1 /* 2 *************************************************************************** 3 * Copyright (C) 1999-2011, International Business Machines Corporation 4 * and others. All Rights Reserved. 5 *************************************************************************** 6 * Date Name Description 7 * 10/20/99 alan Creation. 8 *************************************************************************** 9 */ 10 11 #ifndef UNICODESET_H 12 #define UNICODESET_H 13 14 #include "unicode/unifilt.h" 15 #include "unicode/unistr.h" 16 #include "unicode/uset.h" 17 18 /** 19 * \file 20 * \brief C++ API: Unicode Set 21 */ 22 23 U_NAMESPACE_BEGIN 24 25 class BMPSet; 26 class ParsePosition; 27 class RBBIRuleScanner; 28 class SymbolTable; 29 class UnicodeSetStringSpan; 30 class UVector; 31 class RuleCharacterIterator; 32 33 /** 34 * A mutable set of Unicode characters and multicharacter strings. Objects of this class 35 * represent <em>character classes</em> used in regular expressions. 36 * A character specifies a subset of Unicode code points. Legal 37 * code points are U+0000 to U+10FFFF, inclusive. 38 * 39 * <p>The UnicodeSet class is not designed to be subclassed. 40 * 41 * <p><code>UnicodeSet</code> supports two APIs. The first is the 42 * <em>operand</em> API that allows the caller to modify the value of 43 * a <code>UnicodeSet</code> object. It conforms to Java 2's 44 * <code>java.util.Set</code> interface, although 45 * <code>UnicodeSet</code> does not actually implement that 46 * interface. All methods of <code>Set</code> are supported, with the 47 * modification that they take a character range or single character 48 * instead of an <code>Object</code>, and they take a 49 * <code>UnicodeSet</code> instead of a <code>Collection</code>. The 50 * operand API may be thought of in terms of boolean logic: a boolean 51 * OR is implemented by <code>add</code>, a boolean AND is implemented 52 * by <code>retain</code>, a boolean XOR is implemented by 53 * <code>complement</code> taking an argument, and a boolean NOT is 54 * implemented by <code>complement</code> with no argument. In terms 55 * of traditional set theory function names, <code>add</code> is a 56 * union, <code>retain</code> is an intersection, <code>remove</code> 57 * is an asymmetric difference, and <code>complement</code> with no 58 * argument is a set complement with respect to the superset range 59 * <code>MIN_VALUE-MAX_VALUE</code> 60 * 61 * <p>The second API is the 62 * <code>applyPattern()</code>/<code>toPattern()</code> API from the 63 * <code>java.text.Format</code>-derived classes. Unlike the 64 * methods that add characters, add categories, and control the logic 65 * of the set, the method <code>applyPattern()</code> sets all 66 * attributes of a <code>UnicodeSet</code> at once, based on a 67 * string pattern. 68 * 69 * <p><b>Pattern syntax</b></p> 70 * 71 * Patterns are accepted by the constructors and the 72 * <code>applyPattern()</code> methods and returned by the 73 * <code>toPattern()</code> method. These patterns follow a syntax 74 * similar to that employed by version 8 regular expression character 75 * classes. Here are some simple examples: 76 * 77 * \htmlonly<blockquote>\endhtmlonly 78 * <table> 79 * <tr align="top"> 80 * <td nowrap valign="top" align="left"><code>[]</code></td> 81 * <td valign="top">No characters</td> 82 * </tr><tr align="top"> 83 * <td nowrap valign="top" align="left"><code>[a]</code></td> 84 * <td valign="top">The character 'a'</td> 85 * </tr><tr align="top"> 86 * <td nowrap valign="top" align="left"><code>[ae]</code></td> 87 * <td valign="top">The characters 'a' and 'e'</td> 88 * </tr> 89 * <tr> 90 * <td nowrap valign="top" align="left"><code>[a-e]</code></td> 91 * <td valign="top">The characters 'a' through 'e' inclusive, in Unicode code 92 * point order</td> 93 * </tr> 94 * <tr> 95 * <td nowrap valign="top" align="left"><code>[\\u4E01]</code></td> 96 * <td valign="top">The character U+4E01</td> 97 * </tr> 98 * <tr> 99 * <td nowrap valign="top" align="left"><code>[a{ab}{ac}]</code></td> 100 * <td valign="top">The character 'a' and the multicharacter strings "ab" and 101 * "ac"</td> 102 * </tr> 103 * <tr> 104 * <td nowrap valign="top" align="left"><code>[\\p{Lu}]</code></td> 105 * <td valign="top">All characters in the general category Uppercase Letter</td> 106 * </tr> 107 * </table> 108 * \htmlonly</blockquote>\endhtmlonly 109 * 110 * Any character may be preceded by a backslash in order to remove any special 111 * meaning. White space characters, as defined by UCharacter.isWhitespace(), are 112 * ignored, unless they are escaped. 113 * 114 * <p>Property patterns specify a set of characters having a certain 115 * property as defined by the Unicode standard. Both the POSIX-like 116 * "[:Lu:]" and the Perl-like syntax "\\p{Lu}" are recognized. For a 117 * complete list of supported property patterns, see the User's Guide 118 * for UnicodeSet at 119 * <a href="http://icu-project.org/userguide/unicodeSet.html"> 120 * http://icu-project.org/userguide/unicodeSet.html</a>. 121 * Actual determination of property data is defined by the underlying 122 * Unicode database as implemented by UCharacter. 123 * 124 * <p>Patterns specify individual characters, ranges of characters, and 125 * Unicode property sets. When elements are concatenated, they 126 * specify their union. To complement a set, place a '^' immediately 127 * after the opening '['. Property patterns are inverted by modifying 128 * their delimiters; "[:^foo]" and "\\P{foo}". In any other location, 129 * '^' has no special meaning. 130 * 131 * <p>Ranges are indicated by placing two a '-' between two 132 * characters, as in "a-z". This specifies the range of all 133 * characters from the left to the right, in Unicode order. If the 134 * left character is greater than or equal to the 135 * right character it is a syntax error. If a '-' occurs as the first 136 * character after the opening '[' or '[^', or if it occurs as the 137 * last character before the closing ']', then it is taken as a 138 * literal. Thus "[a\-b]", "[-ab]", and "[ab-]" all indicate the same 139 * set of three characters, 'a', 'b', and '-'. 140 * 141 * <p>Sets may be intersected using the '&' operator or the asymmetric 142 * set difference may be taken using the '-' operator, for example, 143 * "[[:L:]&[\\u0000-\\u0FFF]]" indicates the set of all Unicode letters 144 * with values less than 4096. Operators ('&' and '|') have equal 145 * precedence and bind left-to-right. Thus 146 * "[[:L:]-[a-z]-[\\u0100-\\u01FF]]" is equivalent to 147 * "[[[:L:]-[a-z]]-[\\u0100-\\u01FF]]". This only really matters for 148 * difference; intersection is commutative. 