2 * unistr.c - NTFS Unicode string handling. Part of the Linux-NTFS project.
4 * Copyright (c) 2001-2005 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
30 * All these routines assume that the Unicode characters are in little endian
31 * encoding inside the strings!!!
35 * This is used by the name collation functions to quickly determine what
36 * characters are (in)valid.
38 static const u8 legal_ansi_char_array[0x40] = {
39 0x00, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10,
40 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10,
42 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10,
43 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10,
45 0x17, 0x07, 0x18, 0x17, 0x17, 0x17, 0x17, 0x17,
46 0x17, 0x17, 0x18, 0x16, 0x16, 0x17, 0x07, 0x00,
48 0x17, 0x17, 0x17, 0x17, 0x17, 0x17, 0x17, 0x17,
49 0x17, 0x17, 0x04, 0x16, 0x18, 0x16, 0x18, 0x18,
53 * ntfs_are_names_equal - compare two Unicode names for equality
54 * @s1: name to compare to @s2
55 * @s1_len: length in Unicode characters of @s1
56 * @s2: name to compare to @s1
57 * @s2_len: length in Unicode characters of @s2
58 * @ic: ignore case bool
59 * @upcase: upcase table (only if @ic == IGNORE_CASE)
60 * @upcase_size: length in Unicode characters of @upcase (if present)
62 * Compare the names @s1 and @s2 and return TRUE (1) if the names are
63 * identical, or FALSE (0) if they are not identical. If @ic is IGNORE_CASE,
64 * the @upcase table is used to performa a case insensitive comparison.
66 BOOL ntfs_are_names_equal(const ntfschar *s1, size_t s1_len,
67 const ntfschar *s2, size_t s2_len, const IGNORE_CASE_BOOL ic,
68 const ntfschar *upcase, const u32 upcase_size)
72 if (ic == CASE_SENSITIVE)
73 return !ntfs_ucsncmp(s1, s2, s1_len);
74 return !ntfs_ucsncasecmp(s1, s2, s1_len, upcase, upcase_size);
78 * ntfs_collate_names - collate two Unicode names
79 * @name1: first Unicode name to compare
80 * @name2: second Unicode name to compare
81 * @err_val: if @name1 contains an invalid character return this value
82 * @ic: either CASE_SENSITIVE or IGNORE_CASE
83 * @upcase: upcase table (ignored if @ic is CASE_SENSITIVE)
84 * @upcase_len: upcase table size (ignored if @ic is CASE_SENSITIVE)
86 * ntfs_collate_names collates two Unicode names and returns:
88 * -1 if the first name collates before the second one,
89 * 0 if the names match,
90 * 1 if the second name collates before the first one, or
91 * @err_val if an invalid character is found in @name1 during the comparison.
93 * The following characters are considered invalid: '"', '*', '<', '>' and '?'.
95 int ntfs_collate_names(const ntfschar *name1, const u32 name1_len,
96 const ntfschar *name2, const u32 name2_len,
97 const int err_val, const IGNORE_CASE_BOOL ic,
98 const ntfschar *upcase, const u32 upcase_len)
104 if (name1_len > name2_len)
106 for (cnt = 0; cnt < min_len; ++cnt) {
107 c1 = le16_to_cpu(*name1++);
108 c2 = le16_to_cpu(*name2++);
111 c1 = le16_to_cpu(upcase[c1]);
113 c2 = le16_to_cpu(upcase[c2]);
115 if (c1 < 64 && legal_ansi_char_array[c1] & 8)
122 if (name1_len < name2_len)
124 if (name1_len == name2_len)
126 /* name1_len > name2_len */
127 c1 = le16_to_cpu(*name1);
128 if (c1 < 64 && legal_ansi_char_array[c1] & 8)
134 * ntfs_ucsncmp - compare two little endian Unicode strings
137 * @n: maximum unicode characters to compare
139 * Compare the first @n characters of the Unicode strings @s1 and @s2,
140 * The strings in little endian format and appropriate le16_to_cpu()
141 * conversion is performed on non-little endian machines.
143 * The function returns an integer less than, equal to, or greater than zero
144 * if @s1 (or the first @n Unicode characters thereof) is found, respectively,
145 * to be less than, to match, or be greater than @s2.
147 int ntfs_ucsncmp(const ntfschar *s1, const ntfschar *s2, size_t n)
152 for (i = 0; i < n; ++i) {
153 c1 = le16_to_cpu(s1[i]);
154 c2 = le16_to_cpu(s2[i]);
166 * ntfs_ucsncasecmp - compare two little endian Unicode strings, ignoring case
169 * @n: maximum unicode characters to compare
170 * @upcase: upcase table
171 * @upcase_size: upcase table size in Unicode characters
173 * Compare the first @n characters of the Unicode strings @s1 and @s2,
174 * ignoring case. The strings in little endian format and appropriate
175 * le16_to_cpu() conversion is performed on non-little endian machines.
177 * Each character is uppercased using the @upcase table before the comparison.
179 * The function returns an integer less than, equal to, or greater than zero
180 * if @s1 (or the first @n Unicode characters thereof) is found, respectively,
181 * to be less than, to match, or be greater than @s2.
