| OLD | NEW |
| (Empty) | |
| 1 // This code was auto-generated, is not intended to be edited, and is subject to |
| 2 // significant change. Please see the README file for more information. |
| 3 |
| 4 library engine.html.scanner; |
| 5 |
| 6 import 'dart:collection'; |
| 7 import 'java_core.dart'; |
| 8 import 'source.dart'; |
| 9 import 'error.dart'; |
| 10 import 'instrumentation.dart'; |
| 11 |
| 12 /** |
| 13 * Instances of the class {@code Token} represent a token that was scanned from
the input. Each |
| 14 * token knows which token follows it, acting as the head of a linked list of to
kens. |
| 15 */ |
| 16 class Token { |
| 17 /** |
| 18 * The offset from the beginning of the file to the first character in the tok
en. |
| 19 */ |
| 20 int _offset = 0; |
| 21 /** |
| 22 * The previous token in the token stream. |
| 23 */ |
| 24 Token _previous; |
| 25 /** |
| 26 * The next token in the token stream. |
| 27 */ |
| 28 Token _next; |
| 29 /** |
| 30 * The type of the token. |
| 31 */ |
| 32 TokenType _type; |
| 33 /** |
| 34 * The lexeme represented by this token. |
| 35 */ |
| 36 String _value; |
| 37 /** |
| 38 * Initialize a newly created token. |
| 39 * @param type the token type (not {@code null}) |
| 40 * @param offset the offset from the beginning of the file to the first charac
ter in the token |
| 41 */ |
| 42 Token.con1(TokenType type, int offset) { |
| 43 _jtd_constructor_143_impl(type, offset); |
| 44 } |
| 45 _jtd_constructor_143_impl(TokenType type, int offset) { |
| 46 _jtd_constructor_144_impl(type, offset, type.lexeme); |
| 47 } |
| 48 /** |
| 49 * Initialize a newly created token. |
| 50 * @param type the token type (not {@code null}) |
| 51 * @param offset the offset from the beginning of the file to the first charac
ter in the token |
| 52 * @param value the lexeme represented by this token (not {@code null}) |
| 53 */ |
| 54 Token.con2(TokenType type4, int offset3, String value7) { |
| 55 _jtd_constructor_144_impl(type4, offset3, value7); |
| 56 } |
| 57 _jtd_constructor_144_impl(TokenType type4, int offset3, String value7) { |
| 58 this._type = type4; |
| 59 this._value = value7; |
| 60 this._offset = offset3; |
| 61 } |
| 62 /** |
| 63 * Return the offset from the beginning of the file to the character after las
t character of the |
| 64 * token. |
| 65 * @return the offset from the beginning of the file to the first character af
ter last character |
| 66 * of the token |
| 67 */ |
| 68 int get end => _offset + length; |
| 69 /** |
| 70 * Return the number of characters in the node's source range. |
| 71 * @return the number of characters in the node's source range |
| 72 */ |
| 73 int get length => lexeme.length; |
| 74 /** |
| 75 * Return the lexeme that represents this token. |
| 76 * @return the lexeme (not {@code null}) |
| 77 */ |
| 78 String get lexeme => _value; |
| 79 /** |
| 80 * Return the next token in the token stream. |
| 81 * @return the next token in the token stream |
| 82 */ |
| 83 Token get next => _next; |
| 84 /** |
| 85 * Return the offset from the beginning of the file to the first character in
the token. |
| 86 * @return the offset from the beginning of the file to the first character in
the token |
| 87 */ |
| 88 int get offset => _offset; |
| 89 /** |
| 90 * Return the previous token in the token stream. |
| 91 * @return the previous token in the token stream |
| 92 */ |
| 93 Token get previous => _previous; |
| 94 /** |
| 95 * Answer the token type for the receiver. |
| 96 * @return the token type (not {@code null}) |
| 97 */ |
| 98 TokenType get type => _type; |
| 99 /** |
| 100 * Return {@code true} if this token is a synthetic token. A synthetic token i
s a token that was |
| 101 * introduced by the parser in order to recover from an error in the code. Syn
thetic tokens always |
| 102 * have a length of zero ({@code 0}). |
| 103 * @return {@code true} if this token is a synthetic token |
| 104 */ |
| 105 bool isSynthetic() => length == 0; |
| 106 /** |
| 107 * Set the next token in the token stream to the given token. This has the sid
e-effect of setting |
| 108 * this token to be the previous token for the given token. |
| 109 * @param token the next token in the token stream |
| 110 * @return the token that was passed in |
| 111 */ |
| 112 Token setNext(Token token) { |
| 113 _next = token; |
| 114 token.previous = this; |
| 115 return token; |
| 116 } |
| 117 String toString() => lexeme; |
| 118 /** |
| 119 * Set the previous token in the token stream to the given token. |
| 120 * @param previous the previous token in the token stream |
| 121 */ |
| 122 void set previous(Token previous2) { |
| 123 this._previous = previous2; |
| 124 } |
| 125 } |
| 126 /** |
| 127 * Instances of {@code HtmlParseResult} hold the result of parsing an HTML file. |
| 128 */ |
| 129 class HtmlParseResult extends HtmlScanResult { |
| 130 /** |
| 131 * The unit containing the parsed information (not {@code null}). |
| 132 */ |
| 133 HtmlUnit _unit; |
| 134 HtmlParseResult(Token token, List<int> lineStarts, HtmlUnit unit) : super(toke
n, lineStarts) { |
| 135 this._unit = unit; |
| 136 } |
| 137 /** |
| 138 * Answer the unit generated by parsing the source |
| 139 * @return the unit (not {@code null}) |
| 140 */ |
| 141 HtmlUnit get htmlUnit => _unit; |
| 142 } |
| 143 /** |
| 144 * Instances of the class {@code RecursiveXmlVisitor} implement an XML visitor t
hat will recursively |
| 145 * visit all of the nodes in an XML structure. For example, using an instance of
this class to visit |
| 146 * a {@link XmlTagNode} will also cause all of the contained {@link XmlAttribute
Node}s and{@link XmlTagNode}s to be visited. |
| 147 * <p> |
| 148 * Subclasses that override a visit method must either invoke the overridden vis
it method or must |
| 149 * explicitly ask the visited node to visit its children. Failure to do so will
cause the children |
| 150 * of the visited node to not be visited. |
| 151 */ |
| 152 class RecursiveXmlVisitor<R> implements XmlVisitor<R> { |
| 153 R visitHtmlUnit(HtmlUnit node) { |
| 154 node.visitChildren(this); |
| 155 return null; |
