t *unicode)	/* (Output) uncompressed unicode characters. */
40 {
41 	unsigned int compID;
42 	int returnValue, unicodeIndex, byteIndex;
43 
44 	/* Use UDFCompressed to store current byte being read. */
45 	compID = UDFCompressed[0];
46 
47 	/* First check for valid compID. */
48 	if (compID != 8 && compID != 16) {
49 		returnValue = -1;
50 	} else {
51 		unicodeIndex = 0;
52 		byteIndex = 1;
53 
54 		/* Loop through all the bytes. */
55 		while (byteIndex < numberOfBytes) {
56 			if (compID == 16) {
57 				/* Move the first byte to the high bits of the
58 				 * unicode char.
59 				 */
60 				unicode[unicodeIndex] =
61 				    UDFCompressed[byteIndex++] << 8;
62 			} else {
63 				unicode[unicodeIndex] = 0;
64 			}
65 			if (byteIndex < numberOfBytes) {
66 				/*Then the next byte to the low bits. */
67 				unicode[unicodeIndex] |=
68 				    UDFCompressed[byteIndex++];
69 			}
70 			unicodeIndex++;
71 		}
72 		returnValue = unicodeIndex;
73 	}
74 	return(returnValue);
75 }
76 
77 /*
78  * Almost same as udf_UncompressUnicode(). The difference is that
79  * it keeps byte order of unicode string.
80  */
81 int
82 udf_UncompressUnicodeByte(
83 	int numberOfBytes,	/* (Input) number of bytes read from media. */
84 	byte *UDFCompressed,	/* (Input) bytes read from media. */
85 	byte *unicode)		/* (Output) uncompressed unicode characters. */
86 {
87 	unsigned int compID;
88 	int returnValue, unicodeIndex, byteIndex;
89 
90 	/* Use UDFCompressed to store current byte being read. */
91 	compID = UDFCompressed[0];
92 
93 	/* First check for valid compID. */
94 	if (compID != 8 && compID != 16) {
95 		returnValue = -1;
96 	} else {
97 		unicodeIndex = 0;
98 		byteIndex = 1;
99 
100 		/* Loop through all the bytes. */
101 		while (byteIndex < numberOfBytes) {
102 			if (compID == 16) {
103 				/* Move the first byte to the high bits of the
104 				 * unicode char.
105 				 */
106 				unicode[unicodeIndex++] =
107 				    UDFCompressed[byteIndex++];
108 			} else {
109 				unicode[unicodeIndex++] = 0;
110 			}
111 			if (byteIndex < numberOfBytes) {
112 				/*Then the next byte to the low bits. */
113 				unicode[unicodeIndex++] =
114 				    UDFCompressed[byteIndex++];
115 			}
116 		}
117 		returnValue = unicodeIndex;
118 	}
119 	return(returnValue);
120 }
121 
122 /***********************************************************************
123  * DESCRIPTION:
124  * Takes a string of unicode wide characters and returns an OSTA CS0
125  * compressed unicode string. The unicode MUST be in the byte order of
126  * the compiler in order to obtain correct results. Returns an error
127  * if the compression ID is invalid.
128  *
129  * NOTE: This routine assumes the implementation already knows, by
130  * the local environment, how many bits are appropriate and
131  * therefore does no checking to test if the input characters fit
132  * into that number of bits or not.
133  *
134  * RETURN VALUE
135  *
136  * The total number of bytes in the compressed OSTA CS0 string,
137  * including the compression ID.
138  * A -1 is returned if the compression ID is invalid.
139  */
140 int
141 udf_CompressUnicode(
142 	int numberOfChars,	/* (Input) number of unicode characters. */
143 	int compID,		/* (Input) compression ID to be used. */
144 	unicode_t *unicode,	/* (Input) unicode characters to compress. */
145 	byte *UDFCompressed)	/* (Output) compressed string, as bytes. */
146 {
147 	int byteIndex, unicodeIndex;
148 
149 	if (compID != 8 && compID != 16) {
150 		byteIndex = -1; /* Unsupported compression ID ! */
151 	} else {
152 		/* Place compression code in first byte. */
153 		UDFCompressed[0] = compID;
154 
155 		byteIndex = 1;
156 		unicodeIndex = 0;
157 		while (unicodeIndex < numberOfChars) {
158 			if (compID == 16) {
159 				/* First, place the high bits of the char
160 				 * into the byte stream.
