lh | 9ed821d | 2023-04-07 01:36:19 -0700 | [diff] [blame^] | 1 | /* idna.c Convert to or from IDN strings. |
| 2 | * Copyright (C) 2002, 2003, 2004, 2011 Simon Josefsson |
| 3 | * |
| 4 | * This file is part of GNU Libidn. |
| 5 | * |
| 6 | * GNU Libidn 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. |
| 10 | * |
| 11 | * GNU Libidn 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. |
| 15 | * |
| 16 | * You should have received a copy of the GNU Lesser General Public |
| 17 | * License along with GNU Libidn; if not, see <http://www.gnu.org/licenses/>. |
| 18 | */ |
| 19 | |
| 20 | #if HAVE_CONFIG_H |
| 21 | # include "config.h" |
| 22 | #endif |
| 23 | |
| 24 | #include <stdlib.h> |
| 25 | #include <string.h> |
| 26 | #include <stringprep.h> |
| 27 | #include <punycode.h> |
| 28 | #include <stdint.h> |
| 29 | |
| 30 | #include "idna.h" |
| 31 | |
| 32 | #define DOTP(c) ((c) == 0x002E || (c) == 0x3002 || \ |
| 33 | (c) == 0xFF0E || (c) == 0xFF61) |
| 34 | |
| 35 | /* Core functions */ |
| 36 | |
| 37 | /** |
| 38 | * idna_to_ascii_4i |
| 39 | * @in: input array with unicode code points. |
| 40 | * @inlen: length of input array with unicode code points. |
| 41 | * @out: output zero terminated string that must have room for at |
| 42 | * least 63 characters plus the terminating zero. |
| 43 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 44 | * |
| 45 | * The ToASCII operation takes a sequence of Unicode code points that make |
| 46 | * up one label and transforms it into a sequence of code points in the |
| 47 | * ASCII range (0..7F). If ToASCII succeeds, the original sequence and the |
| 48 | * resulting sequence are equivalent labels. |
| 49 | * |
| 50 | * It is important to note that the ToASCII operation can fail. ToASCII |
| 51 | * fails if any step of it fails. If any step of the ToASCII operation |
| 52 | * fails on any label in a domain name, that domain name MUST NOT be used |
| 53 | * as an internationalized domain name. The method for deadling with this |
| 54 | * failure is application-specific. |
| 55 | * |
| 56 | * The inputs to ToASCII are a sequence of code points, the AllowUnassigned |
| 57 | * flag, and the UseSTD3ASCIIRules flag. The output of ToASCII is either a |
| 58 | * sequence of ASCII code points or a failure condition. |
| 59 | * |
| 60 | * ToASCII never alters a sequence of code points that are all in the ASCII |
| 61 | * range to begin with (although it could fail). Applying the ToASCII |
| 62 | * operation multiple times has exactly the same effect as applying it just |
| 63 | * once. |
| 64 | * |
| 65 | * Return value: Returns 0 on success, or an error code. |
| 66 | */ |
| 67 | int |
| 68 | idna_to_ascii_4i (const uint32_t * in, size_t inlen, char *out, int flags) |
| 69 | { |
| 70 | size_t len, outlen; |
| 71 | uint32_t *src; /* XXX don't need to copy data? */ |
| 72 | int rc; |
| 73 | |
| 74 | /* |
| 75 | * ToASCII consists of the following steps: |
| 76 | * |
| 77 | * 1. If all code points in the sequence are in the ASCII range (0..7F) |
| 78 | * then skip to step 3. |
| 79 | */ |
| 80 | |
| 81 | { |
| 82 | size_t i; |
| 83 | int inasciirange; |
| 84 | |
| 85 | inasciirange = 1; |
| 86 | for (i = 0; i < inlen; i++) |
| 87 | if (in[i] > 0x7F) |
| 88 | inasciirange = 0; |
| 89 | if (inasciirange) |
| 90 | { |
| 91 | src = malloc (sizeof (in[0]) * (inlen + 1)); |
| 92 | if (src == NULL) |
| 93 | return IDNA_MALLOC_ERROR; |
| 94 | |
