/src/openssl/crypto/asn1/a_mbstr.c
Line | Count | Source (jump to first uncovered line) |
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 | | |
65 | | static int traverse_string(const unsigned char *p, int len, int inform, |
66 | | int (*rfunc) (unsigned long value, void *in), |
67 | | void *arg); |
68 | | static int in_utf8(unsigned long value, void *arg); |
69 | | static int out_utf8(unsigned long value, void *arg); |
70 | | static int type_str(unsigned long value, void *arg); |
71 | | static int cpy_asc(unsigned long value, void *arg); |
72 | | static int cpy_bmp(unsigned long value, void *arg); |
73 | | static int cpy_univ(unsigned long value, void *arg); |
74 | | static int cpy_utf8(unsigned long value, void *arg); |
75 | | static 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 | | |
86 | | int ASN1_mbstring_copy(ASN1_STRING **out, const unsigned char *in, int len, |
87 | | int inform, unsigned long mask) |
88 | 1.18M | { |
89 | 1.18M | return ASN1_mbstring_ncopy(out, in, len, inform, mask, 0, 0); |
90 | 1.18M | } |
91 | | |
92 | | int ASN1_mbstring_ncopy(ASN1_STRING **out, const unsigned char *in, int len, |
93 | | int inform, unsigned long mask, |
94 | | long minsize, long maxsize) |
95 | 1.18M | { |
96 | 1.18M | int str_type; |
97 | 1.18M | int ret; |
98 | 1.18M | char free_out; |
99 | 1.18M | int outform, outlen = 0; |
100 | 1.18M | ASN1_STRING *dest; |
101 | 1.18M | unsigned char *p; |
102 | 1.18M | int nchar; |
103 | 1.18M | char strbuf[32]; |
104 | 1.18M | int (*cpyfunc) (unsigned long, void *) = NULL; |
105 | 1.18M | if (len == -1) |
106 | 0 | len = strlen((const char *)in); |
107 | 1.18M | if (!mask) |
108 | 0 | mask = DIRSTRING_TYPE; |
109 | | |
110 | | /* First do a string check and work out the number of characters */ |
111 | 1.18M | switch (inform) { |
112 | | |
113 | 0 | case MBSTRING_BMP: |
114 | 0 | if (len & 1) { |
115 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, |
116 | 0 | ASN1_R_INVALID_BMPSTRING_LENGTH); |
117 | 0 | return -1; |
118 | 0 | } |
119 | 0 | nchar = len >> 1; |
120 | 0 | break; |
121 | | |
122 | 0 | case MBSTRING_UNIV: |
123 | 0 | if (len & 3) { |
124 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, |
125 | 0 | ASN1_R_INVALID_UNIVERSALSTRING_LENGTH); |
126 | 0 | return -1; |
127 | 0 | } |
128 | 0 | nchar = len >> 2; |
129 | 0 | break; |
130 | | |
131 | 295k | case MBSTRING_UTF8: |
132 | 295k | nchar = 0; |
133 | | /* This counts the characters and does utf8 syntax checking */ |
134 | 295k | ret = traverse_string(in, len, MBSTRING_UTF8, in_utf8, &nchar); |
135 | 295k | if (ret < 0) { |
136 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_INVALID_UTF8STRING); |
137 | 0 | return -1; |
138 | 0 | } |
139 | 295k | break; |
140 | | |
141 | 892k | case MBSTRING_ASC: |
142 | 892k | nchar = len; |
143 | 892k | break; |
144 | | |
145 | 0 | default: |
146 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_UNKNOWN_FORMAT); |
