1/*
2                            __  __            _
3                         ___\ \/ /_ __   __ _| |_
4                        / _ \\  /| '_ \ / _` | __|
5                       |  __//  \| |_) | (_| | |_
6                        \___/_/\_\ .__/ \__,_|\__|
7                                 |_| XML parser
8
9   Copyright (c) 1997-2000 Thai Open Source Software Center Ltd
10   Copyright (c) 2002      Fred L. Drake, Jr. <fdrake@users.sourceforge.net>
11   Copyright (c) 2016-2018 Sebastian Pipping <sebastian@pipping.org>
12   Copyright (c) 2018      Marco Maggi <marco.maggi-ipsu@poste.it>
13   Licensed under the MIT license:
14
15   Permission is  hereby granted,  free of charge,  to any  person obtaining
16   a  copy  of  this  software   and  associated  documentation  files  (the
17   "Software"),  to  deal in  the  Software  without restriction,  including
18   without  limitation the  rights  to use,  copy,  modify, merge,  publish,
19   distribute, sublicense, and/or sell copies of the Software, and to permit
20   persons  to whom  the Software  is  furnished to  do so,  subject to  the
21   following conditions:
22
23   The above copyright  notice and this permission notice  shall be included
24   in all copies or substantial portions of the Software.
25
26   THE  SOFTWARE  IS  PROVIDED  "AS  IS",  WITHOUT  WARRANTY  OF  ANY  KIND,
27   EXPRESS  OR IMPLIED,  INCLUDING  BUT  NOT LIMITED  TO  THE WARRANTIES  OF
28   MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
29   NO EVENT SHALL THE AUTHORS OR  COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
30   DAMAGES OR  OTHER LIABILITY, WHETHER  IN AN  ACTION OF CONTRACT,  TORT OR
31   OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
32   USE OR OTHER DEALINGS IN THE SOFTWARE.
33*/
34
35#include <string.h>
36#include "xmlmime.h"
37
38static const char *
39getTok(const char **pp) {
40  /* inComment means one level of nesting; inComment+1 means two levels etc */
41  enum { inAtom, inString, init, inComment };
42  int state = init;
43  const char *tokStart = 0;
44  for (;;) {
45    switch (**pp) {
46    case '\0':
47      if (state == inAtom)
48        return tokStart;
49      return 0;
50    case ' ':
51    case '\r':
52    case '\t':
53    case '\n':
54      if (state == inAtom)
55        return tokStart;
56      break;
57    case '(':
58      if (state == inAtom)
59        return tokStart;
60      if (state != inString)
61        state++;
62      break;
63    case ')':
64      if (state > init)
65        --state;
66      else if (state != inString)
67        return 0;
68      break;
69    case ';':
70    case '/':
71    case '=':
72      if (state == inAtom)
73        return tokStart;
74      if (state == init)
75        return (*pp)++;
76      break;
77    case '\\':
78      ++*pp;
79      if (**pp == '\0')
80        return 0;
81      break;
82    case '"':
83      switch (state) {
84      case inString:
85        ++*pp;
86        return tokStart;
87      case inAtom:
88        return tokStart;
89      case init:
90        tokStart = *pp;
91        state = inString;
92        break;
93      }
94      break;
95    default:
96      if (state == init) {
97        tokStart = *pp;
98        state = inAtom;
99      }
100      break;
101    }
102    ++*pp;
103  }
104  /* not reached */
105}
106
107/* key must be lowercase ASCII */
108
109static int
110matchkey(const char *start, const char *end, const char *key) {
111  if (! start)
112    return 0;
113  for (; start != end; start++, key++)
114    if (*start != *key && *start != 'A' + (*key - 'a'))
115      return 0;
116  return *key == '\0';
117}
118
119void
120getXMLCharset(const char *buf, char *charset) {
121  const char *next, *p;
122
123  charset[0] = '\0';
124  next = buf;
125  p = getTok(&next);
126  if (matchkey(p, next, "text"))
127    strcpy(charset, "us-ascii");
128  else if (! matchkey(p, next, "application"))
129    return;
130  p = getTok(&next);
131  if (! p || *p != '/')
132    return;
133  p = getTok(&next);
134  /* BEGIN disabled code */
135  if (0) {
136    if (! matchkey(p, next, "xml") && charset[0] == '\0')
137      return;
138  }
139  /* END disabled code */
140  p = getTok(&next);
141  while (p) {
142    if (*p == ';') {
143      p = getTok(&next);
144      if (matchkey(p, next, "charset")) {
145        p = getTok(&next);
146        if (p && *p == '=') {
147          p = getTok(&next);
148          if (p) {
149            char *s = charset;
150            if (*p == '"') {
151              while (++p != next - 1) {
152                if (*p == '\\')
153                  ++p;
154                if (s == charset + CHARSET_MAX - 1) {
155                  charset[0] = '\0';
156                  break;
157                }
158                *s++ = *p;
159              }
160              *s++ = '\0';
161            } else {
162              if (next - p > CHARSET_MAX - 1)
163                break;
164              while (p != next)
165                *s++ = *p++;
166              *s = 0;
167              break;
168            }
169          }
170        }
171        break;
172      }
173    } else
174      p = getTok(&next);
175  }
176}
177
178#ifdef TEST
179
180#  include <stdio.h>
181
182int
183main(int argc, char *argv[]) {
184  char buf[CHARSET_MAX];
185  if (argc <= 1)
186    return 1;
187  printf("%s\n", argv[1]);
188  getXMLCharset(argv[1], buf);
189  printf("charset=\"%s\"\n", buf);
190  return 0;
191}
192
193#endif /* TEST */
194