161 				 */
162 				UDFCompressed[byteIndex++] =
163 				    (unicode[unicodeIndex] & 0xFF00) >> 8;
164 			}
165 			/*Then place the low bits into the stream. */
166 			UDFCompressed[byteIndex++] =
167 			    unicode[unicodeIndex] & 0x00FF;
168 			unicodeIndex++;
169 		}
170 	}
171 	return(byteIndex);
172 }
173 
174 /*****************************************************************************/
175 /*
176  * CRC 010041
177  */
178 static unsigned short crc_table[256] = {
179 	0x0000, 0x1021, 0x2042, 0x3063, 0x4084, 0x50A5, 0x60C6, 0x70E7,
180 	0x8108, 0x9129, 0xA14A, 0xB16B, 0xC18C, 0xD1AD, 0xE1CE, 0xF1EF,
181 	0x1231, 0x0210, 0x3273, 0x2252, 0x52B5, 0x4294, 0x72F7, 0x62D6,
182 	0x9339, 0x8318, 0xB37B, 0xA35A, 0xD3BD, 0xC39C, 0xF3FF, 0xE3DE,
183 	0x2462, 0x3443, 0x0420, 0x1401, 0x64E6, 0x74C7, 0x44A4, 0x5485,
184 	0xA56A, 0xB54B, 0x8528, 0x9509, 0xE5EE, 0xF5CF, 0xC5AC, 0xD58D,
185 	0x3653, 0x2672, 0x1611, 0x0630, 0x76D7, 0x66F6, 0x5695, 0x46B4,
186 	0xB75B, 0xA77A, 0x9719, 0x8738, 0xF7DF, 0xE7FE, 0xD79D, 0xC7BC,
187 	0x48C4, 0x58E5, 0x6886, 0x78A7, 0x0840, 0x1861, 0x2802, 0x3823,
188 	0xC9CC, 0xD9ED, 0xE98E, 0xF9AF, 0x8948, 0x9969, 0xA90A, 0xB92B,
189 	0x5AF5, 0x4AD4, 0x7AB7, 0x6A96, 0x1A71, 0x0A50, 0x3A33, 0x2A12,
190 	0xDBFD, 0xCBDC, 0xFBBF, 0xEB9E, 0x9B79, 0x8B58, 0xBB3B, 0xAB1A,
191 	0x6CA6, 0x7C87, 0x4CE4, 0x5CC5, 0x2C22, 0x3C03, 0x0C60, 0x1C41,
192 	0xEDAE, 0xFD8F, 0xCDEC, 0xDDCD, 0xAD2A, 0xBD0B, 0x8D68, 0x9D49,
193 	0x7E97, 0x6EB6, 0x5ED5, 0x4EF4, 0x3E13, 0x2E32, 0x1E51, 0x0E70,
194 	0xFF9F, 0xEFBE, 0xDFDD, 0xCFFC, 0xBF1B, 0xAF3A, 0x9F59, 0x8F78,
195 	0x9188, 0x81A9, 0xB1CA, 0xA1EB, 0xD10C, 0xC12D, 0xF14E, 0xE16F,
196 	0x1080, 0x00A1, 0x30C2, 0x20E3, 0x5004, 0x4025, 0x7046, 0x6067,
197 	0x83B9, 0x9398, 0xA3FB, 0xB3DA, 0xC33D, 0xD31C, 0xE37F, 0xF35E,
198 	0x02B1, 0x1290, 0x22F3, 0x32D2, 0x4235, 0x5214, 0x6277, 0x7256,
199 	0xB5EA, 0xA5CB, 0x95A8, 0x8589, 0xF56E, 0xE54F, 0xD52C, 0xC50D,
200 	0x34E2, 0x24C3, 0x14A0, 0x0481, 0x7466, 0x6447, 0x5424, 0x4405,
201 	0xA7DB, 0xB7FA, 0x8799, 0x97B8, 0xE75F, 0xF77E, 0xC71D, 0xD73C,
202 	0x26D3, 0x36F2, 0x0691, 0x16B0, 0x6657, 0x7676, 0x4615, 0x5634,
203 	0xD94C, 0xC96D, 0xF90E, 0xE92F, 0x99C8, 0x89E9, 0xB98A, 0xA9AB,
204 	0x5844, 0x4865, 0x7806, 0x6827, 0x18C0, 0x08E1, 0x3882, 0x28A3,