| 95 | memcpy (src, in, sizeof (in[0]) * inlen); |
| 96 | src[inlen] = 0; |
| 97 | |
| 98 | goto step3; |
| 99 | } |
| 100 | } |
| 101 | |
| 102 | /* |
| 103 | * 2. Perform the steps specified in [NAMEPREP] and fail if there is |
| 104 | * an error. The AllowUnassigned flag is used in [NAMEPREP]. |
| 105 | */ |
| 106 | |
| 107 | { |
| 108 | char *p; |
| 109 | |
| 110 | p = stringprep_ucs4_to_utf8 (in, inlen, NULL, NULL); |
| 111 | if (p == NULL) |
| 112 | return IDNA_MALLOC_ERROR; |
| 113 | |
| 114 | len = strlen (p); |
| 115 | do |
| 116 | { |
| 117 | char *newp; |
| 118 | |
| 119 | len = 2 * len + 10; /* XXX better guess? */ |
| 120 | newp = realloc (p, len); |
| 121 | if (newp == NULL) |
| 122 | { |
| 123 | free (p); |
| 124 | return IDNA_MALLOC_ERROR; |
| 125 | } |
| 126 | p = newp; |
| 127 | |
| 128 | if (flags & IDNA_ALLOW_UNASSIGNED) |
| 129 | rc = stringprep_nameprep (p, len); |
| 130 | else |
| 131 | rc = stringprep_nameprep_no_unassigned (p, len); |
| 132 | } |
| 133 | while (rc == STRINGPREP_TOO_SMALL_BUFFER); |
| 134 | |
| 135 | if (rc != STRINGPREP_OK) |
| 136 | { |
| 137 | free (p); |
| 138 | return IDNA_STRINGPREP_ERROR; |
| 139 | } |
| 140 | |
| 141 | src = stringprep_utf8_to_ucs4 (p, -1, NULL); |
| 142 | |
| 143 | free (p); |
| 144 | } |
| 145 | |
| 146 | step3: |
| 147 | /* |
| 148 | * 3. If the UseSTD3ASCIIRules flag is set, then perform these checks: |
| 149 | * |
| 150 | * (a) Verify the absence of non-LDH ASCII code points; that is, |
| 151 | * the absence of 0..2C, 2E..2F, 3A..40, 5B..60, and 7B..7F. |
| 152 | * |
| 153 | * (b) Verify the absence of leading and trailing hyphen-minus; |
| 154 | * that is, the absence of U+002D at the beginning and end of |
| 155 | * the sequence. |
| 156 | */ |
| 157 | |
| 158 | if (flags & IDNA_USE_STD3_ASCII_RULES) |
| 159 | { |
| 160 | size_t i; |
| 161 | |
| 162 | for (i = 0; src[i]; i++) |
| 163 | if (src[i] <= 0x2C || src[i] == 0x2E || src[i] == 0x2F || |
| 164 | (src[i] >= 0x3A && src[i] <= 0x40) || |
| 165 | (src[i] >= 0x5B && src[i] <= 0x60) || |
| 166 | (src[i] >= 0x7B && src[i] <= 0x7F)) |
| 167 | { |
| 168 | free (src); |
| 169 | return IDNA_CONTAINS_NON_LDH; |
| 170 | } |
| 171 | |
| 172 | if (src[0] == 0x002D || (i > 0 && src[i - 1] == 0x002D)) |
| 173 | { |
| 174 | free (src); |
| 175 | return IDNA_CONTAINS_MINUS; |
| 176 | } |
| 177 | } |
| 178 | |
| 179 | /* |
| 180 | * 4. If all code points in the sequence are in the ASCII range |
| 181 | * (0..7F), then skip to step 8. |
| 182 | */ |
| 183 | |
| 184 | { |
| 185 | size_t i; |
| 186 | int inasciirange; |
| 187 | |
| 188 | inasciirange = 1; |
| 189 | for (i = 0; src[i]; i++) |
| 190 | { |
| 191 | if (src[i] > 0x7F) |
| 192 | inasciirange = 0; |
| 193 | /* copy string to output buffer if we are about to skip to step8 */ |
| 194 | if (i < 64) |
| 195 | out[i] = src[i]; |
| 196 | } |
| 197 | if (i < 64) |
| 198 | out[i] = '\0'; |
| 199 | if (inasciirange) |
| 200 | goto step8; |
| 201 | } |
| 202 | |
| 203 | /* |
| 204 | * 5. Verify that the sequence does NOT begin with the ACE prefix. |
| 205 | * |
| 206 | */ |
| 207 | |
| 208 | { |
| 209 | size_t i; |
| 210 | int match; |
| 211 | |
| 212 | match = 1; |
| 213 | for (i = 0; match && i < strlen (IDNA_ACE_PREFIX); i++) |
| 214 | if (((uint32_t) IDNA_ACE_PREFIX[i] & 0xFF) != src[i]) |
| 215 | match = 0; |
| 216 | if (match) |
| 217 | { |
| 218 | free (src); |
| 219 | return IDNA_CONTAINS_ACE_PREFIX; |
| 220 | } |
| 221 | } |
| 222 | |
| 223 | /* |