147 | 0 | return -1; |
148 | 1.18M | } |
149 | | |
150 | 1.18M | if ((minsize > 0) && (nchar < minsize)) { |
151 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_STRING_TOO_SHORT); |
152 | 0 | BIO_snprintf(strbuf, sizeof strbuf, "%ld", minsize); |
153 | 0 | ERR_add_error_data(2, "minsize=", strbuf); |
154 | 0 | return -1; |
155 | 0 | } |
156 | | |
157 | 1.18M | if ((maxsize > 0) && (nchar > maxsize)) { |
158 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_STRING_TOO_LONG); |
159 | 0 | BIO_snprintf(strbuf, sizeof strbuf, "%ld", maxsize); |
160 | 0 | ERR_add_error_data(2, "maxsize=", strbuf); |
161 | 0 | return -1; |
162 | 0 | } |
163 | | |
164 | | /* Now work out minimal type (if any) */ |
165 | 1.18M | if (traverse_string(in, len, inform, type_str, &mask) < 0) { |
166 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_ILLEGAL_CHARACTERS); |
167 | 0 | return -1; |
168 | 0 | } |
169 | | |
170 | | /* Now work out output format and string type */ |
171 | 1.18M | outform = MBSTRING_ASC; |
172 | 1.18M | if (mask & B_ASN1_PRINTABLESTRING) |
173 | 0 | str_type = V_ASN1_PRINTABLESTRING; |
174 | 1.18M | else if (mask & B_ASN1_IA5STRING) |
175 | 0 | str_type = V_ASN1_IA5STRING; |
176 | 1.18M | else if (mask & B_ASN1_T61STRING) |
177 | 0 | str_type = V_ASN1_T61STRING; |
178 | 1.18M | else if (mask & B_ASN1_BMPSTRING) { |
179 | 0 | str_type = V_ASN1_BMPSTRING; |
180 | 0 | outform = MBSTRING_BMP; |
181 | 1.18M | } else if (mask & B_ASN1_UNIVERSALSTRING) { |
182 | 0 | str_type = V_ASN1_UNIVERSALSTRING; |
183 | 0 | outform = MBSTRING_UNIV; |
184 | 1.18M | } else { |
185 | 1.18M | str_type = V_ASN1_UTF8STRING; |
186 | 1.18M | outform = MBSTRING_UTF8; |
187 | 1.18M | } |
188 | 1.18M | if (!out) |
189 | 0 | return str_type; |
190 | 1.18M | if (*out) { |
191 | 1.18M | free_out = 0; |
192 | 1.18M | dest = *out; |
193 | 1.18M | if (dest->data) { |
194 | 0 | dest->length = 0; |
195 | 0 | OPENSSL_free(dest->data); |
196 | 0 | dest->data = NULL; |
197 | 0 | } |
198 | 1.18M | dest->type = str_type; |
199 | 1.18M | } else { |
200 | 0 | free_out = 1; |
201 | 0 | dest = ASN1_STRING_type_new(str_type); |
202 | 0 | if (!dest) { |
203 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ERR_R_MALLOC_FAILURE); |
204 | 0 | return -1; |
205 | 0 | } |
206 | 0 | *out = dest; |
207 | 0 | } |
208 | | /* If both the same type just copy across */ |
209 | 1.18M | if (inform == outform) { |
210 | 295k | if (!ASN1_STRING_set(dest, in, len)) { |
211 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ERR_R_MALLOC_FAILURE); |
212 | 0 | return -1; |
213 | 0 | } |
214 | 295k | return str_type; |
215 | 295k | } |
216 | | |
217 | | /* Work out how much space the destination will need */ |
218 | 892k | switch (outform) { |
219 | 0 | case MBSTRING_ASC: |
220 | 0 | outlen = nchar; |
221 | 0 | cpyfunc = cpy_asc; |
222 | 0 | break; |
223 | | |
224 | 0 | case MBSTRING_BMP: |
225 | 0 | outlen = nchar << 1; |
226 | 0 | cpyfunc = cpy_bmp; |