| 156 } |
| 157 R visitXmlAttributeNode(XmlAttributeNode node) { |
| 158 node.visitChildren(this); |
| 159 return null; |
| 160 } |
| 161 R visitXmlTagNode(XmlTagNode node) { |
| 162 node.visitChildren(this); |
| 163 return null; |
| 164 } |
| 165 } |
| 166 /** |
| 167 * The abstract class {@code XmlNode} defines behavior common to all XML/HTML no
des. |
| 168 */ |
| 169 abstract class XmlNode { |
| 170 /** |
| 171 * The parent of the node, or {@code null} if the node is the root of an AST s
tructure. |
| 172 */ |
| 173 XmlNode _parent; |
| 174 /** |
| 175 * Use the given visitor to visit this node. |
| 176 * @param visitor the visitor that will visit this node |
| 177 * @return the value returned by the visitor as a result of visiting this node |
| 178 */ |
| 179 accept(XmlVisitor visitor); |
| 180 /** |
| 181 * Return the first token included in this node's source range. |
| 182 * @return the first token or {@code null} if none |
| 183 */ |
| 184 Token get beginToken; |
| 185 /** |
| 186 * Return the offset of the character immediately following the last character
of this node's |
| 187 * source range. This is equivalent to {@code node.getOffset() + node.getLengt
h()}. For an html |
| 188 * unit this will be equal to the length of the unit's source. |
| 189 * @return the offset of the character just past the node's source range |
| 190 */ |
| 191 int get end => offset + length; |
| 192 /** |
| 193 * Return the last token included in this node's source range. |
| 194 * @return the last token or {@code null} if none |
| 195 */ |
| 196 Token get endToken; |
| 197 /** |
| 198 * Return the number of characters in the node's source range. |
| 199 * @return the number of characters in the node's source range |
| 200 */ |
| 201 int get length { |
| 202 Token beginToken5 = beginToken; |
| 203 Token endToken4 = endToken; |
| 204 if (beginToken5 == null || endToken4 == null) { |
| 205 return -1; |
| 206 } |
| 207 return endToken4.offset + endToken4.length - beginToken5.offset; |
| 208 } |
| 209 /** |
| 210 * Return the offset from the beginning of the file to the first character in
the node's source |
| 211 * range. |
| 212 * @return the offset from the beginning of the file to the first character in
the node's source |
| 213 * range |
| 214 */ |
| 215 int get offset { |
| 216 Token beginToken6 = beginToken; |
| 217 if (beginToken6 == null) { |
| 218 return -1; |
| 219 } |
| 220 return beginToken.offset; |
| 221 } |
| 222 /** |
| 223 * Return this node's parent node, or {@code null} if this node is the root of
an AST structure. |
| 224 * <p> |
| 225 * Note that the relationship between an AST node and its parent node may chan
ge over the lifetime |
| 226 * of a node. |
| 227 * @return the parent of this node, or {@code null} if none |
| 228 */ |
| 229 XmlNode get parent => _parent; |
| 230 /** |
| 231 * Use the given visitor to visit all of the children of this node. The childr
en will be visited |
| 232 * in source order. |
| 233 * @param visitor the visitor that will be used to visit the children of this
node |
| 234 */ |
| 235 void visitChildren(XmlVisitor<Object> visitor); |
| 236 /** |
| 237 * Make this node the parent of the given child nodes. |
| 238 * @param children the nodes that will become the children of this node |
| 239 * @return the nodes that were made children of this node |
| 240 */ |
| 241 List<XmlNode> becomeParentOf(List<XmlNode> children) { |
| 242 if (children != null) { |
| 243 for (JavaIterator<XmlNode> iter = new JavaIterator(children); iter.hasNext
;) { |
| 244 XmlNode node = iter.next(); |
| 245 node.parent = this; |
| 246 } |
| 247 } |
| 248 return children; |
| 249 } |
| 250 /** |
| 251 * Make this node the parent of the given child node. |
| 252 * @param child the node that will become a child of this node |
| 253 * @return the node that was made a child of this node |
| 254 */ |
| 255 XmlNode becomeParentOf2(XmlNode child) { |
| 256 if (child != null) { |
| 257 XmlNode node = child; |
| 258 node.parent = this; |
| 259 } |
| 260 return child; |
| 261 } |
| 262 /** |
| 263 * Set the parent of this node to the given node. |
| 264 * @param newParent the node that is to be made the parent of this node |
| 265 */ |
| 266 void set parent(XmlNode newParent) { |
| 267 _parent = newParent; |
| 268 } |
| 269 } |
| 270 /** |
| 271 * The abstract class {@code AbstractScanner} implements a scanner for HTML code
. Subclasses are |
| 272 * required to implement the interface used to access the characters being scann
ed. |
| 273 */ |
| 274 abstract class AbstractScanner { |
| 275 static List<String> _NO_PASS_THROUGH_ELEMENTS = <String> []; |
| 276 /** |
| 277 * The source being scanned. |
| 278 */ |
| 279 Source _source; |
| 280 /** |
| 281 * The token pointing to the head of the linked list of tokens. |
| 282 */ |
| 283 Token _tokens; |
| 284 /** |
| 285 * The last token that was scanned. |
| 286 */ |
| 287 Token _tail; |
| 288 /** |
| 289 * A list containing the offsets of the first character of each line in the so
urce code. |
| 290 */ |
| 291 List<int> _lineStarts = new List<int>(); |
| 292 /** |
| 293 * An array of element tags for which the content between tags should be consi
der a single token. |
| 294 */ |
| 295 List<String> _passThroughElements = _NO_PASS_THROUGH_ELEMENTS; |
| 296 /** |
| 297 * Initialize a newly created scanner. |
| 298 * @param source the source being scanned |
| 299 */ |
| 300 AbstractScanner(Source source) { |
| 301 this._source = source; |
| 302 _tokens = new Token.con1(TokenType.EOF, -1); |
| 303 _tokens.setNext(_tokens); |
| 304 _tail = _tokens; |
| 305 recordStartOfLine(); |
| 306 } |
| 307 /** |
| 308 * Return an array containing the offsets of the first character of each line
in the source code. |
| 309 * @return an array containing the offsets of the first character of each line
in the source code |
| 310 */ |
| 311 List<int> get lineStarts => _lineStarts; |
| 312 /** |
| 313 * Return the current offset relative to the beginning of the file. Return the
initial offset if |
| 314 * the scanner has not yet scanned the source code, and one (1) past the end o