205 	0xCB7D, 0xDB5C, 0xEB3F, 0xFB1E, 0x8BF9, 0x9BD8, 0xABBB, 0xBB9A,
206 	0x4A75, 0x5A54, 0x6A37, 0x7A16, 0x0AF1, 0x1AD0, 0x2AB3, 0x3A92,
207 	0xFD2E, 0xED0F, 0xDD6C, 0xCD4D, 0xBDAA, 0xAD8B, 0x9DE8, 0x8DC9,
208 	0x7C26, 0x6C07, 0x5C64, 0x4C45, 0x3CA2, 0x2C83, 0x1CE0, 0x0CC1,
209 	0xEF1F, 0xFF3E, 0xCF5D, 0xDF7C, 0xAF9B, 0xBFBA, 0x8FD9, 0x9FF8,
210 	0x6E17, 0x7E36, 0x4E55, 0x5E74, 0x2E93, 0x3EB2, 0x0ED1, 0x1EF0
211 };
212 
213 unsigned short
214 udf_cksum(unsigned char *s, int n)
215 {
216 	unsigned short crc=0;
217 
218 	while (n-- > 0)
219 		crc = crc_table[(crc>>8 ^ *s++) & 0xff] ^ (crc<<8);
220 	return crc;
221 }
222 
223 /* UNICODE Checksum */
224 unsigned short
225 udf_unicode_cksum(unsigned short *s, int n)
226 {
227 	unsigned short crc=0;
228 
229 	while (n-- > 0) {
230 		/* Take high order byte first--corresponds to a big endian
231 		 * byte stream.
232 		 */
233 		crc = crc_table[(crc>>8 ^ (*s>>8)) & 0xff] ^ (crc<<8);
234 		crc = crc_table[(crc>>8 ^ (*s++ & 0xff)) & 0xff] ^ (crc<<8);
235 	}
236 	return crc;
237 }
238 
239 #ifdef MAIN
240 unsigned char bytes[] = { 0x70, 0x6A, 0x77 };
241 
242 main()
243 {
244 	unsigned short x;
245 	x = cksum(bytes, sizeof bytes);
246 	printf("checksum: calculated=%4.4x, correct=%4.4x\en", x, 0x3299);
247 	exit(0);
248 }
249 #endif
250 
251 /*****************************************************************************/
252 #ifdef NEEDS_ISPRINT
253 /*-
254  **********************************************************************
255  * OSTA UDF compliant file name translation routine for OS/2,
256  * Windows 95, Windows NT, Macintosh and UNIX.
257  * Copyright 1995 Micro Design International, Inc.
258  * Written by Jason M. Rinn.
259  * Micro Design International gives permission for the free use of the
260  * following source code.
261  */
262 
263 /***********************************************************************
264  * To use these routines with different operating systems.
265  *
266  * OS/2
267  * Define OS2
268  * Define MAXLEN = 254
269  *
270  * Windows 95
271  * Define WIN_95
272  * Define MAXLEN = 255
273  *
274  * Windows NT
275  * Define WIN_NT
276  * Define MAXLEN = 255
277  *
278  * Macintosh:
279  * Define APPLE_MAC.
280  * Define MAXLEN = 31.
281  *
282  * UNIX
283  * Define UNIX.