227 | 0 | break; |
228 | | |
229 | 0 | case MBSTRING_UNIV: |
230 | 0 | outlen = nchar << 2; |
231 | 0 | cpyfunc = cpy_univ; |
232 | 0 | break; |
233 | | |
234 | 892k | case MBSTRING_UTF8: |
235 | 892k | outlen = 0; |
236 | 892k | traverse_string(in, len, inform, out_utf8, &outlen); |
237 | 892k | cpyfunc = cpy_utf8; |
238 | 892k | break; |
239 | 892k | } |
240 | 892k | if (!(p = OPENSSL_malloc(outlen + 1))) { |
241 | 0 | if (free_out) |
242 | 0 | ASN1_STRING_free(dest); |
243 | 0 | ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ERR_R_MALLOC_FAILURE); |
244 | 0 | return -1; |
245 | 0 | } |
246 | 892k | dest->length = outlen; |
247 | 892k | dest->data = p; |
248 | 892k | p[outlen] = 0; |
249 | 892k | traverse_string(in, len, inform, cpyfunc, &p); |
250 | 892k | return str_type; |
251 | 892k | } |
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 | | |
258 | | static int traverse_string(const unsigned char *p, int len, int inform, |
259 | | int (*rfunc) (unsigned long value, void *in), |
260 | | void *arg) |
261 | 3.26M | { |
262 | 3.26M | unsigned long value; |
263 | 3.26M | int ret; |
264 | 55.6M | while (len) { |
265 | 52.3M | if (inform == MBSTRING_ASC) { |
266 | 40.8M | value = *p++; |
267 | 40.8M | len--; |
268 | 40.8M | } else if (inform == MBSTRING_BMP) { |
269 | 0 | value = *p++ << 8; |
270 | 0 | value |= *p++; |
271 | 0 | len -= 2; |
272 | 11.5M | } else if (inform == MBSTRING_UNIV) { |
273 | 0 | value = ((unsigned long)*p++) << 24; |
274 | 0 | value |= ((unsigned long)*p++) << 16; |
275 | 0 | value |= *p++ << 8; |
276 | 0 | value |= *p++; |
277 | 0 | len -= 4; |
278 | 11.5M | } else { |
279 | 11.5M | ret = UTF8_getc(p, len, &value); |
280 | 11.5M | if (ret < 0) |
281 | 0 | return -1; |
282 | 11.5M | len -= ret; |
283 | 11.5M | p += ret; |
284 | 11.5M | } |
285 | 52.3M | if (rfunc) { |
286 | 52.3M | ret = rfunc(value, arg); |
287 | 52.3M | if (ret <= 0) |
288 | 0 | return ret; |
289 | 52.3M | } |
290 | 52.3M | } |
291 | 3.26M | return 1; |
292 | 3.26M | } |
293 | | |
294 | | /* Various utility functions for traverse_string */ |
295 | | |
296 | | /* Just count number of characters */ |
297 | | |
298 | | static int in_utf8(unsigned long value, void *arg) |
299 | 5.76M | { |
300 | 5.76M | int *nchar; |
301 | 5.76M | nchar = arg; |
302 | 5.76M | (*nchar)++; |
303 | 5.76M | return 1; |
304 | 5.76M | } |
305 | | |
306 | | /* Determine size of output as a UTF8 String */ |
307 | | |
308 | | static int out_utf8(unsigned long value, void *arg) |
309 | 13.6M | { |
310 | 13.6M | int *outlen; |
311 | 13.6M | outlen = arg; |
312 | 13.6M | *outlen += UTF8_putc(NULL, -1, value); |
313 | 13.6M | return 1; |
314 | 13.6M | } |
315 | | |
316 | | /* |
317 | | * Determine the "type" of a string: check each character against a supplied |
318 | | * "mask". |
319 | | */ |
320 | | |
321 | | static int type_str(unsigned long value, void *arg) |
322 | 19.3M | { |
323 | 19.3M | unsigned long types; |
