- support of [Strings.LanguageID]-sections for inf-files added in setupapi
[reactos.git] / reactos / nls / 3rdparty / icu / source / common / unicode / uidna.h
1 /*
2 *******************************************************************************
3 *
4 * Copyright (C) 2003-2007, International Business Machines
5 * Corporation and others. All Rights Reserved.
6 *
7 *******************************************************************************
8 * file name: uidna.h
9 * encoding: US-ASCII
10 * tab size: 8 (not used)
11 * indentation:4
12 *
13 * created on: 2003feb1
14 * created by: Ram Viswanadha
15 */
16
17 #ifndef __UIDNA_H__
18 #define __UIDNA_H__
19
20 #include "unicode/utypes.h"
21
22 #if !UCONFIG_NO_IDNA
23
24 #include "unicode/parseerr.h"
25
26 /**
27 * \file
28 * \brief C API: Internationalized Domain Names in Applications Tranformation
29 *
30 * UIDNA API implements the IDNA protocol as defined in the IDNA RFC
31 * (http://www.ietf.org/rfc/rfc3490.txt).
32 * The RFC defines 2 operations: ToASCII and ToUnicode. Domain labels
33 * containing non-ASCII code points are required to be processed by
34 * ToASCII operation before passing it to resolver libraries. Domain names
35 * that are obtained from resolver libraries are required to be processed by
36 * ToUnicode operation before displaying the domain name to the user.
37 * IDNA requires that implementations process input strings with Nameprep
38 * (http://www.ietf.org/rfc/rfc3491.txt),
39 * which is a profile of Stringprep (http://www.ietf.org/rfc/rfc3454.txt),
40 * and then with Punycode (http://www.ietf.org/rfc/rfc3492.txt).
41 * Implementations of IDNA MUST fully implement Nameprep and Punycode;
42 * neither Nameprep nor Punycode are optional.
43 * The input and output of ToASCII and ToUnicode operations are Unicode
44 * and are designed to be chainable, i.e., applying ToASCII or ToUnicode operations
45 * multiple times to an input string will yield the same result as applying the operation
46 * once.
47 * ToUnicode(ToUnicode(ToUnicode...(ToUnicode(string)))) == ToUnicode(string)
48 * ToASCII(ToASCII(ToASCII...(ToASCII(string))) == ToASCII(string).
49 *
50 */
51
52 /**
53 * Option to prohibit processing of unassigned codepoints in the input and
54 * do not check if the input conforms to STD-3 ASCII rules.
55 *
56 * @see uidna_toASCII uidna_toUnicode
57 * @stable ICU 2.6
58 */
59 #define UIDNA_DEFAULT 0x0000
60 /**
61 * Option to allow processing of unassigned codepoints in the input
62 *
63 * @see uidna_toASCII uidna_toUnicode
64 * @stable ICU 2.6
65 */
66 #define UIDNA_ALLOW_UNASSIGNED 0x0001
67 /**
68 * Option to check if input conforms to STD-3 ASCII rules
69 *
70 * @see uidna_toASCII uidna_toUnicode
71 * @stable ICU 2.6
72 */
73 #define UIDNA_USE_STD3_RULES 0x0002
74
75 /**
76 * This function implements the ToASCII operation as defined in the IDNA RFC.
77 * This operation is done on <b>single labels</b> before sending it to something that expects
78 * ASCII names. A label is an individual part of a domain name. Labels are usually
79 * separated by dots; e.g." "www.example.com" is composed of 3 labels
80 * "www","example", and "com".
81 *
82 *
83 * @param src Input UChar array containing label in Unicode.
84 * @param srcLength Number of UChars in src, or -1 if NUL-terminated.
85 * @param dest Output UChar array with ASCII (ACE encoded) label.
86 * @param destCapacity Size of dest.
87 * @param options A bit set of options:
88 *
89 * - UIDNA_DEFAULT Use default options, i.e., do not process unassigned code points
90 * and do not use STD3 ASCII rules
91 * If unassigned code points are found the operation fails with
92 * U_UNASSIGNED_ERROR error code.
93 *
94 * - UIDNA_ALLOW_UNASSIGNED Unassigned values can be converted to ASCII for query operations
95 * If this option is set, the unassigned code points are in the input
96 * are treated as normal Unicode code points.
97 *
98 * - UIDNA_USE_STD3_RULES Use STD3 ASCII rules for host name syntax restrictions
99 * If this option is set and the input does not satisfy STD3 rules,
100 * the operation will fail with U_IDNA_STD3_ASCII_RULES_ERROR
101 *
102 * @param parseError Pointer to UParseError struct to receive information on position
103 * of error if an error is encountered. Can be NULL.
104 * @param status ICU in/out error code parameter.
105 * U_INVALID_CHAR_FOUND if src contains
106 * unmatched single surrogates.
107 * U_INDEX_OUTOFBOUNDS_ERROR if src contains
108 * too many code points.