| 224 | * 6. Encode the sequence using the encoding algorithm in [PUNYCODE] |
| 225 | * and fail if there is an error. |
| 226 | */ |
| 227 | for (len = 0; src[len]; len++) |
| 228 | ; |
| 229 | src[len] = '\0'; |
| 230 | outlen = 63 - strlen (IDNA_ACE_PREFIX); |
| 231 | rc = punycode_encode (len, src, NULL, |
| 232 | &outlen, &out[strlen (IDNA_ACE_PREFIX)]); |
| 233 | if (rc != PUNYCODE_SUCCESS) |
| 234 | { |
| 235 | free (src); |
| 236 | return IDNA_PUNYCODE_ERROR; |
| 237 | } |
| 238 | out[strlen (IDNA_ACE_PREFIX) + outlen] = '\0'; |
| 239 | |
| 240 | /* |
| 241 | * 7. Prepend the ACE prefix. |
| 242 | */ |
| 243 | |
| 244 | memcpy (out, IDNA_ACE_PREFIX, strlen (IDNA_ACE_PREFIX)); |
| 245 | |
| 246 | /* |
| 247 | * 8. Verify that the number of code points is in the range 1 to 63 |
| 248 | * inclusive (0 is excluded). |
| 249 | */ |
| 250 | |
| 251 | step8: |
| 252 | free (src); |
| 253 | if (strlen (out) < 1 || strlen (out) > 63) |
| 254 | return IDNA_INVALID_LENGTH; |
| 255 | |
| 256 | return IDNA_SUCCESS; |
| 257 | } |
| 258 | |
| 259 | /* ToUnicode(). May realloc() utf8in. */ |
| 260 | static int |
| 261 | idna_to_unicode_internal (char *utf8in, |
| 262 | uint32_t * out, size_t * outlen, int flags) |
| 263 | { |
| 264 | int rc; |
| 265 | char tmpout[64]; |
| 266 | size_t utf8len = strlen (utf8in) + 1; |
| 267 | size_t addlen = 0; |
| 268 | |
| 269 | /* |
| 270 | * ToUnicode consists of the following steps: |
| 271 | * |
| 272 | * 1. If the sequence contains any code points outside the ASCII range |
| 273 | * (0..7F) then proceed to step 2, otherwise skip to step 3. |
| 274 | */ |
| 275 | |
| 276 | { |
| 277 | size_t i; |
| 278 | int inasciirange; |
| 279 | |
| 280 | inasciirange = 1; |
| 281 | for (i = 0; utf8in[i]; i++) |
| 282 | if (utf8in[i] & ~0x7F) |
| 283 | inasciirange = 0; |
| 284 | if (inasciirange) |
| 285 | goto step3; |
| 286 | } |
| 287 | |
| 288 | /* |
| 289 | * 2. Perform the steps specified in [NAMEPREP] and fail if there is an |
| 290 | * error. (If step 3 of ToASCII is also performed here, it will not |
| 291 | * affect the overall behavior of ToUnicode, but it is not |
| 292 | * necessary.) The AllowUnassigned flag is used in [NAMEPREP]. |
| 293 | */ |
| 294 | do |
| 295 | { |
| 296 | char *newp = realloc (utf8in, utf8len + addlen); |
| 297 | if (newp == NULL) |
| 298 | { |
| 299 | free (utf8in); |
| 300 | return IDNA_MALLOC_ERROR; |
| 301 | } |
| 302 | utf8in = newp; |
| 303 | if (flags & IDNA_ALLOW_UNASSIGNED) |
| 304 | rc = stringprep_nameprep (utf8in, utf8len + addlen); |
| 305 | else |
| 306 | rc = stringprep_nameprep_no_unassigned (utf8in, utf8len + addlen); |
| 307 | addlen += 1; |
| 308 | } |
| 309 | while (rc == STRINGPREP_TOO_SMALL_BUFFER); |
| 310 | |
| 311 | if (rc != STRINGPREP_OK) |
| 312 | { |
| 313 | free (utf8in); |
| 314 | return IDNA_STRINGPREP_ERROR; |
| 315 | } |
| 316 | |
| 317 | /* 3. Verify that the sequence begins with the ACE prefix, and save a |
| 318 | * copy of the sequence. |
| 319 | */ |
| 320 | |
| 321 | step3: |
| 322 | if (memcmp (IDNA_ACE_PREFIX, utf8in, strlen (IDNA_ACE_PREFIX)) != 0) |
| 323 | { |
| 324 | free (utf8in); |
| 325 | return IDNA_NO_ACE_PREFIX; |
| 326 | } |
| 327 | |
| 328 | /* 4. Remove the ACE prefix. |
| 329 | */ |
| 330 | |
| 331 | memmove (utf8in, &utf8in[strlen (IDNA_ACE_PREFIX)], |
| 332 | strlen (utf8in) - strlen (IDNA_ACE_PREFIX) + 1); |
| 333 | |
| 334 | /* 5. Decode the sequence using the decoding algorithm in [PUNYCODE] |
