]> git.saurik.com Git - apple/icu.git/blob - icuSources/common/unicode/uidna.h
ICU-8.11.4.tar.gz
[apple/icu.git] / icuSources / common / unicode / uidna.h
1 /*
2 *******************************************************************************
3 *
4 * Copyright (C) 2003-2006, 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 Number of ASCII characters converted.
111 * @stable ICU 2.6
112 */
113 U_STABLE int32_t U_EXPORT2
114 uidna_toASCII(const UChar* src, int32_t srcLength,
115 UChar* dest, int32_t destCapacity,
116 int32_t options,
117 UParseError* parseError,
118 UErrorCode* status);
119
120
121 /**
122 * This function implements the ToUnicode operation as defined in the IDNA RFC.
123 * This operation is done on <b>single labels</b> before sending it to something that expects
124 * Unicode names. A label is an individual part of a domain name. Labels are usually
125 * separated by dots; for e.g." "www.example.com" is composed of 3 labels
126 * "www","example", and "com".
127 *
128 * @param src Input UChar array containing ASCII (ACE encoded) label.
129 * @param srcLength Number of UChars in src, or -1 if NUL-terminated.
130 * @param dest Output Converted UChar array containing Unicode equivalent of label.
131 * @param destCapacity Size of dest.
132 * @param options A bit set of options:
133 *
134 * - UIDNA_DEFAULT Use default options, i.e., do not process unassigned code points
135 * and do not use STD3 ASCII rules
136 * If unassigned code points are found the operation fails with
137 * U_UNASSIGNED_ERROR error code.
138 *
139 * - UIDNA_ALLOW_UNASSIGNED Unassigned values can be converted to ASCII for query operations
140 * If this option is set, the unassigned code points are in the input
141 * are treated as normal Unicode code points. <b> Note: </b> This option is
142 * required on toUnicode operation because the RFC mandates
143 * verification of decoded ACE input by applying toASCII and comparing
144 * its output with source
145 *
146 *
147 *
148 * - UIDNA_USE_STD3_RULES Use STD3 ASCII rules for host name syntax restrictions
149 * If this option is set and the input does not satisfy STD3 rules,
150 * the operation will fail with U_IDNA_STD3_ASCII_RULES_ERROR
151 *
152 * @param parseError Pointer to UParseError struct to receive information on position
153 * of error if an error is encountered. Can be NULL.
154 * @param status ICU in/out error code parameter.
155 * U_INVALID_CHAR_FOUND if src contains
156 * unmatched single surrogates.
157 * U_INDEX_OUTOFBOUNDS_ERROR if src contains
158 * too many code points.
159 * U_BUFFER_OVERFLOW_ERROR if destCapacity is not enough
160 * @return Number of Unicode characters converted.
161 * @stable ICU 2.6
162 */
163 U_STABLE int32_t U_EXPORT2
164 uidna_toUnicode(const UChar* src, int32_t srcLength,
165 UChar* dest, int32_t destCapacity,
166 int32_t options,
167 UParseError* parseError,
168 UErrorCode* status);
169
170
171 /**
172 * Convenience function that implements the IDNToASCII operation as defined in the IDNA RFC.
173 * This operation is done on complete domain names, e.g: "www.example.com".
174 * It is important to note that this operation can fail. If it fails, then the input
175 * domain name cannot be used as an Internationalized Domain Name and the application
176 * should have methods defined to deal with the failure.
177 *
178 * <b>Note:</b> IDNA RFC specifies that a conformant application should divide a domain name
179 * into separate labels, decide whether to apply allowUnassigned and useSTD3ASCIIRules on each,
180 * and then convert. This function does not offer that level of granularity. The options once
181 * set will apply to all labels in the domain name
182 *
183 * @param src Input UChar array containing IDN in Unicode.
184 * @param srcLength Number of UChars in src, or -1 if NUL-terminated.
185 * @param dest Output UChar array with ASCII (ACE encoded) IDN.
186 * @param destCapacity Size of dest.
187 * @param options A bit set of options:
188 *
189 * - UIDNA_DEFAULT Use default options, i.e., do not process unassigned code points
190 * and do not use STD3 ASCII rules
191 * If unassigned code points are found the operation fails with
192 * U_UNASSIGNED_CODE_POINT_FOUND error code.
193 *
194 * - UIDNA_ALLOW_UNASSIGNED Unassigned values can be converted to ASCII for query operations
195 * If this option is set, the unassigned code points are in the input
196 * are treated as normal Unicode code points.
197 *
198 * - UIDNA_USE_STD3_RULES Use STD3 ASCII rules for host name syntax restrictions
199 * If this option is set and the input does not satisfy STD3 rules,
200 * the operation will fail with U_IDNA_STD3_ASCII_RULES_ERROR
201 *
202 * @param parseError Pointer to UParseError struct to receive information on position
203 * of error if an error is encountered. Can be NULL.
204 * @param status ICU in/out error code parameter.
