| OLD | NEW |
| (Empty) |
| 1 part of petitparser; | |
| 2 | |
| 3 /** | |
| 4 * Parser class for individual character classes. | |
| 5 */ | |
| 6 class CharacterParser extends Parser { | |
| 7 final CharacterPredicate _predicate; | |
| 8 | |
| 9 final String _message; | |
| 10 | |
| 11 CharacterParser(this._predicate, this._message); | |
| 12 | |
| 13 @override | |
| 14 Result parseOn(Context context) { | |
| 15 var buffer = context.buffer; | |
| 16 var position = context.position; | |
| 17 if (position < buffer.length && | |
| 18 _predicate.test(buffer.codeUnitAt(position))) { | |
| 19 return context.success(buffer[position], position + 1); | |
| 20 } | |
| 21 return context.failure(_message); | |
| 22 } | |
| 23 | |
| 24 @override | |
| 25 String toString() => '${super.toString()}[$_message]'; | |
| 26 | |
| 27 @override | |
| 28 Parser copy() => new CharacterParser(_predicate, _message); | |
| 29 | |
| 30 @override | |
| 31 bool hasEqualProperties(Parser other) { | |
| 32 return other is CharacterParser | |
| 33 && super.hasEqualProperties(other) | |
| 34 && _predicate == other._predicate | |
| 35 && _message == other._message; | |
| 36 } | |
| 37 } | |
| 38 | |
| 39 /** | |
| 40 * Abstract character predicate class. | |
| 41 */ | |
| 42 abstract class CharacterPredicate { | |
| 43 | |
| 44 /** | |
| 45 * Tests if the character predicate is satisfied. | |
| 46 */ | |
| 47 bool test(int value); | |
| 48 } | |
| 49 | |
| 50 class _NotCharacterPredicate implements CharacterPredicate { | |
| 51 final CharacterPredicate predicate; | |
| 52 | |
| 53 _NotCharacterPredicate(this.predicate); | |
| 54 | |
| 55 @override | |
| 56 bool test(int value) => !predicate.test(value); | |
| 57 } | |
| 58 | |
| 59 /** | |
| 60 * Returns a parser that accepts any of the specified characters. | |
| 61 */ | |
| 62 Parser anyOf(String string, [String message]) { | |
| 63 return new CharacterParser(_optimizedString(string), | |
| 64 message != null ? message : 'any of "$string" expected'); | |
| 65 } | |
| 66 | |
| 67 CharacterPredicate _optimizedString(String string) { | |
| 68 var ranges = | |
| 69 string.codeUnits.map((value) => new _RangeCharPredicate(value, value)); | |
| 70 return _optimizedRanges(ranges); | |
| 71 } | |
| 72 | |
| 73 CharacterPredicate _optimizedRanges(Iterable<_RangeCharPredicate> ranges) { | |
| 74 | |
| 75 // 1. sort the ranges | |
| 76 var sortedRanges = new List.from(ranges, growable: false); | |
| 77 sortedRanges.sort((first, second) { | |
| 78 return first.start != second.start | |
| 79 ? first.start - second.start | |
| 80 : first.stop - second.stop; | |
| 81 }); | |
| 82 | |
| 83 // 2. merge adjacent or overlapping ranges | |
| 84 var mergedRanges = new List(); | |
| 85 for (var thisRange in sortedRanges) { | |
| 86 if (mergedRanges.isEmpty) { | |
| 87 mergedRanges.add(thisRange); | |
| 88 } else { | |
| 89 var lastRange = mergedRanges.last; | |
| 90 if (lastRange.stop + 1 >= thisRange.start) { | |
| 91 var characterRange = new _RangeCharPredicate(lastRange.start, thisRange.
