| 1 | /* |
| 2 | * Copyright (C) 2013 Google, Inc. All Rights Reserved. |
| 3 | * Copyright (C) 2015 Apple Inc. All Rights Reserved. |
| 4 | * |
| 5 | * Redistribution and use in source and binary forms, with or without |
| 6 | * modification, are permitted provided that the following conditions |
| 7 | * are met: |
| 8 | * 1. Redistributions of source code must retain the above copyright |
| 9 | * notice, this list of conditions and the following disclaimer. |
| 10 | * 2. Redistributions in binary form must reproduce the above copyright |
| 11 | * notice, this list of conditions and the following disclaimer in the |
| 12 | * documentation and/or other materials provided with the distribution. |
| 13 | * |
| 14 | * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY |
| 15 | * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| 16 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR |
| 17 | * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR |
| 18 | * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, |
| 19 | * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, |
| 20 | * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR |
| 21 | * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY |
| 22 | * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
| 23 | * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
| 24 | * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| 25 | */ |
| 26 | |
| 27 | #pragma once |
| 28 | |
| 29 | #include "HTMLToken.h" |
| 30 | |
| 31 | namespace WebCore { |
| 32 | |
| 33 | class AtomicHTMLToken { |
| 34 | public: |
| 35 | explicit AtomicHTMLToken(HTMLToken&); |
| 36 | AtomicHTMLToken(HTMLToken::Type, const AtomicString& name, Vector<Attribute>&& = { }); // Only StartTag or EndTag. |
| 37 | |
| 38 | AtomicHTMLToken(const AtomicHTMLToken&) = delete; |
| 39 | AtomicHTMLToken(AtomicHTMLToken&&) = default; |
| 40 | |
| 41 | HTMLToken::Type type() const; |
| 42 | |
| 43 | // StartTag, EndTag, DOCTYPE. |
| 44 | |
| 45 | void setName(const AtomicString&); |
| 46 | |
| 47 | const AtomicString& name() const; |
| 48 | |
| 49 | // DOCTYPE. |
| 50 | |
| 51 | bool forceQuirks() const; |
| 52 | String publicIdentifier() const; |
| 53 | String systemIdentifier() const; |
| 54 | |
| 55 | // StartTag, EndTag. |
| 56 | |
| 57 | Vector<Attribute>& attributes(); |
| 58 | |
| 59 | bool selfClosing() const; |
| 60 | const Vector<Attribute>& attributes() const; |
| 61 | |
| 62 | // Characters |
| 63 | |
| 64 | const UChar* characters() const; |
| 65 | unsigned charactersLength() const; |
| 66 | bool charactersIsAll8BitData() const; |
| 67 | |
| 68 | // Comment |
| 69 | |
| 70 | const String& comment() const; |
| 71 | |
| 72 | private: |
| 73 | HTMLToken::Type m_type; |
| 74 | |
| 75 | void initializeAttributes(const HTMLToken::AttributeList& attributes); |
| 76 | |
| 77 | AtomicString m_name; // StartTag, EndTag, DOCTYPE. |
| 78 | |
| 79 | String m_data; // Comment |
| 80 | |
| 81 | // We don't want to copy the characters out of the HTMLToken, so we keep a pointer to its buffer instead. |
| 82 | // This buffer is owned by the HTMLToken and causes a lifetime dependence between these objects. |
| 83 | // FIXME: Add a mechanism for "internalizing" the characters when the HTMLToken is destroyed. |
| 84 | const UChar* m_externalCharacters; // Character |
| 85 | unsigned m_externalCharactersLength; // Character |
| 86 | bool m_externalCharactersIsAll8BitData; // Character |
| 87 | |
| 88 | std::unique_ptr<DoctypeData> m_doctypeData; // DOCTYPE. |
| 89 | |
| 90 | bool m_selfClosing; // StartTag, EndTag. |
| 91 | Vector<Attribute> m_attributes; // StartTag, EndTag. |
| 92 | }; |
| 93 | |
| 94 | const Attribute* findAttribute(const Vector<Attribute>&, const QualifiedName&); |
| 95 | bool hasAttribute(const Vector<Attribute>&, const AtomicString& localName); |
| 96 | |
| 97 | inline HTMLToken::Type AtomicHTMLToken::type() const |
| 98 | { |
| 99 | return m_type; |
| 100 | } |
| 101 | |
| 102 | inline const AtomicString& AtomicHTMLToken::name() const |
| 103 | { |
| 104 | ASSERT(m_type == HTMLToken::StartTag || m_type == HTMLToken::EndTag || m_type == HTMLToken::DOCTYPE); |
| 105 | return m_name; |
| 106 | } |
| 107 | |
| 108 | inline void AtomicHTMLToken::setName(const AtomicString& name) |
| 109 | { |
| 110 | ASSERT(m_type == HTMLToken::StartTag || m_type == HTMLToken::EndTag || m_type == HTMLToken::DOCTYPE); |
| 111 | m_name = name; |
| 112 | } |
| 113 | |
| 114 | inline bool AtomicHTMLToken::selfClosing() const |
| 115 | { |
| 116 | ASSERT(m_type == HTMLToken::StartTag || m_type == HTMLToken::EndTag); |
| 117 | return m_selfClosing; |
| 118 | } |
| 119 | |
| 120 | inline Vector<Attribute>& AtomicHTMLToken::attributes() |
| 121 | { |
