OLD | NEW |
1 // Copyright 2006-2008 the V8 project authors. All rights reserved. | 1 // Copyright 2006-2008 the V8 project authors. All rights reserved. |
2 // Redistribution and use in source and binary forms, with or without | 2 // Redistribution and use in source and binary forms, with or without |
3 // modification, are permitted provided that the following conditions are | 3 // modification, are permitted provided that the following conditions are |
4 // met: | 4 // met: |
5 // | 5 // |
6 // * Redistributions of source code must retain the above copyright | 6 // * Redistributions of source code must retain the above copyright |
7 // notice, this list of conditions and the following disclaimer. | 7 // notice, this list of conditions and the following disclaimer. |
8 // * Redistributions in binary form must reproduce the above | 8 // * Redistributions in binary form must reproduce the above |
9 // copyright notice, this list of conditions and the following | 9 // copyright notice, this list of conditions and the following |
10 // disclaimer in the documentation and/or other materials provided | 10 // disclaimer in the documentation and/or other materials provided |
(...skipping 101 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
112 } | 112 } |
113 | 113 |
114 | 114 |
115 void UTF16Buffer::PushBack(uc32 ch) { | 115 void UTF16Buffer::PushBack(uc32 ch) { |
116 pushback_buffer()->Add(last_); | 116 pushback_buffer()->Add(last_); |
117 last_ = ch; | 117 last_ = ch; |
118 pos_--; | 118 pos_--; |
119 } | 119 } |
120 | 120 |
121 | 121 |
| 122 static inline bool IsByteOrderMark(uc32 c) { |
| 123 // The Unicode value U+FFFE is guaranteed never to be assigned as a |
| 124 // Unicode character; this implies that in a Unicode context the |
| 125 // 0xFF, 0xFE byte pattern can only be interpreted as the U+FEFF |
| 126 // character expressed in little-endian byte order (since it could |
| 127 // not be a U+FFFE character expressed in big-endian byte |
| 128 // order). Nevertheless, we check for it to be compatible with |
| 129 // Spidermonkey. |
| 130 return c == 0xFEFF || c == 0xFFFE; |
| 131 } |
| 132 |
| 133 |
122 uc32 UTF16Buffer::Advance() { | 134 uc32 UTF16Buffer::Advance() { |
123 // NOTE: It is of importance to Persian / Farsi resources that we do | 135 // NOTE: It is of importance to Persian / Farsi resources that we do |
124 // *not* strip format control characters in the scanner; see | 136 // *not* strip format control characters in the scanner; see |
125 // | 137 // |
126 // https://bugzilla.mozilla.org/show_bug.cgi?id=274152 | 138 // https://bugzilla.mozilla.org/show_bug.cgi?id=274152 |
127 // | 139 // |
128 // So, even though ECMA-262, section 7.1, page 11, dictates that we | 140 // So, even though ECMA-262, section 7.1, page 11, dictates that we |
129 // must remove Unicode format-control characters, we do not. This is | 141 // must remove Unicode format-control characters, we only remove the BOM. |
130 // in line with how IE and SpiderMonkey handles it. | 142 // This is in line with how Safari handles it. |
131 if (!pushback_buffer()->is_empty()) { | 143 if (!pushback_buffer()->is_empty()) { |
132 pos_++; | 144 pos_++; |
133 return last_ = pushback_buffer()->RemoveLast(); | 145 return last_ = pushback_buffer()->RemoveLast(); |
134 } else if (stream_->has_more()) { | |
135 pos_++; | |
136 uc32 next = stream_->GetNext(); | |
137 return last_ = next; | |
138 } else { | 146 } else { |
| 147 while (stream_->has_more()) { |
| 148 pos_++; |
| 149 uc32 next = stream_->GetNext(); |
| 150 if (!IsByteOrderMark(next)) return last_ = next; |
| 151 } |
139 // note: currently the following increment is necessary to avoid a | 152 // note: currently the following increment is necessary to avoid a |
140 // test-parser problem! | 153 // test-parser problem! |
141 pos_++; | 154 pos_++; |
142 return last_ = static_cast<uc32>(-1); | 155 return last_ = static_cast<uc32>(-1); |
143 } | 156 } |
144 } | 157 } |
145 | 158 |
146 | 159 |
147 void UTF16Buffer::SeekForward(int pos) { | 160 void UTF16Buffer::SeekForward(int pos) { |
148 pos_ = pos; | 161 pos_ = pos; |
(...skipping 78 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
227 c0_ = source_.Advance(); | 240 c0_ = source_.Advance(); |
228 } | 241 } |
229 | 242 |
230 | 243 |
231 void Scanner::PushBack(uc32 ch) { | 244 void Scanner::PushBack(uc32 ch) { |
232 source_.PushBack(ch); | 245 source_.PushBack(ch); |
233 c0_ = ch; | 246 c0_ = ch; |
234 } | 247 } |
235 | 248 |
236 | 249 |
237 static inline bool IsByteOrderMark(uc32 c) { | |
238 // The Unicode value U+FFFE is guaranteed never to be assigned as a | |
239 // Unicode character; this implies that in a Unicode context the | |
240 // 0xFF, 0xFE byte pattern can only be interpreted as the U+FEFF | |
241 // character expressed in little-endian byte order (since it could | |
242 // not be a U+FFFE character expressed in big-endian byte | |
243 // order). Nevertheless, we check for it to be compatible with | |
244 // Spidermonkey. | |
245 return c == 0xFEFF || c == 0xFFFE; | |
246 } | |
247 | |
248 | |
249 void Scanner::SkipWhiteSpace(bool initial) { | 250 void Scanner::SkipWhiteSpace(bool initial) { |
250 has_line_terminator_before_next_ = initial; | 251 has_line_terminator_before_next_ = initial; |
251 | 252 |
252 while (true) { | 253 while (true) { |
253 // We treat byte-order marks (BOMs) as whitespace for better | 254 while (kIsWhiteSpace.get(c0_)) { |
254 // compatibility with Spidermonkey and other JavaScript engines. | |
255 while (kIsWhiteSpace.get(c0_) || IsByteOrderMark(c0_)) { | |
256 // IsWhiteSpace() includes line terminators! | 255 // IsWhiteSpace() includes line terminators! |
257 if (kIsLineTerminator.get(c0_)) | 256 if (kIsLineTerminator.get(c0_)) |
258 // Ignore line terminators, but remember them. This is necessary | 257 // Ignore line terminators, but remember them. This is necessary |
259 // for automatic semicolon insertion. | 258 // for automatic semicolon insertion. |
260 has_line_terminator_before_next_ = true; | 259 has_line_terminator_before_next_ = true; |
261 Advance(); | 260 Advance(); |
262 } | 261 } |
263 | 262 |
264 // If there is an HTML comment end '-->' at the beginning of a | 263 // If there is an HTML comment end '-->' at the beginning of a |
265 // line (with only whitespace in front of it), we treat the rest | 264 // line (with only whitespace in front of it), we treat the rest |
(...skipping 565 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
831 StartLiteral(); | 830 StartLiteral(); |
832 while (kIsIdentifierPart.get(c0_)) | 831 while (kIsIdentifierPart.get(c0_)) |
833 AddCharAdvance(); | 832 AddCharAdvance(); |
834 TerminateLiteral(); | 833 TerminateLiteral(); |
835 | 834 |
836 next_.location.end_pos = source_pos() - 1; | 835 next_.location.end_pos = source_pos() - 1; |
837 return true; | 836 return true; |
838 } | 837 } |
839 | 838 |
840 } } // namespace v8::internal | 839 } } // namespace v8::internal |
OLD | NEW |