324 | 19.3M | types = *((unsigned long *)arg); |
325 | 19.3M | if ((types & B_ASN1_PRINTABLESTRING) && !is_printable(value)) |
326 | 0 | types &= ~B_ASN1_PRINTABLESTRING; |
327 | 19.3M | if ((types & B_ASN1_IA5STRING) && (value > 127)) |
328 | 0 | types &= ~B_ASN1_IA5STRING; |
329 | 19.3M | if ((types & B_ASN1_T61STRING) && (value > 0xff)) |
330 | 0 | types &= ~B_ASN1_T61STRING; |
331 | 19.3M | if ((types & B_ASN1_BMPSTRING) && (value > 0xffff)) |
332 | 0 | types &= ~B_ASN1_BMPSTRING; |
333 | 19.3M | if (!types) |
334 | 0 | return -1; |
335 | 19.3M | *((unsigned long *)arg) = types; |
336 | 19.3M | return 1; |
337 | 19.3M | } |
338 | | |
339 | | /* Copy one byte per character ASCII like strings */ |
340 | | |
341 | | static int cpy_asc(unsigned long value, void *arg) |
342 | 0 | { |
343 | 0 | unsigned char **p, *q; |
344 | 0 | p = arg; |
345 | 0 | q = *p; |
346 | 0 | *q = (unsigned char)value; |
347 | 0 | (*p)++; |
348 | 0 | return 1; |
349 | 0 | } |
350 | | |
351 | | /* Copy two byte per character BMPStrings */ |
352 | | |
353 | | static int cpy_bmp(unsigned long value, void *arg) |
354 | 0 | { |
355 | 0 | unsigned char **p, *q; |
356 | 0 | p = arg; |
357 | 0 | q = *p; |
358 | 0 | *q++ = (unsigned char)((value >> 8) & 0xff); |
359 | 0 | *q = (unsigned char)(value & 0xff); |
360 | 0 | *p += 2; |
361 | 0 | return 1; |
362 | 0 | } |
363 | | |
364 | | /* Copy four byte per character UniversalStrings */ |
365 | | |
366 | | static int cpy_univ(unsigned long value, void *arg) |
367 | 0 | { |
368 | 0 | unsigned char **p, *q; |
369 | 0 | p = arg; |
370 | 0 | q = *p; |
371 | 0 | *q++ = (unsigned char)((value >> 24) & 0xff); |
372 | 0 | *q++ = (unsigned char)((value >> 16) & 0xff); |
373 | 0 | *q++ = (unsigned char)((value >> 8) & 0xff); |
374 | 0 | *q = (unsigned char)(value & 0xff); |
375 | 0 | *p += 4; |
376 | 0 | return 1; |
377 | 0 | } |
378 | | |
379 | | /* Copy to a UTF8String */ |
380 | | |
381 | | static int cpy_utf8(unsigned long value, void *arg) |
382 | 13.6M | { |
383 | 13.6M | unsigned char **p; |
384 | 13.6M | int ret; |
385 | 13.6M | p = arg; |
386 | | /* We already know there is enough room so pass 0xff as the length */ |
387 | 13.6M | ret = UTF8_putc(*p, 0xff, value); |
388 | 13.6M | *p += ret; |
389 | 13.6M | return 1; |
390 | 13.6M | } |
391 | | |
392 | | /* Return 1 if the character is permitted in a PrintableString */ |
393 | | static int is_printable(unsigned long value) |
394 | 0 | { |
395 | 0 | int ch; |
396 | 0 | if (value > 0x7f) |
397 | 0 | return 0; |
398 | 0 | 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 | 0 | #ifndef CHARSET_EBCDIC |
404 | 0 | if ((ch >= 'a') && (ch <= 'z')) |
405 | 0 | return 1; |
406 | 0 | if ((ch >= 'A') && (ch <= 'Z')) |
407 | 0 | return 1; |
408 | 0 | if ((ch >= '0') && (ch <= '9')) |
409 | 0 | return 1; |
410 | 0 | if ((ch == ' ') || strchr("'()+,-./:=?", ch)) |
411 | 0 | 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 | 0 | return 0; |
423 | 0 | } |