149 * 150 * <table> 151 * <tr valign=top><td nowrap><code>[a]</code><td>The set containing 'a' 152 * <tr valign=top><td nowrap><code>[a-z]</code><td>The set containing 'a' 153 * through 'z' and all letters in between, in Unicode order 154 * <tr valign=top><td nowrap><code>[^a-z]</code><td>The set containing 155 * all characters but 'a' through 'z', 156 * that is, U+0000 through 'a'-1 and 'z'+1 through U+10FFFF 157 * <tr valign=top><td nowrap><code>[[<em>pat1</em>][<em>pat2</em>]]</code> 158 * <td>The union of sets specified by <em>pat1</em> and <em>pat2</em> 159 * <tr valign=top><td nowrap><code>[[<em>pat1</em>]&[<em>pat2</em>]]</code> 160 * <td>The intersection of sets specified by <em>pat1</em> and <em>pat2</em> 161 * <tr valign=top><td nowrap><code>[[<em>pat1</em>]-[<em>pat2</em>]]</code> 162 * <td>The asymmetric difference of sets specified by <em>pat1</em> and 163 * <em>pat2</em> 164 * <tr valign=top><td nowrap><code>[:Lu:] or \\p{Lu}</code> 165 * <td>The set of characters having the specified 166 * Unicode property; in 167 * this case, Unicode uppercase letters 168 * <tr valign=top><td nowrap><code>[:^Lu:] or \\P{Lu}</code> 169 * <td>The set of characters <em>not</em> having the given 170 * Unicode property 171 * </table> 172 * 173 * <p><b>Warning</b>: you cannot add an empty string ("") to a UnicodeSet.</p> 174 * 175 * <p><b>Formal syntax</b></p> 176 * 177 * \htmlonly<blockquote>\endhtmlonly 178 * <table> 179 * <tr align="top"> 180 * <td nowrap valign="top" align="right"><code>pattern := </code></td> 181 * <td valign="top"><code>('[' '^'? item* ']') | 182 * property</code></td> 183 * </tr> 184 * <tr align="top"> 185 * <td nowrap valign="top" align="right"><code>item := </code></td> 186 * <td valign="top"><code>char | (char '-' char) | pattern-expr<br> 187 * </code></td> 188 * </tr> 189 * <tr align="top"> 190 * <td nowrap valign="top" align="right"><code>pattern-expr := </code></td> 191 * <td valign="top"><code>pattern | pattern-expr pattern | 192 * pattern-expr op pattern<br> 193 * </code></td> 194 * </tr> 195 * <tr align="top"> 196 * <td nowrap valign="top" align="right"><code>op := </code></td> 197 * <td valign="top"><code>'&' | '-'<br> 198 * </code></td> 199 * </tr> 200 * <tr align="top"> 201 * <td nowrap valign="top" align="right"><code>special := </code></td> 202 * <td valign="top"><code>'[' | ']' | '-'<br> 203 * </code></td> 204 * </tr> 205 * <tr align="top"> 206 * <td nowrap valign="top" align="right"><code>char := </code></td> 207 * <td valign="top"><em>any character that is not</em><code> special<br> 208 * | ('\' </code><em>any character</em><code>)<br> 209 * | ('\\u' hex hex hex hex)<br> 210 * </code></td> 211 * </tr> 212 * <tr align="top"> 213 * <td nowrap valign="top" align="right"><code>hex := </code></td> 214 * <td valign="top"><em>any character for which 215 * </em><code>Character.digit(c, 16)</code><em> 216 * returns a non-negative result</em></td> 217 * </tr> 218 * <tr> 219 * <td nowrap valign="top" align="right"><code>property := </code></td> 220 * <td valign="top"><em>a Unicode property set pattern</em></td> 221 * </tr> 222 * </table> 223 * <br> 224 * <table border="1"> 225 * <tr> 226 * <td>Legend: <table> 227 * <tr> 228 * <td nowrap valign="top"><code>a := b</code></td> 229 * <td width="20" valign="top"> </td> 230 * <td valign="top"><code>a</code> may be replaced by <code>b</code> </td> 231 * </tr> 232 * <tr> 233 * <td nowrap valign="top"><code>a?</code></td> 234 * <td valign="top"></td> 235 * <td valign="top">zero or one instance of <code>a</code><br> 236 * </td> 237 * </tr> 238 * <tr> 239 * <td nowrap valign="top"><code>a*</code></td> 240 * <td valign="top"></td> 241 * <td valign="top">one or more instances of <code>a</code><br> 242 * </td> 243 * </tr> 244 * <tr> 245 * <td nowrap valign="top"><code>a | b</code></td> 246 * <td valign="top"></td> 247 * <td valign="top">either <code>a</code> or <code>b</code><br> 248 * </td> 249 * </tr> 250 * <tr> 251 * <td nowrap valign="top"><code>'a'</code></td> 252 * <td valign="top"></td> 253 * <td valign="top">the literal string between the quotes </td> 254 * </tr> 255 * </table> 256 * </td> 257 * </tr> 258 * </table> 259 * \htmlonly</blockquote>\endhtmlonly 260 * 261 * <p>Note: 262 * - Most UnicodeSet methods do not take a UErrorCode parameter because 263 * there are usually very few opportunities for failure other than a shortage 264 * of memory, error codes in low-level C++ string methods would be inconvenient, 265 * and the error code as the last parameter (ICU convention) would prevent 266 * the use of default parameter values. 267 * Instead, such methods set the UnicodeSet into a "bogus" state 268 * (see isBogus()) if an error occurs. 269 * 270 * @author Alan Liu 271 * @stable ICU 2.0 272 */ 273 class U_COMMON_API UnicodeSet : public UnicodeFilter { 274 275 int32_t len; // length of list used; 0 <= len <= capacity 276 int32_t capacity; // capacity of list 277 UChar32* list; // MUST be terminated with HIGH 278 BMPSet *bmpSet; // The set is frozen iff either bmpSet or stringSpan is not NULL. 279 UChar32* buffer; // internal buffer, may be NULL 280 int32_t bufferCapacity; // capacity of buffer 281 int32_t patLen; 282 283 /** 284 * The pattern representation of this set. This may not be the 285 * most economical pattern. It is the pattern supplied to 286 * applyPattern(), with variables substituted and whitespace 287 * removed. For sets constructed without applyPattern(), or 288 * modified using the non-pattern API, this string will be empty, 289 * indicating that toPattern() must generate a pattern 290 * representation from the inversion list. 291 */ 292 UChar *pat; 293 UVector* strings; // maintained in sorted order 294 UnicodeSetStringSpan *stringSpan; 295 296 private: 297 enum { // constants 298 kIsBogus = 1 // This set is bogus (i.e. not valid) 299 }; 300 uint8_t fFlags; // Bit flag (see constants above) 301 public: 302 /** 303 * Determine if this object contains a valid set. 304 * A bogus set has no value. It is different from an empty set. 305 * It can be used to indicate that no set value is available. 306 * 307 * @return TRUE if the set is valid, FALSE otherwise 308 * @see setToBogus() 309 * @stable ICU 4.0 310 */ 311 inline UBool isBogus(void) const; 312 313 /** 314 * Make this UnicodeSet object invalid. 315 * The string will test TRUE with isBogus(). 316 * 317 * A bogus set has no value. It is different from an empty set. 318 * It can be used to indicate that no set value is available. 319 * 320 * This utility function is used throughout the UnicodeSet 321 * implementation to indicate that a UnicodeSet operation failed, 322 * and may be used in other functions, 323 * especially but not exclusively when such functions do not 324 * take a UErrorCode for simplicity. 