1/*-
2 * SPDX-License-Identifier: BSD-4-Clause
3 *
4 * Copyright (C) 1994, 1995, 1997 Wolfgang Solfrank.
5 * Copyright (C) 1994, 1995, 1997 TooLs GmbH.
6 * All rights reserved.
7 * Original code by Paul Popelka (paulp@uts.amdahl.com) (see below).
8 *
9 * Redistribution and use in source and binary forms, with or without
10 * modification, are permitted provided that the following conditions
11 * are met:
12 * 1. Redistributions of source code must retain the above copyright
13 *    notice, this list of conditions and the following disclaimer.
14 * 2. Redistributions in binary form must reproduce the above copyright
15 *    notice, this list of conditions and the following disclaimer in the
16 *    documentation and/or other materials provided with the distribution.
17 * 3. All advertising materials mentioning features or use of this software
18 *    must display the following acknowledgement:
19 *	This product includes software developed by TooLs GmbH.
20 * 4. The name of TooLs GmbH may not be used to endorse or promote products
21 *    derived from this software without specific prior written permission.
22 *
23 * THIS SOFTWARE IS PROVIDED BY TOOLS GMBH ``AS IS'' AND ANY EXPRESS OR
24 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
25 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
26 * IN NO EVENT SHALL TOOLS GMBH BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
27 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
28 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
29 * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
30 * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
31 * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
32 * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
33 */
34/*-
35 * Written by Paul Popelka (paulp@uts.amdahl.com)
36 *
37 * You can do anything you want with this software, just don't say you wrote
38 * it, and don't remove this notice.
39 *
40 * This software is provided "as is".
41 *
42 * The author supplies this software to be publicly redistributed on the
43 * understanding that the author is not responsible for the correct
44 * functioning of this software in any circumstances and is not liable for
45 * any damages caused by this software.
46 *
47 * October 1992
48 */
49
50#include <sys/param.h>
51#include <sys/endian.h>
52
53#include <dirent.h>
54#include <stdio.h>
55#include <string.h>
56
57#include <fs/msdosfs/bpb.h>
58#include "msdos/direntry.h"
59#include <fs/msdosfs/msdosfsmount.h>
60
61#include "makefs.h"
62#include "msdos.h"
63
64static int char8ucs2str(const uint8_t *in, int n, uint16_t *out, int m);
65static void ucs2pad(uint16_t *buf, int len, int size);
66static int char8match(uint16_t *w1, uint16_t *w2, int n);
67
68static const u_char unix2dos[256] = {
69	0,    0,    0,    0,    0,    0,    0,    0,	/* 00-07 */
70	0,    0,    0,    0,    0,    0,    0,    0,	/* 08-0f */
71	0,    0,    0,    0,    0,    0,    0,    0,	/* 10-17 */
72	0,    0,    0,    0,    0,    0,    0,    0,	/* 18-1f */
73	0,    '!',  0,    '#',  '$',  '%',  '&',  '\'',	/* 20-27 */
74	'(',  ')',  0,    '+',  0,    '-',  0,    0,	/* 28-2f */
75	'0',  '1',  '2',  '3',  '4',  '5',  '6',  '7',	/* 30-37 */
76	'8',  '9',  0,    0,    0,    0,    0,    0,	/* 38-3f */
77	'@',  'A',  'B',  'C',  'D',  'E',  'F',  'G',	/* 40-47 */
78	'H',  'I',  'J',  'K',  'L',  'M',  'N',  'O',	/* 48-4f */
79	'P',  'Q',  'R',  'S',  'T',  'U',  'V',  'W',	/* 50-57 */
80	'X',  'Y',  'Z',  0,    0,    0,    '^',  '_',	/* 58-5f */
81	'`',  'A',  'B',  'C',  'D',  'E',  'F',  'G',	/* 60-67 */
82	'H',  'I',  'J',  'K',  'L',  'M',  'N',  'O',	/* 68-6f */