109 * U_BUFFER_OVERFLOW_ERROR if destCapacity is not enough
110 * @return The length of the result string, if successful - or in case of a buffer overflow,
111 * in which case it will be greater than destCapacity.
112 * @stable ICU 2.6
113 */
114 U_STABLE int32_t U_EXPORT2
115 uidna_toASCII(const UChar* src, int32_t srcLength,
116 UChar* dest, int32_t destCapacity,
117 int32_t options,
118 UParseError* parseError,
119 UErrorCode* status);
120
121
122 /**
123 * This function implements the ToUnicode operation as defined in the IDNA RFC.
124 * This operation is done on <b>single labels</b> before sending it to something that expects
125 * Unicode names. A label is an individual part of a domain name. Labels are usually
126 * separated by dots; for e.g." "www.example.com" is composed of 3 labels
127 * "www","example", and "com".
128 *
129 * @param src Input UChar array containing ASCII (ACE encoded) label.
130 * @param srcLength Number of UChars in src, or -1 if NUL-terminated.
131 * @param dest Output Converted UChar array containing Unicode equivalent of label.
132 * @param destCapacity Size of dest.
133 * @param options A bit set of options:
134 *
135 * - UIDNA_DEFAULT Use default options, i.e., do not process unassigned code points
136 * and do not use STD3 ASCII rules
137 * If unassigned code points are found the operation fails with
138 * U_UNASSIGNED_ERROR error code.
139 *
140 * - UIDNA_ALLOW_UNASSIGNED Unassigned values can be converted to ASCII for query operations
141 * If this option is set, the unassigned code points are in the input
142 * are treated as normal Unicode code points. <b> Note: </b> This option is
143 * required on toUnicode operation because the RFC mandates
144 * verification of decoded ACE input by applying toASCII and comparing
145 * its output with source
146 *
147 *
148 *
149 * - UIDNA_USE_STD3_RULES Use STD3 ASCII rules for host name syntax restrictions
150 * If this option is set and the input does not satisfy STD3 rules,
151 * the operation will fail with U_IDNA_STD3_ASCII_RULES_ERROR
152 *
153 * @param parseError Pointer to UParseError struct to receive information on position
154 * of error if an error is encountered. Can be NULL.
155 * @param status ICU in/out error code parameter.
156 * U_INVALID_CHAR_FOUND if src contains
157 * unmatched single surrogates.
158 * U_INDEX_OUTOFBOUNDS_ERROR if src contains
159 * too many code points.
160 * U_BUFFER_OVERFLOW_ERROR if destCapacity is not enough
161 * @return The length of the result string, if successful - or in case of a buffer overflow,
162 * in which case it will be greater than destCapacity.
163 * @stable ICU 2.6
164 */
165 U_STABLE int32_t U_EXPORT2
166 uidna_toUnicode(const UChar* src, int32_t srcLength,
167 UChar* dest, int32_t destCapacity,
168 int32_t options,
169 UParseError* parseError,
170 UErrorCode* status);
171
172
173 /**
174 * Convenience function that implements the IDNToASCII operation as defined in the IDNA RFC.
175 * This operation is done on complete domain names, e.g: "www.example.com".
176 * It is important to note that this operation can fail. If it fails, then the input
177 * domain name cannot be used as an Internationalized Domain Name and the application
178 * should have methods defined to deal with the failure.
179 *
180 * <b>Note:</b> IDNA RFC specifies that a conformant application should divide a domain name
181 * into separate labels, decide whether to apply allowUnassigned and useSTD3ASCIIRules on each,
182 * and then convert. This function does not offer that level of granularity. The options once
183 * set will apply to all labels in the domain name
184 *
185 * @param src Input UChar array containing IDN in Unicode.
186 * @param srcLength Number of UChars in src, or -1 if NUL-terminated.
187 * @param dest Output UChar array with ASCII (ACE encoded) IDN.
188 * @param destCapacity Size of dest.
189 * @param options A bit set of options:
190 *
191 * - UIDNA_DEFAULT Use default options, i.e., do not process unassigned code points
192 * and do not use STD3 ASCII rules
193 * If unassigned code points are found the operation fails with
194 * U_UNASSIGNED_CODE_POINT_FOUND error code.
195 *
196 * - UIDNA_ALLOW_UNASSIGNED Unassigned values can be converted to ASCII for query operations
197 * If this option is set, the unassigned code points are in the input
198 * are treated as normal Unicode code points.
199 *
200 * - UIDNA_USE_STD3_RULES Use STD3 ASCII rules for host name syntax restrictions
201 * If this option is set and the input does not satisfy STD3 rules,
202 * the operation will fail with U_IDNA_STD3_ASCII_RULES_ERROR
203 *
204 * @param parseError Pointer to UParseError struct to receive information on position
205 * of error if an error is encountered. Can be NULL.
206 * @param status ICU in/out error code parameter.
207 * U_INVALID_CHAR_FOUND if src contains
208 * unmatched single surrogates.
209 * U_INDEX_OUTOFBOUNDS_ERROR if src contains
210 * too many code points.