325 * 326 * @see isBogus() 327 * @stable ICU 4.0 328 */ 329 void setToBogus(); 330 331 public: 332 333 enum { 334 /** 335 * Minimum value that can be stored in a UnicodeSet. 336 * @stable ICU 2.4 337 */ 338 MIN_VALUE = 0, 339 340 /** 341 * Maximum value that can be stored in a UnicodeSet. 342 * @stable ICU 2.4 343 */ 344 MAX_VALUE = 0x10ffff 345 }; 346 347 //---------------------------------------------------------------- 348 // Constructors &c 349 //---------------------------------------------------------------- 350 351 public: 352 353 /** 354 * Constructs an empty set. 355 * @stable ICU 2.0 356 */ 357 UnicodeSet(); 358 359 /** 360 * Constructs a set containing the given range. If <code>end > 361 * start</code> then an empty set is created. 362 * 363 * @param start first character, inclusive, of range 364 * @param end last character, inclusive, of range 365 * @stable ICU 2.4 366 */ 367 UnicodeSet(UChar32 start, UChar32 end); 368 369 /** 370 * Constructs a set from the given pattern. See the class 371 * description for the syntax of the pattern language. 372 * @param pattern a string specifying what characters are in the set 373 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 374 * contains a syntax error. 375 * @stable ICU 2.0 376 */ 377 UnicodeSet(const UnicodeString& pattern, 378 UErrorCode& status); 379 380 #ifndef U_HIDE_INTERNAL_API 381 /** 382 * Constructs a set from the given pattern. See the class 383 * description for the syntax of the pattern language. 384 * @param pattern a string specifying what characters are in the set 385 * @param options bitmask for options to apply to the pattern. 386 * Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE. 387 * @param symbols a symbol table mapping variable names to values 388 * and stand-in characters to UnicodeSets; may be NULL 389 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 390 * contains a syntax error. 391 * @internal 392 */ 393 UnicodeSet(const UnicodeString& pattern, 394 uint32_t options, 395 const SymbolTable* symbols, 396 UErrorCode& status); 397 #endif /* U_HIDE_INTERNAL_API */ 398 399 /** 400 * Constructs a set from the given pattern. See the class description 401 * for the syntax of the pattern language. 402 * @param pattern a string specifying what characters are in the set 403 * @param pos on input, the position in pattern at which to start parsing. 404 * On output, the position after the last character parsed. 405 * @param options bitmask for options to apply to the pattern. 406 * Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE. 407 * @param symbols a symbol table mapping variable names to values 408 * and stand-in characters to UnicodeSets; may be NULL 409 * @param status input-output error code 410 * @stable ICU 2.8 411 */ 412 UnicodeSet(const UnicodeString& pattern, ParsePosition& pos, 413 uint32_t options, 414 const SymbolTable* symbols, 415 UErrorCode& status); 416 417 /** 418 * Constructs a set that is identical to the given UnicodeSet. 419 * @stable ICU 2.0 420 */ 421 UnicodeSet(const UnicodeSet& o); 422 423 /** 424 * Destructs the set. 425 * @stable ICU 2.0 426 */ 427 virtual ~UnicodeSet(); 428 429 /** 430 * Assigns this object to be a copy of another. 431 * A frozen set will not be modified. 432 * @stable ICU 2.0 433 */ 434 UnicodeSet& operator=(const UnicodeSet& o); 435 436 /** 437 * Compares the specified object with this set for equality. Returns 438 * <tt>true</tt> if the two sets 439 * have the same size, and every member of the specified set is 440 * contained in this set (or equivalently, every member of this set is 441 * contained in the specified set). 442 * 443 * @param o set to be compared for equality with this set. 444 * @return <tt>true</tt> if the specified set is equal to this set. 445 * @stable ICU 2.0 446 */ 447 virtual UBool operator==(const UnicodeSet& o) const; 448 449 /** 450 * Compares the specified object with this set for equality. Returns 451 * <tt>true</tt> if the specified set is not equal to this set. 452 * @stable ICU 2.0 453 */ 454 UBool operator!=(const UnicodeSet& o) const; 455 456 /** 457 * Returns a copy of this object. All UnicodeFunctor objects have 458 * to support cloning in order to allow classes using 459 * UnicodeFunctors, such as Transliterator, to implement cloning. 460 * If this set is frozen, then the clone will be frozen as well. 461 * Use cloneAsThawed() for a mutable clone of a frozen set. 462 * @see cloneAsThawed 463 * @stable ICU 2.0 464 */ 465 virtual UnicodeFunctor* clone() const; 466 467 /** 468 * Returns the hash code value for this set. 469 * 470 * @return the hash code value for this set. 471 * @see Object#hashCode() 472 * @stable ICU 2.0 473 */ 474 virtual int32_t hashCode(void) const; 475 476 /** 477 * Get a UnicodeSet pointer from a USet 478 * 479 * @param uset a USet (the ICU plain C type for UnicodeSet) 480 * @return the corresponding UnicodeSet pointer. 481 * 482 * @stable ICU 4.2 483 */ 484 inline static UnicodeSet *fromUSet(USet *uset); 485 486 /** 487 * Get a UnicodeSet pointer from a const USet 488 * 489 * @param uset a const USet (the ICU plain C type for UnicodeSet) 490 * @return the corresponding UnicodeSet pointer. 491 * 492 * @stable ICU 4.2 493 */ 494 inline static const UnicodeSet *fromUSet(const USet *uset); 495 496 /** 497 * Produce a USet * pointer for this UnicodeSet. 498 * USet is the plain C type for UnicodeSet 499 * 500 * @return a USet pointer for this UnicodeSet 501 * @stable ICU 4.2 502 */ 503 inline USet *toUSet(); 504 505 506 /** 507 * Produce a const USet * pointer for this UnicodeSet. 508 * USet is the plain C type for UnicodeSet 509 * 510 * @return a const USet pointer for this UnicodeSet 511 * @stable ICU 4.2 512 */ 513 inline const USet * toUSet() const; 514 515 516 //---------------------------------------------------------------- 517 // Freezable API 518 //---------------------------------------------------------------- 519 520 /** 521 * Determines whether the set has been frozen (made immutable) or not. 522 * See the ICU4J Freezable interface for details. 523 * @return TRUE/FALSE for whether the set has been frozen 524 * @see freeze 525 * @see cloneAsThawed 526 * @stable ICU 3.8 527 */ 528 inline UBool isFrozen() const; 529 530 /** 531 * Freeze the set (make it immutable). 532 * Once frozen, it cannot be unfrozen and is therefore thread-safe 533 * until it is deleted. 534 * See the ICU4J Freezable interface for details. 535 * Freezing the set may also make some operations faster, for example 536 * contains() and span(). 537 * A frozen set will not be modified. (It remains frozen.) 538 * @return this set. 539 * @see isFrozen 540 * @see cloneAsThawed 541 * @stable ICU 3.8 542 */ 543 UnicodeFunctor *freeze(); 544 545 /** 546 * Clone the set and make the clone mutable. 