2 * unistr.c - NTFS Unicode string handling. Part of the Linux-NTFS project.
4 * Copyright (c) 2001-2004 Anton Altaparmakov
6 * This program/include file is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU General Public License as published
8 * by the Free Software Foundation; either version 2 of the License, or
9 * (at your option) any later version.
11 * This program/include file is distributed in the hope that it will be
12 * useful, but WITHOUT ANY WARRANTY; without even the implied warranty
13 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 * GNU General Public License for more details.
16 * You should have received a copy of the GNU General Public License
17 * along with this program (in the main directory of the Linux-NTFS
18 * distribution in the file COPYING); if not, write to the Free Software
19 * Foundation,Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
28 * All these routines assume that the Unicode characters are in little endian
29 * encoding inside the strings!!!
33 * This is used by the name collation functions to quickly determine what
34 * characters are (in)valid.
36 static const u8 legal_ansi_char_array[0x40] = {
37 0x00, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10,
38 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10,
40 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10,
41 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10,
43 0x17, 0x07, 0x18, 0x17, 0x17, 0x17, 0x17, 0x17,
44 0x17, 0x17, 0x18, 0x16, 0x16, 0x17, 0x07, 0x00,
46 0x17, 0x17, 0x17, 0x17, 0x17, 0x17, 0x17, 0x17,
47 0x17, 0x17, 0x04, 0x16, 0x18, 0x16, 0x18, 0x18,
51 * ntfs_are_names_equal - compare two Unicode names for equality
52 * @s1: name to compare to @s2
53 * @s1_len: length in Unicode characters of @s1
54 * @s2: name to compare to @s1
55 * @s2_len: length in Unicode characters of @s2
56 * @ic: ignore case bool
57 * @upcase: upcase table (only if @ic == IGNORE_CASE)
58 * @upcase_size: length in Unicode characters of @upcase (if present)
60 * Compare the names @s1 and @s2 and return TRUE (1) if the names are
61 * identical, or FALSE (0) if they are not identical. If @ic is IGNORE_CASE,
62 * the @upcase table is used to performa a case insensitive comparison.
64 BOOL ntfs_are_names_equal(const uchar_t *s1, size_t s1_len,
65 const uchar_t *s2, size_t s2_len,
66 const IGNORE_CASE_BOOL ic,
67 const uchar_t *upcase, const u32 upcase_size)
71 if (ic == CASE_SENSITIVE)
72 return !ntfs_ucsncmp(s1, s2, s1_len);
73 return !ntfs_ucsncasecmp(s1, s2, s1_len, upcase, upcase_size);
77 * ntfs_collate_names - collate two Unicode names
78 * @name1: first Unicode name to compare
79 * @name2: second Unicode name to compare
80 * @err_val: if @name1 contains an invalid character return this value
81 * @ic: either CASE_SENSITIVE or IGNORE_CASE
82 * @upcase: upcase table (ignored if @ic is CASE_SENSITIVE)
83 * @upcase_len: upcase table size (ignored if @ic is CASE_SENSITIVE)
85 * ntfs_collate_names collates two Unicode names and returns:
87 * -1 if the first name collates before the second one,
88 * 0 if the names match,
89 * 1 if the second name collates before the first one, or
90 * @err_val if an invalid character is found in @name1 during the comparison.
92 * The following characters are considered invalid: '"', '*', '<', '>' and '?'.
