xref: /netbsd-src/sys/fs/udf/udf_osta.c (revision d48f14661dda8638fee055ba15d35bdfb29b9fa8)
1 /* $NetBSD: udf_osta.c,v 1.2 2006/06/12 00:20:21 christos Exp $ */
2 
3 /*
4  * Various routines from the OSTA 2.01 specs.  Copyrights are included with
5  * each code segment.  Slight whitespace modifications have been made for
6  * formatting purposes.  Typos/bugs have been fixed.
7  *
8  */
9 
10 #include "udf_osta.h"
11 
12 
13 /*****************************************************************************/
14 /***********************************************************************
15  * OSTA compliant Unicode compression, uncompression routines.
16  * Copyright 1995 Micro Design International, Inc.
17  * Written by Jason M. Rinn.
18  * Micro Design International gives permission for the free use of the
19  * following source code.
20  */
21 
22 /***********************************************************************
23  * Takes an OSTA CS0 compressed unicode name, and converts
24  * it to Unicode.
25  * The Unicode output will be in the byte order
26  * that the local compiler uses for 16-bit values.
27  * NOTE: This routine only performs error checking on the compID.
28  * It is up to the user to ensure that the unicode buffer is large
29  * enough, and that the compressed unicode name is correct.
30  *
31  * RETURN VALUE
32  *
33  * The number of unicode characters which were uncompressed.
34  * A -1 is returned if the compression ID is invalid.
35  */
36 int
37 udf_UncompressUnicode(
38 	int numberOfBytes,	/* (Input) number of bytes read from media. */
39 	byte *UDFCompressed,	/* (Input) bytes read from media. */
40 	unicode_t *unicode)	/* (Output) uncompressed unicode characters. */
41 {
42 	unsigned int compID;
43 	int returnValue, unicodeIndex, byteIndex;
44 
45 	/* Use UDFCompressed to store current byte being read. */
46 	compID = UDFCompressed[0];
47 
48 	/* First check for valid compID. */
49 	if (compID != 8 && compID != 16) {
50 		returnValue = -1;
51 	} else {
52 		unicodeIndex = 0;
53 		byteIndex = 1;
54 
55 		/* Loop through all the bytes. */
56 		while (byteIndex < numberOfBytes) {
57 			if (compID == 16) {
58 				/* Move the first byte to the high bits of the
59 				 * unicode char.
60 				 */
61 				unicode[unicodeIndex] =
62 				    UDFCompressed[byteIndex++] << 8;
63 			} else {
64 				unicode[unicodeIndex] = 0;
65 			}
66 			if (byteIndex < numberOfBytes) {
67 				/*Then the next byte to the low bits. */
68 				unicode[unicodeIndex] |=
69 				    UDFCompressed[byteIndex++];
70 			}
71 			unicodeIndex++;
72 		}
73 		returnValue = unicodeIndex;
74 	}
75 	return(returnValue);
76 }
77 
78 /***********************************************************************
79  * DESCRIPTION:
80  * Takes a string of unicode wide characters and returns an OSTA CS0
81  * compressed unicode string. The unicode MUST be in the byte order of
82  * the compiler in order to obtain correct results. Returns an error
83  * if the compression ID is invalid.
84  *
85  * NOTE: This routine assumes the implementation already knows, by
86  * the local environment, how many bits are appropriate and
87  * therefore does no checking to test if the input characters fit
88  * into that number of bits or not.
89  *
90  * RETURN VALUE
91  *
92  * The total number of bytes in the compressed OSTA CS0 string,
93  * including the compression ID.
94  * A -1 is returned if the compression ID is invalid.
95  */
96 int
97 udf_CompressUnicode(
98 	int numberOfChars,	/* (Input) number of unicode characters. */
99 	int compID,		/* (Input) compression ID to be used. */
100 	unicode_t *unicode,	/* (Input) unicode characters to compress. */
101 	byte *UDFCompressed)	/* (Output) compressed string, as bytes. */
102 {
103 	int byteIndex, unicodeIndex;
104 
105 	if (compID != 8 && compID != 16) {
106 		byteIndex = -1; /* Unsupported compression ID ! */
107 	} else {
108 		/* Place compression code in first byte. */
109 		UDFCompressed[0] = compID;
110 
111 		byteIndex = 1;
112 		unicodeIndex = 0;
113 		while (unicodeIndex < numberOfChars) {
114 			if (compID == 16) {
115 				/* First, place the high bits of the char
116 				 * into the byte stream.