547 * See the ICU4J Freezable interface for details. 548 * @return the mutable clone 549 * @see freeze 550 * @see isFrozen 551 * @stable ICU 3.8 552 */ 553 UnicodeFunctor *cloneAsThawed() const; 554 555 //---------------------------------------------------------------- 556 // Public API 557 //---------------------------------------------------------------- 558 559 /** 560 * Make this object represent the range <code>start - end</code>. 561 * If <code>end > start</code> then this object is set to an 562 * an empty range. 563 * A frozen set will not be modified. 564 * 565 * @param start first character in the set, inclusive 566 * @param end last character in the set, inclusive 567 * @stable ICU 2.4 568 */ 569 UnicodeSet& set(UChar32 start, UChar32 end); 570 571 /** 572 * Return true if the given position, in the given pattern, appears 573 * to be the start of a UnicodeSet pattern. 574 * @stable ICU 2.4 575 */ 576 static UBool resemblesPattern(const UnicodeString& pattern, 577 int32_t pos); 578 579 /** 580 * Modifies this set to represent the set specified by the given 581 * pattern, ignoring Unicode Pattern_White_Space characters. 582 * See the class description for the syntax of the pattern language. 583 * A frozen set will not be modified. 584 * @param pattern a string specifying what characters are in the set 585 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 586 * contains a syntax error. 587 * <em> Empties the set passed before applying the pattern.</em> 588 * @return a reference to this 589 * @stable ICU 2.0 590 */ 591 UnicodeSet& applyPattern(const UnicodeString& pattern, 592 UErrorCode& status); 593 594 #ifndef U_HIDE_INTERNAL_API 595 /** 596 * Modifies this set to represent the set specified by the given 597 * pattern, optionally ignoring Unicode Pattern_White_Space characters. 598 * See the class description for the syntax of the pattern language. 599 * A frozen set will not be modified. 600 * @param pattern a string specifying what characters are in the set 601 * @param options bitmask for options to apply to the pattern. 602 * Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE. 603 * @param symbols a symbol table mapping variable names to 604 * values and stand-ins to UnicodeSets; may be NULL 605 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 606 * contains a syntax error. 607 *<em> Empties the set passed before applying the pattern.</em> 608 * @return a reference to this 609 * @internal 610 */ 611 UnicodeSet& applyPattern(const UnicodeString& pattern, 612 uint32_t options, 613 const SymbolTable* symbols, 614 UErrorCode& status); 615 #endif /* U_HIDE_INTERNAL_API */ 616 617 /** 618 * Parses the given pattern, starting at the given position. The 619 * character at pattern.charAt(pos.getIndex()) must be '[', or the 620 * parse fails. Parsing continues until the corresponding closing 621 * ']'. If a syntax error is encountered between the opening and 622 * closing brace, the parse fails. Upon return from a successful 623 * parse, the ParsePosition is updated to point to the character 624 * following the closing ']', and a StringBuffer containing a 625 * pairs list for the parsed pattern is returned. This method calls 626 * itself recursively to parse embedded subpatterns. 627 *<em> Empties the set passed before applying the pattern.</em> 628 * A frozen set will not be modified. 629 * 630 * @param pattern the string containing the pattern to be parsed. 631 * The portion of the string from pos.getIndex(), which must be a 632 * '[', to the corresponding closing ']', is parsed. 633 * @param pos upon entry, the position at which to being parsing. 634 * The character at pattern.charAt(pos.getIndex()) must be a '['. 635 * Upon return from a successful parse, pos.getIndex() is either 636 * the character after the closing ']' of the parsed pattern, or 637 * pattern.length() if the closing ']' is the last character of 638 * the pattern string. 639 * @param options bitmask for options to apply to the pattern. 640 * Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE. 641 * @param symbols a symbol table mapping variable names to 642 * values and stand-ins to UnicodeSets; may be NULL 643 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 644 * contains a syntax error. 645 * @return a reference to this 646 * @stable ICU 2.8 647 */ 648 UnicodeSet& applyPattern(const UnicodeString& pattern, 649 ParsePosition& pos, 650 uint32_t options, 651 const SymbolTable* symbols, 652 UErrorCode& status); 653 654 /** 655 * Returns a string representation of this set. If the result of 656 * calling this function is passed to a UnicodeSet constructor, it 657 * will produce another set that is equal to this one. 658 * A frozen set will not be modified. 659 * @param result the string to receive the rules. Previous 660 * contents will be deleted. 661 * @param escapeUnprintable if TRUE then convert unprintable 662 * character to their hex escape representations, \\uxxxx or 663 * \\Uxxxxxxxx. Unprintable characters are those other than 664 * U+000A, U+0020..U+007E. 665 * @stable ICU 2.0 666 */ 667 virtual UnicodeString& toPattern(UnicodeString& result, 668 UBool escapeUnprintable = FALSE) const; 669 670 /** 671 * Modifies this set to contain those code points which have the given value 672 * for the given binary or enumerated property, as returned by 673 * u_getIntPropertyValue. Prior contents of this set are lost. 674 * A frozen set will not be modified. 675 * 676 * @param prop a property in the range UCHAR_BIN_START..UCHAR_BIN_LIMIT-1 677 * or UCHAR_INT_START..UCHAR_INT_LIMIT-1 678 * or UCHAR_MASK_START..UCHAR_MASK_LIMIT-1. 679 * 680 * @param value a value in the range u_getIntPropertyMinValue(prop).. 681 * u_getIntPropertyMaxValue(prop), with one exception. If prop is 682 * UCHAR_GENERAL_CATEGORY_MASK, then value should not be a UCharCategory, but 683 * rather a mask value produced by U_GET_GC_MASK(). This allows grouped 684 * categories such as [:L:] to be represented. 685 * 686 * @param ec error code input/output parameter 687 * 688 * @return a reference to this set 689 * 690 * @stable ICU 2.4 691 */ 692 UnicodeSet& applyIntPropertyValue(UProperty prop, 693 int32_t value, 694 UErrorCode& ec); 695 696 /** 697 * Modifies this set to contain those code points which have the 698 * given value for the given property. Prior contents of this 699 * set are lost. 700 * A frozen set will not be modified. 