284  * Define MAXLEN as specified by unix version.
285  */
286 
287 #define	ILLEGAL_CHAR_MARK	0x005F
288 #define	CRC_MARK	0x0023
289 #define	EXT_SIZE	5
290 #define	TRUE	1
291 #define	FALSE	0
292 #define	PERIOD	0x002E
293 #define	SPACE	0x0020
294 
295 /*** PROTOTYPES ***/
296 int IsIllegal(unicode_t ch);
297 
298 /* Define a function or macro which determines if a Unicode character is
299  * printable under your implementation.
300  */
301 int UnicodeIsPrint(unicode_t);
302 
303 /***********************************************************************
304  * Translates a long file name to one using a MAXLEN and an illegal
305  * char set in accord with the OSTA requirements. Assumes the name has
306  * already been translated to Unicode.
307  *
308  * RETURN VALUE
309  *
310  * Number of unicode characters in translated name.
311  */
312 int UDFTransName(
313 	unicode_t *newName,	/* (Output)Translated name. Must be of length
314 				 * MAXLEN */
315 	unicode_t *udfName,	/* (Input) Name from UDF volume.*/
316 	int udfLen)		/* (Input) Length of UDF Name. */
317 {
318 	int index, newIndex = 0, needsCRC = FALSE;
319 	int extIndex = 0, newExtIndex = 0, hasExt = FALSE;
320 #if defined OS2 || defined WIN_95 || defined WIN_NT
321 	int trailIndex = 0;
322 #endif
323 	unsigned short valueCRC;
324 	unicode_t current;
325 	const char hexChar[] = "0123456789ABCDEF";
326 
327 	for (index = 0; index < udfLen; index++) {
328 		current = udfName[index];
329 
330 		if (IsIllegal(current) || !UnicodeIsPrint(current)) {
331 			needsCRC = TRUE;
332 			/* Replace Illegal and non-displayable chars with
333 			 * underscore.
334 			 */
335 			current = ILLEGAL_CHAR_MARK;
336 			/* Skip any other illegal or non-displayable
337 			 * characters.
338 			 */
339 			while(index+1 < udfLen && (IsIllegal(udfName[index+1])
340 			    || !UnicodeIsPrint(udfName[index+1]))) {
341 				index++;
342 			}
343 		}
344 
345 		/* Record position of extension, if one is found. */
346 		if (current == PERIOD && (udfLen - index -1) <= EXT_SIZE) {
347 			if (udfLen == index + 1) {
348 				/* A trailing period is NOT an extension. */
349 				hasExt = FALSE;
350 			} else {
351 				hasExt = TRUE;
352 				extIndex = index;
353 				newExtIndex = newIndex;
354 			}
355 		}
356 
357 #if defined OS2 || defined WIN_95 || defined WIN_NT
358 		/* Record position of last char which is NOT period or space. */
359 		else if (current != PERIOD && current != SPACE) {
360 			trailIndex = newIndex;
361 		}
362 #endif
363 
364 		if (newIndex < MAXLEN) {
365 			newName[newIndex++] = current;
366 		} else {
367 			needsCRC = TRUE;
368 		}
369 	}
370 
371 #if defined OS2 || defined WIN_95 || defined WIN_NT
372 	/* For OS2, 95 & NT, truncate any trailing periods and\or spaces. */
373 	if (trailIndex != newIndex - 1) {
374 		newIndex = trailIndex + 1;
375 		needsCRC = TRUE;
376 		hasExt = FALSE; /* Trailing period does not make an
377 				 * extension. */
378 	}
379 #endif
380 
381 	if (needsCRC) {
382 		unicode_t ext[EXT_SIZE];
383 		int localExtIndex = 0;
384 		if (hasExt) {
385 			int maxFilenameLen;
386 			/* Translate extension, and store it in ext. */
387 			for(index = 0; index<EXT_SIZE &&
388 			    extIndex + index +1 < udfLen; index++ ) {
389 				current = udfName[extIndex + index + 1];
390 				if (IsIllegal(current) ||
391 				    !UnicodeIsPrint(current)) {
392 					needsCRC = 1;
393 					/* Replace Illegal and non-displayable
394 					 * chars with underscore.