117 				 */
118 				UDFCompressed[byteIndex++] =
119 				    (unicode[unicodeIndex] & 0xFF00) >> 8;
120 			}
121 			/*Then place the low bits into the stream. */
122 			UDFCompressed[byteIndex++] =
123 			    unicode[unicodeIndex] & 0x00FF;
124 			unicodeIndex++;
125 		}
126 	}
127 	return(byteIndex);
128 }
129 
130 /*****************************************************************************/
131 /*
132  * CRC 010041
133  */
134 static unsigned short crc_table[256] = {
135 	0x0000, 0x1021, 0x2042, 0x3063, 0x4084, 0x50A5, 0x60C6, 0x70E7,
136 	0x8108, 0x9129, 0xA14A, 0xB16B, 0xC18C, 0xD1AD, 0xE1CE, 0xF1EF,
137 	0x1231, 0x0210, 0x3273, 0x2252, 0x52B5, 0x4294, 0x72F7, 0x62D6,
138 	0x9339, 0x8318, 0xB37B, 0xA35A, 0xD3BD, 0xC39C, 0xF3FF, 0xE3DE,
139 	0x2462, 0x3443, 0x0420, 0x1401, 0x64E6, 0x74C7, 0x44A4, 0x5485,
140 	0xA56A, 0xB54B, 0x8528, 0x9509, 0xE5EE, 0xF5CF, 0xC5AC, 0xD58D,
141 	0x3653, 0x2672, 0x1611, 0x0630, 0x76D7, 0x66F6, 0x5695, 0x46B4,
142 	0xB75B, 0xA77A, 0x9719, 0x8738, 0xF7DF, 0xE7FE, 0xD79D, 0xC7BC,
143 	0x48C4, 0x58E5, 0x6886, 0x78A7, 0x0840, 0x1861, 0x2802, 0x3823,
144 	0xC9CC, 0xD9ED, 0xE98E, 0xF9AF, 0x8948, 0x9969, 0xA90A, 0xB92B,
145 	0x5AF5, 0x4AD4, 0x7AB7, 0x6A96, 0x1A71, 0x0A50, 0x3A33, 0x2A12,
146 	0xDBFD, 0xCBDC, 0xFBBF, 0xEB9E, 0x9B79, 0x8B58, 0xBB3B, 0xAB1A,
147 	0x6CA6, 0x7C87, 0x4CE4, 0x5CC5, 0x2C22, 0x3C03, 0x0C60, 0x1C41,
148 	0xEDAE, 0xFD8F, 0xCDEC, 0xDDCD, 0xAD2A, 0xBD0B, 0x8D68, 0x9D49,
149 	0x7E97, 0x6EB6, 0x5ED5, 0x4EF4, 0x3E13, 0x2E32, 0x1E51, 0x0E70,
150 	0xFF9F, 0xEFBE, 0xDFDD, 0xCFFC, 0xBF1B, 0xAF3A, 0x9F59, 0x8F78,
151 	0x9188, 0x81A9, 0xB1CA, 0xA1EB, 0xD10C, 0xC12D, 0xF14E, 0xE16F,
152 	0x1080, 0x00A1, 0x30C2, 0x20E3, 0x5004, 0x4025, 0x7046, 0x6067,
153 	0x83B9, 0x9398, 0xA3FB, 0xB3DA, 0xC33D, 0xD31C, 0xE37F, 0xF35E,
154 	0x02B1, 0x1290, 0x22F3, 0x32D2, 0x4235, 0x5214, 0x6277, 0x7256,
155 	0xB5EA, 0xA5CB, 0x95A8, 0x8589, 0xF56E, 0xE54F, 0xD52C, 0xC50D,
156 	0x34E2, 0x24C3, 0x14A0, 0x0481, 0x7466, 0x6447, 0x5424, 0x4405,
157 	0xA7DB, 0xB7FA, 0x8799, 0x97B8, 0xE75F, 0xF77E, 0xC71D, 0xD73C,
158 	0x26D3, 0x36F2, 0x0691, 0x16B0, 0x6657, 0x7676, 0x4615, 0x5634,
159 	0xD94C, 0xC96D, 0xF90E, 0xE92F, 0x99C8, 0x89E9, 0xB98A, 0xA9AB,
160 	0x5844, 0x4865, 0x7806, 0x6827, 0x18C0, 0x08E1, 0x3882, 0x28A3,
161 	0xCB7D, 0xDB5C, 0xEB3F, 0xFB1E, 0x8BF9, 0x9BD8, 0xABBB, 0xBB9A,