f the source code if |
| 315 * the source code has been scanned. |
| 316 * @return the current offset of the scanner in the source |
| 317 */ |
| 318 int get offset; |
| 319 /** |
| 320 * Answer the source being scanned. |
| 321 * @return the source or {@code null} if undefined |
| 322 */ |
| 323 Source get source => _source; |
| 324 /** |
| 325 * Set array of element tags for which the content between tags should be cons
ider a single token. |
| 326 */ |
| 327 void set passThroughElements(List<String> passThroughElements2) { |
| 328 this._passThroughElements = passThroughElements2 != null ? passThroughElemen
ts2 : _NO_PASS_THROUGH_ELEMENTS; |
| 329 } |
| 330 /** |
| 331 * Scan the source code to produce a list of tokens representing the source. |
| 332 * @return the first token in the list of tokens that were produced |
| 333 */ |
| 334 Token tokenize() { |
| 335 scan(); |
| 336 appendEofToken(); |
| 337 return firstToken(); |
| 338 } |
| 339 /** |
| 340 * Advance the current position and return the character at the new current po
sition. |
| 341 * @return the character at the new current position |
| 342 */ |
| 343 int advance(); |
| 344 /** |
| 345 * Return the substring of the source code between the start offset and the mo
dified current |
| 346 * position. The current position is modified by adding the end delta. |
| 347 * @param start the offset to the beginning of the string, relative to the sta
rt of the file |
| 348 * @param endDelta the number of character after the current location to be in
cluded in the |
| 349 * string, or the number of characters before the current location to be exclu
ded if the |
| 350 * offset is negative |
| 351 * @return the specified substring of the source code |
| 352 */ |
| 353 String getString(int start, int endDelta); |
| 354 /** |
| 355 * Return the character at the current position without changing the current p
osition. |
| 356 * @return the character at the current position |
| 357 */ |
| 358 int peek(); |
| 359 /** |
| 360 * Record the fact that we are at the beginning of a new line in the source. |
| 361 */ |
| 362 void recordStartOfLine() { |
| 363 _lineStarts.add(offset); |
| 364 } |
| 365 void appendEofToken() { |
| 366 Token eofToken = new Token.con1(TokenType.EOF, offset); |
| 367 eofToken.setNext(eofToken); |
| 368 _tail = _tail.setNext(eofToken); |
| 369 } |
| 370 Token emit(Token token) { |
| 371 _tail.setNext(token); |
| 372 _tail = token; |
| 373 return token; |
| 374 } |
| 375 Token emit2(TokenType type, int start) => emit(new Token.con1(type, start)); |
| 376 Token emit3(TokenType type, int start, int count) => emit(new Token.con2(type,
start, getString(start, count))); |
| 377 Token firstToken() => _tokens.next; |
| 378 int recordStartOfLineAndAdvance(int c) { |
| 379 if (c == 0xD) { |
| 380 c = advance(); |
| 381 if (c == 0xA) { |
| 382 c = advance(); |
| 383 } |
| 384 recordStartOfLine(); |
| 385 } else if (c == 0xA) { |
| 386 c = advance(); |
| 387 recordStartOfLine(); |
| 388 } else { |
| 389 c = advance(); |
| 390 } |
| 391 return c; |
| 392 } |
| 393 void scan() { |
| 394 bool inBrackets = false; |
| 395 bool passThrough = false; |
| 396 int c = advance(); |
| 397 while (c >= 0) { |
| 398 int start = offset; |
| 399 if (c == 0x3C) { |
| 400 c = advance(); |
| 401 if (c == 0x21) { |
| 402 c = advance(); |
| 403 if (c == 0x2D && peek() == 0x2D) { |
| 404 c = advance(); |
| 405 int dashCount = 1; |
| 406 while (c >= 0) { |
| 407 if (c == 0x2D) { |
| 408 dashCount++; |
| 409 } else if (c == 0x3E && dashCount >= 2) { |
| 410 c = advance(); |
| 411 break; |
| 412 } else { |
| 413 dashCount = 0; |
| 414 } |
| 415 c = recordStartOfLineAndAdvance(c); |
| 416 } |
| 417 emit3(TokenType.COMMENT, start, -1); |
| 418 if (_tail.length < 7) { |
| 419 } |
| 420 } else { |
| 421 while (c >= 0) { |
| 422 if (c == 0x3E) { |
| 423 c = advance(); |
| 424 break; |
| 425 } |
| 426 c = recordStartOfLineAndAdvance(c); |
| 427 } |
| 428 emit3(TokenType.DECLARATION, start, -1); |
| 429 if (!_tail.lexeme.endsWith(">")) { |
| 430 } |
| 431 } |
| 432 } else if (c == 0x3F) { |
| 433 while (c >= 0) { |
| 434 if (c == 0x3F) { |
| 435 c = advance(); |
| 436 if (c == 0x3E) { |
| 437 c = advance(); |
| 438 break; |
| 439 } |
| 440 } else { |
| 441 c = recordStartOfLineAndAdvance(c); |
| 442 } |
| 443 } |
| 444 emit3(TokenType.DIRECTIVE, start, -1); |
| 445 if (_tail.length < 4) { |
| 446 } |
| 447 } else if (c == 0x2F) { |
| 448 emit2(TokenType.LT_SLASH, start); |
| 449 inBrackets = true; |
| 450 c = advance(); |
| 451 } else { |
| 452 inBrackets = true; |
| 453 emit2(TokenType.LT, start); |
| 454 while (Character.isWhitespace(c)) { |
| 455 c = recordStartOfLineAndAdvance(c); |
| 456 } |
| 457 if (Character.isLetterOrDigit(c)) { |
| 458 int tagStart = offset; |
| 459 c = advance(); |
| 460 while (Character.isLetterOrDigit(c) || c == 0x2D || c == 0x5F) { |
| 461 c = advance(); |
| 462 } |
| 463 emit3(TokenType.TAG, tagStart, -1); |
| 464 String tag = _tail.lexeme; |
| 465 for (String str in _passThroughElements) { |
| 466 if (str == tag) { |
| 467 passThrough = true; |
| 468 break; |
| 469 } |
| 470 } |
| 471 } |
| 472 } |
| 473 } else if (c == 0x3E) { |
| 474 emit2(TokenType.GT, start); |
| 475 inBrackets = false; |
| 476 c = advance(); |
| 477 if (passThrough) { |
| 478 while (c >= 0 && (c != 0x3C || peek() != 0x2F)) { |
| 479 c = recordStartOfLineAndAdvance(c); |
| 480 } |
| 481 if (start + 1 < offset) { |
| 482 emit3(TokenType.TEXT, start + 1, -1); |
| 483 } |
| 484 passThrough = false; |
| 485 } |
| 486 } else if (c == 0x2F && peek() == 0x3E) { |
| 487 advance(); |
| 488 emit2(TokenType.SLASH_GT, start); |
| 489 inBrackets = false; |
| 490 c = advance(); |
| 491 } else if (!inBrackets) { |
| 492 c = recordStartOfLineAndAdvance(c); |
| 493 while (c != 0x3C && c >= 0) { |
| 494 c = recordStartOfLineAndAdvance(c); |
| 495 } |
| 496 emit3(TokenType.TEXT, start, -1); |
| 497 } else if (c == 0x22 || c == 0x27) { |
| 498 int endQuote = c; |
| 499 c = advance(); |
| 500 while (c >= 0) { |
| 501 if (c == endQuote) { |
| 502 c = advance(); |
| 503 break; |
| 504 } |