83	'P',  'Q',  'R',  'S',  'T',  'U',  'V',  'W',	/* 70-77 */
84	'X',  'Y',  'Z',  '{',  0,    '}',  '~',  0,	/* 78-7f */
85	0,    0,    0,    0,    0,    0,    0,    0,	/* 80-87 */
86	0,    0,    0,    0,    0,    0,    0,    0,	/* 88-8f */
87	0,    0,    0,    0,    0,    0,    0,    0,	/* 90-97 */
88	0,    0,    0,    0,    0,    0,    0,    0,	/* 98-9f */
89	0,    0xad, 0xbd, 0x9c, 0xcf, 0xbe, 0xdd, 0xf5,	/* a0-a7 */
90	0xf9, 0xb8, 0xa6, 0xae, 0xaa, 0xf0, 0xa9, 0xee,	/* a8-af */
91	0xf8, 0xf1, 0xfd, 0xfc, 0xef, 0xe6, 0xf4, 0xfa,	/* b0-b7 */
92	0xf7, 0xfb, 0xa7, 0xaf, 0xac, 0xab, 0xf3, 0xa8,	/* b8-bf */
93	0xb7, 0xb5, 0xb6, 0xc7, 0x8e, 0x8f, 0x92, 0x80,	/* c0-c7 */
94	0xd4, 0x90, 0xd2, 0xd3, 0xde, 0xd6, 0xd7, 0xd8,	/* c8-cf */
95	0xd1, 0xa5, 0xe3, 0xe0, 0xe2, 0xe5, 0x99, 0x9e,	/* d0-d7 */
96	0x9d, 0xeb, 0xe9, 0xea, 0x9a, 0xed, 0xe8, 0xe1,	/* d8-df */
97	0xb7, 0xb5, 0xb6, 0xc7, 0x8e, 0x8f, 0x92, 0x80,	/* e0-e7 */
98	0xd4, 0x90, 0xd2, 0xd3, 0xde, 0xd6, 0xd7, 0xd8,	/* e8-ef */
99	0xd1, 0xa5, 0xe3, 0xe0, 0xe2, 0xe5, 0x99, 0xf6,	/* f0-f7 */
100	0x9d, 0xeb, 0xe9, 0xea, 0x9a, 0xed, 0xe8, 0x98,	/* f8-ff */
101};
102
103static const u_char u2l[256] = {
104	0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, /* 00-07 */
105	0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, /* 08-0f */
106	0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, /* 10-17 */
107	0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, /* 18-1f */
108	' ',  '!',  '"',  '#',  '$',  '%',  '&', '\'', /* 20-27 */
109	'(',  ')',  '*',  '+',  ',',  '-',  '.',  '/', /* 28-2f */
110	'0',  '1',  '2',  '3',  '4',  '5',  '6',  '7', /* 30-37 */
111	'8',  '9',  ':',  ';',  '<',  '=',  '>',  '?', /* 38-3f */
112	'@',  'a',  'b',  'c',  'd',  'e',  'f',  'g', /* 40-47 */
113	'h',  'i',  'j',  'k',  'l',  'm',  'n',  'o', /* 48-4f */
114	'p',  'q',  'r',  's',  't',  'u',  'v',  'w', /* 50-57 */
115	'x',  'y',  'z',  '[', '\\',  ']',  '^',  '_', /* 58-5f */
116	'`',  'a',  'b',  'c',  'd',  'e',  'f',  'g', /* 60-67 */
117	'h',  'i',  'j',  'k',  'l',  'm',  'n',  'o', /* 68-6f */
118	'p',  'q',  'r',  's',  't',  'u',  'v',  'w', /* 70-77 */
119	'x',  'y',  'z',  '{',  '|',  '}',  '~', 0x7f, /* 78-7f */
120	0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, /* 80-87 */
121	0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f, /* 88-8f */
122	0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, /* 90-97 */
123	0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, /* 98-9f */
124	0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, /* a0-a7 */
125	0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf, /* a8-af */
126	0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, /* b0-b7 */
127	0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf, /* b8-bf */
128	0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, /* c0-c7 */
129	0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, /* c8-cf */
130	0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xd7, /* d0-d7 */
131	0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xdf, /* d8-df */
132	0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, /* e0-e7 */
133	0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, /* e8-ef */
134	0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, /* f0-f7 */
135	0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff, /* f8-ff */
136};
137
138/*
139 * Determine the number of slots necessary for Win95 names
140 */
141int
142winSlotCnt(const u_char *un, size_t unlen)
143{
144	const u_char *cp;
145
146	/*
147	 * Drop trailing blanks and dots
148	 */
149	for (cp = un + unlen; unlen > 0; unlen--)
150		if (*--cp != ' ' && *cp != '.')
