xref: /openbmc/linux/fs/smb/server/unicode.h (revision 404e077a)
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 #include <linux/unicode.h>
28 
29 #define  UNIUPR_NOLOWER		/* Example to not expand lower case tables */
30 
31 /*
32  * Windows maps these to the user defined 16 bit Unicode range since they are
33  * reserved symbols (along with \ and /), otherwise illegal to store
34  * in filenames in NTFS
35  */
36 #define UNI_ASTERISK    ((__u16)('*' + 0xF000))
37 #define UNI_QUESTION    ((__u16)('?' + 0xF000))
38 #define UNI_COLON       ((__u16)(':' + 0xF000))
39 #define UNI_GRTRTHAN    ((__u16)('>' + 0xF000))
40 #define UNI_LESSTHAN    ((__u16)('<' + 0xF000))
41 #define UNI_PIPE        ((__u16)('|' + 0xF000))
42 #define UNI_SLASH       ((__u16)('\\' + 0xF000))
43 
44 /* Just define what we want from uniupr.h.  We don't want to define the tables
45  * in each source file.
46  */
47 #ifndef	UNICASERANGE_DEFINED
48 struct UniCaseRange {
49 	wchar_t start;
50 	wchar_t end;
51 	signed char *table;
52 };
53 #endif				/* UNICASERANGE_DEFINED */
54 
55 #ifndef UNIUPR_NOUPPER
56 extern signed char SmbUniUpperTable[512];
57 extern const struct UniCaseRange SmbUniUpperRange[];
58 #endif				/* UNIUPR_NOUPPER */
59 
60 #ifndef UNIUPR_NOLOWER
61 extern signed char CifsUniLowerTable[512];
62 extern const struct UniCaseRange CifsUniLowerRange[];
63 #endif				/* UNIUPR_NOLOWER */
64 
65 #ifdef __KERNEL__
66 int smb_strtoUTF16(__le16 *to, const char *from, int len,
67 		   const struct nls_table *codepage);
68 char *smb_strndup_from_utf16(const char *src, const int maxlen,
69 			     const bool is_unicode,
70 			     const struct nls_table *codepage);
71 int smbConvertToUTF16(__le16 *target, const char *source, int srclen,
72 		      const struct nls_table *cp, int mapchars);
73 char *ksmbd_extract_sharename(struct unicode_map *um, const char *treename);
74 #endif
75 
76 /*
77  * UniStrcat:  Concatenate the second string to the first
78  *
79  * Returns:
80  *     Address of the first string
81  */
82 static inline wchar_t *UniStrcat(wchar_t *ucs1, const wchar_t *ucs2)
83 {
84 	wchar_t *anchor = ucs1;	/* save a pointer to start of ucs1 */
85 
86 	while (*ucs1++)
87 	/*NULL*/;	/* To end of first string */
88 	ucs1--;			/* Return to the null */
89 	while ((*ucs1++ = *ucs2++))
90 	/*NULL*/;	/* copy string 2 over */
91 	return anchor;
92 }
93 
94 /*
95  * UniStrchr:  Find a character in a string
96  *
97  * Returns:
98  *     Address of first occurrence of character in string
99  *     or NULL if the character is not in the string
100  */
101 static inline wchar_t *UniStrchr(const wchar_t *ucs, wchar_t uc)
102 {
103 	while ((*ucs != uc) && *ucs)
104 		ucs++;
105 
106 	if (*ucs == uc)
107 		return (wchar_t *)ucs;
108 	return NULL;
109 }
110 
111 /*
112  * UniStrcmp:  Compare two strings
113  *
114  * Returns:
115  *     < 0:  First string is less than second
116  *     = 0:  Strings are equal
117  *     > 0:  First string is greater than second
118  */
119 static inline int UniStrcmp(const wchar_t *ucs1, const wchar_t *ucs2)
120 {
121 	while ((*ucs1 == *ucs2) && *ucs1) {
122 		ucs1++;
123 		ucs2++;
124 	}
125 	return (int)*ucs1 - (int)*ucs2;
126 }
127 
128 /*
129  * UniStrcpy:  Copy a string
130  */
131 static inline wchar_t *UniStrcpy(wchar_t *ucs1, const wchar_t *ucs2)
132 {
133 	wchar_t *anchor = ucs1;	/* save the start of result string */
134 
135 	while ((*ucs1++ = *ucs2++))
136 	/*NULL*/;
137 	return anchor;
138 }
139 
140 /*
141  * UniStrlen:  Return the length of a string (in 16 bit Unicode chars not bytes)
142  */
143 static inline size_t UniStrlen(const wchar_t *ucs1)
144 {
145 	int i = 0;
146 
147 	while (*ucs1++)
148 		i++;
149 	return i;
150 }
151 
152 /*
153  * UniStrnlen:  Return the length (in 16 bit Unicode chars not bytes) of a
154  *		string (length limited)
155  */
156 static inline size_t UniStrnlen(const wchar_t *ucs1, int maxlen)
157 {
158 	int i = 0;
159 
160 	while (*ucs1++) {
161 		i++;
162 		if (i >= maxlen)
163 			break;
164 	}
165 	return i;
166 }
167 
168 /*
169  * UniStrncat:  Concatenate length limited string
170  */
171 static inline wchar_t *UniStrncat(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
172 {
173 	wchar_t *anchor = ucs1;	/* save pointer to string 1 */
174 
175 	while (*ucs1++)
176 	/*NULL*/;
177 	ucs1--;			/* point to null terminator of s1 */
178 	while (n-- && (*ucs1 = *ucs2)) {	/* copy s2 after s1 */
179 		ucs1++;
180 		ucs2++;
181 	}
182 	*ucs1 = 0;		/* Null terminate the result */
183 	return anchor;
