1 | /* |
2 | * Copyright (C) 2004-2017 Apple 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 "TextCodecLatin1.h" |
28 | |
29 | #include "TextCodecASCIIFastPath.h" |
30 | #include <array> |
31 | #include <wtf/text/CString.h> |
32 | #include <wtf/text/WTFString.h> |
33 | |
34 | namespace WebCore { |
35 | |
36 | static const UChar latin1ConversionTable[256] = { |
37 | 0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0006, 0x0007, // 00-07 |
38 | 0x0008, 0x0009, 0x000A, 0x000B, 0x000C, 0x000D, 0x000E, 0x000F, // 08-0F |
39 | 0x0010, 0x0011, 0x0012, 0x0013, 0x0014, 0x0015, 0x0016, 0x0017, // 10-17 |
40 | 0x0018, 0x0019, 0x001A, 0x001B, 0x001C, 0x001D, 0x001E, 0x001F, // 18-1F |
41 | 0x0020, 0x0021, 0x0022, 0x0023, 0x0024, 0x0025, 0x0026, 0x0027, // 20-27 |
42 | 0x0028, 0x0029, 0x002A, 0x002B, 0x002C, 0x002D, 0x002E, 0x002F, // 28-2F |
43 | 0x0030, 0x0031, 0x0032, 0x0033, 0x0034, 0x0035, 0x0036, 0x0037, // 30-37 |
44 | 0x0038, 0x0039, 0x003A, 0x003B, 0x003C, 0x003D, 0x003E, 0x003F, // 38-3F |
45 | 0x0040, 0x0041, 0x0042, 0x0043, 0x0044, 0x0045, 0x0046, 0x0047, // 40-47 |
46 | 0x0048, 0x0049, 0x004A, 0x004B, 0x004C, 0x004D, 0x004E, 0x004F, // 48-4F |
47 | 0x0050, 0x0051, 0x0052, 0x0053, 0x0054, 0x0055, 0x0056, 0x0057, // 50-57 |
48 | 0x0058, 0x0059, 0x005A, 0x005B, 0x005C, 0x005D, 0x005E, 0x005F, // 58-5F |
49 | 0x0060, 0x0061, 0x0062, 0x0063, 0x0064, 0x0065, 0x0066, 0x0067, // 60-67 |
50 | 0x0068, 0x0069, 0x006A, 0x006B, 0x006C, 0x006D, 0x006E, 0x006F, // 68-6F |
51 | 0x0070, 0x0071, 0x0072, 0x0073, 0x0074, 0x0075, 0x0076, 0x0077, // 70-77 |
52 | 0x0078, 0x0079, 0x007A, 0x007B, 0x007C, 0x007D, 0x007E, 0x007F, // 78-7F |
53 | 0x20AC, 0x0081, 0x201A, 0x0192, 0x201E, 0x2026, 0x2020, 0x2021, // 80-87 |
54 | 0x02C6, 0x2030, 0x0160, 0x2039, 0x0152, 0x008D, 0x017D, 0x008F, // 88-8F |
55 | 0x0090, 0x2018, 0x2019, 0x201C, 0x201D, 0x2022, 0x2013, 0x2014, // 90-97 |
56 | 0x02DC, 0x2122, 0x0161, 0x203A, 0x0153, 0x009D, 0x017E, 0x0178, // 98-9F |
57 | 0x00A0, 0x00A1, 0x00A2, 0x00A3, 0x00A4, 0x00A5, 0x00A6, 0x00A7, // A0-A7 |
58 | 0x00A8, 0x00A9, 0x00AA, 0x00AB, 0x00AC, 0x00AD, 0x00AE, 0x00AF, // A8-AF |
59 | 0x00B0, 0x00B1, 0x00B2, 0x00B3, 0x00B4, 0x00B5, 0x00B6, 0x00B7, // B0-B7 |
60 | 0x00B8, 0x00B9, 0x00BA, 0x00BB, 0x00BC, 0x00BD, 0x00BE, 0x00BF, // B8-BF |
61 | 0x00C0, 0x00C1, 0x00C2, 0x00C3, 0x00C4, 0x00C5, 0x00C6, 0x00C7, // C0-C7 |
62 | 0x00C8, 0x00C9, 0x00CA, 0x00CB, 0x00CC, 0x00CD, 0x00CE, 0x00CF, // C8-CF |
63 | 0x00D0, 0x00D1, 0x00D2, 0x00D3, 0x00D4, 0x00D5, 0x00D6, 0x00D7, // D0-D7 |
64 | 0x00D8, 0x00D9, 0x00DA, 0x00DB, 0x00DC, 0x00DD, 0x00DE, 0x00DF, // D8-DF |
65 | 0x00E0, 0x00E1, 0x00E2, 0x00E3, 0x00E4, 0x00E5, 0x00E6, 0x00E7, // E0-E7 |
66 | 0x00E8, 0x00E9, 0x00EA, 0x00EB, 0x00EC, 0x00ED, 0x00EE, 0x00EF, // E8-EF |
67 | 0x00F0, 0x00F1, 0x00F2, 0x00F3, 0x00F4, 0x00F5, 0x00F6, 0x00F7, // F0-F7 |
68 | 0x00F8, 0x00F9, 0x00FA, 0x00FB, 0x00FC, 0x00FD, 0x00FE, 0x00FF // F8-FF |
69 | }; |
70 | |
71 | void TextCodecLatin1::registerEncodingNames(EncodingNameRegistrar registrar) |
