1 /* $OpenBSD: a_mbstr.c,v 1.19 2014/07/11 08:44:47 jsing Exp $ */ 2 /* Written by Dr Stephen N Henson (steve@openssl.org) for the OpenSSL 3 * project 1999. 4 */ 5 /* ==================================================================== 6 * Copyright (c) 1999 The OpenSSL Project. All rights reserved. 7 * 8 * Redistribution and use in source and binary forms, with or without 9 * modification, are permitted provided that the following conditions 10 * are met: 11 * 12 * 1. Redistributions of source code must retain the above copyright 13 * notice, this list of conditions and the following disclaimer. 14 * 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in 17 * the documentation and/or other materials provided with the 18 * distribution. 19 * 20 * 3. All advertising materials mentioning features or use of this 21 * software must display the following acknowledgment: 22 * "This product includes software developed by the OpenSSL Project 23 * for use in the OpenSSL Toolkit. (http://www.OpenSSL.org/)" 24 * 25 * 4. The names "OpenSSL Toolkit" and "OpenSSL Project" must not be used to 26 * endorse or promote products derived from this software without 27 * prior written permission. For written permission, please contact 28 * licensing@OpenSSL.org. 29 * 30 * 5. Products derived from this software may not be called "OpenSSL" 31 * nor may "OpenSSL" appear in their names without prior written 32 * permission of the OpenSSL Project. 33 * 34 * 6. Redistributions of any form whatsoever must retain the following 35 * acknowledgment: 36 * "This product includes software developed by the OpenSSL Project 37 * for use in the OpenSSL Toolkit (http://www.OpenSSL.org/)" 38 * 39 * THIS SOFTWARE IS PROVIDED BY THE OpenSSL PROJECT ``AS IS'' AND ANY 40 * EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 41 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 42 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE OpenSSL PROJECT OR 43 * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, 44 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 45 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; 46 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 47 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, 48 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 49 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED 50 * OF THE POSSIBILITY OF SUCH DAMAGE. 51 * ==================================================================== 52 * 53 * This product includes cryptographic software written by Eric Young 54 * (eay@cryptsoft.com). This product includes software written by Tim 55 * Hudson (tjh@cryptsoft.com). 56 * 57 */ 58 59 #include <ctype.h> 60 #include <stdio.h> 61 #include <string.h> 62 63 #include <openssl/asn1.h> 64 #include <openssl/err.h> 65 66 #include "asn1_locl.h" 67 68 static int traverse_string(const unsigned char *p, int len, int inform, 69 int (*rfunc)(unsigned long value, void *in), void *arg); 70 static int in_utf8(unsigned long value, void *arg); 71 static int out_utf8(unsigned long value, void *arg); 72 static int type_str(unsigned long value, void *arg); 73 static int cpy_asc(unsigned long value, void *arg); 74 static int cpy_bmp(unsigned long value, void *arg); 75 static int cpy_univ(unsigned long value, void *arg); 76 static int cpy_utf8(unsigned long value, void *arg); 77 static int is_printable(unsigned long value); 78 79 /* These functions take a string in UTF8, ASCII or multibyte form and 80 * a mask of permissible ASN1 string types. It then works out the minimal 81 * type (using the order Printable < IA5 < T61 < BMP < Universal < UTF8) 82 * and creates a string of the correct type with the supplied data. 83 * Yes this is horrible: it has to be :-( 84 * The 'ncopy' form checks minimum and maximum size limits too. 85 */ 86 87 int 88 ASN1_mbstring_copy(ASN1_STRING **out, const unsigned char *in, int len, 89 int inform, unsigned long mask) 90 { 91 return ASN1_mbstring_ncopy(out, in, len, inform, mask, 0, 0); 92 } 93 94 int 95 ASN1_mbstring_ncopy(ASN1_STRING **out, const unsigned char *in, int len, 96 int inform, unsigned long mask, long minsize, long maxsize) 97 { 98 int str_type; 99 int ret; 100 char free_out; 101 int outform, outlen = 0; 102 ASN1_STRING *dest; 103 unsigned char *p; 104 int nchar; 105 int (*cpyfunc)(unsigned long, void *) = NULL; 106 107 if (len == -1) 108 len = strlen((const char *)in); 109 if (!mask) 110 mask = DIRSTRING_TYPE; 111 112 /* First do a string check and work out the number of characters */ 113 switch (inform) { 114 case MBSTRING_BMP: 115 if (len & 1) { 116 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, 117 ASN1_R_INVALID_BMPSTRING_LENGTH); 118 return -1; 119 } 120 nchar = len >> 1; 121 break; 122 123 case MBSTRING_UNIV: 124 if (len & 3) { 125 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, 126 ASN1_R_INVALID_UNIVERSALSTRING_LENGTH); 127 return -1; 128 } 129 nchar = len >> 2; 130 break; 131 132 case MBSTRING_UTF8: 133 nchar = 0; 134 /* This counts the characters and does utf8 syntax checking */ 135 ret = traverse_string(in, len, MBSTRING_UTF8, in_utf8, &nchar); 136 if (ret < 0) { 137 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, 138 ASN1_R_INVALID_UTF8STRING); 139 return -1; 140 } 141 break; 142 143 case MBSTRING_ASC: 144 nchar = len; 145 break; 146 147 default: 148 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_UNKNOWN_FORMAT); 149 return -1; 150 } 151 152 if ((minsize > 0) && (nchar < minsize)) { 153 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_STRING_TOO_SHORT); 154 ERR_asprintf_error_data("minsize=%ld", minsize); 155 return -1; 156 } 157 158 if ((maxsize > 0) && (nchar > maxsize)) { 159 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_STRING_TOO_LONG); 160 ERR_asprintf_error_data("maxsize=%ld", maxsize); 161 return -1; 162 } 163 164 /* Now work out minimal type (if any) */ 165 if (traverse_string(in, len, inform, type_str, &mask) < 0) { 166 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_ILLEGAL_CHARACTERS); 167 return -1; 168 } 169 170 171 /* Now work out output format and string type */ 172 outform = MBSTRING_ASC; 173 if (mask & B_ASN1_PRINTABLESTRING) 174 str_type = V_ASN1_PRINTABLESTRING; 175 else if (mask & B_ASN1_IA5STRING) 176 str_type = V_ASN1_IA5STRING; 177 else if (mask & B_ASN1_T61STRING) 178 str_type = V_ASN1_T61STRING; 179 else if (mask & B_ASN1_BMPSTRING) { 180 str_type = V_ASN1_BMPSTRING; 181 outform = MBSTRING_BMP; 182 } else if (mask & B_ASN1_UNIVERSALSTRING) { 183 str_type = V_ASN1_UNIVERSALSTRING; 184 outform = MBSTRING_UNIV; 185 } else { 186 str_type = V_ASN1_UTF8STRING; 187 outform = MBSTRING_UTF8; 188 } 189 if (!out) 190 return str_type; 191 if (*out) { 192 free_out = 0; 193 dest = *out; 194 if (dest->data) { 195 dest->length = 0; 196 free(dest->data); 197 dest->data = NULL; 198 } 199 dest->type = str_type; 200 } else { 201 free_out = 1; 202 dest = ASN1_STRING_type_new(str_type); 203 if (!dest) { 204 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, 205 ERR_R_MALLOC_FAILURE); 206 return -1; 207 } 208 *out = dest; 209 } 210 /* If both the same type just copy across */ 211 if (inform == outform) { 212 if (!ASN1_STRING_set(dest, in, len)) { 213 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, 214 ERR_R_MALLOC_FAILURE); 215 return -1; 216 } 217 return str_type; 218 } 219 220 /* Work out how much space the destination will need */ 221 switch (outform) { 222 case MBSTRING_ASC: 223 outlen = nchar; 224 cpyfunc = cpy_asc; 225 break; 226 227 case MBSTRING_BMP: 228 outlen = nchar << 1; 229 cpyfunc = cpy_bmp; 230 break; 231 232 case MBSTRING_UNIV: 233 outlen = nchar << 2; 234 cpyfunc = cpy_univ; 235 break; 236 237 case MBSTRING_UTF8: 238 outlen = 0; 239 if (traverse_string(in, len, inform, out_utf8, &outlen) < 0) { 240 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, 241 ASN1_R_ILLEGAL_CHARACTERS); 242 return -1; 243 } 244 cpyfunc = cpy_utf8; 245 break; 246 } 247 if (!(p = malloc(outlen + 1))) { 248 if (free_out) 249 ASN1_STRING_free(dest); 250 ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ERR_R_MALLOC_FAILURE); 251 return -1; 252 } 253 dest->length = outlen; 254 dest->data = p; 255 p[outlen] = 0; 256 traverse_string(in, len, inform, cpyfunc, &p); 257 return str_type; 258 } 259 260 /* This function traverses a string and passes the value of each character 261 * to an optional function along with a void * argument. 