1 /* SPDX-License-Identifier: GPL-2.0-or-later */
2 /*
3 * Some of the source code in this file came from fs/cifs/cifs_unicode.c
4 * cifs_unicode: Unicode kernel case support
5 *
6 * Function:
7 * Convert a unicode character to upper or lower case using
8 * compressed tables.
9 *
10 * Copyright (c) International Business Machines Corp., 2000,2009
11 *
12 *
13 * Notes:
14 * These APIs are based on the C library functions. The semantics
15 * should match the C functions but with expanded size operands.
16 *
17 * The upper/lower functions are based on a table created by mkupr.
18 * This is a compressed table of upper and lower case conversion.
19 *
20 */
21 #ifndef _CIFS_UNICODE_H
22 #define _CIFS_UNICODE_H
23
24 #include <asm/byteorder.h>
25 #include <linux/types.h>
26 #include <linux/nls.h>
27
28 #define UNIUPR_NOLOWER /* Example to not expand lower case tables */
29
30 /*
31 * Windows maps these to the user defined 16 bit Unicode range since they are
32 * reserved symbols (along with \ and /), otherwise illegal to store
33 * in filenames in NTFS
34 */
35 #define UNI_ASTERISK ((__u16)('*' + 0xF000))
36 #define UNI_QUESTION ((__u16)('?' + 0xF000))
37 #define UNI_COLON ((__u16)(':' + 0xF000))
38 #define UNI_GRTRTHAN ((__u16)('>' + 0xF000))
39 #define UNI_LESSTHAN ((__u16)('<' + 0xF000))
40 #define UNI_PIPE ((__u16)('|' + 0xF000))
41 #define UNI_SLASH ((__u16)('\\' + 0xF000))
42
43 /* Just define what we want from uniupr.h. We don't want to define the tables
44 * in each source file.
45 */
46 #ifndef UNICASERANGE_DEFINED
47 struct UniCaseRange {
48 wchar_t start;
49 wchar_t end;
50 signed char *table;
51 };
52 #endif /* UNICASERANGE_DEFINED */
53
54 #ifndef UNIUPR_NOUPPER
55 extern signed char SmbUniUpperTable[512];
56 extern const struct UniCaseRange SmbUniUpperRange[];
57 #endif /* UNIUPR_NOUPPER */
58
59 #ifndef UNIUPR_NOLOWER
60 extern signed char CifsUniLowerTable[512];
61 extern const struct UniCaseRange CifsUniLowerRange[];
62 #endif /* UNIUPR_NOLOWER */
63
64 #ifdef __KERNEL__
65 int smb_strtoUTF16(__le16 *to, const char *from, int len,
66 const struct nls_table *codepage);
67 char *smb_strndup_from_utf16(const char *src, const int maxlen,
68 const bool is_unicode,
69 const struct nls_table *codepage);
70 int smbConvertToUTF16(__le16 *target, const char *source, int srclen,
71 const struct nls_table *cp, int mapchars);
72 char *ksmbd_extract_sharename(char *treename);
73 #endif
74
75 /*
76 * UniStrcat: Concatenate the second string to the first
77 *
78 * Returns:
79 * Address of the first string
80 */
UniStrcat(wchar_t * ucs1,const wchar_t * ucs2)81 static inline wchar_t *UniStrcat(wchar_t *ucs1, const wchar_t *ucs2)
82 {
83 wchar_t *anchor = ucs1; /* save a pointer to start of ucs1 */
84
85 while (*ucs1++)
86 /*NULL*/; /* To end of first string */
87 ucs1--; /* Return to the null */
88 while ((*ucs1++ = *ucs2++))
89 /*NULL*/; /* copy string 2 over */
90 return anchor;
91 }
92
93 /*
94 * UniStrchr: Find a character in a string
95 *
96 * Returns:
97 * Address of first occurrence of character in string
98 * or NULL if the character is not in the string
99 */
UniStrchr(const wchar_t * ucs,wchar_t uc)100 static inline wchar_t *UniStrchr(const wchar_t *ucs, wchar_t uc)
101 {
102 while ((*ucs != uc) && *ucs)
103 ucs++;
104
105 if (*ucs == uc)
106 return (wchar_t *)ucs;
