1 /*
2 * Copyright (C) 2008 Apple Inc. All Rights Reserved.
3 * Copyright (C) 2009 Torch Mobile, Inc. http://www.torchmobile.com/
4 * Copyright (C) 2010 Google Inc. All Rights Reserved.
5 *
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions
8 * are met:
9 * 1. Redistributions of source code must retain the above copyright
10 * notice, this list of conditions and the following disclaimer.
11 * 2. Redistributions in binary form must reproduce the above copyright
12 * notice, this list of conditions and the following disclaimer in the
13 * documentation and/or other materials provided with the distribution.
14 *
15 * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
16 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
17 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
18 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR
19 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
20 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
21 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
22 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
23 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
24 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
25 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
26 */
27
28 #include "config.h"
29 #include "HTMLPreloadScanner.h"
30
31 #include "CachedResourceLoader.h"
32 #include "Document.h"
33 #include "InputType.h"
34 #include "HTMLDocumentParser.h"
35 #include "HTMLTokenizer.h"
36 #include "HTMLLinkElement.h"
37 #include "HTMLNames.h"
38 #include "HTMLParserIdioms.h"
39 #include "MediaList.h"
40 #include "MediaQueryEvaluator.h"
41
42 namespace WebCore {
43
44 using namespace HTMLNames;
45
46 namespace {
47
48 class PreloadTask {
49 public:
PreloadTask(const HTMLToken & token)50 PreloadTask(const HTMLToken& token)
51 : m_tagName(token.name().data(), token.name().size())
52 , m_linkIsStyleSheet(false)
53 , m_linkMediaAttributeIsScreen(true)
54 , m_inputIsImage(false)
55 {
56 processAttributes(token.attributes());
57 }
58
processAttributes(const HTMLToken::AttributeList & attributes)59 void processAttributes(const HTMLToken::AttributeList& attributes)
60 {
61 if (m_tagName != imgTag
62 && m_tagName != inputTag
63 && m_tagName != linkTag
64 && m_tagName != scriptTag)
65 return;
66
67 for (HTMLToken::AttributeList::const_iterator iter = attributes.begin();
68 iter != attributes.end(); ++iter) {
69 AtomicString attributeName(iter->m_name.data(), iter->m_name.size());
70 String attributeValue(iter->m_value.data(), iter->m_value.size());
71
72 if (attributeName == charsetAttr)
73 m_charset = attributeValue;
74
75 if (m_tagName == scriptTag || m_tagName == imgTag) {
76 if (attributeName == srcAttr)
77 setUrlToLoad(attributeValue);
78 } else if (m_tagName == linkTag) {
79 if (attributeName == hrefAttr)
80 setUrlToLoad(attributeValue);
81 else if (attributeName == relAttr)
82 m_linkIsStyleSheet = relAttributeIsStyleSheet(attributeValue);
83 else if (attributeName == mediaAttr)
84 m_linkMediaAttributeIsScreen = linkMediaAttributeIsScreen(attributeValue);
85 } else if (m_tagName == inputTag) {
86 if (attributeName == srcAttr)
87 setUrlToLoad(attributeValue);
88 else if (attributeName == typeAttr)
89 m_inputIsImage = equalIgnoringCase(attributeValue, InputTypeNames::image());
90 }
91 }
92 }
93
relAttributeIsStyleSheet(const String & attributeValue)94 static bool relAttributeIsStyleSheet(const String& attributeValue)
95 {
96 HTMLLinkElement::RelAttribute rel;
97 HTMLLinkElement::tokenizeRelAttribute(attributeValue, rel);
98 return rel.m_isStyleSheet && !rel.m_isAlternate && !rel.m_isIcon && !rel.m_isDNSPrefetch;
99 }
100
linkMediaAttributeIsScreen(const String & attributeValue)101 static bool linkMediaAttributeIsScreen(const String& attributeValue)
102 {
103 if (attributeValue.isEmpty())
104 return true;
105 RefPtr<MediaList> mediaList = MediaList::createAllowingDescriptionSyntax(attributeValue);
106
107 // Only preload screen media stylesheets. Used this way, the evaluator evaluates to true for any
108 // rules containing complex queries (full evaluation is possible but it requires a frame and a style selector which
109 // may be problematic here).
