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