162 	0x4A75, 0x5A54, 0x6A37, 0x7A16, 0x0AF1, 0x1AD0, 0x2AB3, 0x3A92,
163 	0xFD2E, 0xED0F, 0xDD6C, 0xCD4D, 0xBDAA, 0xAD8B, 0x9DE8, 0x8DC9,
164 	0x7C26, 0x6C07, 0x5C64, 0x4C45, 0x3CA2, 0x2C83, 0x1CE0, 0x0CC1,
165 	0xEF1F, 0xFF3E, 0xCF5D, 0xDF7C, 0xAF9B, 0xBFBA, 0x8FD9, 0x9FF8,
166 	0x6E17, 0x7E36, 0x4E55, 0x5E74, 0x2E93, 0x3EB2, 0x0ED1, 0x1EF0
167 };
168 
169 unsigned short
170 udf_cksum(s, n)
171 	unsigned char *s;
172 	int n;
173 {
174 	unsigned short crc=0;
175 
176 	while (n-- > 0)
177 		crc = crc_table[(crc>>8 ^ *s++) & 0xff] ^ (crc<<8);
178 	return crc;
179 }
180 
181 /* UNICODE Checksum */
182 unsigned short
183 udf_unicode_cksum(s, n)
184 	unsigned short *s;
185 	int n;
186 {
187 	unsigned short crc=0;
188 
189 	while (n-- > 0) {
190 		/* Take high order byte first--corresponds to a big endian
191 		 * byte stream.
192 		 */
193 		crc = crc_table[(crc>>8 ^ (*s>>8)) & 0xff] ^ (crc<<8);
194 		crc = crc_table[(crc>>8 ^ (*s++ & 0xff)) & 0xff] ^ (crc<<8);
195 	}
196 	return crc;
197 }
198 
199 #ifdef MAIN
200 unsigned char bytes[] = { 0x70, 0x6A, 0x77 };
201 
202 main()
203 {
204 	unsigned short x;
205 	x = cksum(bytes, sizeof bytes);
206 	printf("checksum: calculated=%4.4x, correct=%4.4x\en", x, 0x3299);
207 	exit(0);
208 }
209 #endif
210 
211 /*****************************************************************************/
212 /* #ifdef NEEDS_ISPRINT */
213 /***********************************************************************
214  * OSTA UDF compliant file name translation routine for OS/2,
215  * Windows 95, Windows NT, Macintosh and UNIX.
216  * Copyright 1995 Micro Design International, Inc.
217  * Written by Jason M. Rinn.
218  * Micro Design International gives permission for the free use of the
219  * following source code.
220  */
221 
222 /***********************************************************************
223  * To use these routines with different operating systems.
224  *
225  * OS/2
226  * Define OS2
227  * Define MAXLEN = 254
228  *
229  * Windows 95
230  * Define WIN_95
231  * Define MAXLEN = 255
232  *
233  * Windows NT
234  * Define WIN_NT
235  * Define MAXLEN = 255
236  *
237  * Macintosh:
238  * Define MAC.
239  * Define MAXLEN = 31.
240  *
241  * UNIX
242  * Define UNIX.
243  * Define MAXLEN as specified by unix version.