110 MediaQueryEvaluator mediaQueryEvaluator("screen");
111 return mediaQueryEvaluator.eval(mediaList.get());
112 }
113
setUrlToLoad(const String & attributeValue)114 void setUrlToLoad(const String& attributeValue)
115 {
116 // We only respect the first src/href, per HTML5:
117 // http://www.whatwg.org/specs/web-apps/current-work/multipage/tokenization.html#attribute-name-state
118 if (!m_urlToLoad.isEmpty())
119 return;
120 m_urlToLoad = stripLeadingAndTrailingHTMLSpaces(attributeValue);
121 }
122
preload(Document * document,bool scanningBody)123 void preload(Document* document, bool scanningBody)
124 {
125 if (m_urlToLoad.isEmpty())
126 return;
127
128 CachedResourceLoader* cachedResourceLoader = document->cachedResourceLoader();
129 if (m_tagName == scriptTag)
130 cachedResourceLoader->preload(CachedResource::Script, m_urlToLoad, m_charset, scanningBody);
131 else if (m_tagName == imgTag || (m_tagName == inputTag && m_inputIsImage))
132 cachedResourceLoader->preload(CachedResource::ImageResource, m_urlToLoad, String(), scanningBody);
133 else if (m_tagName == linkTag && m_linkIsStyleSheet && m_linkMediaAttributeIsScreen)
134 cachedResourceLoader->preload(CachedResource::CSSStyleSheet, m_urlToLoad, m_charset, scanningBody);
135 }
136
tagName() const137 const AtomicString& tagName() const { return m_tagName; }
138
139 private:
140 AtomicString m_tagName;
141 String m_urlToLoad;
142 String m_charset;
143 bool m_linkIsStyleSheet;
144 bool m_linkMediaAttributeIsScreen;
145 bool m_inputIsImage;
146 };
147
148 } // namespace
149
HTMLPreloadScanner(Document * document)150 HTMLPreloadScanner::HTMLPreloadScanner(Document* document)
151 : m_document(document)
152 , m_cssScanner(document)
153 , m_tokenizer(HTMLTokenizer::create(HTMLDocumentParser::usePreHTML5ParserQuirks(document)))
154 , m_bodySeen(false)
155 , m_inStyle(false)
156 {
157 }
158
appendToEnd(const SegmentedString & source)159 void HTMLPreloadScanner::appendToEnd(const SegmentedString& source)
160 {
161 m_source.append(source);
162 }
163
scan()164 void HTMLPreloadScanner::scan()
165 {
166 // FIXME: We should save and re-use these tokens in HTMLDocumentParser if
167 // the pending script doesn't end up calling document.write.
168 while (m_tokenizer->nextToken(m_source, m_token)) {
169 processToken();
170 m_token.clear();
171 }
172 }
173
processToken()174 void HTMLPreloadScanner::processToken()
175 {
176 if (m_inStyle) {
177 if (m_token.type() == HTMLToken::Character)
178 m_cssScanner.scan(m_token, scanningBody());
179 else if (m_token.type() == HTMLToken::EndTag) {
180 m_inStyle = false;
181 m_cssScanner.reset();
182 }
183 }
184
185 if (m_token.type() != HTMLToken::StartTag)
186 return;
187
188 PreloadTask task(m_token);
189 m_tokenizer->updateStateFor(task.tagName(), m_document->frame());
190
191 if (task.tagName() == bodyTag)
192 m_bodySeen = true;
193
194 if (task.tagName() == styleTag)
195 m_inStyle = true;
196
197 task.preload(m_document, scanningBody());
198 }
199
scanningBody() const200 bool HTMLPreloadScanner::scanningBody() const
201 {
202 return m_document->body() || m_bodySeen;
203 }
204
205 }
206