262 */ 263 264 static int 265 traverse_string(const unsigned char *p, int len, int inform, 266 int (*rfunc)(unsigned long value, void *in), void *arg) 267 { 268 unsigned long value; 269 int ret; 270 271 while (len) { 272 if (inform == MBSTRING_ASC) { 273 value = *p++; 274 len--; 275 } else if (inform == MBSTRING_BMP) { 276 value = *p++ << 8; 277 value |= *p++; 278 /* BMP is explictly defined to not support surrogates */ 279 if (UNICODE_IS_SURROGATE(value)) 280 return -1; 281 len -= 2; 282 } else if (inform == MBSTRING_UNIV) { 283 value = ((unsigned long)*p++) << 24; 284 value |= ((unsigned long)*p++) << 16; 285 value |= *p++ << 8; 286 value |= *p++; 287 if (value > UNICODE_MAX || UNICODE_IS_SURROGATE(value)) 288 return -1; 289 len -= 4; 290 } else { 291 ret = UTF8_getc(p, len, &value); 292 if (ret < 0) 293 return -1; 294 len -= ret; 295 p += ret; 296 } 297 if (rfunc) { 298 ret = rfunc(value, arg); 299 if (ret <= 0) 300 return ret; 301 } 302 } 303 return 1; 304 } 305 306 /* Various utility functions for traverse_string */ 307 308 /* Just count number of characters */ 309 310 static int 311 in_utf8(unsigned long value, void *arg) 312 { 313 int *nchar; 314 315 nchar = arg; 316 (*nchar)++; 317 return 1; 318 } 319 320 /* Determine size of output as a UTF8 String */ 321 322 static int 323 out_utf8(unsigned long value, void *arg) 324 { 325 int *outlen; 326 int ret; 327 328 outlen = arg; 329 ret = UTF8_putc(NULL, -1, value); 330 if (ret < 0) 331 return ret; 332 *outlen += ret; 333 return 1; 334 } 335 336 /* Determine the "type" of a string: check each character against a 337 * supplied "mask". 338 */ 339 340 static int 341 type_str(unsigned long value, void *arg) 342 { 343 unsigned long types; 344 345 types = *((unsigned long *)arg); 346 if ((types & B_ASN1_PRINTABLESTRING) && !is_printable(value)) 347 types &= ~B_ASN1_PRINTABLESTRING; 348 if ((types & B_ASN1_IA5STRING) && (value > 127)) 349 types &= ~B_ASN1_IA5STRING; 350 if ((types & B_ASN1_T61STRING) && (value > 0xff)) 351 types &= ~B_ASN1_T61STRING; 352 if ((types & B_ASN1_BMPSTRING) && (value > 0xffff)) 353 types &= ~B_ASN1_BMPSTRING; 354 if (!types) 355 return -1; 356 *((unsigned long *)arg) = types; 357 return 1; 358 } 359 360 /* Copy one byte per character ASCII like strings */ 361 362 static int 363 cpy_asc(unsigned long value, void *arg) 364 { 365 unsigned char **p, *q; 366 367 p = arg; 368 q = *p; 369 *q = (unsigned char) value; 370 (*p)++; 371 return 1; 372 } 373 374 /* Copy two byte per character BMPStrings */ 375 376 static int 377 cpy_bmp(unsigned long value, void *arg) 378 { 379 unsigned char **p, *q; 380 381 p = arg; 382 q = *p; 383 *q++ = (unsigned char) ((value >> 8) & 0xff); 384 *q = (unsigned char) (value & 0xff); 385 *p += 2; 386 return 1; 387 } 388 389 /* Copy four byte per character UniversalStrings */ 390 391 static int 392 cpy_univ(unsigned long value, void *arg) 393 { 394 unsigned char **p, *q; 395 396 p = arg; 397 q = *p; 398 *q++ = (unsigned char) ((value >> 24) & 0xff); 399 *q++ = (unsigned char) ((value >> 16) & 0xff); 400 *q++ = (unsigned char) ((value >> 8) & 0xff); 401 *q = (unsigned char) (value & 0xff); 402 *p += 4; 403 return 1; 404 } 405 406 /* Copy to a UTF8String */ 407 408 static int 409 cpy_utf8(unsigned long value, void *arg) 410 { 411 unsigned char **p; 412 413 int ret; 414 p = arg; 415 /* We already know there is enough room so pass 0xff as the length */ 416 ret = UTF8_putc(*p, 0xff, value); 417 *p += ret; 418 return 1; 419 } 420 421 /* Return 1 if the character is permitted in a PrintableString */ 422 static int 423 is_printable(unsigned long value) 424 { 425 int ch; 426 427 if (value > 0x7f) 428 return 0; 429 ch = (int)value; 430 431 /* Note: we can't use 'isalnum' because certain accented 432 * characters may count as alphanumeric in some environments. 433 */ 434 if ((ch >= 'a') && (ch <= 'z')) 435 return 1; 436 if ((ch >= 'A') && (ch <= 'Z')) 437 return 1; 438 if ((ch >= '0') && (ch <= '9')) 439 return 1; 440 if ((ch == ' ') || strchr("'()+,-./:=?", ch)) 441 return 1; 442 return 0; 443 } 444