1 /*
2 __ __ _
3 ___\ \/ /_ __ __ _| |_
4 / _ \\ /| '_ \ / _` | __|
5 | __// \| |_) | (_| | |_
6 \___/_/\_\ .__/ \__,_|\__|
7 |_| XML parser
8
9 Copyright (c) 1997-2000 Thai Open Source Software Center Ltd
10 Copyright (c) 2000-2017 Expat development team
11 Licensed under the MIT license:
12
13 Permission is hereby granted, free of charge, to any person obtaining
14 a copy of this software and associated documentation files (the
15 "Software"), to deal in the Software without restriction, including
16 without limitation the rights to use, copy, modify, merge, publish,
17 distribute, sublicense, and/or sell copies of the Software, and to permit
18 persons to whom the Software is furnished to do so, subject to the
19 following conditions:
20
21 The above copyright notice and this permission notice shall be included
22 in all copies or substantial portions of the Software.
23
24 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
25 EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
26 MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
27 NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
28 DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
29 OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
30 USE OR OTHER DEALINGS IN THE SOFTWARE.
31 */
32
33 #define CHARSET_MAX 41
34
35 static const char *
getTok(const char ** pp)36 getTok(const char **pp)
37 {
38 enum { inAtom, inString, init, inComment };
39 int state = init;
40 const char *tokStart = 0;
41 for (;;) {
42 switch (**pp) {
43 case '\0':
44 return 0;
45 case ' ':
46 case '\r':
47 case '\t':
48 case '\n':
49 if (state == inAtom)
50 return tokStart;
51 break;
52 case '(':
53 if (state == inAtom)
54 return tokStart;
55 if (state != inString)
56 state++;
57 break;
58 case ')':
59 if (state > init)
60 --state;
61 else if (state != inString)
62 return 0;
63 break;
64 case ';':
65 case '/':
66 case '=':
67 if (state == inAtom)
68 return tokStart;
69 if (state == init)
70 return (*pp)++;
71 break;
72 case '\\':
73 ++*pp;
74 if (**pp == '\0')
75 return 0;
76 break;
77 case '"':
78 switch (state) {
79 case inString:
80 ++*pp;
81 return tokStart;
82 case inAtom:
83 return tokStart;
84 case init:
85 tokStart = *pp;
86 state = inString;
87 break;
88 }
89 break;
90 default:
91 if (state == init) {
92 tokStart = *pp;
93 state = inAtom;
94 }
95 break;
96 }
97 ++*pp;
98 }
99 /* not reached */
100 }
101
102 /* key must be lowercase ASCII */
103
104 static int
matchkey(const char * start,const char * end,const char * key)105 matchkey(const char *start, const char *end, const char *key)
106 {
107 if (!start)
108 return 0;
109 for (; start != end; start++, key++)
110 if (*start != *key && *start != 'A' + (*key - 'a'))
111 return 0;
112 return *key == '\0';
113 }
114
115 void
getXMLCharset(const char * buf,char * charset)116 getXMLCharset(const char *buf, char *charset)
117 {
118 const char *next, *p;
119
120 charset[0] = '\0';
121 next = buf;
122 p = getTok(&next);
123 if (matchkey(p, next, "text"))
124 strcpy(charset, "us-ascii");
125 else if (!matchkey(p, next, "application"))
126 return;
127 p = getTok(&next);
128 if (!p || *p != '/')
129 return;
130 p = getTok(&next);
131 if (matchkey(p, next, "xml"))
132 isXml = 1;
133 p = getTok(&next);
134 while (p) {
135 if (*p == ';') {
136 p = getTok(&next);
137 if (matchkey(p, next, "charset")) {
138 p = getTok(&next);
139 if (p && *p == '=') {
140 p = getTok(&next);
141 if (p) {
142 char *s = charset;
143 if (*p == '"') {
144 while (++p != next - 1) {
145 if (*p == '\\')
146 ++p;
147 if (s == charset + CHARSET_MAX - 1) {
148 charset[0] = '\0';
149 break;
150 }
151 *s++ = *p;
152 }
153 *s++ = '\0';
154 }
155 else {
156 if (next - p > CHARSET_MAX - 1)
157 break;
158 while (p != next)
159 *s++ = *p++;
160 *s = 0;
161 break;
162 }
163 }
164 }
165 }
166 }
167 else
168 p = getTok(&next);
169 }
170 }
171
172 int
main(int argc,char ** argv)173 main(int argc, char **argv)
174 {
175 char buf[CHARSET_MAX];
176 getXMLCharset(argv[1], buf);
177 printf("charset = \"%s\"\n", buf);
178 return 0;
179 }
180