1/* a_mbstr.c */
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 <stdio.h>
60#include <ctype.h>
61#include "cryptlib.h"
62#include <openssl/asn1.h>
63
64static int traverse_string(const unsigned char *p, int len, int inform,
65		 int (*rfunc)(unsigned long value, void *in), void *arg);
66static int in_utf8(unsigned long value, void *arg);
67static int out_utf8(unsigned long value, void *arg);
68static int type_str(unsigned long value, void *arg);
69static int cpy_asc(unsigned long value, void *arg);
70static int cpy_bmp(unsigned long value, void *arg);
71static int cpy_univ(unsigned long value, void *arg);
72static int cpy_utf8(unsigned long value, void *arg);
73static int is_printable(unsigned long value);
74
75/* These functions take a string in UTF8, ASCII or multibyte form and
76 * a mask of permissible ASN1 string types. It then works out the minimal
77 * type (using the order Printable < IA5 < T61 < BMP < Universal < UTF8)
78 * and creates a string of the correct type with the supplied data.
79 * Yes this is horrible: it has to be :-(
80 * The 'ncopy' form checks minimum and maximum size limits too.
81 */
82
83int ASN1_mbstring_copy(ASN1_STRING **out, const unsigned char *in, int len,
84					int inform, unsigned long mask)
85{
86	return ASN1_mbstring_ncopy(out, in, len, inform, mask, 0, 0);
87}
88
89int ASN1_mbstring_ncopy(ASN1_STRING **out, const unsigned char *in, int len,
90					int inform, unsigned long mask,
91					long minsize, long maxsize)
92{
93	int str_type;
94	int ret;
95	char free_out;
96	int outform, outlen = 0;
97	ASN1_STRING *dest;
98	unsigned char *p;
99	int nchar;
100	char strbuf[32];
101	int (*cpyfunc)(unsigned long,void *) = NULL;
102	if(len == -1) len = strlen((const char *)in);
103	if(!mask) mask = DIRSTRING_TYPE;
104
105	/* First do a string check and work out the number of characters */
106	switch(inform) {
107
108		case MBSTRING_BMP:
109		if(len & 1) {
110			ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY,
111					 ASN1_R_INVALID_BMPSTRING_LENGTH);
112			return -1;
113		}
114		nchar = len >> 1;
115		break;
116
117		case MBSTRING_UNIV:
118		if(len & 3) {
119			ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY,
120					 ASN1_R_INVALID_UNIVERSALSTRING_LENGTH);
121			return -1;
122		}
123		nchar = len >> 2;
124		break;
125
126		case MBSTRING_UTF8:
127		nchar = 0;
128		/* This counts the characters and does utf8 syntax checking */
129		ret = traverse_string(in, len, MBSTRING_UTF8, in_utf8, &nchar);
130		if(ret < 0) {
131			ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY,
132						 ASN1_R_INVALID_UTF8STRING);
133			return -1;
134		}
135		break;
136
137		case MBSTRING_ASC:
138		nchar = len;
139		break;
140
141		default:
142		ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_UNKNOWN_FORMAT);
143		return -1;
144	}
145
146	if((minsize > 0) && (nchar < minsize)) {
147		ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_STRING_TOO_SHORT);
148		BIO_snprintf(strbuf, sizeof strbuf, "%ld", minsize);
149		ERR_add_error_data(2, "minsize=", strbuf);
150		return -1;
151	}
152
153	if((maxsize > 0) && (nchar > maxsize)) {
154		ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_STRING_TOO_LONG);
155		BIO_snprintf(strbuf, sizeof strbuf, "%ld", maxsize);
156		ERR_add_error_data(2, "maxsize=", strbuf);
157		return -1;
158	}
159
160	/* Now work out minimal type (if any) */
161	if(traverse_string(in, len, inform, type_str, &mask) < 0) {
162		ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY, ASN1_R_ILLEGAL_CHARACTERS);
163		return -1;
164	}
165
166