244  */
245 
246 #define	ILLEGAL_CHAR_MARK	0x005F
247 #define	CRC_MARK	0x0023
248 #define	EXT_SIZE	5
249 #define	TRUE	1
250 #define	FALSE	0
251 #define	PERIOD	0x002E
252 #define	SPACE	0x0020
253 
254 /*** PROTOTYPES ***/
255 int IsIllegal(unicode_t ch);
256 
257 /* Define a function or macro which determines if a Unicode character is
258  * printable under your implementation.
259  */
260 
261 
262 /* #include <stdio.h> */
263 static int UnicodeIsPrint(unicode_t ch) {
264 	return (ch >=' ') && (ch < 127);
265 }
266 
267 
268 int UnicodeLength(unicode_t *string) {
269 	int length;
270 	length = 0;
271 	while (*string++) length++;
272 
273 	return length;
274 }
275 
276 
277 static int isprint(unsigned char c) {
278 	return (c >= ' ') && (c != 127);
279 }
280 
281 
282 /***********************************************************************
283  * Translates a long file name to one using a MAXLEN and an illegal
284  * char set in accord with the OSTA requirements. Assumes the name has
285  * already been translated to Unicode.
286  *
287  * RETURN VALUE
288  *
289  * Number of unicode characters in translated name.
290  */
291 int UDFTransName(
292 	unicode_t *newName,	/* (Output)Translated name. Must be of length
293 				 * MAXLEN */
294 	unicode_t *udfName,	/* (Input) Name from UDF volume.*/
295 	int udfLen)		/* (Input) Length of UDF Name. */
296 {
297 	int index, newIndex = 0, needsCRC = FALSE;
298 	int extIndex = 0, newExtIndex = 0, hasExt = FALSE;
299 #if defined OS2 || defined WIN_95 || defined WIN_NT
300 	int trailIndex = 0;
301 #endif
302 	unsigned short valueCRC;
303 	unicode_t current;
304 	const char hexChar[] = "0123456789ABCDEF";
305 
306 	for (index = 0; index < udfLen; index++) {
307 		current = udfName[index];
308 
309 		if (IsIllegal(current) || !UnicodeIsPrint(current)) {
310 			needsCRC = TRUE;
311 			/* Replace Illegal and non-displayable chars with
312 			 * underscore.
313 			 */
314 			current = ILLEGAL_CHAR_MARK;
315 			/* Skip any other illegal or non-displayable
316 			 * characters.
317 			 */
318 			while(index+1 < udfLen && (IsIllegal(udfName[index+1])
319 			    || !UnicodeIsPrint(udfName[index+1]))) {
320 				index++;
321 			}
322 		}
323 
324 		/* Record position of extension, if one is found. */
325 		if (current == PERIOD && (udfLen - index -1) <= EXT_SIZE) {
326 			if (udfLen == index + 1) {
327 				/* A trailing period is NOT an extension. */
328 				hasExt = FALSE;
329 			} else {
330 				hasExt = TRUE;
331 				extIndex = index;
332 				newExtIndex = newIndex;
333 			}
334 		}
335 
336 #if defined OS2 || defined WIN_95 || defined WIN_NT
337 		/* Record position of last char which is NOT period or space. */
338 		else if (current != PERIOD && current != SPACE) {
339 			trailIndex = newIndex;
340 		}
341 #endif
342 
343 		if (newIndex < MAXLEN) {
344 			newName[newIndex++] = current;
345 		} else {
346 			needsCRC = TRUE;
347 		}
348 	}
349 
350 #if defined OS2 || defined WIN_95 || defined WIN_NT
351 	/* For OS2, 95 & NT, truncate any trailing periods and\or spaces. */
352 	if (trailIndex != newIndex - 1) {
353 		newIndex = trailIndex + 1;
354 		needsCRC = TRUE;
355 		hasExt = FALSE; /* Trailing period does not make an
356 				 * extension. */
357 	}
358 #endif
359 
360 	if (needsCRC) {
361 		unicode_t ext[EXT_SIZE];
362 		int localExtIndex = 0;
363 		if (hasExt) {
364 			int maxFilenameLen;
365 			/* Translate extension, and store it in ext. */
366 			for(index = 0; index<EXT_SIZE &&
367 			    extIndex + index +1 < udfLen; index++ ) {
368 				current = udfName[extIndex + index + 1];
369 				if (IsIllegal(current) ||
370 				    !UnicodeIsPrint(current)) {
371 					needsCRC = 1;
372 					/* Replace Illegal and non-displayable
373 					 * chars with underscore.