| 505 c = recordStartOfLineAndAdvance(c); |
| 506 } |
| 507 emit3(TokenType.STRING, start, -1); |
| 508 } else if (c == 0x3D) { |
| 509 emit2(TokenType.EQ, start); |
| 510 c = advance(); |
| 511 } else if (Character.isWhitespace(c)) { |
| 512 do { |
| 513 c = recordStartOfLineAndAdvance(c); |
| 514 } while (Character.isWhitespace(c)); |
| 515 } else if (Character.isLetterOrDigit(c)) { |
| 516 c = advance(); |
| 517 while (Character.isLetterOrDigit(c) || c == 0x2D || c == 0x5F) { |
| 518 c = advance(); |
| 519 } |
| 520 emit3(TokenType.TAG, start, -1); |
| 521 } else { |
| 522 emit3(TokenType.TEXT, start, 0); |
| 523 c = advance(); |
| 524 } |
| 525 } |
| 526 } |
| 527 } |
| 528 /** |
| 529 * Instances of {@code HtmlScanResult} hold the result of scanning an HTML file. |
| 530 */ |
| 531 class HtmlScanResult { |
| 532 /** |
| 533 * The first token in the token stream (not {@code null}). |
| 534 */ |
| 535 Token _token; |
| 536 /** |
| 537 * The line start information that was produced. |
| 538 */ |
| 539 List<int> _lineStarts; |
| 540 HtmlScanResult(Token token, List<int> lineStarts) { |
| 541 this._token = token; |
| 542 this._lineStarts = lineStarts; |
| 543 } |
| 544 /** |
| 545 * Answer the line start information that was produced. |
| 546 * @return an array of line starts (not {@code null}) |
| 547 */ |
| 548 List<int> get lineStarts => _lineStarts; |
| 549 /** |
| 550 * Answer the first token in the token stream. |
| 551 * @return the token (not {@code null}) |
| 552 */ |
| 553 Token get token => _token; |
| 554 } |
| 555 /** |
| 556 * Instances of the class {@code StringScanner} implement a scanner that reads f
rom a string. The |
| 557 * scanning logic is in the superclass. |
| 558 */ |
| 559 class StringScanner extends AbstractScanner { |
| 560 /** |
| 561 * The string from which characters will be read. |
| 562 */ |
| 563 String _string; |
| 564 /** |
| 565 * The number of characters in the string. |
| 566 */ |
| 567 int _stringLength = 0; |
| 568 /** |
| 569 * The index, relative to the string, of the last character that was read. |
| 570 */ |
| 571 int _charOffset = 0; |
| 572 /** |
| 573 * Initialize a newly created scanner to scan the characters in the given stri
ng. |
| 574 * @param source the source being scanned |
| 575 * @param string the string from which characters will be read |
| 576 */ |
| 577 StringScanner(Source source, String string) : super(source) { |
| 578 this._string = string; |
| 579 this._stringLength = string.length; |
| 580 this._charOffset = -1; |
| 581 } |
| 582 int get offset => _charOffset; |
| 583 void set offset(int offset12) { |
| 584 _charOffset = offset12; |
| 585 } |
| 586 int advance() { |
| 587 if (++_charOffset < _stringLength) { |
| 588 return _string.codeUnitAt(_charOffset); |
| 589 } |
| 590 _charOffset = _stringLength; |
| 591 return -1; |
| 592 } |
| 593 String getString(int start, int endDelta) => _string.substring(start, _charOff
set + 1 + endDelta); |
| 594 int peek() { |
| 595 if (_charOffset + 1 < _stringLength) { |
| 596 return _string.codeUnitAt(_charOffset + 1); |
| 597 } |
| 598 return -1; |
| 599 } |
| 600 } |
| 601 /** |
| 602 * Instances of the class {@code CharBufferScanner} implement a scanner that rea
ds from a character |
| 603 * buffer. The scanning logic is in the superclass. |
| 604 */ |
| 605 class CharBufferScanner extends AbstractScanner { |
| 606 /** |
| 607 * The buffer from which characters will be read. |
| 608 */ |
| 609 CharBuffer _buffer; |
| 610 /** |
| 611 * The number of characters in the buffer. |
| 612 */ |
| 613 int _bufferLength = 0; |
| 614 /** |
| 615 * The index of the last character that was read. |
| 616 */ |
| 617 int _charOffset = 0; |
| 618 /** |
| 619 * Initialize a newly created scanner to scan the characters in the given char
acter buffer. |
| 620 * @param source the source being scanned |
| 621 * @param buffer the buffer from which characters will be read |
| 622 */ |
| 623 CharBufferScanner(Source source, CharBuffer buffer) : super(source) { |
| 624 this._buffer = buffer; |
| 625 this._bufferLength = buffer.length(); |
| 626 this._charOffset = -1; |
| 627 } |
| 628 int get offset => _charOffset; |
| 629 int advance() { |
| 630 if (++_charOffset < _bufferLength) { |
| 631 return _buffer.charAt(_charOffset); |
| 632 } |
| 633 _charOffset = _bufferLength; |
| 634 return -1; |
| 635 } |
| 636 String getString(int start, int endDelta) => _buffer.subSequence(start, _charO
ffset + 1 + endDelta).toString(); |
| 637 int peek() { |
| 638 if (_charOffset + 1 < _bufferLength) { |
| 639 return _buffer.charAt(_charOffset + 1); |
| 640 } |
| 641 return -1; |
| 642 } |
| 643 } |
| 644 /** |
| 645 * The enumeration {@code TokenType} defines the types of tokens that can be ret
urned by the |
| 646 * scanner. |
| 647 */ |
| 648 class TokenType { |
| 649 /** |
| 650 * The type of the token that marks the end of the input. |
| 651 */ |
| 652 static final TokenType EOF = new TokenType_EOF('EOF', 0, ""); |
| 653 static final TokenType EQ = new TokenType('EQ', 1, "="); |
| 654 static final TokenType GT = new TokenType('GT', 2, ">"); |
| 655 static final TokenType LT_SLASH = new TokenType('LT_SLASH', 3, "</"); |
| 656 static final TokenType LT = new TokenType('LT', 4, "<"); |
| 657 static final TokenType SLASH_GT = new TokenType('SLASH_GT', 5, "/>"); |
| 658 static final TokenType COMMENT = new TokenType('COMMENT', 6, null); |
| 659 static final TokenType DECLARATION = new TokenType('DECLARATION', 7, null); |
| 660 static final TokenType DIRECTIVE = new TokenType('DIRECTIVE', 8, null); |
| 661 static final TokenType STRING = new TokenType('STRING', 9, null); |
| 662 static final TokenType TAG = new TokenType('TAG', 10, null); |
| 663 static final TokenType TEXT = new TokenType('TEXT', 11, null); |
| 664 static final List<TokenType> values = [EOF, EQ, GT, LT_SLASH, LT, SLASH_GT, CO
MMENT, DECLARATION, DIRECTIVE, STRING, TAG, TEXT]; |
| 665 final String __name; |
| 666 final int __ordinal; |
| 667 /** |
| 668 * The lexeme that defines this type of token, or {@code null} if there is mor
e than one possible |