| 335 | * and fail if there is an error. Save a copy of the result of |
| 336 | * this step. |
| 337 | */ |
| 338 | |
| 339 | (*outlen)--; /* reserve one for the zero */ |
| 340 | |
| 341 | rc = punycode_decode (strlen (utf8in), utf8in, outlen, out, NULL); |
| 342 | if (rc != PUNYCODE_SUCCESS) |
| 343 | { |
| 344 | free (utf8in); |
| 345 | return IDNA_PUNYCODE_ERROR; |
| 346 | } |
| 347 | |
| 348 | out[*outlen] = 0; /* add zero */ |
| 349 | |
| 350 | /* 6. Apply ToASCII. |
| 351 | */ |
| 352 | |
| 353 | rc = idna_to_ascii_4i (out, *outlen, tmpout, flags); |
| 354 | if (rc != IDNA_SUCCESS) |
| 355 | { |
| 356 | free (utf8in); |
| 357 | return rc; |
| 358 | } |
| 359 | |
| 360 | /* 7. Verify that the result of step 6 matches the saved copy from |
| 361 | * step 3, using a case-insensitive ASCII comparison. |
| 362 | */ |
| 363 | |
| 364 | if (strcasecmp (utf8in, tmpout + strlen (IDNA_ACE_PREFIX)) != 0) |
| 365 | { |
| 366 | free (utf8in); |
| 367 | return IDNA_ROUNDTRIP_VERIFY_ERROR; |
| 368 | } |
| 369 | |
| 370 | /* 8. Return the saved copy from step 5. |
| 371 | */ |
| 372 | |
| 373 | free (utf8in); |
| 374 | return IDNA_SUCCESS; |
| 375 | } |
| 376 | |
| 377 | /** |
| 378 | * idna_to_unicode_44i |
| 379 | * @in: input array with unicode code points. |
| 380 | * @inlen: length of input array with unicode code points. |
| 381 | * @out: output array with unicode code points. |
| 382 | * @outlen: on input, maximum size of output array with unicode code points, |
| 383 | * on exit, actual size of output array with unicode code points. |
| 384 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 385 | * |
| 386 | * The ToUnicode operation takes a sequence of Unicode code points |
| 387 | * that make up one label and returns a sequence of Unicode code |
| 388 | * points. If the input sequence is a label in ACE form, then the |
| 389 | * result is an equivalent internationalized label that is not in ACE |
| 390 | * form, otherwise the original sequence is returned unaltered. |
| 391 | * |
| 392 | * ToUnicode never fails. If any step fails, then the original input |
| 393 | * sequence is returned immediately in that step. |
| 394 | * |
| 395 | * The Punycode decoder can never output more code points than it |
| 396 | * inputs, but Nameprep can, and therefore ToUnicode can. Note that |
| 397 | * the number of octets needed to represent a sequence of code points |
| 398 | * depends on the particular character encoding used. |
| 399 | * |
| 400 | * The inputs to ToUnicode are a sequence of code points, the |
| 401 | * AllowUnassigned flag, and the UseSTD3ASCIIRules flag. The output of |
| 402 | * ToUnicode is always a sequence of Unicode code points. |
| 403 | * |
| 404 | * Return value: Returns error condition, but it must only be used for |
| 405 | * debugging purposes. The output buffer is always |
| 406 | * guaranteed to contain the correct data according to |
| 407 | * the specification (sans malloc induced errors). NB! |
| 408 | * This means that you normally ignore the return code |
| 409 | * from this function, as checking it means breaking the |
| 410 | * standard. |
| 411 | */ |
| 412 | int |
| 413 | idna_to_unicode_44i (const uint32_t * in, size_t inlen, |
| 414 | uint32_t * out, size_t * outlen, int flags) |
| 415 | { |
| 416 | int rc; |
| 417 | size_t outlensave = *outlen; |
| 418 | char *p; |
| 419 | |
| 420 | p = stringprep_ucs4_to_utf8 (in, inlen, NULL, NULL); |
| 421 | if (p == NULL) |