374 					 */
375 					current = ILLEGAL_CHAR_MARK;
376 					/* Skip any other illegal or
377 					 * non-displayable characters.
378 					 */
379 					while(index + 1 < EXT_SIZE
380 					    && (IsIllegal(udfName[extIndex +
381 					    index + 2]) ||
382 					    !isprint(udfName[extIndex +
383 					    index + 2]))) {
384 						index++;
385 					}
386 				}
387 				ext[localExtIndex++] = current;
388 			}
389 
390 			/* Truncate filename to leave room for extension and
391 			 * CRC.
392 			 */
393 			maxFilenameLen = ((MAXLEN - 5) - localExtIndex - 1);
394 			if (newIndex > maxFilenameLen) {
395 				newIndex = maxFilenameLen;
396 			} else {
397 				newIndex = newExtIndex;
398 			}
399 		} else if (newIndex > MAXLEN - 5) {
400 			/*If no extension, make sure to leave room for CRC. */
401 			newIndex = MAXLEN - 5;
402 		}
403 		newName[newIndex++] = CRC_MARK; /* Add mark for CRC. */
404 
405 		/*Calculate CRC from original filename from FileIdentifier. */
406 		valueCRC = udf_unicode_cksum(udfName, udfLen);
407 		/* Convert 16-bits of CRC to hex characters. */
408 		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
409 		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
410 		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
411 		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
412 
413 		/* Place a translated extension at end, if found. */
414 		if (hasExt) {
415 			newName[newIndex++] = PERIOD;
416 			for (index = 0;index < localExtIndex ;index++ ) {
417 				newName[newIndex++] = ext[index];
418 			}
419 		}
420 	}
421 	return(newIndex);
422 }
423 
424 #if defined OS2 || defined WIN_95 || defined WIN_NT
425 /***********************************************************************
426  * Decides if a Unicode character matches one of a list
427  * of ASCII characters.
428  * Used by OS2 version of IsIllegal for readability, since all of the
429  * illegal characters above 0x0020 are in the ASCII subset of Unicode.
430  * Works very similarly to the standard C function strchr().
431  *
432  * RETURN VALUE
433  *
434  * Non-zero if the Unicode character is in the given ASCII string.
435  */
436 int UnicodeInString(
437 	unsigned char *string,	/* (Input) String to search through. */
438 	unicode_t ch)		/* (Input) Unicode char to search for. */
439 {
440 	int found = FALSE;
441 	while (*string != '\0' && found == FALSE) {
442 		/* These types should compare, since both are unsigned
443 		 * numbers. */
444 		if (*string == ch) {
445 			found = TRUE;
446 		}
447 		string++;
448 	}
449 	return(found);
450 }
451 #endif /* OS2 */
452 
453 /***********************************************************************
454  * Decides whether the given character is illegal for a given OS.
455  *
456  * RETURN VALUE
457  *
458  * Non-zero if char is illegal.
459  */
460 int IsIllegal(unicode_t ch)
461 {
462 #ifdef MAC
463 	/* Only illegal character on the MAC is the colon. */
464 	if (ch == 0x003A) {
465 		return(1);
466 	} else {
467 		return(0);
468 	}
469 
470 #elif defined UNIX
471 	/* Illegal UNIX characters are NULL and slash. */
472 	if (ch == 0x0000 || ch == 0x002F) {
473 		return(1);
474 	} else {
475 		return(0);
476 	}
477 
478 #elif defined OS2 || defined WIN_95 || defined WIN_NT
479 	/* Illegal char's for OS/2 according to WARP toolkit. */
480 	if (ch < 0x0020 || UnicodeInString("\\/:*?\"<>|", ch)) {
481 		return(1);
482 	} else {
483 		return(0);
484 	}
485 #endif
486 }
487 /* #endif*/	/* NEEDS_ISPRINT */
488 
489