395 					 */
396 					current = ILLEGAL_CHAR_MARK;
397 					/* Skip any other illegal or
398 					 * non-displayable characters.
399 					 */
400 					while(index + 1 < EXT_SIZE
401 					    && (IsIllegal(udfName[extIndex +
402 					    index + 2]) ||
403 					    !isprint(udfName[extIndex +
404 					    index + 2]))) {
405 						index++;
406 					}
407 				}
408 				ext[localExtIndex++] = current;
409 			}
410 
411 			/* Truncate filename to leave room for extension and
412 			 * CRC.
413 			 */
414 			maxFilenameLen = ((MAXLEN - 5) - localExtIndex - 1);
415 			if (newIndex > maxFilenameLen) {
416 				newIndex = maxFilenameLen;
417 			} else {
418 				newIndex = newExtIndex;
419 			}
420 		} else if (newIndex > MAXLEN - 5) {
421 			/*If no extension, make sure to leave room for CRC. */
422 			newIndex = MAXLEN - 5;
423 		}
424 		newName[newIndex++] = CRC_MARK; /* Add mark for CRC. */
425 
426 		/*Calculate CRC from original filename from FileIdentifier. */
427 		valueCRC = udf_unicode_cksum(udfName, udfLen);
428 		/* Convert 16-bits of CRC to hex characters. */
429 		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
430 		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
431 		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
432 		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
433 
434 		/* Place a translated extension at end, if found. */
435 		if (hasExt) {
436 			newName[newIndex++] = PERIOD;
437 			for (index = 0;index < localExtIndex ;index++ ) {
438 				newName[newIndex++] = ext[index];
439 			}
440 		}
441 	}
442 	return(newIndex);
443 }
444 
445 #if defined OS2 || defined WIN_95 || defined WIN_NT
446 /***********************************************************************
447  * Decides if a Unicode character matches one of a list
448  * of ASCII characters.
449  * Used by OS2 version of IsIllegal for readability, since all of the
450  * illegal characters above 0x0020 are in the ASCII subset of Unicode.
451  * Works very similarly to the standard C function strchr().
452  *
453  * RETURN VALUE
454  *
455  * Non-zero if the Unicode character is in the given ASCII string.
456  */
457 int UnicodeInString(
458 	unsigned char *string,	/* (Input) String to search through. */
459 	unicode_t ch)		/* (Input) Unicode char to search for. */
460 {
461 	int found = FALSE;
462 	while (*string != '\0' && found == FALSE) {
463 		/* These types should compare, since both are unsigned
464 		 * numbers. */
465 		if (*string == ch) {
466 			found = TRUE;
467 		}
468 		string++;
469 	}
470 	return(found);
471 }
472 #endif /* OS2 */
473 
474 /***********************************************************************
475  * Decides whether the given character is illegal for a given OS.
476  *
477  * RETURN VALUE
478  *
479  * Non-zero if char is illegal.
480  */
481 int IsIllegal(unicode_t ch)
482 {
483 #ifdef APPLE_MAC
484 	/* Only illegal character on the MAC is the colon. */
485 	if (ch == 0x003A) {
486 		return(1);
487 	} else {
488 		return(0);
489 	}
490 
491 #elif defined UNIX
492 	/* Illegal UNIX characters are NULL and slash. */
493 	if (ch == 0x0000 || ch == 0x002F) {
494 		return(1);
495 	} else {
496 		return(0);
497 	}
498 
499 #elif defined OS2 || defined WIN_95 || defined WIN_NT
500 	/* Illegal char's for OS/2 according to WARP toolkit. */
501 	if (ch < 0x0020 || UnicodeInString("\\/:*?\"<>|", ch)) {
502 		return(1);
503 	} else {
504 		return(0);
505 	}
506 #endif
507 }
508 #endif
509