xref: /netbsd/sys/fs/cd9660/cd9660_util.c (revision 6550d01e)
1 /*	$NetBSD: cd9660_util.c,v 1.9 2008/02/27 19:43:36 matt Exp $	*/
2 
3 /*-
4  * Copyright (c) 1994
5  *	The Regents of the University of California.  All rights reserved.
6  *
7  * This code is derived from software contributed to Berkeley
8  * by Pace Willisson (pace@blitz.com).  The Rock Ridge Extension
9  * Support code is derived from software contributed to Berkeley
10  * by Atsushi Murai (amurai@spec.co.jp).
11  *
12  * Redistribution and use in source and binary forms, with or without
13  * modification, are permitted provided that the following conditions
14  * are met:
15  * 1. Redistributions of source code must retain the above copyright
16  *    notice, this list of conditions and the following disclaimer.
17  * 2. Redistributions in binary form must reproduce the above copyright
18  *    notice, this list of conditions and the following disclaimer in the
19  *    documentation and/or other materials provided with the distribution.
20  * 3. Neither the name of the University nor the names of its contributors
21  *    may be used to endorse or promote products derived from this software
22  *    without specific prior written permission.
23  *
24  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
25  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
26  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
27  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
28  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
29  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
30  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
31  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
32  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
33  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
34  * SUCH DAMAGE.
35  *
36  *	@(#)cd9660_util.c	8.3 (Berkeley) 12/5/94
37  */
38 
39 #include <sys/cdefs.h>
40 __KERNEL_RCSID(0, "$NetBSD: cd9660_util.c,v 1.9 2008/02/27 19:43:36 matt Exp $");
41 
42 #include <sys/param.h>
43 #include <sys/systm.h>
44 #include <sys/namei.h>
45 #include <sys/resourcevar.h>
46 #include <sys/kernel.h>
47 #include <sys/file.h>
48 #include <sys/stat.h>
49 #include <sys/buf.h>
50 #include <sys/proc.h>
51 #include <sys/mount.h>
52 #include <sys/vnode.h>
53 #include <sys/malloc.h>
54 #include <sys/dirent.h>
55 
56 #include <fs/cd9660/iso.h>
57 #include <fs/cd9660/cd9660_extern.h>
58 
59 #include <fs/unicode.h>
60 
61 static u_int16_t wget(const u_char **, size_t *, int);
62 static int wput(u_char *, size_t, u_int16_t, int);
63 
64 int cd9660_utf8_joliet = 1;
65 
66 /*
67  * Get one character out of an iso filename
68  * Return number of bytes consumed
69  */
70 int
71 isochar(const u_char *isofn, const u_char *isoend, int joliet_level,
72 	u_int16_t *c)
73 {
74 	*c = isofn[0];
75 	if (joliet_level == 0 || isofn + 1 == isoend) {
76 		/* (00) and (01) are one byte in Joliet, too */
77 		return 1;
78 	}
79 
80 	if (cd9660_utf8_joliet) {
81 		*c = (*c << 8) + isofn[1];
82 	} else {
83 		/* characters outside ISO-8859-1 subset replaced with '?' */
84 		if (*c != 0)
85 			*c = '?';
86 		else
87 			*c = isofn[1];
88 	}
89 
90 	return 2;
91 }
92 
93 /*
94  * translate and compare a filename
95  * Note: Version number plus ';' may be omitted.