211 * U_BUFFER_OVERFLOW_ERROR if destCapacity is not enough
212 * @return The length of the result string, if successful - or in case of a buffer overflow,
213 * in which case it will be greater than destCapacity.
214 * @stable ICU 2.6
215 */
216 U_STABLE int32_t U_EXPORT2
217 uidna_IDNToASCII( const UChar* src, int32_t srcLength,
218 UChar* dest, int32_t destCapacity,
219 int32_t options,
220 UParseError* parseError,
221 UErrorCode* status);
222
223 /**
224 * Convenience function that implements the IDNToUnicode operation as defined in the IDNA RFC.
225 * This operation is done on complete domain names, e.g: "www.example.com".
226 *
227 * <b>Note:</b> IDNA RFC specifies that a conformant application should divide a domain name
228 * into separate labels, decide whether to apply allowUnassigned and useSTD3ASCIIRules on each,
229 * and then convert. This function does not offer that level of granularity. The options once
230 * set will apply to all labels in the domain name
231 *
232 * @param src Input UChar array containing IDN in ASCII (ACE encoded) form.
233 * @param srcLength Number of UChars in src, or -1 if NUL-terminated.
234 * @param dest Output UChar array containing Unicode equivalent of source IDN.
235 * @param destCapacity Size of dest.
236 * @param options A bit set of options:
237 *
238 * - UIDNA_DEFAULT Use default options, i.e., do not process unassigned code points
239 * and do not use STD3 ASCII rules
240 * If unassigned code points are found the operation fails with
241 * U_UNASSIGNED_CODE_POINT_FOUND error code.
242 *
243 * - UIDNA_ALLOW_UNASSIGNED Unassigned values can be converted to ASCII for query operations
244 * If this option is set, the unassigned code points are in the input
245 * are treated as normal Unicode code points.
246 *
247 * - UIDNA_USE_STD3_RULES Use STD3 ASCII rules for host name syntax restrictions
248 * If this option is set and the input does not satisfy STD3 rules,
249 * the operation will fail with U_IDNA_STD3_ASCII_RULES_ERROR
250 *
251 * @param parseError Pointer to UParseError struct to receive information on position
252 * of error if an error is encountered. Can be NULL.
253 * @param status ICU in/out error code parameter.
254 * U_INVALID_CHAR_FOUND if src contains
255 * unmatched single surrogates.
256 * U_INDEX_OUTOFBOUNDS_ERROR if src contains
257 * too many code points.
258 * U_BUFFER_OVERFLOW_ERROR if destCapacity is not enough
259 * @return The length of the result string, if successful - or in case of a buffer overflow,
260 * in which case it will be greater than destCapacity.
261 * @stable ICU 2.6
262 */
263 U_STABLE int32_t U_EXPORT2
264 uidna_IDNToUnicode( const UChar* src, int32_t srcLength,
265 UChar* dest, int32_t destCapacity,
266 int32_t options,
267 UParseError* parseError,
268 UErrorCode* status);
269
270 /**
271 * Compare two IDN strings for equivalence.
272 * This function splits the domain names into labels and compares them.
273 * According to IDN RFC, whenever two labels are compared, they are
274 * considered equal if and only if their ASCII forms (obtained by
275 * applying toASCII) match using an case-insensitive ASCII comparison.
276 * Two domain names are considered a match if and only if all labels
277 * match regardless of whether label separators match.
278 *
279 * @param s1 First source string.
280 * @param length1 Length of first source string, or -1 if NUL-terminated.
281 *
282 * @param s2 Second source string.
283 * @param length2 Length of second source string, or -1 if NUL-terminated.
284 * @param options A bit set of options:
285 *
286 * - UIDNA_DEFAULT Use default options, i.e., do not process unassigned code points
287 * and do not use STD3 ASCII rules
288 * If unassigned code points are found the operation fails with
289 * U_UNASSIGNED_CODE_POINT_FOUND error code.
290 *
291 * - UIDNA_ALLOW_UNASSIGNED Unassigned values can be converted to ASCII for query operations
292 * If this option is set, the unassigned code points are in the input
293 * are treated as normal Unicode code points.
294 *
295 * - UIDNA_USE_STD3_RULES Use STD3 ASCII rules for host name syntax restrictions
296 * If this option is set and the input does not satisfy STD3 rules,
297 * the operation will fail with U_IDNA_STD3_ASCII_RULES_ERROR
298 *
299 * @param status ICU error code in/out parameter.
300 * Must fulfill U_SUCCESS before the function call.
301 * @return <0 or 0 or >0 as usual for string comparisons
302 * @stable ICU 2.6
303 */
304 U_STABLE int32_t U_EXPORT2
305 uidna_compare( const UChar *s1, int32_t length1,
306 const UChar *s2, int32_t length2,
307 int32_t options,
308 UErrorCode* status);
309
310 #endif /* #if !UCONFIG_NO_IDNA */
311
312 #endif