151			break;
152
153	return howmany(unlen, WIN_CHARS);
154}
155
156/*
157 * Compare our filename to the one in the Win95 entry
158 * Returns the checksum or -1 if no match
159 */
160int
161winChkName(const u_char *un, size_t unlen, struct winentry *wep, int chksum)
162{
163	uint16_t wn[WIN_MAXLEN], *p;
164	uint16_t buf[WIN_CHARS];
165	int i, len;
166
167	/*
168	 * First compare checksums
169	 */
170	if (wep->weCnt & WIN_LAST)
171		chksum = wep->weChksum;
172	else if (chksum != wep->weChksum)
173		chksum = -1;
174	if (chksum == -1)
175		return -1;
176
177	/*
178	 * Offset of this entry
179	 */
180	i = ((wep->weCnt & WIN_CNT) - 1) * WIN_CHARS;
181
182	/*
183	 * Translate UNIX name to ucs-2
184	 */
185	len = char8ucs2str(un, unlen, wn, WIN_MAXLEN);
186	ucs2pad(wn, len, WIN_MAXLEN);
187
188	if (i >= len + 1)
189		return -1;
190	if ((wep->weCnt & WIN_LAST) && (len - i > WIN_CHARS))
191		return -1;
192
193	/*
194	 * Fetch name segment from directory entry
195	 */
196	p = &buf[0];
197	memcpy(p, wep->wePart1, sizeof(wep->wePart1));
198	p += sizeof(wep->wePart1) / sizeof(*p);
199	memcpy(p, wep->wePart2, sizeof(wep->wePart2));
200	p += sizeof(wep->wePart2) / sizeof(*p);
201	memcpy(p, wep->wePart3, sizeof(wep->wePart3));
202
203	/*
204	 * And compare name segment
205	 */
206	if (!(char8match(&wn[i], buf, WIN_CHARS)))
207		return -1;
208
209	return chksum;
210}
211
212/*
213 * Compute the checksum of a DOS filename for Win95 use
214 */
215uint8_t
216winChksum(uint8_t *name)
217{
218	int i;
219	uint8_t s;
220
221	for (s = 0, i = 11; --i >= 0; s += *name++)
222		s = (s << 7) | (s >> 1);
223	return s;
224}
225
226/*
227 * Create a Win95 long name directory entry
228 * Note: assumes that the filename is valid,
229 *	 i.e. doesn't consist solely of blanks and dots
230 */
231int
232unix2winfn(const u_char *un, size_t unlen, struct winentry *wep, int cnt,
233    int chksum)
234{
235	uint16_t wn[WIN_MAXLEN], *p;
236	int i, len;
237	const u_char *cp;
238
239	/*
240	 * Drop trailing blanks and dots
241	 */
242	for (cp = un + unlen; unlen > 0; unlen--)
243		if (*--cp != ' ' && *cp != '.')
244			break;
245
246	/*
247	 * Offset of this entry
248	 */
249	i = (cnt - 1) * WIN_CHARS;
250
251	/*
252	 * Translate UNIX name to ucs-2
253	 */
254	len = char8ucs2str(un, unlen, wn, WIN_MAXLEN);
255	ucs2pad(wn, len, WIN_MAXLEN);
256
257	/*
258	 * Initialize winentry to some useful default
259	 */
260	memset(wep, 0xff, sizeof(*wep));
261	wep->weCnt = cnt;
262	wep->weAttributes = ATTR_WIN95;
263	wep->weReserved1 = 0;
264	wep->weChksum = chksum;
265	wep->weReserved2 = 0;
266
267	/*
268	 * Store name segment into directory entry
269	 */
270	p = &wn[i];
271	memcpy(wep->wePart1, p, sizeof(wep->wePart1));
272	p += sizeof(wep->wePart1) / sizeof(*p);
273	memcpy(wep->wePart2, p, sizeof(wep->wePart2));
274	p += sizeof(wep->wePart2) / sizeof(*p);
275	memcpy(wep->wePart3, p, sizeof(wep->wePart3));
276
277	if (len > i + WIN_CHARS)
278		return 1;
279
280	wep->weCnt |= WIN_LAST;
281	return 0;
282}
283
284/*
285 * Convert a unix filename to a DOS filename according to Win95 rules.
286 * If applicable and gen is not 0, it is inserted into the converted
287 * filename as a generation number.
288 * Returns
289 *	0 if name couldn't be converted
290 *	1 if the converted name is the same as the original
291 *	  (no long filename entry necessary for Win95)
292 *	2 if conversion was successful
293 *	3 if conversion was successful and generation number was inserted
294 */
295int
296unix2dosfn(const u_char *un, u_char dn[12], size_t unlen, u_int gen)
297{
298	int i, j, l;
299	int conv = 1;
300	const u_char *cp, *dp, *dp1;
301	u_char gentext[6], *wcp;
302	int shortlen;
303
304	/*
305	 * Fill the dos filename string with blanks. These are DOS's pad
306	 * characters.
307	 */
308	for (i = 0; i < 11; i++)
309		dn[i] = ' ';
310	dn[11] = 0;
311
312	/*
313	 * The filenames "." and ".." are handled specially, since they
314	 * don't follow dos filename rules.
315	 */
316	if (un[0] == '.' && unlen == 1) {
317		dn[0] = '.';
318		return gen <= 1;
319	}
320	if (un[0] == '.' && un[1] == '.' && unlen == 2) {
321		dn[0] = '.';
322		dn[1] = '.';
323		return gen <= 1;
324	}
325
326	/*
327	 * Filenames with only blanks and dots are not allowed!