701 * 702 * @param prop a property alias, either short or long. The name is matched 703 * loosely. See PropertyAliases.txt for names and a description of loose 704 * matching. If the value string is empty, then this string is interpreted 705 * as either a General_Category value alias, a Script value alias, a binary 706 * property alias, or a special ID. Special IDs are matched loosely and 707 * correspond to the following sets: 708 * 709 * "ANY" = [\\u0000-\\U0010FFFF], 710 * "ASCII" = [\\u0000-\\u007F], 711 * "Assigned" = [:^Cn:]. 712 * 713 * @param value a value alias, either short or long. The name is matched 714 * loosely. See PropertyValueAliases.txt for names and a description of 715 * loose matching. In addition to aliases listed, numeric values and 716 * canonical combining classes may be expressed numerically, e.g., ("nv", 717 * "0.5") or ("ccc", "220"). The value string may also be empty. 718 * 719 * @param ec error code input/output parameter 720 * 721 * @return a reference to this set 722 * 723 * @stable ICU 2.4 724 */ 725 UnicodeSet& applyPropertyAlias(const UnicodeString& prop, 726 const UnicodeString& value, 727 UErrorCode& ec); 728 729 /** 730 * Returns the number of elements in this set (its cardinality). 731 * Note than the elements of a set may include both individual 732 * codepoints and strings. 733 * 734 * @return the number of elements in this set (its cardinality). 735 * @stable ICU 2.0 736 */ 737 virtual int32_t size(void) const; 738 739 /** 740 * Returns <tt>true</tt> if this set contains no elements. 741 * 742 * @return <tt>true</tt> if this set contains no elements. 743 * @stable ICU 2.0 744 */ 745 virtual UBool isEmpty(void) const; 746 747 /** 748 * Returns true if this set contains the given character. 749 * This function works faster with a frozen set. 750 * @param c character to be checked for containment 751 * @return true if the test condition is met 752 * @stable ICU 2.0 753 */ 754 virtual UBool contains(UChar32 c) const; 755 756 /** 757 * Returns true if this set contains every character 758 * of the given range. 759 * @param start first character, inclusive, of the range 760 * @param end last character, inclusive, of the range 761 * @return true if the test condition is met 762 * @stable ICU 2.0 763 */ 764 virtual UBool contains(UChar32 start, UChar32 end) const; 765 766 /** 767 * Returns <tt>true</tt> if this set contains the given 768 * multicharacter string. 769 * @param s string to be checked for containment 770 * @return <tt>true</tt> if this set contains the specified string 771 * @stable ICU 2.4 772 */ 773 UBool contains(const UnicodeString& s) const; 774 775 /** 776 * Returns true if this set contains all the characters and strings 777 * of the given set. 778 * @param c set to be checked for containment 779 * @return true if the test condition is met 780 * @stable ICU 2.4 781 */ 782 virtual UBool containsAll(const UnicodeSet& c) const; 783 784 /** 785 * Returns true if this set contains all the characters 786 * of the given string. 787 * @param s string containing characters to be checked for containment 788 * @return true if the test condition is met 789 * @stable ICU 2.4 790 */ 791 UBool containsAll(const UnicodeString& s) const; 792 793 /** 794 * Returns true if this set contains none of the characters 795 * of the given range. 796 * @param start first character, inclusive, of the range 797 * @param end last character, inclusive, of the range 798 * @return true if the test condition is met 799 * @stable ICU 2.4 800 */ 801 UBool containsNone(UChar32 start, UChar32 end) const; 802 803 /** 804 * Returns true if this set contains none of the characters and strings 805 * of the given set. 806 * @param c set to be checked for containment 807 * @return true if the test condition is met 808 * @stable ICU 2.4 809 */ 810 UBool containsNone(const UnicodeSet& c) const; 811 812 /** 813 * Returns true if this set contains none of the characters 814 * of the given string. 815 * @param s string containing characters to be checked for containment 816 * @return true if the test condition is met 817 * @stable ICU 2.4 818 */ 819 UBool containsNone(const UnicodeString& s) const; 820 821 /** 822 * Returns true if this set contains one or more of the characters 823 * in the given range. 824 * @param start first character, inclusive, of the range 825 * @param end last character, inclusive, of the range 826 * @return true if the condition is met 827 * @stable ICU 2.4 828 */ 829 inline UBool containsSome(UChar32 start, UChar32 end) const; 830 831 /** 832 * Returns true if this set contains one or more of the characters 833 * and strings of the given set. 834 * @param s The set to be checked for containment 835 * @return true if the condition is met 836 * @stable ICU 2.4 837 */ 838 inline UBool containsSome(const UnicodeSet& s) const; 839 840 /** 841 * Returns true if this set contains one or more of the characters 842 * of the given string. 843 * @param s string containing characters to be checked for containment 844 * @return true if the condition is met 845 * @stable ICU 2.4 846 */ 847 inline UBool containsSome(const UnicodeString& s) const; 848 849 /** 850 * Returns the length of the initial substring of the input string which 851 * consists only of characters and strings that are contained in this set 852 * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE), 853 * or only of characters and strings that are not contained 854 * in this set (USET_SPAN_NOT_CONTAINED). 855 * See USetSpanCondition for details. 856 * Similar to the strspn() C library function. 857 * Unpaired surrogates are treated according to contains() of their surrogate code points. 858 * This function works faster with a frozen set and with a non-negative string length argument. 859 * @param s start of the string 860 * @param length of the string; can be -1 for NUL-terminated 861 * @param spanCondition specifies the containment condition 862 * @return the length of the initial substring according to the spanCondition; 863 * 0 if the start of the string does not fit the spanCondition 864 * @stable ICU 3.8 865 * @see USetSpanCondition 866 */ 867 int32_t span(const UChar *s, int32_t length, USetSpanCondition spanCondition) const; 868 869 /** 870 * Returns the end of the substring of the input string according to the USetSpanCondition. 871 * Same as <code>start+span(s.getBuffer()+start, s.length()-start, spanCondition)</code> 872 * after pinning start to 0<=start<=s.length(). 873 * @param s the string 874 * @param start the start index in the string for the span operation 875 * @param spanCondition specifies the containment condition 876 * @return the exclusive end of the substring according to the spanCondition; 877 * the substring s.tempSubStringBetween(start, end) fulfills the spanCondition 878 * @stable ICU 4.4 879 * @see USetSpanCondition 880 */ 881 inline int32_t span(const UnicodeString &s, int32_t start, USetSpanCondition spanCondition) const; 882 883 /** 884 * Returns the start of the trailing substring of the input string which 885 * consists only of characters and strings that are contained in this set 886 * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE), 887 * or only of characters and strings that are not contained 888 * in this set (USET_SPAN_NOT_CONTAINED). 