| 422 | return IDNA_MALLOC_ERROR; |
| 423 | |
| 424 | rc = idna_to_unicode_internal (p, out, outlen, flags); |
| 425 | if (rc != IDNA_SUCCESS) |
| 426 | { |
| 427 | memcpy (out, in, sizeof (in[0]) * (inlen < outlensave ? |
| 428 | inlen : outlensave)); |
| 429 | *outlen = inlen; |
| 430 | } |
| 431 | |
| 432 | /* p is freed in idna_to_unicode_internal. */ |
| 433 | |
| 434 | return rc; |
| 435 | } |
| 436 | |
| 437 | /* Wrappers that handle several labels */ |
| 438 | |
| 439 | /** |
| 440 | * idna_to_ascii_4z: |
| 441 | * @input: zero terminated input Unicode string. |
| 442 | * @output: pointer to newly allocated output string. |
| 443 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 444 | * |
| 445 | * Convert UCS-4 domain name to ASCII string. The domain name may |
| 446 | * contain several labels, separated by dots. The output buffer must |
| 447 | * be deallocated by the caller. |
| 448 | * |
| 449 | * Return value: Returns IDNA_SUCCESS on success, or error code. |
| 450 | **/ |
| 451 | int |
| 452 | idna_to_ascii_4z (const uint32_t * input, char **output, int flags) |
| 453 | { |
| 454 | const uint32_t *start = input; |
| 455 | const uint32_t *end = input; |
| 456 | char buf[64]; |
| 457 | char *out = NULL; |
| 458 | int rc; |
| 459 | |
| 460 | /* 1) Whenever dots are used as label separators, the following |
| 461 | characters MUST be recognized as dots: U+002E (full stop), |
| 462 | U+3002 (ideographic full stop), U+FF0E (fullwidth full stop), |
| 463 | U+FF61 (halfwidth ideographic full stop). */ |
| 464 | |
| 465 | if (input[0] == 0) |
| 466 | { |
| 467 | /* Handle implicit zero-length root label. */ |
| 468 | *output = malloc (1); |
| 469 | if (!*output) |
| 470 | return IDNA_MALLOC_ERROR; |
| 471 | strcpy (*output, ""); |
| 472 | return IDNA_SUCCESS; |
| 473 | } |
| 474 | |
| 475 | if (DOTP (input[0]) && input[1] == 0) |
| 476 | { |
| 477 | /* Handle explicit zero-length root label. */ |
| 478 | *output = malloc (2); |
| 479 | if (!*output) |
| 480 | return IDNA_MALLOC_ERROR; |
| 481 | strcpy (*output, "."); |
| 482 | return IDNA_SUCCESS; |
| 483 | } |
| 484 | |
| 485 | *output = NULL; |
| 486 | do |
| 487 | { |
| 488 | end = start; |
| 489 | |
| 490 | for (; *end && !DOTP (*end); end++) |
| 491 | ; |
| 492 | |
| 493 | if (*end == '\0' && start == end) |
| 494 | { |
| 495 | /* Handle explicit zero-length root label. */ |
| 496 | buf[0] = '\0'; |
| 497 | } |
| 498 | else |
| 499 | { |
| 500 | rc = idna_to_ascii_4i (start, end - start, buf, flags); |
| 501 | if (rc != IDNA_SUCCESS) |
| 502 | return rc; |
| 503 | } |
| 504 | |
| 505 | if (out) |
| 506 | { |
| 507 | char *newp = realloc (out, strlen (out) + 1 + strlen (buf) + 1); |
| 508 | if (!newp) |
| 509 | { |
| 510 | free (out); |
| 511 | return IDNA_MALLOC_ERROR; |
| 512 | } |
| 513 | out = newp; |
| 514 | strcat (out, "."); |
| 515 | strcat (out, buf); |
| 516 | } |
| 517 | else |
| 518 | { |
| 519 | out = (char *) malloc (strlen (buf) + 1); |
| 520 | if (!out) |
| 521 | return IDNA_MALLOC_ERROR; |
| 522 | strcpy (out, buf); |
| 523 | } |
| 524 | |
| 525 | start = end + 1; |
| 526 | } |
| 527 | while (*end); |
| 528 | |
| 529 | *output = out; |
| 530 | |
| 531 | return IDNA_SUCCESS; |
| 532 | } |
| 533 | |
| 534 | /** |
| 535 | * idna_to_ascii_8z: |
| 536 | * @input: zero terminated input UTF-8 string. |
| 537 | * @output: pointer to newly allocated output string. |
| 538 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 539 | * |