205 * U_INVALID_CHAR_FOUND if src contains
206 * unmatched single surrogates.
207 * U_INDEX_OUTOFBOUNDS_ERROR if src contains
208 * too many code points.
209 * U_BUFFER_OVERFLOW_ERROR if destCapacity is not enough
210 * @return Number of ASCII characters converted.
211 * @stable ICU 2.6
212 */
213 U_STABLE int32_t U_EXPORT2
214 uidna_IDNToASCII( const UChar* src, int32_t srcLength,
215 UChar* dest, int32_t destCapacity,
216 int32_t options,
217 UParseError* parseError,
218 UErrorCode* status);
219
220 /**
221 * Convenience function that implements the IDNToUnicode operation as defined in the IDNA RFC.
222 * This operation is done on complete domain names, e.g: "www.example.com".
223 *
224 * <b>Note:</b> IDNA RFC specifies that a conformant application should divide a domain name
225 * into separate labels, decide whether to apply allowUnassigned and useSTD3ASCIIRules on each,
226 * and then convert. This function does not offer that level of granularity. The options once
227 * set will apply to all labels in the domain name
228 *
229 * @param src Input UChar array containing IDN in ASCII (ACE encoded) form.
230 * @param srcLength Number of UChars in src, or -1 if NUL-terminated.
231 * @param dest Output UChar array containing Unicode equivalent of source IDN.
232 * @param destCapacity Size of dest.
233 * @param options A bit set of options:
234 *
235 * - UIDNA_DEFAULT Use default options, i.e., do not process unassigned code points
236 * and do not use STD3 ASCII rules
237 * If unassigned code points are found the operation fails with
238 * U_UNASSIGNED_CODE_POINT_FOUND error code.
239 *
240 * - UIDNA_ALLOW_UNASSIGNED Unassigned values can be converted to ASCII for query operations
241 * If this option is set, the unassigned code points are in the input
242 * are treated as normal Unicode code points.
243 *
244 * - UIDNA_USE_STD3_RULES Use STD3 ASCII rules for host name syntax restrictions
245 * If this option is set and the input does not satisfy STD3 rules,
246 * the operation will fail with U_IDNA_STD3_ASCII_RULES_ERROR
247 *
248 * @param parseError Pointer to UParseError struct to receive information on position
249 * of error if an error is encountered. Can be NULL.
250 * @param status ICU in/out error code parameter.
251 * U_INVALID_CHAR_FOUND if src contains
252 * unmatched single surrogates.
253 * U_INDEX_OUTOFBOUNDS_ERROR if src contains
254 * too many code points.
255 * U_BUFFER_OVERFLOW_ERROR if destCapacity is not enough
256 * @return Number of ASCII characters converted.
257 * @stable ICU 2.6
258 */
259 U_STABLE int32_t U_EXPORT2
260 uidna_IDNToUnicode( const UChar* src, int32_t srcLength,
261 UChar* dest, int32_t destCapacity,
262 int32_t options,
263 UParseError* parseError,
264 UErrorCode* status);
265
266 /**
267 * Compare two IDN strings for equivalence.
268 * This function splits the domain names into labels and compares them.
269 * According to IDN RFC, whenever two labels are compared, they are
270 * considered equal if and only if their ASCII forms (obtained by
271 * applying toASCII) match using an case-insensitive ASCII comparison.
272 * Two domain names are considered a match if and only if all labels
273 * match regardless of whether label separators match.
274 *
275 * @param s1 First source string.
276 * @param length1 Length of first source string, or -1 if NUL-terminated.
277 *
278 * @param s2 Second source string.
279 * @param length2 Length of second source string, or -1 if NUL-terminated.
280 * @param options A bit set of options:
281 *
282 * - UIDNA_DEFAULT Use default options, i.e., do not process unassigned code points
283 * and do not use STD3 ASCII rules
284 * If unassigned code points are found the operation fails with
285 * U_UNASSIGNED_CODE_POINT_FOUND error code.
286 *
287 * - UIDNA_ALLOW_UNASSIGNED Unassigned values can be converted to ASCII for query operations
288 * If this option is set, the unassigned code points are in the input
289 * are treated as normal Unicode code points.
290 *
291 * - UIDNA_USE_STD3_RULES Use STD3 ASCII rules for host name syntax restrictions
292 * If this option is set and the input does not satisfy STD3 rules,
293 * the operation will fail with U_IDNA_STD3_ASCII_RULES_ERROR
294 *
295 * @param status ICU error code in/out parameter.
296 * Must fulfill U_SUCCESS before the function call.
297 * @return <0 or 0 or >0 as usual for string comparisons
298 * @stable ICU 2.6
299 */
300 U_STABLE int32_t U_EXPORT2
301 uidna_compare( const UChar *s1, int32_t length1,
302 const UChar *s2, int32_t length2,
303 int32_t options,
304 UErrorCode* status);
305
306 #endif /* #if !UCONFIG_NO_IDNA */
307
308 #endif