107 return NULL;
108 }
109
110 /*
111 * UniStrcmp: Compare two strings
112 *
113 * Returns:
114 * < 0: First string is less than second
115 * = 0: Strings are equal
116 * > 0: First string is greater than second
117 */
UniStrcmp(const wchar_t * ucs1,const wchar_t * ucs2)118 static inline int UniStrcmp(const wchar_t *ucs1, const wchar_t *ucs2)
119 {
120 while ((*ucs1 == *ucs2) && *ucs1) {
121 ucs1++;
122 ucs2++;
123 }
124 return (int)*ucs1 - (int)*ucs2;
125 }
126
127 /*
128 * UniStrcpy: Copy a string
129 */
UniStrcpy(wchar_t * ucs1,const wchar_t * ucs2)130 static inline wchar_t *UniStrcpy(wchar_t *ucs1, const wchar_t *ucs2)
131 {
132 wchar_t *anchor = ucs1; /* save the start of result string */
133
134 while ((*ucs1++ = *ucs2++))
135 /*NULL*/;
136 return anchor;
137 }
138
139 /*
140 * UniStrlen: Return the length of a string (in 16 bit Unicode chars not bytes)
141 */
UniStrlen(const wchar_t * ucs1)142 static inline size_t UniStrlen(const wchar_t *ucs1)
143 {
144 int i = 0;
145
146 while (*ucs1++)
147 i++;
148 return i;
149 }
150
151 /*
152 * UniStrnlen: Return the length (in 16 bit Unicode chars not bytes) of a
153 * string (length limited)
154 */
UniStrnlen(const wchar_t * ucs1,int maxlen)155 static inline size_t UniStrnlen(const wchar_t *ucs1, int maxlen)
156 {
157 int i = 0;
158
159 while (*ucs1++) {
160 i++;
161 if (i >= maxlen)
162 break;
163 }
164 return i;
165 }
166
167 /*
168 * UniStrncat: Concatenate length limited string
169 */
UniStrncat(wchar_t * ucs1,const wchar_t * ucs2,size_t n)170 static inline wchar_t *UniStrncat(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
171 {
172 wchar_t *anchor = ucs1; /* save pointer to string 1 */
173
174 while (*ucs1++)
175 /*NULL*/;
176 ucs1--; /* point to null terminator of s1 */
177 while (n-- && (*ucs1 = *ucs2)) { /* copy s2 after s1 */
178 ucs1++;
179 ucs2++;
180 }
181 *ucs1 = 0; /* Null terminate the result */
182 return anchor;
183 }
184
185 /*
186 * UniStrncmp: Compare length limited string
187 */
UniStrncmp(const wchar_t * ucs1,const wchar_t * ucs2,size_t n)188 static inline int UniStrncmp(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
189 {
190 if (!n)
191 return 0; /* Null strings are equal */
192 while ((*ucs1 == *ucs2) && *ucs1 && --n) {
193 ucs1++;
194 ucs2++;
195 }
196 return (int)*ucs1 - (int)*ucs2;
197 }
198
199 /*
200 * UniStrncmp_le: Compare length limited string - native to little-endian
201 */
202 static inline int
UniStrncmp_le(const wchar_t * ucs1,const wchar_t * ucs2,size_t n)203 UniStrncmp_le(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
204 {
205 if (!n)
206 return 0; /* Null strings are equal */
207 while ((*ucs1 == __le16_to_cpu(*ucs2)) && *ucs1 && --n) {
208 ucs1++;
209 ucs2++;
210 }
211 return (int)*ucs1 - (int)__le16_to_cpu(*ucs2);
212 }
213
214 /*
215 * UniStrncpy: Copy length limited string with pad
216 */
UniStrncpy(wchar_t * ucs1,const wchar_t * ucs2,size_t n)217 static inline wchar_t *UniStrncpy(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
218 {
219 wchar_t *anchor = ucs1;
220
221 while (n-- && *ucs2) /* Copy the strings */
222 *ucs1++ = *ucs2++;
223
224 n++;
225 while (n--) /* Pad with nulls */
226 *ucs1++ = 0;
227 return anchor;
228 }
229
230 /*
231 * UniStrncpy_le: Copy length limited string with pad to little-endian
232 */