96  */
97 int
98 isofncmp(const u_char *fn, size_t fnlen, const u_char *isofn, size_t isolen,
99 	int joliet_level)
100 {
101 	int i, j;
102 	u_int16_t fc, ic;
103 	const u_char *isoend = isofn + isolen;
104 
105 	while (fnlen > 0) {
106 		fc = wget(&fn, &fnlen, joliet_level);
107 
108 		if (isofn == isoend)
109 			return fc;
110 		isofn += isochar(isofn, isoend, joliet_level, &ic);
111 		if (ic == ';') {
112 			switch (fc) {
113 			default:
114 				return fc;
115 			case 0:
116 				return 0;
117 			case ';':
118 				break;
119 			}
120 			fn++;
121 			for (i = 0; fnlen-- != 0; i = i * 10 + *fn++ - '0') {
122 				if (*fn < '0' || *fn > '9') {
123 					return -1;
124 				}
125 			}
126 			for (j = 0; isofn != isoend; j = j * 10 + ic - '0')
127 				isofn += isochar(isofn, isoend,
128 						 joliet_level, &ic);
129 			return i - j;
130 		}
131 		if (ic != fc) {
132 			if (ic >= 'A' && ic <= 'Z') {
133 				if (ic + ('a' - 'A') != fc) {
134 					if (fc >= 'a' && fc <= 'z')
135 						fc -= 'a' - 'A';
136 
137 					return (int) fc - (int) ic;
138 				}
139 			} else
140 				return (int) fc - (int) ic;
141 		}
142 	}
143 	if (isofn != isoend) {
144 		isofn += isochar(isofn, isoend, joliet_level, &ic);
145 		switch (ic) {
146 		default:
147 			return -1;
148 		case '.':
149 			if (isofn != isoend) {
150 				isochar(isofn, isoend, joliet_level, &ic);
151 				if (ic == ';')
152 					return 0;
153 			}
154 			return -1;
155 		case ';':
156 			return 0;
157 		}
158 	}
159 	return 0;
160 }
161 
162 /*
163  * translate a filename
164  */
165 void
166 isofntrans(const u_char *infn, int infnlen, u_char *outfn, u_short *outfnlen,
167 	int original, int casetrans, int assoc, int joliet_level)
168 {
169 	int fnidx = 0;
170 	const u_char *infnend = infn + infnlen;
171 	u_int16_t c;
172 	int sz;
173 
174 	if (assoc) {
175 		*outfn++ = ASSOCCHAR;
176 		fnidx++;
177 	}
178 
179 	for(; infn != infnend; fnidx += sz) {
180 		infn += isochar(infn, infnend, joliet_level, &c);
181 
182 		if (casetrans && joliet_level == 0 && c >= 'A' && c <= 'Z')
183 			c = c + ('a' - 'A');
184 		else if (!original && c == ';') {
185 			if (fnidx > 0 && outfn[-1] == '.')
186 				fnidx--;
187 			break;
188 		}
189 
190 		sz = wput(outfn, MAXNAMLEN - fnidx, c, joliet_level);
191 		if (sz == 0) {
192 			/* not enough space to write the character */
193 			if (fnidx < MAXNAMLEN) {
194 				*outfn = '?';
195 				fnidx++;
196 			}
197 			break;
198 		}
199 		outfn += sz;
200 	}
201 	*outfnlen = fnidx;
202 }
203 
204 static u_int16_t
205 wget(const u_char **str, size_t *sz, int joliet_level)
206 {
207 	if (joliet_level > 0 && cd9660_utf8_joliet) {
208 		/* decode UTF-8 sequence */
209 		return wget_utf8((const char **) str, sz);
210 	} else {
211 		/*
212 		 * Raw 8-bit characters without any conversion. For Joliet,
213 		 * this effectively assumes provided file name is using
214 		 * ISO-8859-1 subset.
215 		 */
216 		u_int16_t c = *str[0];
217 		(*str)++;
218 		(*sz)--;
219 
220 		return c;
221 	}
222 }
223 
224 static int
225 wput(u_char *s, size_t n, u_int16_t c, int joliet_level)
226 {
227 	if (joliet_level > 0 && cd9660_utf8_joliet) {
228 		/* Store Joliet file name encoded into UTF-8 */
229 		return wput_utf8((char *)s, n, c);
230 	} else {
231 		/*
232 		 * Store raw 8-bit characters without any conversion.
233 		 * For Joliet case, this filters the Unicode characters
234 		 * to ISO-8859-1 subset.
235 		 */
236 		*s = (u_char)c;
237 		return 1;
238 	}
239 }
240