94 int ntfs_collate_names(const uchar_t *name1, const u32 name1_len,
95 const uchar_t *name2, const u32 name2_len,
96 const int err_val, const IGNORE_CASE_BOOL ic,
97 const uchar_t *upcase, const u32 upcase_len)
103 if (name1_len > name2_len)
105 for (cnt = 0; cnt < min_len; ++cnt) {
106 c1 = le16_to_cpu(*name1++);
107 c2 = le16_to_cpu(*name2++);
110 c1 = le16_to_cpu(upcase[c1]);
112 c2 = le16_to_cpu(upcase[c2]);
114 if (c1 < 64 && legal_ansi_char_array[c1] & 8)
121 if (name1_len < name2_len)
123 if (name1_len == name2_len)
125 /* name1_len > name2_len */
126 c1 = le16_to_cpu(*name1);
127 if (c1 < 64 && legal_ansi_char_array[c1] & 8)
133 * ntfs_ucsncmp - compare two little endian Unicode strings
136 * @n: maximum unicode characters to compare
138 * Compare the first @n characters of the Unicode strings @s1 and @s2,
139 * The strings in little endian format and appropriate le16_to_cpu()
140 * conversion is performed on non-little endian machines.
142 * The function returns an integer less than, equal to, or greater than zero
143 * if @s1 (or the first @n Unicode characters thereof) is found, respectively,
144 * to be less than, to match, or be greater than @s2.
146 int ntfs_ucsncmp(const uchar_t *s1, const uchar_t *s2, size_t n)
151 for (i = 0; i < n; ++i) {
152 c1 = le16_to_cpu(s1[i]);
153 c2 = le16_to_cpu(s2[i]);
165 * ntfs_ucsncasecmp - compare two little endian Unicode strings, ignoring case
168 * @n: maximum unicode characters to compare
169 * @upcase: upcase table
170 * @upcase_size: upcase table size in Unicode characters
172 * Compare the first @n characters of the Unicode strings @s1 and @s2,
173 * ignoring case. The strings in little endian format and appropriate
174 * le16_to_cpu() conversion is performed on non-little endian machines.
176 * Each character is uppercased using the @upcase table before the comparison.
178 * The function returns an integer less than, equal to, or greater than zero
179 * if @s1 (or the first @n Unicode characters thereof) is found, respectively,
180 * to be less than, to match, or be greater than @s2.
182 int ntfs_ucsncasecmp(const uchar_t *s1, const uchar_t *s2, size_t n,
183 const uchar_t *upcase, const u32 upcase_size)
188 for (i = 0; i < n; ++i) {
189 if ((c1 = le16_to_cpu(s1[i])) < upcase_size)
190 c1 = le16_to_cpu(upcase[c1]);
191 if ((c2 = le16_to_cpu(s2[i])) < upcase_size)
192 c2 = le16_to_cpu(upcase[c2]);
203 void ntfs_upcase_name(uchar_t *name, u32 name_len, const uchar_t *upcase,
204 const u32 upcase_len)
209 for (i = 0; i < name_len; i++)
210 if ((u = le16_to_cpu(name[i])) < upcase_len)
214 void ntfs_file_upcase_value(FILE_NAME_ATTR *file_name_attr,
215 const uchar_t *upcase, const u32 upcase_len)
217 ntfs_upcase_name((uchar_t*)&file_name_attr->file_name,
218 file_name_attr->file_name_length, upcase, upcase_len);
221 int ntfs_file_compare_values(FILE_NAME_ATTR *file_name_attr1,
222 FILE_NAME_ATTR *file_name_attr2,
223 const int err_val, const IGNORE_CASE_BOOL ic,
224 const uchar_t *upcase, const u32 upcase_len)
226 return ntfs_collate_names((uchar_t*)&file_name_attr1->file_name,
227 file_name_attr1->file_name_length,
228 (uchar_t*)&file_name_attr2->file_name,
229 file_name_attr2->file_name_length,
230 err_val, ic, upcase, upcase_len);
234 * ntfs_nlstoucs - convert NLS string to little endian Unicode string
235 * @vol: ntfs volume which we are working with
236 * @ins: input NLS string buffer
237 * @ins_len: length of input string in bytes
238 * @outs: on return contains the allocated output Unicode string buffer
240 * Convert the input string @ins, which is in whatever format the loaded NLS
241 * map dictates, into a little endian, 2-byte Unicode string.