UniStrncpy_le(wchar_t * ucs1,const wchar_t * ucs2,size_t n)233 static inline wchar_t *UniStrncpy_le(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
234 {
235 wchar_t *anchor = ucs1;
236
237 while (n-- && *ucs2) /* Copy the strings */
238 *ucs1++ = __le16_to_cpu(*ucs2++);
239
240 n++;
241 while (n--) /* Pad with nulls */
242 *ucs1++ = 0;
243 return anchor;
244 }
245
246 /*
247 * UniStrstr: Find a string in a string
248 *
249 * Returns:
250 * Address of first match found
251 * NULL if no matching string is found
252 */
UniStrstr(const wchar_t * ucs1,const wchar_t * ucs2)253 static inline wchar_t *UniStrstr(const wchar_t *ucs1, const wchar_t *ucs2)
254 {
255 const wchar_t *anchor1 = ucs1;
256 const wchar_t *anchor2 = ucs2;
257
258 while (*ucs1) {
259 if (*ucs1 == *ucs2) {
260 /* Partial match found */
261 ucs1++;
262 ucs2++;
263 } else {
264 if (!*ucs2) /* Match found */
265 return (wchar_t *)anchor1;
266 ucs1 = ++anchor1; /* No match */
267 ucs2 = anchor2;
268 }
269 }
270
271 if (!*ucs2) /* Both end together */
272 return (wchar_t *)anchor1; /* Match found */
273 return NULL; /* No match */
274 }
275
276 #ifndef UNIUPR_NOUPPER
277 /*
278 * UniToupper: Convert a unicode character to upper case
279 */
UniToupper(register wchar_t uc)280 static inline wchar_t UniToupper(register wchar_t uc)
281 {
282 register const struct UniCaseRange *rp;
283
284 if (uc < sizeof(SmbUniUpperTable)) {
285 /* Latin characters */
286 return uc + SmbUniUpperTable[uc]; /* Use base tables */
287 }
288
289 rp = SmbUniUpperRange; /* Use range tables */
290 while (rp->start) {
291 if (uc < rp->start) /* Before start of range */
292 return uc; /* Uppercase = input */
293 if (uc <= rp->end) /* In range */
294 return uc + rp->table[uc - rp->start];
295 rp++; /* Try next range */
296 }
297 return uc; /* Past last range */
298 }
299
300 /*
301 * UniStrupr: Upper case a unicode string
302 */
UniStrupr(register __le16 * upin)303 static inline __le16 *UniStrupr(register __le16 *upin)
304 {
305 register __le16 *up;
306
307 up = upin;
308 while (*up) { /* For all characters */
309 *up = cpu_to_le16(UniToupper(le16_to_cpu(*up)));
310 up++;
311 }
312 return upin; /* Return input pointer */
313 }
314 #endif /* UNIUPR_NOUPPER */
315
316 #ifndef UNIUPR_NOLOWER
317 /*
318 * UniTolower: Convert a unicode character to lower case
319 */
UniTolower(register wchar_t uc)320 static inline wchar_t UniTolower(register wchar_t uc)
321 {
322 register const struct UniCaseRange *rp;
323
324 if (uc < sizeof(CifsUniLowerTable)) {
325 /* Latin characters */
326 return uc + CifsUniLowerTable[uc]; /* Use base tables */
327 }
328
329 rp = CifsUniLowerRange; /* Use range tables */
330 while (rp->start) {
331 if (uc < rp->start) /* Before start of range */
332 return uc; /* Uppercase = input */
333 if (uc <= rp->end) /* In range */
334 return uc + rp->table[uc - rp->start];
335 rp++; /* Try next range */
336 }
337 return uc; /* Past last range */
338 }
339
340 /*
341 * UniStrlwr: Lower case a unicode string
342 */
UniStrlwr(register wchar_t * upin)343 static inline wchar_t *UniStrlwr(register wchar_t *upin)
344 {
345 register wchar_t *up;
346
347 up = upin;
348 while (*up) { /* For all characters */
349 *up = UniTolower(*up);
350 up++;
351 }
352 return upin; /* Return input pointer */
353 }
354
355 #endif
356
357 #endif /* _CIFS_UNICODE_H */
358