| 669 * lexeme for this type of token. |
| 670 */ |
| 671 String _lexeme; |
| 672 TokenType(this.__name, this.__ordinal, String lexeme) { |
| 673 this._lexeme = lexeme; |
| 674 } |
| 675 /** |
| 676 * Return the lexeme that defines this type of token, or {@code null} if there
is more than one |
| 677 * possible lexeme for this type of token. |
| 678 * @return the lexeme that defines this type of token |
| 679 */ |
| 680 String get lexeme => _lexeme; |
| 681 String toString() => __name; |
| 682 } |
| 683 class TokenType_EOF extends TokenType { |
| 684 TokenType_EOF(String ___name, int ___ordinal, String arg0) : super(___name, __
_ordinal, arg0); |
| 685 String toString() => "-eof-"; |
| 686 } |
| 687 /** |
| 688 * Instances of {@code XmlAttributeNode} represent name/value pairs owned by an
{@link XmlTagNode}. |
| 689 */ |
| 690 class XmlAttributeNode extends XmlNode { |
| 691 Token _name; |
| 692 Token _equals; |
| 693 Token _value; |
| 694 /** |
| 695 * Construct a new instance representing an XML attribute. |
| 696 * @param name the name token (not {@code null}). This may be a zero length to
ken if the attribute |
| 697 * is badly formed. |
| 698 * @param equals the equals sign or {@code null} if none |
| 699 * @param value the value token (not {@code null}) |
| 700 */ |
| 701 XmlAttributeNode(Token name, Token equals, Token value) { |
| 702 this._name = name; |
| 703 this._equals = equals; |
| 704 this._value = value; |
| 705 } |
| 706 accept(XmlVisitor visitor) => visitor.visitXmlAttributeNode(this); |
| 707 Token get beginToken => _name; |
| 708 Token get endToken => _value; |
| 709 /** |
| 710 * Answer the equals sign token that appears between the name and value tokens
. This may be{@code null} if the attribute is badly formed. |
| 711 * @return the token or {@code null} if there is no equals sign between the na
me and value |
| 712 */ |
| 713 Token get equals => _equals; |
| 714 /** |
| 715 * Answer the attribute name. This may be a zero length token if the attribute
is badly formed. |
| 716 * @return the name (not {@code null}) |
| 717 */ |
| 718 Token get name => _name; |
| 719 /** |
| 720 * Answer the attribute value. A properly formed value will start and end with
matching quote |
| 721 * characters, but the value returned may not be properly formed. |
| 722 * @return the value or {@code null} if this represents a badly formed attribu
te |
| 723 */ |
| 724 Token get value => _value; |
| 725 void visitChildren(XmlVisitor<Object> visitor) { |
| 726 } |
| 727 } |
| 728 /** |
| 729 * The interface {@code XmlVisitor} defines the behavior of objects that can be
used to visit an{@link XmlNode} structure. |
| 730 */ |
| 731 abstract class XmlVisitor<R> { |
| 732 R visitHtmlUnit(HtmlUnit htmlUnit); |
| 733 R visitXmlAttributeNode(XmlAttributeNode xmlAttributeNode); |
| 734 R visitXmlTagNode(XmlTagNode xmlTagNode); |
| 735 } |
| 736 /** |
| 737 * Instances of {@code HtmlScanner} receive and scan HTML content from a {@link
Source}.<br/> |
| 738 * For example, the following code scans HTML source and returns the result: |
| 739 * <pre> |
| 740 * HtmlScanner scanner = new HtmlScanner(source); |
| 741 * source.getContents(scanner); |
| 742 * return scanner.getResult(); |
| 743 * </pre> |
| 744 */ |
| 745 class HtmlScanner implements Source_ContentReceiver { |
| 746 List<String> _SCRIPT_TAG = <String> ["script"]; |
| 747 /** |
| 748 * The source being scanned (not {@code null}) |
| 749 */ |
| 750 Source _source; |
| 751 /** |
| 752 * The scanner used to scan the source |
| 753 */ |
| 754 AbstractScanner _scanner; |
| 755 /** |
| 756 * The first token in the token stream. |
| 757 */ |
| 758 Token _token; |
| 759 /** |
| 760 * Construct a new instance to scan the specified source. |
| 761 * @param source the source to be scanned (not {@code null}) |
| 762 */ |
| 763 HtmlScanner(Source source) { |
| 764 this._source = source; |
| 765 } |
| 766 accept(CharBuffer contents) { |
| 767 _scanner = new CharBufferScanner(_source, contents); |
| 768 _scanner.passThroughElements = _SCRIPT_TAG; |
| 769 _token = _scanner.tokenize(); |
| 770 } |
| 771 void accept2(String contents) { |
| 772 _scanner = new StringScanner(_source, contents); |
| 773 _scanner.passThroughElements = _SCRIPT_TAG; |
| 774 _token = _scanner.tokenize(); |
| 775 } |
| 776 /** |
| 777 * Answer the result of scanning the source |
| 778 * @return the result (not {@code null}) |
| 779 */ |
| 780 HtmlScanResult get result => new HtmlScanResult(_token, _scanner.lineStarts); |
| 781 } |
| 782 /** |
| 783 * Instances of the class {@code XmlParser} are used to parse tokens into a AST
structure comprised |
| 784 * of {@link XmlNode}s. |
| 785 */ |
| 786 class XmlParser { |
| 787 /** |
| 788 * The source being parsed. |
| 789 */ |
| 790 Source _source; |
| 791 /** |
| 792 * The next token to be parsed. |
| 793 */ |
| 794 Token _currentToken; |
| 795 /** |
| 796 * Construct a parser for the specified source. |
| 797 * @param source the source being parsed |
| 798 */ |
| 799 XmlParser(Source source) { |
| 800 this._source = source; |
| 801 } |
| 802 /** |
| 803 * Answer the source being parsed. |
| 804 * @return the source |
| 805 */ |
| 806 Source get source => _source; |
| 807 /** |
| 808 * Answer {@code true} if the specified tag is self closing and thus should ne
ver have content or |
| 809 * child tag nodes. |
| 810 * @param tag the tag (not {@code null}) |
| 811 * @return {@code true} if self closing |
| 812 */ |
| 813 bool isSelfClosing(Token tag) => false; |
| 814 /** |
| 815 * Parse the entire token stream and in the process, advance the current token
to the end of the |
| 816 * token stream. |
| 817 * @return the list of tag nodes found (not {@code null}, contains no {@code n
ull}) |
| 818 */ |
| 819 List<XmlTagNode> parseTopTagNodes(Token firstToken) { |
| 820 _currentToken = firstToken; |
| 821 List<XmlTagNode> tagNodes = new List<XmlTagNode>(); |
| 822 while (true) { |
| 823 while (true) { |
| 824 if (_currentToken.type == TokenType.LT) { |
| 825 tagNodes.add(parseTagNode()); |
| 826 } else if (_currentToken.type == TokenType.DECLARATION || _currentToken.