889 * See USetSpanCondition for details. 890 * Unpaired surrogates are treated according to contains() of their surrogate code points. 891 * This function works faster with a frozen set and with a non-negative string length argument. 892 * @param s start of the string 893 * @param length of the string; can be -1 for NUL-terminated 894 * @param spanCondition specifies the containment condition 895 * @return the start of the trailing substring according to the spanCondition; 896 * the string length if the end of the string does not fit the spanCondition 897 * @stable ICU 3.8 898 * @see USetSpanCondition 899 */ 900 int32_t spanBack(const UChar *s, int32_t length, USetSpanCondition spanCondition) const; 901 902 /** 903 * Returns the start of the substring of the input string according to the USetSpanCondition. 904 * Same as <code>spanBack(s.getBuffer(), limit, spanCondition)</code> 905 * after pinning limit to 0<=end<=s.length(). 906 * @param s the string 907 * @param limit the exclusive-end index in the string for the span operation 908 * (use s.length() or INT32_MAX for spanning back from the end of the string) 909 * @param spanCondition specifies the containment condition 910 * @return the start of the substring according to the spanCondition; 911 * the substring s.tempSubStringBetween(start, limit) fulfills the spanCondition 912 * @stable ICU 4.4 913 * @see USetSpanCondition 914 */ 915 inline int32_t spanBack(const UnicodeString &s, int32_t limit, USetSpanCondition spanCondition) const; 916 917 /** 918 * Returns the length of the initial substring of the input string which 919 * consists only of characters and strings that are contained in this set 920 * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE), 921 * or only of characters and strings that are not contained 922 * in this set (USET_SPAN_NOT_CONTAINED). 923 * See USetSpanCondition for details. 924 * Similar to the strspn() C library function. 925 * Malformed byte sequences are treated according to contains(0xfffd). 926 * This function works faster with a frozen set and with a non-negative string length argument. 927 * @param s start of the string (UTF-8) 928 * @param length of the string; can be -1 for NUL-terminated 929 * @param spanCondition specifies the containment condition 930 * @return the length of the initial substring according to the spanCondition; 931 * 0 if the start of the string does not fit the spanCondition 932 * @stable ICU 3.8 933 * @see USetSpanCondition 934 */ 935 int32_t spanUTF8(const char *s, int32_t length, USetSpanCondition spanCondition) const; 936 937 /** 938 * Returns the start of the trailing substring of the input string which 939 * consists only of characters and strings that are contained in this set 940 * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE), 941 * or only of characters and strings that are not contained 942 * in this set (USET_SPAN_NOT_CONTAINED). 943 * See USetSpanCondition for details. 944 * Malformed byte sequences are treated according to contains(0xfffd). 945 * This function works faster with a frozen set and with a non-negative string length argument. 946 * @param s start of the string (UTF-8) 947 * @param length of the string; can be -1 for NUL-terminated 948 * @param spanCondition specifies the containment condition 949 * @return the start of the trailing substring according to the spanCondition; 950 * the string length if the end of the string does not fit the spanCondition 951 * @stable ICU 3.8 952 * @see USetSpanCondition 953 */ 954 int32_t spanBackUTF8(const char *s, int32_t length, USetSpanCondition spanCondition) const; 955 956 /** 957 * Implement UnicodeMatcher::matches() 958 * @stable ICU 2.4 959 */ 960 virtual UMatchDegree matches(const Replaceable& text, 961 int32_t& offset, 962 int32_t limit, 963 UBool incremental); 964 965 private: 966 /** 967 * Returns the longest match for s in text at the given position. 968 * If limit > start then match forward from start+1 to limit 969 * matching all characters except s.charAt(0). If limit < start, 970 * go backward starting from start-1 matching all characters 971 * except s.charAt(s.length()-1). This method assumes that the 972 * first character, text.charAt(start), matches s, so it does not 973 * check it. 974 * @param text the text to match 975 * @param start the first character to match. In the forward 976 * direction, text.charAt(start) is matched against s.charAt(0). 977 * In the reverse direction, it is matched against 978 * s.charAt(s.length()-1). 979 * @param limit the limit offset for matching, either last+1 in 980 * the forward direction, or last-1 in the reverse direction, 981 * where last is the index of the last character to match. 982 * @param s 983 * @return If part of s matches up to the limit, return |limit - 984 * start|. If all of s matches before reaching the limit, return 985 * s.length(). If there is a mismatch between s and text, return 986 * 0 987 */ 988 static int32_t matchRest(const Replaceable& text, 989 int32_t start, int32_t limit, 990 const UnicodeString& s); 991 992 /** 993 * Returns the smallest value i such that c < list[i]. Caller 994 * must ensure that c is a legal value or this method will enter 995 * an infinite loop. This method performs a binary search. 996 * @param c a character in the range MIN_VALUE..MAX_VALUE 997 * inclusive 998 * @return the smallest integer i in the range 0..len-1, 999 * inclusive, such that c < list[i] 1000 */ 1001 int32_t findCodePoint(UChar32 c) const; 1002 1003 public: 1004 1005 /** 1006 * Implementation of UnicodeMatcher API. Union the set of all 1007 * characters that may be matched by this object into the given 1008 * set. 1009 * @param toUnionTo the set into which to union the source characters 1010 * @stable ICU 2.4 1011 */ 1012 virtual void addMatchSetTo(UnicodeSet& toUnionTo) const; 1013 1014 /** 1015 * Returns the index of the given character within this set, where 1016 * the set is ordered by ascending code point. If the character 1017 * is not in this set, return -1. The inverse of this method is 1018 * <code>charAt()</code>. 1019 * @return an index from 0..size()-1, or -1 1020 * @stable ICU 2.4 1021 */ 1022 int32_t indexOf(UChar32 c) const; 1023 1024 /** 1025 * Returns the character at the given index within this set, where 1026 * the set is ordered by ascending code point. If the index is 1027 * out of range, return (UChar32)-1. The inverse of this method is 1028 * <code>indexOf()</code>. 