stop); | |
| 92 mergedRanges[mergedRanges.length - 1] = characterRange; | |
| 93 } else { | |
| 94 mergedRanges.add(thisRange); | |
| 95 } | |
| 96 } | |
| 97 } | |
| 98 | |
| 99 // 3. build the best resulting predicates | |
| 100 if (mergedRanges.length == 1) { | |
| 101 return mergedRanges[0].start == mergedRanges[0].stop | |
| 102 ? new _SingleCharPredicate(mergedRanges[0].start) | |
| 103 : mergedRanges[0]; | |
| 104 } else { | |
| 105 return new _RangesCharPredicate(mergedRanges.length, | |
| 106 mergedRanges.map((range) => range.start).toList(growable: false), | |
| 107 mergedRanges.map((range) => range.stop).toList(growable: false)); | |
| 108 } | |
| 109 } | |
| 110 | |
| 111 /** | |
| 112 * Returns a parser that accepts none of the specified characters. | |
| 113 */ | |
| 114 Parser noneOf(String string, [String message]) { | |
| 115 return new CharacterParser( | |
| 116 new _NotCharacterPredicate(_optimizedString(string)), | |
| 117 message != null ? message : 'none of "$string" expected'); | |
| 118 } | |
| 119 | |
| 120 /** | |
| 121 * Returns a parser that accepts a specific character only. | |
| 122 */ | |
| 123 Parser char(element, [String message]) { | |
| 124 return new CharacterParser(new _SingleCharPredicate(_toCharCode(element)), | |
| 125 message != null ? message : '"$element" expected'); | |
| 126 } | |
| 127 | |
| 128 class _SingleCharPredicate implements CharacterPredicate { | |
| 129 final int value; | |
| 130 | |
| 131 const _SingleCharPredicate(this.value); | |
| 132 | |
| 133 @override | |
| 134 bool test(int value) => identical(this.value, value); | |
| 135 } | |
| 136 | |
| 137 /** | |
| 138 * Returns a parser that accepts any digit character. | |
| 139 */ | |
| 140 Parser digit([String message]) { | |
| 141 return new CharacterParser( | |
| 142 _digitCharPredicate, message != null ? message : 'digit expected'); | |
| 143 } | |
| 144 | |
| 145 class _DigitCharPredicate implements CharacterPredicate { | |
| 146 const _DigitCharPredicate(); | |
| 147 | |
| 148 @override | |
| 149 bool test(int value) => 48 <= value && value <= 57; | |
| 150 } | |
| 151 | |
| 152 const _digitCharPredicate = const _DigitCharPredicate(); | |
| 153 | |
| 154 /** | |
| 155 * Returns a parser that accepts any letter character. | |
| 156 */ | |
| 157 Parser letter([String message]) { | |
| 158 return new CharacterParser( | |
| 159 _letterCharPredicate, message != null ? message : 'letter expected'); | |
| 160 } | |
| 161 | |
| 162 class _LetterCharPredicate implements CharacterPredicate { | |
| 163 const _LetterCharPredicate(); | |
| 164 | |
| 165 @override | |
| 166 bool test(int value) => | |
| 167 (65 <= value && value <= 90) || (97 <= value && value <= 122); | |
| 168 } | |
| 169 | |
| 170 const _letterCharPredicate = const _LetterCharPredicate(); | |
| 171 | |
| 172 /** | |
| 173 * Returns a parser that accepts any lowercase character. | |
| 174 */ | |
| 175 Parser lowercase([String message]) { | |
| 176 return new CharacterParser(_lowercaseCharPredicate, | |
| 177 message != null ? message : 'lowercase letter expected'); | |
| 178 } | |
| 179 | |
| 180 class _LowercaseCharPredicate implements CharacterPredicate { | |
| 181 const _LowercaseCharPredicate(); | |
| 182 | |
| 183 @override | |
| 184 bool test(int value) => 97 <= value && value <= 122; | |
| 185 } | |
| 186 | |
| 187 const _lowercaseCharPredicate = const _LowercaseCharPredicate(); | |
| 188 | |
| 189 /** | |
| 190 * Returns a parser that accepts the given character class pattern. | |
| 191 */ | |
| 192 Parser pattern(String element, [String message]) { | |
| 193 return new CharacterParser(_patternParser.parse(element).value, | |
| 194 message != null ? message : '[$element] expected'); | |
| 195 } | |
| 196 | |
| 197 Parser _createPatternParser() { | |
| 198 var single = any().map( | |
| 199 (each) => new _RangeCharPredicate(_toCharCode(each), _toCharCode(each))); | |
| 200 var multiple = any().seq(char('-')).seq(any()).map((each) => | |
| 201 new _RangeCharPredicate(_toCharCode(each[0]), _toCharCode(each[2]))); | |
| 202 var positive = | |
| 203 multiple.or(single).plus().map((each) => _optimizedRanges(each)); | |
| 204 return char('^').optional().seq(positive).map((each) => | |
| 205 each[0] == null ? each[1] : new _NotCharacterPredicate(each[1])); | |
| 206 } | |
| 207 | |
| 208 final _patternParser = _createPatternParser(); | |
| 209 | |
| 210 class _RangesCharPredicate implements CharacterPredicate { | |