type == TokenType.DIRECTIVE || _currentToken.type == TokenType.COMMENT) { |
| 827 _currentToken = _currentToken.next; |
| 828 } else if (_currentToken.type == TokenType.EOF) { |
| 829 return tagNodes; |
| 830 } else { |
| 831 reportUnexpectedToken(); |
| 832 _currentToken = _currentToken.next; |
| 833 } |
| 834 break; |
| 835 } |
| 836 } |
| 837 } |
| 838 /** |
| 839 * Answer the current token. |
| 840 * @return the current token |
| 841 */ |
| 842 Token get currentToken => _currentToken; |
| 843 /** |
| 844 * Insert a synthetic token of the specified type before the current token |
| 845 * @param type the type of token to be inserted (not {@code null}) |
| 846 * @return the synthetic token that was inserted (not {@code null}) |
| 847 */ |
| 848 Token insertSyntheticToken(TokenType type) { |
| 849 Token token = new Token.con2(type, _currentToken.offset, ""); |
| 850 _currentToken.previous.setNext(token); |
| 851 token.setNext(_currentToken); |
| 852 return token; |
| 853 } |
| 854 /** |
| 855 * Parse the token stream for an attribute. This method advances the current t
oken over the |
| 856 * attribute, but should not be called if the {@link #currentToken} is not {@l
ink TokenType#TAG}. |
| 857 * @return the attribute (not {@code null}) |
| 858 */ |
| 859 XmlAttributeNode parseAttribute() { |
| 860 Token name = _currentToken; |
| 861 _currentToken = _currentToken.next; |
| 862 Token equals; |
| 863 if (identical(_currentToken.type, TokenType.EQ)) { |
| 864 equals = _currentToken; |
| 865 _currentToken = _currentToken.next; |
| 866 } else { |
| 867 reportUnexpectedToken(); |
| 868 equals = insertSyntheticToken(TokenType.EQ); |
| 869 } |
| 870 Token value; |
| 871 if (identical(_currentToken.type, TokenType.STRING)) { |
| 872 value = _currentToken; |
| 873 _currentToken = _currentToken.next; |
| 874 } else { |
| 875 reportUnexpectedToken(); |
| 876 value = insertSyntheticToken(TokenType.STRING); |
| 877 } |
| 878 return new XmlAttributeNode(name, equals, value); |
| 879 } |
| 880 /** |
| 881 * Parse the stream for a sequence of attributes. This method advances the cur
rent token to the |
| 882 * next {@link TokenType#GT}, {@link TokenType#SLASH_GT}, or {@link TokenType#
EOF}. |
| 883 * @return a collection of zero or more attributes (not {@code null}, contains
no {@code null}s) |
| 884 */ |
| 885 List<XmlAttributeNode> parseAttributes() { |
| 886 TokenType type11 = _currentToken.type; |
| 887 if (identical(type11, TokenType.GT) || identical(type11, TokenType.SLASH_GT)
|| identical(type11, TokenType.EOF)) { |
| 888 return XmlTagNode.NO_ATTRIBUTES; |
| 889 } |
| 890 List<XmlAttributeNode> attributes = new List<XmlAttributeNode>(); |
| 891 while (true) { |
| 892 while (true) { |
| 893 if (_currentToken.type == TokenType.GT || _currentToken.type == TokenTyp
e.SLASH_GT || _currentToken.type == TokenType.EOF) { |
| 894 return attributes; |
| 895 } else if (_currentToken.type == TokenType.TAG) { |
| 896 attributes.add(parseAttribute()); |
| 897 } else { |
| 898 reportUnexpectedToken(); |
| 899 _currentToken = _currentToken.next; |
| 900 } |
| 901 break; |
| 902 } |
| 903 } |
| 904 } |
| 905 /** |
| 906 * Parse the stream for a sequence of tag nodes existing within a parent tag n
ode. This method |
| 907 * advances the current token to the next {@link TokenType#LT_SLASH} or {@link
TokenType#EOF}. |
| 908 * @return a list of nodes (not {@code null}, contains no {@code null}s) |
| 909 */ |
| 910 List<XmlTagNode> parseChildTagNodes() { |
| 911 TokenType type12 = _currentToken.type; |
| 912 if (identical(type12, TokenType.LT_SLASH) || identical(type12, TokenType.EOF
)) { |
| 913 return XmlTagNode.NO_TAG_NODES; |
| 914 } |
| 915 List<XmlTagNode> nodes = new List<XmlTagNode>(); |
| 916 while (true) { |
| 917 while (true) { |
| 918 if (_currentToken.type == TokenType.LT) { |
| 919 nodes.add(parseTagNode()); |
| 920 } else if (_currentToken.type == TokenType.LT_SLASH || _currentToken.typ
e == TokenType.EOF) { |
| 921 return nodes; |
| 922 } else if (_currentToken.type == TokenType.COMMENT) { |
| 923 _currentToken = _currentToken.next; |
| 924 } else { |
| 925 reportUnexpectedToken(); |
| 926 _currentToken = _currentToken.next; |
| 927 } |
| 928 break; |
| 929 } |
| 930 } |
| 931 } |
| 932 /** |
| 933 * Parse the token stream for the next tag node. This method advances current
token over the |
| 934 * parsed tag node, but should only be called if the current token is {@link T
okenType#LT} |
| 935 * @return the tag node or {@code null} if none found |
| 936 */ |
| 937 XmlTagNode parseTagNode() { |
| 938 Token nodeStart = _currentToken; |
| 939 _currentToken = _currentToken.next; |
| 940 Token tag; |
| 941 if (identical(_currentToken.type, TokenType.TAG)) { |
| 942 tag = _currentToken; |
| 943 _currentToken = _currentToken.next; |
| 944 } else { |
| 945 reportUnexpectedToken(); |
| 946 tag = insertSyntheticToken(TokenType.TAG); |
| 947 } |
| 948 List<XmlAttributeNode> attributes = parseAttributes(); |
| 949 Token attributeEnd; |
| 950 if (identical(_currentToken.type, TokenType.GT) || identical(_currentToken.t
ype, TokenType.SLASH_GT)) { |
| 951 attributeEnd = _currentToken; |
| 952 _currentToken = _currentToken.next; |
| 953 } else { |
| 954 reportUnexpectedToken(); |
| 955 attributeEnd = insertSyntheticToken(TokenType.SLASH_GT); |
| 956 } |
| 957 if (identical(attributeEnd.type, TokenType.SLASH_GT) || isSelfClosing(tag))
{ |
| 958 return new XmlTagNode(nodeStart, tag, attributes, attributeEnd, XmlTagNode
.NO_TAG_NODES, _currentToken, null, attributeEnd); |
| 959 } |
| 960 List<XmlTagNode> tagNodes = parseChildTagNodes(); |
| 961 Token contentEnd; |
| 962 if (identical(_currentToken.type, TokenType.LT_SLASH)) { |
| 963 contentEnd = _currentToken; |
| 964 _currentToken = _currentToken.next; |
| 965 } else { |
| 966 reportUnexpectedToken(); |
| 967 contentEnd = insertSyntheticToken(TokenType.LT_SLASH); |
| 968 } |
| 969 Token closingTag; |
| 970 if (identical(_currentToken.type, TokenType.TAG)) { |
| 971 closingTag = _currentToken; |
| 972 _currentToken = _currentToken.next; |
| 973 } else { |
| 974 reportUnexpectedToken(); |
| 975 closingTag = insertSyntheticToken(TokenType.TAG); |
| 976 } |
| 977 Token nodeEnd; |
| 978 if (identical(_currentToken.type, TokenType.GT)) { |
| 979 nodeEnd = _currentToken; |
| 980 _currentToken = _currentToken.next; |
| 981 } else { |
| 982 reportUnexpectedToken(); |
| 983 nodeEnd = insertSyntheticToken(TokenType.GT); |
| 984 } |
