1 /*
2 * Copyright (C) 2010 Google Inc. All Rights Reserved.
3 *
4 * Redistribution and use in source and binary forms, with or without
5 * modification, are permitted provided that the following conditions
6 * are met:
7 * 1. Redistributions of source code must retain the above copyright
8 * notice, this list of conditions and the following disclaimer.
9 * 2. Redistributions in binary form must reproduce the above copyright
10 * notice, this list of conditions and the following disclaimer in the
11 * documentation and/or other materials provided with the distribution.
12 *
13 * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
14 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
15 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
16 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR
17 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
18 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
19 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
20 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
21 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
22 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
23 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
24 */
25
26 #include "config.h"
27 #include "HTMLMetaCharsetParser.h"
28
29 #include "HTMLNames.h"
30 #include "HTMLParserIdioms.h"
31 #include "HTMLTokenizer.h"
32 #include "PlatformString.h"
33 #include "TextCodec.h"
34 #include "TextEncodingRegistry.h"
35
36 using namespace WTF;
37
38 namespace WebCore {
39
40 using namespace HTMLNames;
41
HTMLMetaCharsetParser()42 HTMLMetaCharsetParser::HTMLMetaCharsetParser()
43 : m_tokenizer(HTMLTokenizer::create(false)) // No pre-HTML5 parser quirks.
44 , m_assumedCodec(newTextCodec(Latin1Encoding()))
45 , m_inHeadSection(true)
46 , m_doneChecking(false)
47 {
48 }
49
~HTMLMetaCharsetParser()50 HTMLMetaCharsetParser::~HTMLMetaCharsetParser()
51 {
52 }
53
54 static const char charsetString[] = "charset";
55 static const size_t charsetLength = sizeof("charset") - 1;
56
extractCharset(const String & value)57 String HTMLMetaCharsetParser::extractCharset(const String& value)
58 {
59 size_t pos = 0;
60 unsigned length = value.length();
61
62 while (pos < length) {
63 pos = value.find(charsetString, pos, false);
64 if (pos == notFound)
65 break;
66
67 pos += charsetLength;
68
69 // Skip whitespace.
70 while (pos < length && value[pos] <= ' ')
71 ++pos;
72
73 if (value[pos] != '=')
74 continue;
75
76 ++pos;
77
78 while (pos < length && value[pos] <= ' ')
79 ++pos;
80
81 char quoteMark = 0;
82 if (pos < length && (value[pos] == '"' || value[pos] == '\'')) {
83 quoteMark = static_cast<char>(value[pos++]);
84 ASSERT(!(quoteMark & 0x80));
85 }
86
87 if (pos == length)
88 break;
89
90 unsigned end = pos;
91 while (end < length && ((quoteMark && value[end] != quoteMark) || (!quoteMark && value[end] > ' ' && value[end] != '"' && value[end] != '\'' && value[end] != ';')))
92 ++end;
93
94 if (quoteMark && (end == length))
95 break; // Close quote not found.
96
97 return value.substring(pos, end - pos);
98 }
99
100 return "";
101 }
102
processMeta()103 bool HTMLMetaCharsetParser::processMeta()
104 {
105 bool gotPragma = false;
106 Mode mode = None;
107 String charset;
108
109 const HTMLToken::AttributeList& attributes = m_token.attributes();
110 for (HTMLToken::AttributeList::const_iterator iter = attributes.begin();
111 iter != attributes.end(); ++iter) {
112 AtomicString attributeName(iter->m_name.data(), iter->m_name.size());
113 String attributeValue(iter->m_value.data(), iter->m_value.size());
114
115 if (attributeName == http_equivAttr) {
116 if (equalIgnoringCase(attributeValue, "content-type"))
117 gotPragma = true;
118 } else if (charset.isEmpty()) {
119 if (attributeName == charsetAttr) {
120 charset = attributeValue;
121 mode = Charset;
122 } else if (attributeName == contentAttr) {
123 charset = extractCharset(attributeValue);
124 if (charset.length())
125 mode = Pragma;
126 }
127 }
128 }
129
130 if (mode == Charset || (mode == Pragma && gotPragma)) {
131 m_encoding = TextEncoding(stripLeadingAndTrailingHTMLSpaces(charset));
132 if (m_encoding.isValid())
133 return true;
134 }
135
136 return false;
137 }
138
139 static const int bytesToCheckUnconditionally = 1024; // That many input bytes will be checked for meta charset even if <head> section is over.
140
checkForMetaCharset(const char * data,size_t length)141 bool HTMLMetaCharsetParser::checkForMetaCharset(const char* data, size_t length)
142 {
143 if (m_doneChecking)
144 return true;
145
146 ASSERT(!m_encoding.isValid());
147
148 // We still don't have an encoding, and are in the head.
149 // The following tags are allowed in <head>:
150 // SCRIPT|STYLE|META|LINK|OBJECT|TITLE|BASE
151
152 // We stop scanning when a tag that is not permitted in <head>
153 // is seen, rather when </head> is seen, because that more closely
154 // matches behavior in other browsers; more details in
155 // <http://bugs.webkit.org/show_bug.cgi?id=3590>.
156
157 // Additionally, we ignore things that looks like tags in <title>, <script>
158 // and <noscript>; see <http://bugs.webkit.org/show_bug.cgi?id=4560>,
159 // <http://bugs.webkit.org/show_bug.cgi?id=12165> and
160 // <http://bugs.webkit.org/show_bug.cgi?id=12389>.
161
162 // Since many sites have charset declarations after <body> or other tags
163 // that are disallowed in <head>, we don't bail out until we've checked at
164 // least bytesToCheckUnconditionally bytes of input.
165
166 m_input.append(SegmentedString(m_assumedCodec->decode(data, length)));
167
168 while (m_tokenizer->nextToken(m_input, m_token)) {
169 bool end = m_token.type() == HTMLToken::EndTag;
170 if (end || m_token.type() == HTMLToken::StartTag) {
171 AtomicString tagName(m_token.name().data(), m_token.name().size());
172 if (!end) {
173 m_tokenizer->updateStateFor(tagName, 0);
174 if (tagName == metaTag && processMeta()) {
175 m_doneChecking = true;
176 return true;
177 }
178 }
179
180 if (tagName != scriptTag && tagName != noscriptTag
181 && tagName != styleTag && tagName != linkTag
182 && tagName != metaTag && tagName != objectTag
183 && tagName != titleTag && tagName != baseTag
184 && (end || tagName != htmlTag) && (end || tagName != headTag)) {
185 m_inHeadSection = false;
186 }
187 }
188
189 if (!m_inHeadSection && m_input.numberOfCharactersConsumed() >= bytesToCheckUnconditionally) {
190 m_doneChecking = true;
191 return true;
192 }
193
194 m_token.clear();
195 }
196
197 return false;
198 }
199
200 }
201