| 540 | * Convert UTF-8 domain name to ASCII string. The domain name may |
| 541 | * contain several labels, separated by dots. The output buffer must |
| 542 | * be deallocated by the caller. |
| 543 | * |
| 544 | * Return value: Returns IDNA_SUCCESS on success, or error code. |
| 545 | **/ |
| 546 | int |
| 547 | idna_to_ascii_8z (const char *input, char **output, int flags) |
| 548 | { |
| 549 | uint32_t *ucs4; |
| 550 | size_t ucs4len; |
| 551 | int rc; |
| 552 | |
| 553 | ucs4 = stringprep_utf8_to_ucs4 (input, -1, &ucs4len); |
| 554 | if (!ucs4) |
| 555 | return IDNA_ICONV_ERROR; |
| 556 | |
| 557 | rc = idna_to_ascii_4z (ucs4, output, flags); |
| 558 | |
| 559 | free (ucs4); |
| 560 | |
| 561 | return rc; |
| 562 | |
| 563 | } |
| 564 | |
| 565 | /** |
| 566 | * idna_to_ascii_lz: |
| 567 | * @input: zero terminated input UTF-8 string. |
| 568 | * @output: pointer to newly allocated output string. |
| 569 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 570 | * |
| 571 | * Convert domain name in the locale's encoding to ASCII string. The |
| 572 | * domain name may contain several labels, separated by dots. The |
| 573 | * output buffer must be deallocated by the caller. |
| 574 | * |
| 575 | * Return value: Returns IDNA_SUCCESS on success, or error code. |
| 576 | **/ |
| 577 | int |
| 578 | idna_to_ascii_lz (const char *input, char **output, int flags) |
| 579 | { |
| 580 | char *utf8; |
| 581 | int rc; |
| 582 | |
| 583 | utf8 = stringprep_locale_to_utf8 (input); |
| 584 | if (!utf8) |
| 585 | return IDNA_ICONV_ERROR; |
| 586 | |
| 587 | rc = idna_to_ascii_8z (utf8, output, flags); |
| 588 | |
| 589 | free (utf8); |
| 590 | |
| 591 | return rc; |
| 592 | } |
| 593 | |
| 594 | /** |
| 595 | * idna_to_unicode_4z4z: |
| 596 | * @input: zero-terminated Unicode string. |
| 597 | * @output: pointer to newly allocated output Unicode string. |
| 598 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 599 | * |
| 600 | * Convert possibly ACE encoded domain name in UCS-4 format into a |
| 601 | * UCS-4 string. The domain name may contain several labels, |
| 602 | * separated by dots. The output buffer must be deallocated by the |
| 603 | * caller. |
| 604 | * |
| 605 | * Return value: Returns IDNA_SUCCESS on success, or error code. |
| 606 | **/ |
| 607 | int |
| 608 | idna_to_unicode_4z4z (const uint32_t * input, uint32_t ** output, int flags) |
| 609 | { |
| 610 | const uint32_t *start = input; |
| 611 | const uint32_t *end = input; |
| 612 | uint32_t *buf; |
| 613 | size_t buflen; |
| 614 | uint32_t *out = NULL; |
| 615 | size_t outlen = 0; |
| 616 | |
| 617 | *output = NULL; |
| 618 | |
| 619 | do |
| 620 | { |
| 621 | end = start; |
| 622 | |
| 623 | for (; *end && !DOTP (*end); end++) |
| 624 | ; |
| 625 | |
| 626 | buflen = end - start; |
| 627 | buf = malloc (sizeof (buf[0]) * (buflen + 1)); |
| 628 | if (!buf) |
| 629 | return IDNA_MALLOC_ERROR; |
| 630 | |
| 631 | idna_to_unicode_44i (start, end - start, buf, &buflen, flags); |
| 632 | /* don't check return value as per specification! */ |
| 633 | |
| 634 | if (out) |
| 635 | { |
| 636 | uint32_t *newp = realloc (out, |
| 637 | sizeof (out[0]) |
| 638 | * (outlen + 1 + buflen + 1)); |
| 639 | if (!newp) |
| 640 | { |
| 641 | free (buf); |
| 642 | free (out); |
| 643 | return IDNA_MALLOC_ERROR; |
| 644 | } |
| 645 | out = newp; |
| 646 | out[outlen++] = 0x002E; /* '.' (full stop) */ |