243 * This function allocates the string and the caller is responsible for
244 * calling kmem_cache_free(ntfs_name_cache, @outs); when finished with it.
246 * On success the function returns the number of Unicode characters written to
247 * the output string *@outs (>= 0), not counting the terminating Unicode NULL
248 * character. *@outs is set to the allocated output string buffer.
250 * On error, a negative number corresponding to the error code is returned. In
251 * that case the output string is not allocated. Both *@outs and *@outs_len
252 * are then undefined.
254 * This might look a bit odd due to fast path optimization...
256 int ntfs_nlstoucs(const ntfs_volume *vol, const char *ins,
257 const int ins_len, uchar_t **outs)
259 struct nls_table *nls = vol->nls_map;
264 /* We don't trust outside sources. */
266 ucs = (uchar_t*)kmem_cache_alloc(ntfs_name_cache, SLAB_NOFS);
268 for (i = o = 0; i < ins_len; i += wc_len) {
269 wc_len = nls->char2uni(ins + i, ins_len - i,
273 ucs[o++] = cpu_to_le16(wc);
277 } /* else (wc_len < 0) */
280 ucs[o] = cpu_to_le16('\0');
284 ntfs_error(vol->sb, "Failed to allocate name from "
288 ntfs_error(NULL, "Received NULL pointer.");
291 ntfs_error(vol->sb, "Name using character set %s contains characters "
292 "that cannot be converted to Unicode.", nls->charset);
293 kmem_cache_free(ntfs_name_cache, ucs);
298 * ntfs_ucstonls - convert little endian Unicode string to NLS string
299 * @vol: ntfs volume which we are working with
300 * @ins: input Unicode string buffer
301 * @ins_len: length of input string in Unicode characters
302 * @outs: on return contains the (allocated) output NLS string buffer
303 * @outs_len: length of output string buffer in bytes
305 * Convert the input little endian, 2-byte Unicode string @ins, of length
306 * @ins_len into the string format dictated by the loaded NLS.
308 * If *@outs is NULL, this function allocates the string and the caller is
309 * responsible for calling kfree(*@outs); when finished with it. In this case
310 * @outs_len is ignored and can be 0.
312 * On success the function returns the number of bytes written to the output
313 * string *@outs (>= 0), not counting the terminating NULL byte. If the output
314 * string buffer was allocated, *@outs is set to it.
316 * On error, a negative number corresponding to the error code is returned. In
317 * that case the output string is not allocated. The contents of *@outs are
320 * This might look a bit odd due to fast path optimization...
322 int ntfs_ucstonls(const ntfs_volume *vol, const uchar_t *ins,
323 const int ins_len, unsigned char **outs, int outs_len)
325 struct nls_table *nls = vol->nls_map;
327 int i, o, ns_len, wc;
329 /* We don't trust outside sources. */
338 ns_len = ins_len * NLS_MAX_CHARSET_SIZE;
339 ns = (unsigned char*)kmalloc(ns_len + 1, GFP_NOFS);
343 for (i = o = 0; i < ins_len; i++) {
344 retry: wc = nls->uni2char(le16_to_cpu(ins[i]), ns + o,
351 else if (wc == -ENAMETOOLONG && ns != *outs) {
353 /* Grow in multiples of 64 bytes. */
354 tc = (unsigned char*)kmalloc((ns_len + 64) &
357 memcpy(tc, ns, ns_len);
358 ns_len = ((ns_len + 64) & ~63) - 1;
362 } /* No memory so goto conversion_error; */
363 } /* wc < 0, real error. */
370 ntfs_error(vol->sb, "Received NULL pointer.");
373 ntfs_error(vol->sb, "Unicode name contains characters that cannot be "
374 "converted to character set %s.", nls->charset);
377 if (wc != -ENAMETOOLONG)
381 ntfs_error(vol->sb, "Failed to allocate name!");