1029 * @param index an index from 0..size()-1 1030 * @return the character at the given index, or (UChar32)-1. 1031 * @stable ICU 2.4 1032 */ 1033 UChar32 charAt(int32_t index) const; 1034 1035 /** 1036 * Adds the specified range to this set if it is not already 1037 * present. If this set already contains the specified range, 1038 * the call leaves this set unchanged. If <code>end > start</code> 1039 * then an empty range is added, leaving the set unchanged. 1040 * This is equivalent to a boolean logic OR, or a set UNION. 1041 * A frozen set will not be modified. 1042 * 1043 * @param start first character, inclusive, of range to be added 1044 * to this set. 1045 * @param end last character, inclusive, of range to be added 1046 * to this set. 1047 * @stable ICU 2.0 1048 */ 1049 virtual UnicodeSet& add(UChar32 start, UChar32 end); 1050 1051 /** 1052 * Adds the specified character to this set if it is not already 1053 * present. If this set already contains the specified character, 1054 * the call leaves this set unchanged. 1055 * A frozen set will not be modified. 1056 * @stable ICU 2.0 1057 */ 1058 UnicodeSet& add(UChar32 c); 1059 1060 /** 1061 * Adds the specified multicharacter to this set if it is not already 1062 * present. If this set already contains the multicharacter, 1063 * the call leaves this set unchanged. 1064 * Thus "ch" => {"ch"} 1065 * <br><b>Warning: you cannot add an empty string ("") to a UnicodeSet.</b> 1066 * A frozen set will not be modified. 1067 * @param s the source string 1068 * @return this object, for chaining 1069 * @stable ICU 2.4 1070 */ 1071 UnicodeSet& add(const UnicodeString& s); 1072 1073 private: 1074 /** 1075 * @return a code point IF the string consists of a single one. 1076 * otherwise returns -1. 1077 * @param s string to test 1078 */ 1079 static int32_t getSingleCP(const UnicodeString& s); 1080 1081 void _add(const UnicodeString& s); 1082 1083 public: 1084 /** 1085 * Adds each of the characters in this string to the set. Thus "ch" => {"c", "h"} 1086 * If this set already any particular character, it has no effect on that character. 1087 * A frozen set will not be modified. 1088 * @param s the source string 1089 * @return this object, for chaining 1090 * @stable ICU 2.4 1091 */ 1092 UnicodeSet& addAll(const UnicodeString& s); 1093 1094 /** 1095 * Retains EACH of the characters in this string. Note: "ch" == {"c", "h"} 1096 * If this set already any particular character, it has no effect on that character. 1097 * A frozen set will not be modified. 1098 * @param s the source string 1099 * @return this object, for chaining 1100 * @stable ICU 2.4 1101 */ 1102 UnicodeSet& retainAll(const UnicodeString& s); 1103 1104 /** 1105 * Complement EACH of the characters in this string. Note: "ch" == {"c", "h"} 1106 * If this set already any particular character, it has no effect on that character. 1107 * A frozen set will not be modified. 1108 * @param s the source string 1109 * @return this object, for chaining 1110 * @stable ICU 2.4 1111 */ 1112 UnicodeSet& complementAll(const UnicodeString& s); 1113 1114 /** 1115 * Remove EACH of the characters in this string. Note: "ch" == {"c", "h"} 1116 * If this set already any particular character, it has no effect on that character. 1117 * A frozen set will not be modified. 1118 * @param s the source string 1119 * @return this object, for chaining 1120 * @stable ICU 2.4 1121 */ 1122 UnicodeSet& removeAll(const UnicodeString& s); 1123 1124 /** 1125 * Makes a set from a multicharacter string. Thus "ch" => {"ch"} 1126 * <br><b>Warning: you cannot add an empty string ("") to a UnicodeSet.</b> 1127 * @param s the source string 1128 * @return a newly created set containing the given string. 1129 * The caller owns the return object and is responsible for deleting it. 1130 * @stable ICU 2.4 1131 */ 1132 static UnicodeSet* U_EXPORT2 createFrom(const UnicodeString& s); 1133 1134 1135 /** 1136 * Makes a set from each of the characters in the string. Thus "ch" => {"c", "h"} 1137 * @param s the source string 1138 * @return a newly created set containing the given characters 1139 * The caller owns the return object and is responsible for deleting it. 1140 * @stable ICU 2.4 1141 */ 1142 static UnicodeSet* U_EXPORT2 createFromAll(const UnicodeString& s); 1143 1144 /** 1145 * Retain only the elements in this set that are contained in the 1146 * specified range. If <code>end > start</code> then an empty range is 1147 * retained, leaving the set empty. This is equivalent to 1148 * a boolean logic AND, or a set INTERSECTION. 1149 * A frozen set will not be modified. 1150 * 1151 * @param start first character, inclusive, of range to be retained 1152 * to this set. 1153 * @param end last character, inclusive, of range to be retained 1154 * to this set. 1155 * @stable ICU 2.0 1156 */ 1157 virtual UnicodeSet& retain(UChar32 start, UChar32 end); 1158 1159 1160 /** 1161 * Retain the specified character from this set if it is present. 1162 * A frozen set will not be modified. 1163 * @stable ICU 2.0 1164 */ 1165 UnicodeSet& retain(UChar32 c); 1166 1167 /** 1168 * Removes the specified range from this set if it is present. 1169 * The set will not contain the specified range once the call 1170 * returns. If <code>end > start</code> then an empty range is 1171 * removed, leaving the set unchanged. 1172 * A frozen set will not be modified. 1173 * 1174 * @param start first character, inclusive, of range to be removed 1175 * from this set. 1176 * @param end last character, inclusive, of range to be removed 1177 * from this set. 1178 * @stable ICU 2.0 1179 */ 1180 virtual UnicodeSet& remove(UChar32 start, UChar32 end); 1181 1182 /** 1183 * Removes the specified character from this set if it is present. 1184 * The set will not contain the specified range once the call 1185 * returns. 1186 * A frozen set will not be modified. 1187 * @stable ICU 2.0 1188 */ 1189 UnicodeSet& remove(UChar32 c); 1190 1191 /** 1192 * Removes the specified string from this set if it is present. 1193 * The set will not contain the specified character once the call 1194 * returns. 1195 * A frozen set will not be modified. 1196 * @param s the source string 1197 * @return this object, for chaining 1198 * @stable ICU 2.4 1199 */ 1200 UnicodeSet& remove(const UnicodeString& s); 1201 1202 /** 1203 * Inverts this set. This operation modifies this set so that 1204 * its value is its complement. This is equivalent to 1205 * <code>complement(MIN_VALUE, MAX_VALUE)</code>. 1206 * A frozen set will not be modified. 