| 122 | ASSERT(m_type == HTMLToken::StartTag || m_type == HTMLToken::EndTag); |
| 123 | return m_attributes; |
| 124 | } |
| 125 | |
| 126 | inline const Vector<Attribute>& AtomicHTMLToken::attributes() const |
| 127 | { |
| 128 | ASSERT(m_type == HTMLToken::StartTag || m_type == HTMLToken::EndTag); |
| 129 | return m_attributes; |
| 130 | } |
| 131 | |
| 132 | inline const UChar* AtomicHTMLToken::characters() const |
| 133 | { |
| 134 | ASSERT(m_type == HTMLToken::Character); |
| 135 | return m_externalCharacters; |
| 136 | } |
| 137 | |
| 138 | inline unsigned AtomicHTMLToken::charactersLength() const |
| 139 | { |
| 140 | ASSERT(m_type == HTMLToken::Character); |
| 141 | return m_externalCharactersLength; |
| 142 | } |
| 143 | |
| 144 | inline bool AtomicHTMLToken::charactersIsAll8BitData() const |
| 145 | { |
| 146 | return m_externalCharactersIsAll8BitData; |
| 147 | } |
| 148 | |
| 149 | inline const String& AtomicHTMLToken::() const |
| 150 | { |
| 151 | ASSERT(m_type == HTMLToken::Comment); |
| 152 | return m_data; |
| 153 | } |
| 154 | |
| 155 | inline bool AtomicHTMLToken::forceQuirks() const |
| 156 | { |
| 157 | ASSERT(m_type == HTMLToken::DOCTYPE); |
| 158 | return m_doctypeData->forceQuirks; |
| 159 | } |
| 160 | |
| 161 | inline String AtomicHTMLToken::publicIdentifier() const |
| 162 | { |
| 163 | ASSERT(m_type == HTMLToken::DOCTYPE); |
| 164 | if (!m_doctypeData->hasPublicIdentifier) |
| 165 | return String(); |
| 166 | return StringImpl::create8BitIfPossible(m_doctypeData->publicIdentifier); |
| 167 | } |
| 168 | |
| 169 | inline String AtomicHTMLToken::systemIdentifier() const |
| 170 | { |
| 171 | if (!m_doctypeData->hasSystemIdentifier) |
| 172 | return String(); |
| 173 | return StringImpl::create8BitIfPossible(m_doctypeData->systemIdentifier); |
| 174 | } |
| 175 | |
| 176 | inline const Attribute* findAttribute(const Vector<Attribute>& attributes, const QualifiedName& name) |
| 177 | { |
| 178 | for (auto& attribute : attributes) { |
| 179 | if (attribute.name().matches(name)) |
| 180 | return &attribute; |
| 181 | } |
| 182 | return nullptr; |
| 183 | } |
| 184 | |
| 185 | inline bool hasAttribute(const Vector<Attribute>& attributes, const AtomicString& localName) |
| 186 | { |
| 187 | for (auto& attribute : attributes) { |
| 188 | if (attribute.localName() == localName) |
| 189 | return true; |
| 190 | } |
| 191 | return false; |
| 192 | } |
| 193 | |
| 194 | inline void AtomicHTMLToken::initializeAttributes(const HTMLToken::AttributeList& attributes) |
| 195 | { |
| 196 | unsigned size = attributes.size(); |
| 197 | if (!size) |
| 198 | return; |
| 199 | |
| 200 | m_attributes.reserveInitialCapacity(size); |
| 201 | for (auto& attribute : attributes) { |
| 202 | if (attribute.name.isEmpty()) |
| 203 | continue; |
| 204 | |
| 205 | AtomicString localName(attribute.name); |
| 206 | |
| 207 | // FIXME: This is N^2 for the number of attributes. |
| 208 | if (!hasAttribute(m_attributes, localName)) |
| 209 | m_attributes.uncheckedAppend(Attribute(QualifiedName(nullAtom(), localName, nullAtom()), AtomicString(attribute.value))); |
| 210 | } |
| 211 | } |
| 212 | |
| 213 | inline AtomicHTMLToken::AtomicHTMLToken(HTMLToken& token) |
| 214 | : m_type(token.type()) |
| 215 | { |
| 216 | switch (m_type) { |
| 217 | case HTMLToken::Uninitialized: |
| 218 | ASSERT_NOT_REACHED(); |
| 219 | return; |
| 220 | case HTMLToken::DOCTYPE: |
| 221 | m_name = AtomicString(token.name()); |
| 222 | m_doctypeData = token.releaseDoctypeData(); |
| 223 | return; |
| 224 | case HTMLToken::EndOfFile: |
| 225 | return; |
| 226 | case HTMLToken::StartTag: |
| 227 | case HTMLToken::EndTag: |
| 228 | m_selfClosing = token.selfClosing(); |
| 229 | m_name = AtomicString(token.name()); |
| 230 | initializeAttributes(token.attributes()); |
| 231 | return; |
| 232 | case HTMLToken::Comment: |
| 233 | if (token.commentIsAll8BitData()) |
| 234 | m_data = String::make8BitFrom16BitSource(token.comment()); |
| 235 | else |
| 236 | m_data = String(token.comment()); |
| 237 | return; |
| 238 | case HTMLToken::Character: |
| 239 | m_externalCharacters = token.characters().data(); |
| 240 | m_externalCharactersLength = token.characters().size(); |
| 241 | m_externalCharactersIsAll8BitData = token.charactersIsAll8BitData(); |
| 242 | return; |
| 243 | } |
| 244 | ASSERT_NOT_REACHED(); |
| 245 | } |
| 246 | |
| 247 | inline AtomicHTMLToken::AtomicHTMLToken(HTMLToken::Type type, const AtomicString& name, Vector<Attribute>&& attributes) |
| 248 | : m_type(type) |
| 249 | , m_name(name) |
| 250 | , m_selfClosing(false) |
| 251 | , m_attributes(WTFMove(attributes)) |
| 252 | { |
| 253 | ASSERT(type == HTMLToken::StartTag || type == HTMLToken::EndTag); |
| 254 | } |
| 255 | |
| 256 | } // namespace WebCore |
| 257 | |