1 /*
2 * Copyright (C) 2004, 2006, 2007, 2008, 2009 Apple Inc. All rights reserved.
3 * Copyright (C) 2006 Alexey Proskuryakov <ap@nypop.com>
4 * Copyright (C) 2007-2009 Torch Mobile, Inc.
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 COMPUTER, 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 COMPUTER, 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 "wtf/text/TextEncoding.h"
30
31 #include "wtf/text/TextEncodingRegistry.h"
32 #include <unicode/unorm.h>
33 #include "wtf/OwnPtr.h"
34 #include "wtf/StdLibExtras.h"
35 #include "wtf/text/CString.h"
36 #include "wtf/text/WTFString.h"
37
38 namespace WTF {
39
UTF7Encoding()40 static const TextEncoding& UTF7Encoding()
41 {
42 static TextEncoding globalUTF7Encoding("UTF-7");
43 return globalUTF7Encoding;
44 }
45
TextEncoding(const char * name)46 TextEncoding::TextEncoding(const char* name)
47 : m_name(atomicCanonicalTextEncodingName(name))
48 , m_backslashAsCurrencySymbol(backslashAsCurrencySymbol())
49 {
50 }
51
TextEncoding(const String & name)52 TextEncoding::TextEncoding(const String& name)
53 : m_name(atomicCanonicalTextEncodingName(name))
54 , m_backslashAsCurrencySymbol(backslashAsCurrencySymbol())
55 {
56 }
57
decode(const char * data,size_t length,bool stopOnError,bool & sawError) const58 String TextEncoding::decode(const char* data, size_t length, bool stopOnError, bool& sawError) const
59 {
60 if (!m_name)
61 return String();
62
63 return newTextCodec(*this)->decode(data, length, true, stopOnError, sawError);
64 }
65
encode(const String & string,UnencodableHandling handling) const66 CString TextEncoding::encode(const String& string, UnencodableHandling handling) const
67 {
68 if (!m_name)
69 return CString();
70
71 if (string.isEmpty())
72 return "";
73
74 OwnPtr<TextCodec> textCodec = newTextCodec(*this);
75 CString encodedString;
76 if (string.is8Bit())
77 encodedString = textCodec->encode(string.characters8(), string.length(), handling);
78 else
79 encodedString = textCodec->encode(string.characters16(), string.length(), handling);
80 return encodedString;
81 }
82
normalizeAndEncode(const String & string,UnencodableHandling handling) const83 CString TextEncoding::normalizeAndEncode(const String& string, UnencodableHandling handling) const
84 {
85 if (!m_name)
86 return CString();
87
88 if (string.isEmpty())
89 return "";
90
91 // Text exclusively containing Latin-1 characters (U+0000..U+00FF) is left
92 // unaffected by NFC. This is effectively the same as saying that all
93 // Latin-1 text is already normalized to NFC.
94 // Source: http://unicode.org/reports/tr15/
95 if (string.is8Bit())
96 return newTextCodec(*this)->encode(string.characters8(), string.length(), handling);
97
98 const UChar* source = string.characters16();
99 size_t length = string.length();
100
101 Vector<UChar> normalizedCharacters;
102
103 UErrorCode err = U_ZERO_ERROR;
104 if (unorm_quickCheck(source, length, UNORM_NFC, &err) != UNORM_YES) {
105 // First try using the length of the original string, since normalization to NFC rarely increases length.