| 211 final int length; | |
| 212 final List<int> starts; | |
| 213 final List<int> stops; | |
| 214 | |
| 215 _RangesCharPredicate(this.length, this.starts, this.stops); | |
| 216 | |
| 217 @override | |
| 218 bool test(int value) { | |
| 219 var min = 0; | |
| 220 var max = length; | |
| 221 while (min < max) { | |
| 222 var mid = min + ((max - min) >> 1); | |
| 223 var comp = starts[mid] - value; | |
| 224 if (comp == 0) { | |
| 225 return true; | |
| 226 } else if (comp < 0) { | |
| 227 min = mid + 1; | |
| 228 } else { | |
| 229 max = mid; | |
| 230 } | |
| 231 } | |
| 232 return 0 < min && value <= stops[min - 1]; | |
| 233 } | |
| 234 } | |
| 235 | |
| 236 /** | |
| 237 * Returns a parser that accepts any character in the range | |
| 238 * between [start] and [stop]. | |
| 239 */ | |
| 240 Parser range(start, stop, [String message]) { | |
| 241 return new CharacterParser( | |
| 242 new _RangeCharPredicate(_toCharCode(start), _toCharCode(stop)), | |
| 243 message != null ? message : '$start..$stop expected'); | |
| 244 } | |
| 245 | |
| 246 class _RangeCharPredicate implements CharacterPredicate { | |
| 247 final int start; | |
| 248 final int stop; | |
| 249 | |
| 250 _RangeCharPredicate(this.start, this.stop); | |
| 251 | |
| 252 @override | |
| 253 bool test(int value) => start <= value && value <= stop; | |
| 254 } | |
| 255 | |
| 256 /** | |
| 257 * Returns a parser that accepts any uppercase character. | |
| 258 */ | |
| 259 Parser uppercase([String message]) { | |
| 260 return new CharacterParser(_uppercaseCharPredicate, | |
| 261 message != null ? message : 'uppercase letter expected'); | |
| 262 } | |
| 263 | |
| 264 class _UppercaseCharPredicate implements CharacterPredicate { | |
| 265 const _UppercaseCharPredicate(); | |
| 266 | |
| 267 @override | |
| 268 bool test(int value) => 65 <= value && value <= 90; | |
| 269 } | |
| 270 | |
| 271 const _uppercaseCharPredicate = const _UppercaseCharPredicate(); | |
| 272 | |
| 273 /** | |
| 274 * Returns a parser that accepts any whitespace character. | |
| 275 */ | |
| 276 Parser whitespace([String message]) { | |
| 277 return new CharacterParser(_whitespaceCharPredicate, | |
| 278 message != null ? message : 'whitespace expected'); | |
| 279 } | |
| 280 | |
| 281 class _WhitespaceCharPredicate implements CharacterPredicate { | |
| 282 const _WhitespaceCharPredicate(); | |
| 283 | |
| 284 @override | |
| 285 bool test(int value) { | |
| 286 if (value < 256) { | |
| 287 return value == 0x09 || | |
| 288 value == 0x0A || | |
| 289 value == 0x0B || | |
| 290 value == 0x0C || | |
| 291 value == 0x0D || | |
| 292 value == 0x20 || | |
| 293 value == 0x85 || | |
| 294 value == 0xA0; | |
| 295 } else { | |
| 296 return value == 0x1680 || | |
| 297 value == 0x180E || | |
| 298 value == 0x2000 || | |
| 299 value == 0x2001 || | |
| 300 value == 0x2002 || | |
| 301 value == 0x2003 || | |
| 302 value == 0x2004 || | |
| 303 value == 0x2005 || | |
| 304 value == 0x2006 || | |
| 305 value == 0x2007 || | |
| 306 value == 0x2008 || | |
| 307 value == 0x2009 || | |
| 308 value == 0x200A || | |
| 309 value == 0x2028 || | |
| 310 value == 0x2029 || | |
| 311 value == 0x202F || | |
| 312 value == 0x205F || | |
| 313 value == 0x3000 || | |
| 314 value == 0xFEFF; | |
| 315 } | |
| 316 } | |
| 317 } | |
| 318 | |
| 319 const _whitespaceCharPredicate = const _WhitespaceCharPredicate(); | |
| 320 | |
| 321 /** | |
| 322 * Returns a parser that accepts any word character. | |
| 323 */ | |
| 324 Parser word([String message]) { | |
| 325 return new CharacterParser(_wordCharPredicate, | |
| 326 message != null ? message : 'letter or digit expected'); | |
| 327 } | |
| 328 | |
| 329 class _WordCharPredicate implements CharacterPredicate { | |
| 330 const _WordCharPredicate(); | |
| 331 | |
| 332 @override | |
| 333 bool test(int value) => (65 <= value && value <= 90) || | |
| 334 (97 <= value && value <= 122) || | |
| 335 (48 <= value && value <= 57) || | |
| 336 (value == 95); | |
| 337 } | |
| 338 | |
| 339 const _wordCharPredicate = const _WordCharPredicate(); | |
| 340 | |
| 341 // internal converter for character codes | |
| 342 int _toCharCode(element) { | |
| 343 if (element is num) { | |
| 344 return element.round(); | |
| 345 } | |
| 346 var value = element.toString(); | |
| 347 if (value.length != 1) { | |
| 348 throw new ArgumentError('$value is not a character'); | |
| 349 } | |
| 350 return value.codeUnitAt(0); | |
| 351 } | |
| OLD | NEW |