72 | { |
73 | // From https://encoding.spec.whatwg.org. |
74 | registrar("windows-1252" , "windows-1252" ); |
75 | registrar("ansi_x3.4-1968" , "windows-1252" ); |
76 | registrar("ascii" , "windows-1252" ); |
77 | registrar("cp1252" , "windows-1252" ); |
78 | registrar("cp819" , "windows-1252" ); |
79 | registrar("csisolatin1" , "windows-1252" ); |
80 | registrar("ibm819" , "windows-1252" ); |
81 | registrar("iso-8859-1" , "windows-1252" ); |
82 | registrar("iso-ir-100" , "windows-1252" ); |
83 | registrar("iso8859-1" , "windows-1252" ); |
84 | registrar("iso88591" , "windows-1252" ); |
85 | registrar("iso_8859-1" , "windows-1252" ); |
86 | registrar("iso_8859-1:1987" , "windows-1252" ); |
87 | registrar("l1" , "windows-1252" ); |
88 | registrar("latin1" , "windows-1252" ); |
89 | registrar("us-ascii" , "windows-1252" ); |
90 | registrar("x-cp1252" , "windows-1252" ); |
91 | } |
92 | |
93 | void TextCodecLatin1::registerCodecs(TextCodecRegistrar registrar) |
94 | { |
95 | registrar("windows-1252" , [] { |
96 | return std::make_unique<TextCodecLatin1>(); |
97 | }); |
98 | } |
99 | |
100 | String TextCodecLatin1::decode(const char* bytes, size_t length, bool, bool, bool&) |
101 | { |
102 | LChar* characters; |
103 | if (!length) |
104 | return emptyString(); |
105 | String result = String::createUninitialized(length, characters); |
106 | |
107 | const uint8_t* source = reinterpret_cast<const uint8_t*>(bytes); |
108 | const uint8_t* end = reinterpret_cast<const uint8_t*>(bytes + length); |
109 | const uint8_t* alignedEnd = WTF::alignToMachineWord(end); |
110 | LChar* destination = characters; |
111 | |
112 | while (source < end) { |
113 | if (isASCII(*source)) { |
114 | // Fast path for ASCII. Most Latin-1 text will be ASCII. |
115 | if (WTF::isAlignedToMachineWord(source)) { |
116 | while (source < alignedEnd) { |
117 | auto chunk = *reinterpret_cast_ptr<const WTF::MachineWord*>(source); |
118 | |
119 | if (!WTF::isAllASCII<LChar>(chunk)) |
120 | goto useLookupTable; |
121 | |
122 | copyASCIIMachineWord(destination, source); |
123 | source += sizeof(WTF::MachineWord); |
124 | destination += sizeof(WTF::MachineWord); |
125 | } |
126 | |
127 | if (source == end) |
128 | break; |
129 | |
130 | // *source may not be ASCII anymore if source moves inside the loop of the fast code path |
131 | if (!isASCII(*source)) |
132 | goto useLookupTable; |
133 | } |
134 | *destination = *source; |
135 | } else { |
136 | useLookupTable: |
137 | if (latin1ConversionTable[*source] > 0xff) |
138 | goto upConvertTo16Bit; |
139 | |
140 | *destination = latin1ConversionTable[*source]; |
141 | } |
142 | |
143 | ++source; |
144 | ++destination; |
145 | } |
146 | |
147 | return result; |
148 | |
149 | upConvertTo16Bit: |
150 | UChar* characters16; |
151 | String result16 = String::createUninitialized(length, characters16); |
152 | |
153 | UChar* destination16 = characters16; |
154 | |
155 | // Zero extend and copy already processed 8 bit data |
156 | LChar* ptr8 = characters; |