167	/* Now work out output format and string type */
168	outform = MBSTRING_ASC;
169	if(mask & B_ASN1_PRINTABLESTRING) str_type = V_ASN1_PRINTABLESTRING;
170	else if(mask & B_ASN1_IA5STRING) str_type = V_ASN1_IA5STRING;
171	else if(mask & B_ASN1_T61STRING) str_type = V_ASN1_T61STRING;
172	else if(mask & B_ASN1_BMPSTRING) {
173		str_type = V_ASN1_BMPSTRING;
174		outform = MBSTRING_BMP;
175	} else if(mask & B_ASN1_UNIVERSALSTRING) {
176		str_type = V_ASN1_UNIVERSALSTRING;
177		outform = MBSTRING_UNIV;
178	} else {
179		str_type = V_ASN1_UTF8STRING;
180		outform = MBSTRING_UTF8;
181	}
182	if(!out) return str_type;
183	if(*out) {
184		free_out = 0;
185		dest = *out;
186		if(dest->data) {
187			dest->length = 0;
188			OPENSSL_free(dest->data);
189			dest->data = NULL;
190		}
191		dest->type = str_type;
192	} else {
193		free_out = 1;
194		dest = ASN1_STRING_type_new(str_type);
195		if(!dest) {
196			ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY,
197							ERR_R_MALLOC_FAILURE);
198			return -1;
199		}
200		*out = dest;
201	}
202	/* If both the same type just copy across */
203	if(inform == outform) {
204		if(!ASN1_STRING_set(dest, in, len)) {
205			ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY,ERR_R_MALLOC_FAILURE);
206			return -1;
207		}
208		return str_type;
209	}
210
211	/* Work out how much space the destination will need */
212	switch(outform) {
213		case MBSTRING_ASC:
214		outlen = nchar;
215		cpyfunc = cpy_asc;
216		break;
217
218		case MBSTRING_BMP:
219		outlen = nchar << 1;
220		cpyfunc = cpy_bmp;
221		break;
222
223		case MBSTRING_UNIV:
224		outlen = nchar << 2;
225		cpyfunc = cpy_univ;
226		break;
227
228		case MBSTRING_UTF8:
229		outlen = 0;
230		traverse_string(in, len, inform, out_utf8, &outlen);
231		cpyfunc = cpy_utf8;
232		break;
233	}
234	if(!(p = OPENSSL_malloc(outlen + 1))) {
235		if(free_out) ASN1_STRING_free(dest);
236		ASN1err(ASN1_F_ASN1_MBSTRING_NCOPY,ERR_R_MALLOC_FAILURE);
237		return -1;
238	}
239	dest->length = outlen;
240	dest->data = p;
241	p[outlen] = 0;
242	traverse_string(in, len, inform, cpyfunc, &p);
243	return str_type;
244}
245
246/* This function traverses a string and passes the value of each character
247 * to an optional function along with a void * argument.
248 */
249
250static int traverse_string(const unsigned char *p, int len, int inform,
251		 int (*rfunc)(unsigned long value, void *in), void *arg)
252{
253	unsigned long value;
254	int ret;
255	while(len) {
256		if(inform == MBSTRING_ASC) {
257			value = *p++;
258			len--;
259		} else if(inform == MBSTRING_BMP) {
260			value = *p++ << 8;
261			value |= *p++;
262			len -= 2;
263		} else if(inform == MBSTRING_UNIV) {
264			value = ((unsigned long)*p++) << 24;
265			value |= ((unsigned long)*p++) << 16;
266			value |= *p++ << 8;
267			value |= *p++;
268			len -= 4;
269		} else {
270			ret = UTF8_getc(p, len, &value);
271			if(ret < 0) return -1;
272			len -= ret;
273			p += ret;
274		}
275		if(rfunc) {
276			ret = rfunc(value, arg);
277			if(ret <= 0) return ret;
278		}
279	}
280	return 1;
281}
282
283/* Various utility functions for traverse_string */
284
285/* Just count number of characters */
286
287static int in_utf8(unsigned long value, void *arg)
288{
289	int *nchar;
290	nchar = arg;
291	(*nchar)++;
292	return 1;
293}
294
295/* Determine size of output as a UTF8 String */
296
297static int out_utf8(unsigned long value, void *arg)
298{
299	int *outlen;
300	outlen = arg;
301	*outlen += UTF8_putc(NULL, -1, value);
302	return 1;
303}
304
305/* Determine the "type" of a string: check each character against a
306 * supplied "mask".
