1 /* idna.c Convert to or from IDN strings.
2 * Copyright (C) 2002 Simon Josefsson
4 * This file is part of Libstringprep.
6 * Libstringprep is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2.1 of the License, or (at your option) any later version.
11 * Libstringprep is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 * Lesser General Public License for more details.
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with Libstringprep; if not, write to the Free Software
18 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
26 * @in: input array with unicode code points.
27 * @inlen: length of input array with unicode code points.
28 * @out: output zero terminated string that must have room for at
29 * least 63 characters plus the terminating zero.
30 * @allowunassigned: boolean value as per IDNA specification.
31 * @usestd3asciirules: boolean value as per IDNA specification.
33 * The ToASCII operation takes a sequence of Unicode code points that make
34 * up one label and transforms it into a sequence of code points in the
35 * ASCII range (0..7F). If ToASCII succeeds, the original sequence and the
36 * resulting sequence are equivalent labels.
38 * It is important to note that the ToASCII operation can fail. ToASCII
39 * fails if any step of it fails. If any step of the ToASCII operation
40 * fails on any label in a domain name, that domain name MUST NOT be used
41 * as an internationalized domain name. The method for deadling with this
42 * failure is application-specific.
44 * The inputs to ToASCII are a sequence of code points, the AllowUnassigned
45 * flag, and the UseSTD3ASCIIRules flag. The output of ToASCII is either a
46 * sequence of ASCII code points or a failure condition.
48 * ToASCII never alters a sequence of code points that are all in the ASCII
49 * range to begin with (although it could fail). Applying the ToASCII
50 * operation multiple times has exactly the same effect as applying it just
53 * Return value: Returns 0 on success, or an error code.
56 idna_to_ascii (const unsigned long *in
, size_t inlen
,
57 char *out
, int allowunassigned
, int usestd3asciirules
)
63 src
= malloc (sizeof (in
[0]) * inlen
+ 1);
65 return IDNA_MALLOC_ERROR
;
67 memcpy (src
, in
, sizeof (in
[0]) * inlen
);
71 * ToASCII consists of the following steps:
73 * 1. If all code points in the sequence are in the ASCII range (0..7F)
74 * then skip to step 3.
82 for (i
= 0; src
[i
]; i
++)
90 * 2. Perform the steps specified in [NAMEPREP] and fail if there is
91 * an error. The AllowUnassigned flag is used in [NAMEPREP].
97 p
= stringprep_ucs4_to_utf8 (src
, inlen
, NULL
, NULL
);
99 return IDNA_MALLOC_ERROR
;
101 p
= realloc (p
, BUFSIZ
);
103 return IDNA_MALLOC_ERROR
;
106 rc
= stringprep_nameprep (p
, BUFSIZ
);
108 rc
= stringprep_nameprep_no_unassigned (p
, BUFSIZ
);
110 if (rc
!= STRINGPREP_OK
)
111 return IDNA_STRINGPREP_ERROR
;
115 src
= stringprep_utf8_to_ucs4 (p
, -1, NULL
);
120 * 3. If the UseSTD3ASCIIRules flag is set, then perform these checks:
122 * (a) Verify the absence of non-LDH ASCII code points; that is,
123 * the absence of 0..2C, 2E..2F, 3A..40, 5B..60, and 7B..7F.
125 * (b) Verify the absence of leading and trailing hyphen-minus;
126 * that is, the absence of U+002D at the beginning and end of
130 if (usestd3asciirules
)
134 for (i
= 0; src
[i
]; i
++)
135 if (src
[i
] <= 0x2C || src
[i
] == 0x2E || src
[i
] == 0x2F ||
136 (src
[i
] >= 0x3A && src
[i
] <= 0x40) ||
137 (src
[i
] >= 0x5B && src
[i
] <= 0x60) ||
138 (src
[i
] >= 0x7B && src
[i
] <= 0x7F))
139 return IDNA_CONTAINS_LDH
;
141 if (src
[0] == 0x002D || (i
> 0 && src
[i
- 1] == 0x002D))
142 return IDNA_CONTAINS_MINUS
;
146 * 4. If all code points in the sequence are in the ASCII range
147 * (0..7F), then skip to step 8.
155 for (i
= 0; src
[i
]; i
++)
163 * 5. Verify that the sequence does NOT begin with the ACE prefix.
172 * 6. Encode the sequence using the encoding algorithm in [PUNYCODE]
173 * and fail if there is an error.
175 for (len
= 0; src
[len
]; len
++)
178 outlen
= 63 - strlen (IDNA_ACE_PREFIX
);
179 rc
= punycode_encode (len
, src
, NULL
,
180 &outlen
, &out
[strlen (IDNA_ACE_PREFIX
)]);
181 if (rc
!= PUNYCODE_SUCCESS
)
182 return IDNA_PUNYCODE_ERROR
;
183 out
[strlen (IDNA_ACE_PREFIX
) + outlen
] = '\0';
186 * 7. Prepend the ACE prefix.
