1 /*	$NetBSD: selectiveencode.c,v 1.4 2014/12/10 04:37:56 christos Exp $	*/
2 
3 #ifndef lint
4 static char *rcsid = "Id: selectiveencode.c,v 1.1 2003/06/04 00:27:07 marka Exp ";
5 #endif
6 
7 /*
8  * Copyright (c) 2000,2002 Japan Network Information Center.
9  * All rights reserved.
10  *
11  * By using this file, you agree to the terms and conditions set forth bellow.
12  *
13  * 			LICENSE TERMS AND CONDITIONS
14  *
15  * The following License Terms and Conditions apply, unless a different
16  * license is obtained from Japan Network Information Center ("JPNIC"),
17  * a Japanese association, Kokusai-Kougyou-Kanda Bldg 6F, 2-3-4 Uchi-Kanda,
18  * Chiyoda-ku, Tokyo 101-0047, Japan.
19  *
20  * 1. Use, Modification and Redistribution (including distribution of any
21  *    modified or derived work) in source and/or binary forms is permitted
22  *    under this License Terms and Conditions.
23  *
24  * 2. Redistribution of source code must retain the copyright notices as they
25  *    appear in each source code file, this License Terms and Conditions.
26  *
27  * 3. Redistribution in binary form must reproduce the Copyright Notice,
28  *    this License Terms and Conditions, in the documentation and/or other
29  *    materials provided with the distribution.  For the purposes of binary
30  *    distribution the "Copyright Notice" refers to the following language:
31  *    "Copyright (c) 2000-2002 Japan Network Information Center.  All rights reserved."
32  *
33  * 4. The name of JPNIC may not be used to endorse or promote products
34  *    derived from this Software without specific prior written approval of
35  *    JPNIC.
36  *
37  * 5. Disclaimer/Limitation of Liability: THIS SOFTWARE IS PROVIDED BY JPNIC
38  *    "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
39  *    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
40  *    PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL JPNIC BE LIABLE
41  *    FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
42  *    CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
43  *    SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
44  *    BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
45  *    WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
46  *    OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
47  *    ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
48  */
49 
50 #include <config.h>
51 
52 #include <stddef.h>
53 
54 #include <idn/assert.h>
55 #include <idn/logmacro.h>
56 #include <idn/result.h>
57 #include <idn/utf8.h>
58 #include <idn/debug.h>
59 
60 #include "selectiveencode.h"
61 
62 static int	is_domain_delimiter(char c);
63 static char	*find_nonascii(const char *s);
64 
65 idn_result_t
66 idn_selectiveencode_findregion(const char *s,
67 			       char **startp, char **endp)
68 {
69 	char *non_ascii;
70 	char *start, *end;
71 
72 	assert(s != NULL && startp != NULL && endp != NULL);
73 
74 	TRACE(("idn_selectiveencode_findregion(s=\"%s\")\n",
75 	      idn__debug_xstring(s, 20)));
76 
77 	/*
78 	 * Scan the specified string looking for non-ascii character.
79 	 */
80 	if ((non_ascii = find_nonascii(s)) == NULL)
81 		return (idn_notfound);
82 
83 	/*
84 	 * Non-ascii character found.
85 	 * Determine the region to encode.
86 	 */
87 
88 	/*
89 	 * First, we scan backwards to find the beginning of the region
90 	 * that should be converted.
91 	 */
92 	start = non_ascii;
93 	while (start > s) {
94 		char *prev = idn_utf8_findfirstbyte(start - 1, s);
95 		if (is_domain_delimiter(*prev))
96 			break;			/* Found */
97 		start = prev;
98 	}
99 	*startp = start;
100 
101 	/*
102 	 * Next we scan forwards looking for the end of the region.
103 	 */
104 	end = non_ascii + idn_utf8_mblen(non_ascii);
105 	while (!is_domain_delimiter(*end))
106 		end += idn_utf8_mblen(end);
107 	*endp = end;
108 
109 	return (idn_success);
110 }
111 
112 static int
113 is_domain_delimiter(char c) {
114 	return ((unsigned char)c < 0x80 &&
115 		!('A' <= c && c <= 'Z') &&
116 		!('a' <= c && c <= 'z') &&
117 		!('0' <= c && c <= '9') &&
118 		c != '-' && c != '.');
119 }
120 
121 static char *
122 find_nonascii(const char *s) {
123 	while (*s != '\0' && (unsigned char)*s < 0x80)
124 		s++;
125 	if (*s == '\0')
126 		return (NULL);
127 	else
128 		return ((char *)s);
129 }
130