| 647 | memcpy (out + outlen, buf, sizeof (buf[0]) * buflen); |
| 648 | outlen += buflen; |
| 649 | out[outlen] = 0x0; |
| 650 | free (buf); |
| 651 | } |
| 652 | else |
| 653 | { |
| 654 | out = buf; |
| 655 | outlen = buflen; |
| 656 | out[outlen] = 0x0; |
| 657 | } |
| 658 | |
| 659 | start = end + 1; |
| 660 | } |
| 661 | while (*end); |
| 662 | |
| 663 | *output = out; |
| 664 | |
| 665 | return IDNA_SUCCESS; |
| 666 | } |
| 667 | |
| 668 | /** |
| 669 | * idna_to_unicode_8z4z: |
| 670 | * @input: zero-terminated UTF-8 string. |
| 671 | * @output: pointer to newly allocated output Unicode string. |
| 672 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 673 | * |
| 674 | * Convert possibly ACE encoded domain name in UTF-8 format into a |
| 675 | * UCS-4 string. The domain name may contain several labels, |
| 676 | * separated by dots. The output buffer must be deallocated by the |
| 677 | * caller. |
| 678 | * |
| 679 | * Return value: Returns IDNA_SUCCESS on success, or error code. |
| 680 | **/ |
| 681 | int |
| 682 | idna_to_unicode_8z4z (const char *input, uint32_t ** output, int flags) |
| 683 | { |
| 684 | uint32_t *ucs4; |
| 685 | size_t ucs4len; |
| 686 | int rc; |
| 687 | |
| 688 | ucs4 = stringprep_utf8_to_ucs4 (input, -1, &ucs4len); |
| 689 | if (!ucs4) |
| 690 | return IDNA_ICONV_ERROR; |
| 691 | |
| 692 | rc = idna_to_unicode_4z4z (ucs4, output, flags); |
| 693 | free (ucs4); |
| 694 | |
| 695 | return rc; |
| 696 | } |
| 697 | |
| 698 | /** |
| 699 | * idna_to_unicode_8z8z: |
| 700 | * @input: zero-terminated UTF-8 string. |
| 701 | * @output: pointer to newly allocated output UTF-8 string. |
| 702 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 703 | * |
| 704 | * Convert possibly ACE encoded domain name in UTF-8 format into a |
| 705 | * UTF-8 string. The domain name may contain several labels, |
| 706 | * separated by dots. The output buffer must be deallocated by the |
| 707 | * caller. |
| 708 | * |
| 709 | * Return value: Returns IDNA_SUCCESS on success, or error code. |
| 710 | **/ |
| 711 | int |
| 712 | idna_to_unicode_8z8z (const char *input, char **output, int flags) |
| 713 | { |
| 714 | uint32_t *ucs4; |
| 715 | int rc; |
| 716 | |
| 717 | rc = idna_to_unicode_8z4z (input, &ucs4, flags); |
| 718 | *output = stringprep_ucs4_to_utf8 (ucs4, -1, NULL, NULL); |
| 719 | free (ucs4); |
| 720 | |
| 721 | if (!*output) |
| 722 | return IDNA_ICONV_ERROR; |
| 723 | |
| 724 | return rc; |
| 725 | } |
| 726 | |
| 727 | /** |
| 728 | * idna_to_unicode_8zlz: |
| 729 | * @input: zero-terminated UTF-8 string. |
| 730 | * @output: pointer to newly allocated output string encoded in the |
| 731 | * current locale's character set. |
| 732 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 733 | * |
| 734 | * Convert possibly ACE encoded domain name in UTF-8 format into a |
| 735 | * string encoded in the current locale's character set. The domain |
| 736 | * name may contain several labels, separated by dots. The output |
| 737 | * buffer must be deallocated by the caller. |
| 738 | * |
| 739 | * Return value: Returns IDNA_SUCCESS on success, or error code. |
| 740 | **/ |
| 741 | int |
| 742 | idna_to_unicode_8zlz (const char *input, char **output, int flags) |
| 743 | { |
| 744 | char *utf8; |
| 745 | int rc; |
| 746 | |
| 747 | rc = idna_to_unicode_8z8z (input, &utf8, flags); |
| 748 | *output = stringprep_utf8_to_locale (utf8); |
| 749 | free (utf8); |
| 750 | |
| 751 | if (!*output) |