| 985 return new XmlTagNode(nodeStart, tag, attributes, attributeEnd, tagNodes, co
ntentEnd, closingTag, nodeEnd); |
| 986 } |
| 987 /** |
| 988 * Report the current token as unexpected |
| 989 */ |
| 990 void reportUnexpectedToken() { |
| 991 } |
| 992 } |
| 993 /** |
| 994 * Instances of {@code XmlTagNode} represent XML or HTML elements such as {@code
<p>} and{@code <body foo="bar"> ... </body>}. |
| 995 */ |
| 996 class XmlTagNode extends XmlNode { |
| 997 /** |
| 998 * Constant representing empty list of attributes. |
| 999 */ |
| 1000 static List<XmlAttributeNode> NO_ATTRIBUTES = new UnmodifiableListView(new Lis
t<XmlAttributeNode>()); |
| 1001 /** |
| 1002 * Constant representing empty list of tag nodes. |
| 1003 */ |
| 1004 static List<XmlTagNode> NO_TAG_NODES = new UnmodifiableListView(new List<XmlTa
gNode>()); |
| 1005 /** |
| 1006 * The starting {@link TokenType#LT} token (not {@code null}). |
| 1007 */ |
| 1008 Token _nodeStart; |
| 1009 /** |
| 1010 * The {@link TokenType#TAG} token after the starting '<' (not {@code null}
). |
| 1011 */ |
| 1012 Token _tag; |
| 1013 /** |
| 1014 * The attributes contained by the receiver (not {@code null}, contains no {@c
ode null}s). |
| 1015 */ |
| 1016 List<XmlAttributeNode> _attributes; |
| 1017 /** |
| 1018 * The {@link TokenType#GT} or {@link TokenType#SLASH_GT} token after the attr
ibutes (not{@code null}). The token may be the same token as {@link #nodeEnd} if
there are no child{@link #tagNodes}. |
| 1019 */ |
| 1020 Token _attributeEnd; |
| 1021 /** |
| 1022 * The tag nodes contained in the receiver (not {@code null}, contains no {@co
de null}s). |
| 1023 */ |
| 1024 List<XmlTagNode> _tagNodes; |
| 1025 /** |
| 1026 * The token (not {@code null}) after the content, which may be |
| 1027 * <ul> |
| 1028 * <li>(1) {@link TokenType#LT_SLASH} for nodes with open and close tags, or</
li> |
| 1029 * <li>(2) the {@link TokenType#LT} nodeStart of the next sibling node if this
node is self |
| 1030 * closing or the attributeEnd is {@link TokenType#SLASH_GT}, or</li> |
| 1031 * <li>(3) {@link TokenType#EOF} if the node does not have a closing tag and i
s the last node in |
| 1032 * the stream {@link TokenType#LT_SLASH} token after the content, or {@code nu
ll} if there is no |
| 1033 * content and the attributes ended with {@link TokenType#SLASH_GT}.</li> |
| 1034 * </ul> |
| 1035 */ |
| 1036 Token _contentEnd; |
| 1037 /** |
| 1038 * The closing {@link TokenType#TAG} after the child elements or {@code null}
if there is no |
| 1039 * content and the attributes ended with {@link TokenType#SLASH_GT} |
| 1040 */ |
| 1041 Token _closingTag; |
| 1042 /** |
| 1043 * The ending {@link TokenType#GT} or {@link TokenType#SLASH_GT} token (not {@
code null}). |
| 1044 */ |
| 1045 Token _nodeEnd; |
| 1046 /** |
| 1047 * Construct a new instance representing an XML or HTML element |
| 1048 * @param nodeStart the starting {@link TokenType#LT} token (not {@code null}) |
| 1049 * @param tag the {@link TokenType#TAG} token after the starting '<' (not {
@code null}). |
| 1050 * @param attributes the attributes associated with this element or {@link #NO
_ATTRIBUTES} (not{@code null}, contains no {@code null}s) |
| 1051 * @param attributeEnd The {@link TokenType#GT} or {@link TokenType#SLASH_GT}
token after the |
| 1052 * attributes (not {@code null}). The token may be the same token as {@link #n
odeEnd} if |
| 1053 * there are no child {@link #tagNodes}. |
| 1054 * @param tagNodes child tag nodes of the receiver or {@link #NO_TAG_NODES} (n
ot {@code null}, |
| 1055 * contains no {@code null}s) |
| 1056 * @param contentEnd the token (not {@code null}) after the content, which may
be |
| 1057 * <ul> |
| 1058 * <li>(1) {@link TokenType#LT_SLASH} for nodes with open and close tags, or</
li> |
| 1059 * <li>(2) the {@link TokenType#LT} nodeStart of the next sibling node if this
node is |
| 1060 * self closing or the attributeEnd is {@link TokenType#SLASH_GT}, or</li> |
| 1061 * <li>(3) {@link TokenType#EOF} if the node does not have a closing tag and i
s the last |
| 1062 * node in the stream {@link TokenType#LT_SLASH} token after the content, or {
@code null}if there is no content and the attributes ended with {@link TokenType
#SLASH_GT}.</li> |
| 1063 * </ul> |
| 1064 * @param closingTag the closing {@link TokenType#TAG} after the child element
s or {@code null} if |
| 1065 * there is no content and the attributes ended with {@link TokenType#SLASH_GT
} |
| 1066 * @param nodeEnd the ending {@link TokenType#GT} or {@link TokenType#SLASH_GT
} token (not{@code null}) |
| 1067 */ |
| 1068 XmlTagNode(Token nodeStart, Token tag, List<XmlAttributeNode> attributes, Toke
n attributeEnd, List<XmlTagNode> tagNodes, Token contentEnd, Token closingTag, T
oken nodeEnd) { |
| 1069 this._nodeStart = nodeStart; |
| 1070 this._tag = tag; |
| 1071 this._attributes = becomeParentOf(attributes); |
| 1072 this._attributeEnd = attributeEnd; |
| 1073 this._tagNodes = becomeParentOf(tagNodes); |
| 1074 this._contentEnd = contentEnd; |
| 1075 this._closingTag = closingTag; |
| 1076 this._nodeEnd = nodeEnd; |
| 1077 } |
| 1078 accept(XmlVisitor visitor) => visitor.visitXmlTagNode(this); |
| 1079 /** |
| 1080 * The {@link TokenType#GT} or {@link TokenType#SLASH_GT} token after the attr
ibutes (not{@code null}). The token may be the same token as {@link #nodeEnd} if
there are no child{@link #tagNodes}. |
| 1081 * @return the token (not {@code null}) |
| 1082 */ |
| 1083 Token get attributeEnd => _attributeEnd; |
| 1084 /** |
| 1085 * Answer the receiver's attributes. Callers should not manipulate the returne
d list to edit the |
| 1086 * AST structure. |
| 1087 * @return the attributes (not {@code null}, contains no {@code null}s) |
| 1088 */ |
| 1089 List<XmlAttributeNode> get attributes => _attributes; |
| 1090 Token get beginToken => _nodeStart; |
| 1091 /** |
| 1092 * The the closing {@link TokenType#TAG} after the child elements or {@code nu
ll} if there is no |
| 1093 * content and the attributes ended with {@link TokenType#SLASH_GT} |
| 1094 * @return the closing tag or {@code null} |
| 1095 */ |
| 1096 Token get closingTag => _closingTag; |
| 1097 /** |
| 1098 * Answer a string representing the content contained in the receiver. This in
cludes the textual |
| 1099 * representation of any child tag nodes ({@link #getTagNodes()}). Whitespace
between '<', |
| 1100 * '</', and '>', '/>' is discarded, but all other whitespace is preserved. |
| 1101 * @return the content (not {@code null}) |