183 int ntfs_ucsncasecmp(const ntfschar *s1, const ntfschar *s2, size_t n,
184 const ntfschar *upcase, const u32 upcase_size)
189 for (i = 0; i < n; ++i) {
190 if ((c1 = le16_to_cpu(s1[i])) < upcase_size)
191 c1 = le16_to_cpu(upcase[c1]);
192 if ((c2 = le16_to_cpu(s2[i])) < upcase_size)
193 c2 = le16_to_cpu(upcase[c2]);
204 void ntfs_upcase_name(ntfschar *name, u32 name_len, const ntfschar *upcase,
205 const u32 upcase_len)
210 for (i = 0; i < name_len; i++)
211 if ((u = le16_to_cpu(name[i])) < upcase_len)
215 void ntfs_file_upcase_value(FILE_NAME_ATTR *file_name_attr,
216 const ntfschar *upcase, const u32 upcase_len)
218 ntfs_upcase_name((ntfschar*)&file_name_attr->file_name,
219 file_name_attr->file_name_length, upcase, upcase_len);
222 int ntfs_file_compare_values(FILE_NAME_ATTR *file_name_attr1,
223 FILE_NAME_ATTR *file_name_attr2,
224 const int err_val, const IGNORE_CASE_BOOL ic,
225 const ntfschar *upcase, const u32 upcase_len)
227 return ntfs_collate_names((ntfschar*)&file_name_attr1->file_name,
228 file_name_attr1->file_name_length,
229 (ntfschar*)&file_name_attr2->file_name,
230 file_name_attr2->file_name_length,
231 err_val, ic, upcase, upcase_len);
235 * ntfs_nlstoucs - convert NLS string to little endian Unicode string
236 * @vol: ntfs volume which we are working with
237 * @ins: input NLS string buffer
238 * @ins_len: length of input string in bytes
239 * @outs: on return contains the allocated output Unicode string buffer
241 * Convert the input string @ins, which is in whatever format the loaded NLS
242 * map dictates, into a little endian, 2-byte Unicode string.
244 * This function allocates the string and the caller is responsible for
245 * calling kmem_cache_free(ntfs_name_cache, @outs); when finished with it.
247 * On success the function returns the number of Unicode characters written to
248 * the output string *@outs (>= 0), not counting the terminating Unicode NULL
249 * character. *@outs is set to the allocated output string buffer.
251 * On error, a negative number corresponding to the error code is returned. In
252 * that case the output string is not allocated. Both *@outs and *@outs_len
253 * are then undefined.
255 * This might look a bit odd due to fast path optimization...
257 int ntfs_nlstoucs(const ntfs_volume *vol, const char *ins,
258 const int ins_len, ntfschar **outs)
260 struct nls_table *nls = vol->nls_map;
265 /* We don't trust outside sources. */
267 ucs = kmem_cache_alloc(ntfs_name_cache, SLAB_NOFS);
269 for (i = o = 0; i < ins_len; i += wc_len) {
270 wc_len = nls->char2uni(ins + i, ins_len - i,
274 ucs[o++] = cpu_to_le16(wc);
278 } /* else (wc_len < 0) */
285 ntfs_error(vol->sb, "Failed to allocate name from "
289 ntfs_error(NULL, "Received NULL pointer.");
292 ntfs_error(vol->sb, "Name using character set %s contains characters "
293 "that cannot be converted to Unicode.", nls->charset);
294 kmem_cache_free(ntfs_name_cache, ucs);
299 * ntfs_ucstonls - convert little endian Unicode string to NLS string
300 * @vol: ntfs volume which we are working with
301 * @ins: input Unicode string buffer
302 * @ins_len: length of input string in Unicode characters
303 * @outs: on return contains the (allocated) output NLS string buffer
304 * @outs_len: length of output string buffer in bytes
306 * Convert the input little endian, 2-byte Unicode string @ins, of length
307 * @ins_len into the string format dictated by the loaded NLS.
309 * If *@outs is NULL, this function allocates the string and the caller is
310 * responsible for calling kfree(*@outs); when finished with it. In this case
311 * @outs_len is ignored and can be 0.
313 * On success the function returns the number of bytes written to the output
314 * string *@outs (>= 0), not counting the terminating NULL byte. If the output
315 * string buffer was allocated, *@outs is set to it.
317 * On error, a negative number corresponding to the error code is returned. In
318 * that case the output string is not allocated. The contents of *@outs are
321 * This might look a bit odd due to fast path optimization...
323 int ntfs_ucstonls(const ntfs_volume *vol, const ntfschar *ins,
324 const int ins_len, unsigned char **outs, int outs_len)
326 struct nls_table *nls = vol->nls_map;
328 int i, o, ns_len, wc;
330 /* We don't trust outside sources. */
339 ns_len = ins_len * NLS_MAX_CHARSET_SIZE;
340 ns = (unsigned char*)kmalloc(ns_len + 1, GFP_NOFS);
344 for (i = o = 0; i < ins_len; i++) {
345 retry: wc = nls->uni2char(le16_to_cpu(ins[i]), ns + o,
352 else if (wc == -ENAMETOOLONG && ns != *outs) {
354 /* Grow in multiples of 64 bytes. */
355 tc = (unsigned char*)kmalloc((ns_len + 64) &
358 memcpy(tc, ns, ns_len);
359 ns_len = ((ns_len + 64) & ~63) - 1;
363 } /* No memory so goto conversion_error; */
364 } /* wc < 0, real error. */
371 ntfs_error(vol->sb, "Received NULL pointer.");
374 ntfs_error(vol->sb, "Unicode name contains characters that cannot be "
375 "converted to character set %s. You might want to "
376 "try to use the mount option nls=utf8.", nls->charset);
379 if (wc != -ENAMETOOLONG)
383 ntfs_error(vol->sb, "Failed to allocate name!");