189 memcpy (out
, IDNA_ACE_PREFIX
, strlen (IDNA_ACE_PREFIX
));
192 * 8. Verify that the number of code points is in the range 1 to 63
197 if (strlen (out
) < 1 || strlen (out
) > 63)
198 return IDNA_INVALID_LENGTH
;
204 idna_to_unicode_internal (const unsigned long *in
, size_t inlen
,
205 unsigned long *out
, size_t * outlen
,
206 int allowunassigned
, int usestd3asciirules
,
207 char *utf8in
, size_t utf8len
)
213 * 1. If all code points in the sequence are in the ASCII range (0..7F)
214 * then skip to step 3.
222 for (i
= 0; in
[i
]; i
++)
230 * 2. Perform the steps specified in [NAMEPREP] and fail if there is an
231 * error. (If step 3 of ToASCII is also performed here, it will not
232 * affect the overall behavior of ToUnicode, but it is not
233 * necessary.) The AllowUnassigned flag is used in [NAMEPREP].
237 rc
= stringprep_nameprep (utf8in
, utf8len
);
239 rc
= stringprep_nameprep_no_unassigned (utf8in
, utf8len
);
241 if (rc
!= STRINGPREP_OK
)
242 return IDNA_STRINGPREP_ERROR
;
244 /* 3. Verify that the sequence begins with the ACE prefix, and save a
245 * copy of the sequence.
249 if (memcmp (IDNA_ACE_PREFIX
, utf8in
, strlen (IDNA_ACE_PREFIX
)) != 0)
250 return IDNA_NO_ACE_PREFIX
;
252 /* 4. Remove the ACE prefix.
255 memmove (utf8in
, &utf8in
[strlen (IDNA_ACE_PREFIX
)],
256 strlen (utf8in
) - strlen (IDNA_ACE_PREFIX
) + 1);
258 /* 5. Decode the sequence using the decoding algorithm in [PUNYCODE]
259 * and fail if there is an error. Save a copy of the result of
263 rc
= punycode_decode (strlen(utf8in
), utf8in
, outlen
, out
, NULL
);
264 if (rc
!= PUNYCODE_SUCCESS
)
265 return IDNA_PUNYCODE_ERROR
;
270 rc
= idna_to_ascii (out
, *outlen
, tmpout
,
271 allowunassigned
, usestd3asciirules
);
272 if (rc
!= IDNA_SUCCESS
)
275 /* 7. Verify that the result of step 6 matches the saved copy from
276 * step 3, using a case-insensitive ASCII comparison.
279 if (strcasecmp(utf8in
, tmpout
+ strlen(IDNA_ACE_PREFIX
)) != 0)
280 return IDNA_ROUNDTRIP_VERIFY_ERROR
;
282 /* 8. Return the saved copy from step 5.
290 * @in: input array with unicode code points.
291 * @inlen: length of input array with unicode code points.
292 * @out: output array with unicode code points.
293 * @outlen: on input, maximum size of output array with unicode code points,
294 * on exit, actual size of output array with unicode code points.
295 * @allowunassigned: boolean value as per IDNA specification.
296 * @usestd3asciirules: boolean value as per IDNA specification.
298 * The ToUnicode operation takes a sequence of Unicode code points
299 * that make up one label and returns a sequence of Unicode code
300 * points. If the input sequence is a label in ACE form, then the
301 * result is an equivalent internationalized label that is not in ACE
302 * form, otherwise the original sequence is returned unaltered.
304 * ToUnicode never fails. If any step fails, then the original input
305 * sequence is returned immediately in that step.
307 * The ToUnicode output never contains more code points than its
308 * input. Note that the number of octets needed to represent a
309 * sequence of code points depends on the particular character
312 * The inputs to ToUnicode are a sequence of code points, the
313 * AllowUnassigned flag, and the UseSTD3ASCIIRules flag. The output of
314 * ToUnicode is always a sequence of Unicode code points.
316 * Return value: Returns error condition, but it must only be used for
317 * debugging purposes. The output buffer is always
318 * guaranteed to contain the correct data according to
319 * the specification (sans malloc induced errors). NB!
320 * This means that you normally ignore the return code
321 * from this function, as checking it means breaking the
325 idna_to_unicode (const unsigned long *in
, size_t inlen
,
326 unsigned long *out
, size_t * outlen
,
327 int allowunassigned
, int usestd3asciirules
)
330 int outlensave
= *outlen
;
333 p
= stringprep_ucs4_to_utf8 (in
, inlen
, NULL
, NULL
);
335 return IDNA_MALLOC_ERROR
;
337 p
= realloc (p
, BUFSIZ
);
339 return IDNA_MALLOC_ERROR
;
341 rc
= idna_to_unicode_internal (in
, inlen
, out
, outlen
,
342 allowunassigned
, usestd3asciirules
,
344 if (rc
!= IDNA_SUCCESS
)
347 sizeof (in
[0]) * (inlen
< outlensave
? inlen
: outlensave
));