106 normalizedCharacters.grow(length);
107 int32_t normalizedLength = unorm_normalize(source, length, UNORM_NFC, 0, normalizedCharacters.data(), length, &err);
108 if (err == U_BUFFER_OVERFLOW_ERROR) {
109 err = U_ZERO_ERROR;
110 normalizedCharacters.resize(normalizedLength);
111 normalizedLength = unorm_normalize(source, length, UNORM_NFC, 0, normalizedCharacters.data(), normalizedLength, &err);
112 }
113 ASSERT(U_SUCCESS(err));
114
115 source = normalizedCharacters.data();
116 length = normalizedLength;
117 }
118
119 return newTextCodec(*this)->encode(source, length, handling);
120 }
121
usesVisualOrdering() const122 bool TextEncoding::usesVisualOrdering() const
123 {
124 if (noExtendedTextEncodingNameUsed())
125 return false;
126
127 static const char* const a = atomicCanonicalTextEncodingName("ISO-8859-8");
128 return m_name == a;
129 }
130
backslashAsCurrencySymbol() const131 UChar TextEncoding::backslashAsCurrencySymbol() const
132 {
133 return shouldShowBackslashAsCurrencySymbolIn(m_name) ? 0x00A5 : '\\';
134 }
135
isNonByteBasedEncoding() const136 bool TextEncoding::isNonByteBasedEncoding() const
137 {
138 if (noExtendedTextEncodingNameUsed()) {
139 return *this == UTF16LittleEndianEncoding()
140 || *this == UTF16BigEndianEncoding();
141 }
142
143 return *this == UTF16LittleEndianEncoding()
144 || *this == UTF16BigEndianEncoding()
145 || *this == UTF32BigEndianEncoding()
146 || *this == UTF32LittleEndianEncoding();
147 }
148
isUTF7Encoding() const149 bool TextEncoding::isUTF7Encoding() const
150 {
151 if (noExtendedTextEncodingNameUsed())
152 return false;
153
154 return *this == UTF7Encoding();
155 }
156
closestByteBasedEquivalent() const157 const TextEncoding& TextEncoding::closestByteBasedEquivalent() const
158 {
159 if (isNonByteBasedEncoding())
160 return UTF8Encoding();
161 return *this;
162 }
163
164 // HTML5 specifies that UTF-8 be used in form submission when a form is
165 // is a part of a document in UTF-16 probably because UTF-16 is not a
166 // byte-based encoding and can contain 0x00. By extension, the same
167 // should be done for UTF-32. In case of UTF-7, it is a byte-based encoding,
168 // but it's fraught with problems and we'd rather steer clear of it.
encodingForFormSubmission() const169 const TextEncoding& TextEncoding::encodingForFormSubmission() const
170 {
171 if (isNonByteBasedEncoding() || isUTF7Encoding())
172 return UTF8Encoding();
173 return *this;
174 }
175
ASCIIEncoding()176 const TextEncoding& ASCIIEncoding()
177 {
178 static TextEncoding globalASCIIEncoding("ASCII");
179 return globalASCIIEncoding;
180 }
181
Latin1Encoding()182 const TextEncoding& Latin1Encoding()
183 {
184 static TextEncoding globalLatin1Encoding("latin1");
185 return globalLatin1Encoding;
186 }
187
UTF16BigEndianEncoding()188 const TextEncoding& UTF16BigEndianEncoding()
189 {
190 static TextEncoding globalUTF16BigEndianEncoding("UTF-16BE");
191 return globalUTF16BigEndianEncoding;
192 }
193
UTF16LittleEndianEncoding()194 const TextEncoding& UTF16LittleEndianEncoding()
195 {
196 static TextEncoding globalUTF16LittleEndianEncoding("UTF-16LE");
197 return globalUTF16LittleEndianEncoding;
198 }
199
UTF32BigEndianEncoding()200 const TextEncoding& UTF32BigEndianEncoding()
201 {
202 static TextEncoding globalUTF32BigEndianEncoding("UTF-32BE");
203 return globalUTF32BigEndianEncoding;
204 }
205
UTF32LittleEndianEncoding()206 const TextEncoding& UTF32LittleEndianEncoding()
207 {
208 static TextEncoding globalUTF32LittleEndianEncoding("UTF-32LE");
209 return globalUTF32LittleEndianEncoding;
210 }
211
UTF8Encoding()212 const TextEncoding& UTF8Encoding()
213 {
214 static TextEncoding globalUTF8Encoding("UTF-8");
215 ASSERT(globalUTF8Encoding.isValid());
216 return globalUTF8Encoding;
217 }
218
WindowsLatin1Encoding()219 const TextEncoding& WindowsLatin1Encoding()
220 {
221 static TextEncoding globalWindowsLatin1Encoding("WinLatin1");
222 return globalWindowsLatin1Encoding;
223 }
224
225 } // namespace WTF
226