| 752 | return IDNA_ICONV_ERROR; |
| 753 | |
| 754 | return rc; |
| 755 | } |
| 756 | |
| 757 | /** |
| 758 | * idna_to_unicode_lzlz: |
| 759 | * @input: zero-terminated string encoded in the current locale's |
| 760 | * character set. |
| 761 | * @output: pointer to newly allocated output string encoded in the |
| 762 | * current locale's character set. |
| 763 | * @flags: IDNA flags, e.g. IDNA_ALLOW_UNASSIGNED or IDNA_USE_STD3_ASCII_RULES. |
| 764 | * |
| 765 | * Convert possibly ACE encoded domain name in the locale's character |
| 766 | * set into a string encoded in the current locale's character set. |
| 767 | * The domain name may contain several labels, separated by dots. The |
| 768 | * output buffer must be deallocated by the caller. |
| 769 | * |
| 770 | * Return value: Returns IDNA_SUCCESS on success, or error code. |
| 771 | **/ |
| 772 | int |
| 773 | idna_to_unicode_lzlz (const char *input, char **output, int flags) |
| 774 | { |
| 775 | char *utf8; |
| 776 | int rc; |
| 777 | |
| 778 | utf8 = stringprep_locale_to_utf8 (input); |
| 779 | if (!utf8) |
| 780 | return IDNA_ICONV_ERROR; |
| 781 | |
| 782 | rc = idna_to_unicode_8zlz (utf8, output, flags); |
| 783 | free (utf8); |
| 784 | |
| 785 | return rc; |
| 786 | } |
| 787 | |
| 788 | /** |
| 789 | * IDNA_ACE_PREFIX |
| 790 | * |
| 791 | * The IANA allocated prefix to use for IDNA. "xn--" |
| 792 | */ |
| 793 | |
| 794 | /** |
| 795 | * Idna_rc: |
| 796 | * @IDNA_SUCCESS: Successful operation. This value is guaranteed to |
| 797 | * always be zero, the remaining ones are only guaranteed to hold |
| 798 | * non-zero values, for logical comparison purposes. |
| 799 | * @IDNA_STRINGPREP_ERROR: Error during string preparation. |
| 800 | * @IDNA_PUNYCODE_ERROR: Error during punycode operation. |
| 801 | * @IDNA_CONTAINS_NON_LDH: For IDNA_USE_STD3_ASCII_RULES, indicate that |
| 802 | * the string contains non-LDH ASCII characters. |
| 803 | * @IDNA_CONTAINS_MINUS: For IDNA_USE_STD3_ASCII_RULES, indicate that |
| 804 | * the string contains a leading or trailing hyphen-minus (U+002D). |
| 805 | * @IDNA_INVALID_LENGTH: The final output string is not within the |
| 806 | * (inclusive) range 1 to 63 characters. |
| 807 | * @IDNA_NO_ACE_PREFIX: The string does not contain the ACE prefix |
| 808 | * (for ToUnicode). |
| 809 | * @IDNA_ROUNDTRIP_VERIFY_ERROR: The ToASCII operation on output |
| 810 | * string does not equal the input. |
| 811 | * @IDNA_CONTAINS_ACE_PREFIX: The input contains the ACE prefix (for |
| 812 | * ToASCII). |
| 813 | * @IDNA_ICONV_ERROR: Could not convert string in locale encoding. |
| 814 | * @IDNA_MALLOC_ERROR: Could not allocate buffer (this is typically a |
| 815 | * fatal error). |
| 816 | * @IDNA_DLOPEN_ERROR: Could not dlopen the libcidn DSO (only used |
| 817 | * internally in libc). |
| 818 | * |
| 819 | * Enumerated return codes of idna_to_ascii_4i(), |
| 820 | * idna_to_unicode_44i() functions (and functions derived from those |
| 821 | * functions). The value 0 is guaranteed to always correspond to |
| 822 | * success. |
| 823 | */ |
| 824 | |
| 825 | |
| 826 | /** |
| 827 | * Idna_flags: |
| 828 | * @IDNA_ALLOW_UNASSIGNED: Don't reject strings containing unassigned |
| 829 | * Unicode code points. |
| 830 | * @IDNA_USE_STD3_ASCII_RULES: Validate strings according to STD3 |
| 831 | * rules (i.e., normal host name rules). |
| 832 | * |
| 833 | * Flags to pass to idna_to_ascii_4i(), idna_to_unicode_44i() etc. |
| 834 | */ |