184 }
185 
186 /*
187  * UniStrncmp:  Compare length limited string
188  */
189 static inline int UniStrncmp(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
190 {
191 	if (!n)
192 		return 0;	/* Null strings are equal */
193 	while ((*ucs1 == *ucs2) && *ucs1 && --n) {
194 		ucs1++;
195 		ucs2++;
196 	}
197 	return (int)*ucs1 - (int)*ucs2;
198 }
199 
200 /*
201  * UniStrncmp_le:  Compare length limited string - native to little-endian
202  */
203 static inline int
204 UniStrncmp_le(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
205 {
206 	if (!n)
207 		return 0;	/* Null strings are equal */
208 	while ((*ucs1 == __le16_to_cpu(*ucs2)) && *ucs1 && --n) {
209 		ucs1++;
210 		ucs2++;
211 	}
212 	return (int)*ucs1 - (int)__le16_to_cpu(*ucs2);
213 }
214 
215 /*
216  * UniStrncpy:  Copy length limited string with pad
217  */
218 static inline wchar_t *UniStrncpy(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
219 {
220 	wchar_t *anchor = ucs1;
221 
222 	while (n-- && *ucs2)	/* Copy the strings */
223 		*ucs1++ = *ucs2++;
224 
225 	n++;
226 	while (n--)		/* Pad with nulls */
227 		*ucs1++ = 0;
228 	return anchor;
229 }
230 
231 /*
232  * UniStrncpy_le:  Copy length limited string with pad to little-endian
233  */
234 static inline wchar_t *UniStrncpy_le(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
235 {
236 	wchar_t *anchor = ucs1;
237 
238 	while (n-- && *ucs2)	/* Copy the strings */
239 		*ucs1++ = __le16_to_cpu(*ucs2++);
240 
241 	n++;
242 	while (n--)		/* Pad with nulls */
243 		*ucs1++ = 0;
244 	return anchor;
245 }
246 
247 /*
248  * UniStrstr:  Find a string in a string
249  *
250  * Returns:
251  *     Address of first match found
252  *     NULL if no matching string is found
253  */
254 static inline wchar_t *UniStrstr(const wchar_t *ucs1, const wchar_t *ucs2)
255 {
256 	const wchar_t *anchor1 = ucs1;
257 	const wchar_t *anchor2 = ucs2;
258 
259 	while (*ucs1) {
260 		if (*ucs1 == *ucs2) {
261 			/* Partial match found */
262 			ucs1++;
263 			ucs2++;
264 		} else {
265 			if (!*ucs2)	/* Match found */
266 				return (wchar_t *)anchor1;
267 			ucs1 = ++anchor1;	/* No match */
268 			ucs2 = anchor2;
269 		}
270 	}
271 
272 	if (!*ucs2)		/* Both end together */
273 		return (wchar_t *)anchor1;	/* Match found */
274 	return NULL;		/* No match */
275 }
276 
277 #ifndef UNIUPR_NOUPPER
278 /*
279  * UniToupper:  Convert a unicode character to upper case
280  */
281 static inline wchar_t UniToupper(register wchar_t uc)
282 {
283 	register const struct UniCaseRange *rp;
284 
285 	if (uc < sizeof(SmbUniUpperTable)) {
286 		/* Latin characters */
287 		return uc + SmbUniUpperTable[uc];	/* Use base tables */
288 	}
289 
290 	rp = SmbUniUpperRange;	/* Use range tables */
291 	while (rp->start) {
292 		if (uc < rp->start)	/* Before start of range */
293 			return uc;	/* Uppercase = input */
294 		if (uc <= rp->end)	/* In range */
295 			return uc + rp->table[uc - rp->start];
296 		rp++;	/* Try next range */
297 	}
298 	return uc;		/* Past last range */
299 }
300 
301 /*
302  * UniStrupr:  Upper case a unicode string
303  */
304 static inline __le16 *UniStrupr(register __le16 *upin)
305 {
306 	register __le16 *up;
307 
308 	up = upin;
309 	while (*up) {		/* For all characters */
310 		*up = cpu_to_le16(UniToupper(le16_to_cpu(*up)));
311 		up++;
312 	}
313 	return upin;		/* Return input pointer */
314 }
315 #endif				/* UNIUPR_NOUPPER */
316 
317 #ifndef UNIUPR_NOLOWER
318 /*
319  * UniTolower:  Convert a unicode character to lower case
320  */
321 static inline wchar_t UniTolower(register wchar_t uc)
322 {
323 	register const struct UniCaseRange *rp;
324 
325 	if (uc < sizeof(CifsUniLowerTable)) {
326 		/* Latin characters */
327 		return uc + CifsUniLowerTable[uc];	/* Use base tables */
328 	}
329 
330 	rp = CifsUniLowerRange;	/* Use range tables */
331 	while (rp->start) {
332 		if (uc < rp->start)	/* Before start of range */
333 			return uc;	/* Uppercase = input */
334 		if (uc <= rp->end)	/* In range */
335 			return uc + rp->table[uc - rp->start];
336 		rp++;	/* Try next range */
337 	}
338 	return uc;		/* Past last range */
339 }
340 
341 /*
342  * UniStrlwr:  Lower case a unicode string
343  */
344 static inline wchar_t *UniStrlwr(register wchar_t *upin)
345 {
346 	register wchar_t *up;
347 
348 	up = upin;
349 	while (*up) {		/* For all characters */
350 		*up = UniTolower(*up);
351 		up++;
352 	}
353 	return upin;		/* Return input pointer */
354 }
355 
356 #endif
357 
358 #endif /* _CIFS_UNICODE_H */
359