| 1102 */ |
| 1103 String get content { |
| 1104 Token token = _attributeEnd.next; |
| 1105 if (identical(token, _contentEnd)) { |
| 1106 return ""; |
| 1107 } |
| 1108 String content = token.lexeme; |
| 1109 token = token.next; |
| 1110 if (identical(token, _contentEnd)) { |
| 1111 return content; |
| 1112 } |
| 1113 StringBuffer buffer = new StringBuffer(); |
| 1114 while (token != _contentEnd) { |
| 1115 buffer.write(token.lexeme); |
| 1116 token = token.next; |
| 1117 } |
| 1118 return buffer.toString(); |
| 1119 } |
| 1120 /** |
| 1121 * Answer the token (not {@code null}) after the content, which may be |
| 1122 * <ul> |
| 1123 * <li>(1) {@link TokenType#LT_SLASH} for nodes with open and close tags, or</
li> |
| 1124 * <li>(2) the {@link TokenType#LT} nodeStart of the next sibling node if this
node is self |
| 1125 * closing or the attributeEnd is {@link TokenType#SLASH_GT}, or</li> |
| 1126 * <li>(3) {@link TokenType#EOF} if the node does not have a closing tag and i
s the last node in |
| 1127 * the stream {@link TokenType#LT_SLASH} token after the content, or {@code nu
ll} if there is no |
| 1128 * content and the attributes ended with {@link TokenType#SLASH_GT}.</li> |
| 1129 * </ul> |
| 1130 * @return the token (not {@code null}) |
| 1131 */ |
| 1132 Token get contentEnd => _contentEnd; |
| 1133 Token get endToken { |
| 1134 if (_nodeEnd != null) { |
| 1135 return _nodeEnd; |
| 1136 } |
| 1137 if (_closingTag != null) { |
| 1138 return _closingTag; |
| 1139 } |
| 1140 if (_contentEnd != null) { |
| 1141 return _contentEnd; |
| 1142 } |
| 1143 if (!_tagNodes.isEmpty) { |
| 1144 return _tagNodes[_tagNodes.length - 1].endToken; |
| 1145 } |
| 1146 if (_attributeEnd != null) { |
| 1147 return _attributeEnd; |
| 1148 } |
| 1149 if (!_attributes.isEmpty) { |
| 1150 return _attributes[_attributes.length - 1].endToken; |
| 1151 } |
| 1152 return _tag; |
| 1153 } |
| 1154 /** |
| 1155 * Answer the ending {@link TokenType#GT} or {@link TokenType#SLASH_GT} token. |
| 1156 * @return the token (not {@code null}) |
| 1157 */ |
| 1158 Token get nodeEnd => _nodeEnd; |
| 1159 /** |
| 1160 * Answer the starting {@link TokenType#LT} token. |
| 1161 * @return the token (not {@code null}) |
| 1162 */ |
| 1163 Token get nodeStart => _nodeStart; |
| 1164 /** |
| 1165 * Answer the {@link TokenType#TAG} token after the starting '<'. |
| 1166 * @return the token (not {@code null}) |
| 1167 */ |
| 1168 Token get tag => _tag; |
| 1169 /** |
| 1170 * Answer the tag nodes contained in the receiver. Callers should not manipula
te the returned list |
| 1171 * to edit the AST structure. |
| 1172 * @return the children (not {@code null}, contains no {@code null}s) |
| 1173 */ |
| 1174 List<XmlTagNode> get tagNodes => _tagNodes; |
| 1175 void visitChildren(XmlVisitor<Object> visitor) { |
| 1176 for (XmlAttributeNode node in _attributes) { |
| 1177 node.accept(visitor); |
| 1178 } |
| 1179 for (XmlTagNode node in _tagNodes) { |
| 1180 node.accept(visitor); |
| 1181 } |
| 1182 } |
| 1183 } |
| 1184 /** |
| 1185 * Instances of the class {@code HtmlParser} are used to parse tokens into a AST
structure comprised |
| 1186 * of {@link XmlNode}s. |
| 1187 */ |
| 1188 class HtmlParser extends XmlParser { |
| 1189 static Set<String> SELF_CLOSING = new Set<String>(); |
| 1190 /** |
| 1191 * Construct a parser for the specified source. |
| 1192 * @param source the source being parsed |
| 1193 */ |
| 1194 HtmlParser(Source source) : super(source) { |
| 1195 } |
| 1196 /** |
| 1197 * Parse the tokens specified by the given scan result. |
| 1198 * @param scanResult the result of scanning an HTML source (not {@code null}) |
| 1199 * @return the parse result (not {@code null}) |
| 1200 */ |
| 1201 HtmlParseResult parse(HtmlScanResult scanResult) { |
| 1202 Token firstToken = scanResult.token; |
| 1203 List<XmlTagNode> tagNodes = parseTopTagNodes(firstToken); |
| 1204 HtmlUnit unit = new HtmlUnit(firstToken, tagNodes, currentToken); |
| 1205 return new HtmlParseResult(firstToken, scanResult.lineStarts, unit); |
| 1206 } |
| 1207 /** |
| 1208 * Scan then parse the specified source. |
| 1209 * @param source the source to be scanned and parsed (not {@code null}) |
| 1210 * @return the parse result (not {@code null}) |
| 1211 */ |
| 1212 HtmlParseResult parse2(Source source) { |
| 1213 HtmlScanner scanner = new HtmlScanner(source); |
| 1214 source.getContents(scanner); |
| 1215 return parse(scanner.result); |
| 1216 } |
| 1217 bool isSelfClosing(Token tag) => SELF_CLOSING.contains(tag.lexeme); |
| 1218 } |
| 1219 /** |
| 1220 * Instances of the class {@code HtmlUnit} represent the contents of an HTML fil
e. |
| 1221 */ |
| 1222 class HtmlUnit extends XmlNode { |
| 1223 /** |
| 1224 * The first token in the token stream that was parsed to form this HTML unit. |
| 1225 */ |
| 1226 Token _beginToken; |
| 1227 /** |
| 1228 * The last token in the token stream that was parsed to form this compilation
unit. This token |
| 1229 * should always have a type of {@link TokenType.EOF}. |
| 1230 */ |
| 1231 Token _endToken; |
| 1232 /** |
| 1233 * The tag nodes contained in the receiver (not {@code null}, contains no {@co
de null}s). |
| 1234 */ |
| 1235 List<XmlTagNode> _tagNodes; |
| 1236 /** |
| 1237 * Construct a new instance representing the content of an HTML file. |
| 1238 * @param beginToken the first token in the file (not {@code null}) |
| 1239 * @param tagNodes child tag nodes of the receiver (not {@code null}, contains
no {@code null}s) |
| 1240 * @param endToken the last token in the token stream which should be of type{
@link TokenType.EOF} |
| 1241 */ |
| 1242 HtmlUnit(Token beginToken, List<XmlTagNode> tagNodes, Token endToken) { |
| 1243 this._beginToken = beginToken; |
| 1244 this._tagNodes = becomeParentOf(tagNodes); |
| 1245 this._endToken = endToken; |
| 1246 } |
| 1247 accept(XmlVisitor visitor) => visitor.visitHtmlUnit(this); |
| 1248 Token get beginToken => _beginToken; |
| 1249 Token get endToken => _endToken; |
| 1250 /** |
| 1251 * Answer the tag nodes contained in the receiver. Callers should not manipula
te the returned list |
| 1252 * to edit the AST structure. |
| 1253 * @return the children (not {@code null}, contains no {@code null}s) |
| 1254 */ |
| 1255 List<XmlTagNode> get tagNodes => _tagNodes; |
| 1256 void visitChildren(XmlVisitor<Object> visitor) { |
| 1257 for (XmlTagNode node in _tagNodes) { |
| 1258 node.accept(visitor); |
| 1259 } |
| 1260 } |
| 1261 } |
| OLD | NEW |