1 /*
2 * Copyright (C) 2003 Lars Knoll (knoll@kde.org)
3 * Copyright (C) 2004, 2005, 2006, 2008, 2009, 2010 Apple Inc. All rights reserved.
4 * Copyright (C) 2008 Eric Seidel <eric@webkit.org>
5 * Copyright (C) 2009 - 2010 Torch Mobile (Beijing) Co. Ltd. All rights reserved.
6 *
7 * This library is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Library General Public
9 * License as published by the Free Software Foundation; either
10 * version 2 of the License, or (at your option) any later version.
11 *
12 * This library is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15 * Library General Public License for more details.
16 *
17 * You should have received a copy of the GNU Library General Public License
18 * along with this library; see the file COPYING.LIB. If not, write to
19 * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
20 * Boston, MA 02110-1301, USA.
21 */
22
23 #ifndef CSSTokenizer_h
24 #define CSSTokenizer_h
25
26 #include "wtf/Noncopyable.h"
27 #include "wtf/OwnPtr.h"
28 #include "wtf/text/WTFString.h"
29
30 namespace WebCore {
31
32 class CSSParser;
33 struct CSSParserLocation;
34 struct CSSParserString;
35
36 class CSSTokenizer {
37 WTF_MAKE_NONCOPYABLE(CSSTokenizer);
38 public:
39 // FIXME: This should not be needed but there are still some ties between the 2 classes.
40 friend class CSSParser;
41
CSSTokenizer(CSSParser & parser)42 CSSTokenizer(CSSParser& parser)
43 : m_parser(parser)
44 , m_parsedTextPrefixLength(0)
45 , m_parsedTextSuffixLength(0)
46 , m_parsingMode(NormalMode)
47 , m_is8BitSource(false)
48 , m_length(0)
49 , m_token(0)
50 , m_lineNumber(0)
51 , m_tokenStartLineNumber(0)
52 , m_internal(true)
53 {
54 m_tokenStart.ptr8 = 0;
55 }
56
57 void setupTokenizer(const char* prefix, unsigned prefixLength, const String&, const char* suffix, unsigned suffixLength);
58
59 CSSParserLocation currentLocation();
60
lex(void * yylval)61 inline int lex(void* yylval) { return (this->*m_lexFunc)(yylval); }
62
safeUserStringTokenOffset()63 inline unsigned safeUserStringTokenOffset()
64 {
65 return std::min(tokenStartOffset(), static_cast<unsigned>(m_length - 1 - m_parsedTextSuffixLength)) - m_parsedTextPrefixLength;
66 }
67
is8BitSource()68 bool is8BitSource() const { return m_is8BitSource; }
69
70 // FIXME: These 2 functions should be private so that we don't need the definitions below.
71 template <typename CharacterType>
72 inline CharacterType* tokenStart();
73
74 inline unsigned tokenStartOffset();
75
76 private:
77 UChar*& currentCharacter16();
78
79 template <typename CharacterType>
80 inline CharacterType*& currentCharacter();
81
82 template <typename CharacterType>
83 inline CharacterType* dataStart();
84
85 template <typename CharacterType>
86 inline void setTokenStart(CharacterType*);
87
88 template <typename CharacterType>
89 inline bool isIdentifierStart();
90
91 template <typename CharacterType>
92 inline CSSParserLocation tokenLocation();
93
94 template <typename CharacterType>
95 unsigned parseEscape(CharacterType*&);
96 template <typename DestCharacterType>
97 inline void UnicodeToChars(DestCharacterType*&, unsigned);
98 template <typename SrcCharacterType, typename DestCharacterType>
99 inline bool parseIdentifierInternal(SrcCharacterType*&, DestCharacterType*&, bool&);
100
101 template <typename CharacterType>
102 inline void parseIdentifier(CharacterType*&, CSSParserString&, bool&);
103
104 template <typename SrcCharacterType, typename DestCharacterType>
105 inline bool parseStringInternal(SrcCharacterType*&, DestCharacterType*&, UChar);
106
107 template <typename CharacterType>
108 inline void parseString(CharacterType*&, CSSParserString& resultString, UChar);
109
110 template <typename CharacterType>
111 inline bool findURI(CharacterType*& start, CharacterType*& end, UChar& quote);
112
113 template <typename SrcCharacterType, typename DestCharacterType>
114 inline bool parseURIInternal(SrcCharacterType*&, DestCharacterType*&, UChar quote);
115
116 template <typename CharacterType>
117 inline void parseURI(CSSParserString&);
118 template <typename CharacterType>
119 inline bool parseUnicodeRange();
120 template <typename CharacterType>
121 bool parseNthChild();
122 template <typename CharacterType>
123 bool parseNthChildExtra();
124 template <typename CharacterType>
125 inline bool detectFunctionTypeToken(int);
126 template <typename CharacterType>
127 inline void detectMediaQueryToken(int);
128 template <typename CharacterType>
129 inline void detectNumberToken(CharacterType*, int);
130 template <typename CharacterType>
131 inline void detectDashToken(int);
132 template <typename CharacterType>
133 inline void detectAtToken(int, bool);
134 template <typename CharacterType>
135 inline void detectSupportsToken(int);
136 template <typename CharacterType>
137 inline void detectCSSVariableDefinitionToken(int);
138
139 template <typename SourceCharacterType>
140 int realLex(void* yylval);
141
142 CSSParser& m_parser;
143
144 size_t m_parsedTextPrefixLength;
145 size_t m_parsedTextSuffixLength;
146
147 enum ParsingMode {
148 NormalMode,
149 MediaQueryMode,
150 SupportsMode,
151 NthChildMode
152 };
153
154 ParsingMode m_parsingMode;
155 bool m_is8BitSource;
156 OwnPtr<LChar[]> m_dataStart8;
157 OwnPtr<UChar[]> m_dataStart16;
158 LChar* m_currentCharacter8;
159 UChar* m_currentCharacter16;
160 union {
161 LChar* ptr8;
162 UChar* ptr16;
163 } m_tokenStart;
164 unsigned m_length;
165 int m_token;
166 int m_lineNumber;
167 int m_tokenStartLineNumber;
168
169 // FIXME: This boolean is misnamed. Also it would be nice if we could consolidate it
170 // with the CSSParserMode logic to determine if internal properties are allowed.
171 bool m_internal;
172
173 int (CSSTokenizer::*m_lexFunc)(void*);
174 };
175
tokenStartOffset()176 inline unsigned CSSTokenizer::tokenStartOffset()
177 {
178 if (is8BitSource())
179 return m_tokenStart.ptr8 - m_dataStart8.get();
180 return m_tokenStart.ptr16 - m_dataStart16.get();
181 }
182
183 template <>
184 inline LChar* CSSTokenizer::tokenStart<LChar>()
185 {
186 return m_tokenStart.ptr8;
187 }
188
189 template <>
190 inline UChar* CSSTokenizer::tokenStart<UChar>()
191 {
192 return m_tokenStart.ptr16;
193 }
194
195 } // namespace WebCore
196
197 #endif // CSSTokenizer_h
198