157 | LChar* endPtr8 = destination; |
158 | |
159 | while (ptr8 < endPtr8) |
160 | *destination16++ = *ptr8++; |
161 | |
162 | // Handle the character that triggered the 16 bit path |
163 | *destination16 = latin1ConversionTable[*source]; |
164 | ++source; |
165 | ++destination16; |
166 | |
167 | while (source < end) { |
168 | if (isASCII(*source)) { |
169 | // Fast path for ASCII. Most Latin-1 text will be ASCII. |
170 | if (WTF::isAlignedToMachineWord(source)) { |
171 | while (source < alignedEnd) { |
172 | auto chunk = *reinterpret_cast_ptr<const WTF::MachineWord*>(source); |
173 | |
174 | if (!WTF::isAllASCII<LChar>(chunk)) |
175 | goto useLookupTable16; |
176 | |
177 | copyASCIIMachineWord(destination16, source); |
178 | source += sizeof(WTF::MachineWord); |
179 | destination16 += sizeof(WTF::MachineWord); |
180 | } |
181 | |
182 | if (source == end) |
183 | break; |
184 | |
185 | // *source may not be ASCII anymore if source moves inside the loop of the fast code path |
186 | if (!isASCII(*source)) |
187 | goto useLookupTable16; |
188 | } |
189 | *destination16 = *source; |
190 | } else { |
191 | useLookupTable16: |
192 | *destination16 = latin1ConversionTable[*source]; |
193 | } |
194 | |
195 | ++source; |
196 | ++destination16; |
197 | } |
198 | |
199 | return result16; |
200 | } |
201 | |
202 | static Vector<uint8_t> encodeComplexWindowsLatin1(StringView string, UnencodableHandling handling) |
203 | { |
204 | Vector<uint8_t> result; |
205 | |
206 | for (auto character : string.codePoints()) { |
207 | uint8_t b = character; |
208 | // Do an efficient check to detect characters other than 00-7F and A0-FF. |
209 | if (b != character || (character & 0xE0) == 0x80) { |
210 | // Look for a way to encode this with Windows Latin-1. |
211 | for (b = 0x80; b < 0xA0; ++b) { |
212 | if (latin1ConversionTable[b] == character) |
213 | goto gotByte; |
214 | } |
215 | // No way to encode this character with Windows Latin-1. |
216 | UnencodableReplacementArray replacement; |
217 | int replacementLength = TextCodec::getUnencodableReplacement(character, handling, replacement); |
218 | result.append(replacement.data(), replacementLength); |
219 | continue; |
220 | } |
221 | gotByte: |
222 | result.append(b); |
223 | } |
224 | |
225 | return result; |
226 | } |
227 | |
228 | Vector<uint8_t> TextCodecLatin1::encode(StringView string, UnencodableHandling handling) |
229 | { |
230 | { |
231 | Vector<uint8_t> result(string.length()); |
232 | auto* bytes = result.data(); |
233 | |
234 | // Convert and simultaneously do a check to see if it's all ASCII. |
235 | UChar ored = 0; |
236 | for (auto character : string.codeUnits()) { |
237 | *bytes++ = character; |
238 | ored |= character; |
239 | } |
240 | |
241 | if (!(ored & 0xFF80)) |
242 | return result; |
243 | } |
244 | |
245 | // If it wasn't all ASCII, call the function that handles more-complex cases. |
246 | return encodeComplexWindowsLatin1(string, handling); |
247 | } |
248 | |
249 | } // namespace WebCore |
250 | |