328	 */
329	for (cp = un, i = unlen; --i >= 0; cp++)
330		if (*cp != ' ' && *cp != '.')
331			break;
332	if (i < 0)
333		return 0;
334
335	/*
336	 * Now find the extension
337	 * Note: dot as first char doesn't start extension
338	 *	 and trailing dots and blanks are ignored
339	 */
340	dp = dp1 = 0;
341	for (cp = un + 1, i = unlen - 1; --i >= 0;) {
342		switch (*cp++) {
343		case '.':
344			if (!dp1)
345				dp1 = cp;
346			break;
347		case ' ':
348			break;
349		default:
350			if (dp1)
351				dp = dp1;
352			dp1 = 0;
353			break;
354		}
355	}
356
357	/*
358	 * Now convert it
359	 */
360	if (dp) {
361		if (dp1)
362			l = dp1 - dp;
363		else
364			l = unlen - (dp - un);
365		for (i = 0, j = 8; i < l && j < 11; i++, j++) {
366			if (dp[i] != (dn[j] = unix2dos[dp[i]])
367			    && conv != 3)
368				conv = 2;
369			if (!dn[j]) {
370				conv = 3;
371				dn[j--] = ' ';
372			}
373		}
374		if (i < l)
375			conv = 3;
376		dp--;
377	} else {
378		for (dp = cp; *--dp == ' ' || *dp == '.';);
379		dp++;
380	}
381
382	shortlen = (dp - un) <= 8;
383
384	/*
385	 * Now convert the rest of the name
386	 */
387	for (i = j = 0; un < dp && j < 8; i++, j++, un++) {
388		if ((*un == ' ') && shortlen)
389			dn[j] = ' ';
390		else
391			dn[j] = unix2dos[*un];
392		if ((*un != dn[j])
393		    && conv != 3)
394			conv = 2;
395		if (!dn[j]) {
396			conv = 3;
397			dn[j--] = ' ';
398		}
399	}
400	if (un < dp)
401		conv = 3;
402	/*
403	 * If we didn't have any chars in filename,
404	 * generate a default
405	 */
406	if (!j)
407		dn[0] = '_';
408
409	/*
410	 * The first character cannot be E5,
411	 * because that means a deleted entry
412	 */
413	if (dn[0] == 0xe5)
414		dn[0] = SLOT_E5;
415
416	/*
417	 * If there wasn't any char dropped,
418	 * there is no place for generation numbers
419	 */
420	if (conv != 3) {
421		if (gen > 1)
422			return 0;
423		return conv;
424	}
425
426	/*
427	 * Now insert the generation number into the filename part
428	 */
429	for (wcp = gentext + sizeof(gentext); wcp > gentext && gen; gen /= 10)
430		*--wcp = gen % 10 + '0';
431	if (gen)
432		return 0;
433	for (i = 8; dn[--i] == ' ';);
434	i++;
435	if (gentext + sizeof(gentext) - wcp + 1 > 8 - i)
436		i = 8 - (gentext + sizeof(gentext) - wcp + 1);
437	dn[i++] = '~';
438	while (wcp < gentext + sizeof(gentext))
439		dn[i++] = *wcp++;
440	return 3;
441}
442
443/*
444 * Convert 8bit character string into UCS-2 string
445 * return total number of output chacters
446 */
447static int
448char8ucs2str(const uint8_t *in, int n, uint16_t *out, int m)
449{
450	uint16_t *p;
451
452	p = out;
453	while (n > 0 && in[0] != 0) {
454		if (m < 1)
455			break;
456		if (p)
457			p[0] = htole16(in[0]);
458		p += 1;
459		m -= 1;
460		in += 1;
461		n -= 1;
462	}
463
464	return p - out;
465}
466
467static void
468ucs2pad(uint16_t *buf, int len, int size)
469{
470
471	if (len < size-1)
472		buf[len++] = 0x0000;
473	while (len < size)
474		buf[len++] = 0xffff;
475}
476
477/*
478 * Compare two 8bit char conversions case-insensitive
479 *
480 * uses the DOS case folding table
481 */
482static int
483char8match(uint16_t *w1, uint16_t *w2, int n)
484{
485	uint16_t u1, u2;
486
487	while (n > 0) {
488		u1 = le16toh(*w1);
489		u2 = le16toh(*w2);
490		if (u1 == 0 || u2 == 0)
491			return u1 == u2;
492		if (u1 > 255 || u2 > 255)
493			return 0;
494		u1 = u2l[u1 & 0xff];
495		u2 = u2l[u2 & 0xff];
496		if (u1 != u2)
497			return 0;
498		++w1;
499		++w2;
500		--n;
501	}
502
503	return 1;
504}
505