307 */
308
309static int type_str(unsigned long value, void *arg)
310{
311	unsigned long types;
312	types = *((unsigned long *)arg);
313	if((types & B_ASN1_PRINTABLESTRING) && !is_printable(value))
314					types &= ~B_ASN1_PRINTABLESTRING;
315	if((types & B_ASN1_IA5STRING) && (value > 127))
316					types &= ~B_ASN1_IA5STRING;
317	if((types & B_ASN1_T61STRING) && (value > 0xff))
318					types &= ~B_ASN1_T61STRING;
319	if((types & B_ASN1_BMPSTRING) && (value > 0xffff))
320					types &= ~B_ASN1_BMPSTRING;
321	if(!types) return -1;
322	*((unsigned long *)arg) = types;
323	return 1;
324}
325
326/* Copy one byte per character ASCII like strings */
327
328static int cpy_asc(unsigned long value, void *arg)
329{
330	unsigned char **p, *q;
331	p = arg;
332	q = *p;
333	*q = (unsigned char) value;
334	(*p)++;
335	return 1;
336}
337
338/* Copy two byte per character BMPStrings */
339
340static int cpy_bmp(unsigned long value, void *arg)
341{
342	unsigned char **p, *q;
343	p = arg;
344	q = *p;
345	*q++ = (unsigned char) ((value >> 8) & 0xff);
346	*q = (unsigned char) (value & 0xff);
347	*p += 2;
348	return 1;
349}
350
351/* Copy four byte per character UniversalStrings */
352
353static int cpy_univ(unsigned long value, void *arg)
354{
355	unsigned char **p, *q;
356	p = arg;
357	q = *p;
358	*q++ = (unsigned char) ((value >> 24) & 0xff);
359	*q++ = (unsigned char) ((value >> 16) & 0xff);
360	*q++ = (unsigned char) ((value >> 8) & 0xff);
361	*q = (unsigned char) (value & 0xff);
362	*p += 4;
363	return 1;
364}
365
366/* Copy to a UTF8String */
367
368static int cpy_utf8(unsigned long value, void *arg)
369{
370	unsigned char **p;
371	int ret;
372	p = arg;
373	/* We already know there is enough room so pass 0xff as the length */
374	ret = UTF8_putc(*p, 0xff, value);
375	*p += ret;
376	return 1;
377}
378
379/* Return 1 if the character is permitted in a PrintableString */
380static int is_printable(unsigned long value)
381{
382	int ch;
383	if(value > 0x7f) return 0;
384	ch = (int) value;
385	/* Note: we can't use 'isalnum' because certain accented
386	 * characters may count as alphanumeric in some environments.
387	 */
388#ifndef CHARSET_EBCDIC
389	if((ch >= 'a') && (ch <= 'z')) return 1;
390	if((ch >= 'A') && (ch <= 'Z')) return 1;
391	if((ch >= '0') && (ch <= '9')) return 1;
392	if ((ch == ' ') || strchr("'()+,-./:=?", ch)) return 1;
393#else /*CHARSET_EBCDIC*/
394	if((ch >= os_toascii['a']) && (ch <= os_toascii['z'])) return 1;
395	if((ch >= os_toascii['A']) && (ch <= os_toascii['Z'])) return 1;
396	if((ch >= os_toascii['0']) && (ch <= os_toascii['9'])) return 1;
397	if ((ch == os_toascii[' ']) || strchr("'()+,-./:=?", os_toebcdic[ch])) return 1;
398#endif /*CHARSET_EBCDIC*/
399	return 0;
400}
401