1207 * @stable ICU 2.0 1208 */ 1209 virtual UnicodeSet& complement(void); 1210 1211 /** 1212 * Complements the specified range in this set. Any character in 1213 * the range will be removed if it is in this set, or will be 1214 * added if it is not in this set. If <code>end > start</code> 1215 * then an empty range is complemented, leaving the set unchanged. 1216 * This is equivalent to a boolean logic XOR. 1217 * A frozen set will not be modified. 1218 * 1219 * @param start first character, inclusive, of range to be removed 1220 * from this set. 1221 * @param end last character, inclusive, of range to be removed 1222 * from this set. 1223 * @stable ICU 2.0 1224 */ 1225 virtual UnicodeSet& complement(UChar32 start, UChar32 end); 1226 1227 /** 1228 * Complements the specified character in this set. The character 1229 * will be removed if it is in this set, or will be added if it is 1230 * not in this set. 1231 * A frozen set will not be modified. 1232 * @stable ICU 2.0 1233 */ 1234 UnicodeSet& complement(UChar32 c); 1235 1236 /** 1237 * Complement the specified string in this set. 1238 * The set will not contain the specified string once the call 1239 * returns. 1240 * <br><b>Warning: you cannot add an empty string ("") to a UnicodeSet.</b> 1241 * A frozen set will not be modified. 1242 * @param s the string to complement 1243 * @return this object, for chaining 1244 * @stable ICU 2.4 1245 */ 1246 UnicodeSet& complement(const UnicodeString& s); 1247 1248 /** 1249 * Adds all of the elements in the specified set to this set if 1250 * they're not already present. This operation effectively 1251 * modifies this set so that its value is the <i>union</i> of the two 1252 * sets. The behavior of this operation is unspecified if the specified 1253 * collection is modified while the operation is in progress. 1254 * A frozen set will not be modified. 1255 * 1256 * @param c set whose elements are to be added to this set. 1257 * @see #add(UChar32, UChar32) 1258 * @stable ICU 2.0 1259 */ 1260 virtual UnicodeSet& addAll(const UnicodeSet& c); 1261 1262 /** 1263 * Retains only the elements in this set that are contained in the 1264 * specified set. In other words, removes from this set all of 1265 * its elements that are not contained in the specified set. This 1266 * operation effectively modifies this set so that its value is 1267 * the <i>intersection</i> of the two sets. 1268 * A frozen set will not be modified. 1269 * 1270 * @param c set that defines which elements this set will retain. 1271 * @stable ICU 2.0 1272 */ 1273 virtual UnicodeSet& retainAll(const UnicodeSet& c); 1274 1275 /** 1276 * Removes from this set all of its elements that are contained in the 1277 * specified set. This operation effectively modifies this 1278 * set so that its value is the <i>asymmetric set difference</i> of 1279 * the two sets. 1280 * A frozen set will not be modified. 1281 * 1282 * @param c set that defines which elements will be removed from 1283 * this set. 1284 * @stable ICU 2.0 1285 */ 1286 virtual UnicodeSet& removeAll(const UnicodeSet& c); 1287 1288 /** 1289 * Complements in this set all elements contained in the specified 1290 * set. Any character in the other set will be removed if it is 1291 * in this set, or will be added if it is not in this set. 1292 * A frozen set will not be modified. 1293 * 1294 * @param c set that defines which elements will be xor'ed from 1295 * this set. 1296 * @stable ICU 2.4 1297 */ 1298 virtual UnicodeSet& complementAll(const UnicodeSet& c); 1299 1300 /** 1301 * Removes all of the elements from this set. This set will be 1302 * empty after this call returns. 1303 * A frozen set will not be modified. 1304 * @stable ICU 2.0 1305 */ 1306 virtual UnicodeSet& clear(void); 1307 1308 /** 1309 * Close this set over the given attribute. For the attribute 1310 * USET_CASE, the result is to modify this set so that: 1311 * 1312 * 1. For each character or string 'a' in this set, all strings or 1313 * characters 'b' such that foldCase(a) == foldCase(b) are added 1314 * to this set. 1315 * 1316 * 2. For each string 'e' in the resulting set, if e != 1317 * foldCase(e), 'e' will be removed. 1318 * 1319 * Example: [aq\\u00DF{Bc}{bC}{Fi}] => [aAqQ\\u00DF\\uFB01{ss}{bc}{fi}] 1320 * 1321 * (Here foldCase(x) refers to the operation u_strFoldCase, and a 1322 * == b denotes that the contents are the same, not pointer 1323 * comparison.) 1324 * 1325 * A frozen set will not be modified. 1326 * 1327 * @param attribute bitmask for attributes to close over. 1328 * Currently only the USET_CASE bit is supported. Any undefined bits 1329 * are ignored. 1330 * @return a reference to this set. 1331 * @stable ICU 4.2 1332 */ 1333 UnicodeSet& closeOver(int32_t attribute); 1334 1335 /** 1336 * Remove all strings from this set. 1337 * 1338 * @return a reference to this set. 1339 * @stable ICU 4.2 1340 */ 1341 virtual UnicodeSet &removeAllStrings(); 1342 1343 /** 1344 * Iteration method that returns the number of ranges contained in 1345 * this set. 1346 * @see #getRangeStart 1347 * @see #getRangeEnd 1348 * @stable ICU 2.4 1349 */ 1350 virtual int32_t getRangeCount(void) const; 1351 1352 /** 1353 * Iteration method that returns the first character in the 1354 * specified range of this set. 1355 * @see #getRangeCount 1356 * @see #getRangeEnd 1357 * @stable ICU 2.4 1358 */ 1359 virtual UChar32 getRangeStart(int32_t index) const; 1360 1361 /** 1362 * Iteration method that returns the last character in the 1363 * specified range of this set. 1364 * @see #getRangeStart 1365 * @see #getRangeEnd 1366 * @stable ICU 2.4 1367 */ 1368 virtual UChar32 getRangeEnd(int32_t index) const; 1369 1370 /** 1371 * Serializes this set into an array of 16-bit integers. Serialization 1372 * (currently) only records the characters in the set; multicharacter 1373 * strings are ignored. 1374 * 1375 * The array has following format (each line is one 16-bit 1376 * integer): 1377 * 1378 * length = (n+2*m) | (m!=0?0x8000:0) 1379 * bmpLength = n; present if m!=0 1380 * bmp[0] 1381 * bmp[1] 1382 * ... 1383 * bmp[n-1] 1384 * supp-high[0] 1385 * supp-low[0] 1386 * supp-high[1] 1387 * supp-low[1] 1388 * ... 1389 * supp-high[m-1] 1390 * supp-low[m-1] 1391 * 1392 * The array starts with a header. After the header are n bmp 1393 * code points, then m supplementary code points. Either n or m 1394 * or both may be zero. n+2*m is always <= 0x7FFF. 1395 * 1396 * If there are no supplementary characters (if m==0) then the 1397 * header is one 16-bit integer, 'length', with value n. 1398 * 1399 * If there are supplementary characters (if m!=0) then the header 1400 * is two 16-bit integers. The first, 'length', has value 1401 * (n+2*m)|0x8000. The second, 'bmpLength', has value n. 1402 * 1403 * After the header the code points are stored in ascending order. 1404 * Supplementary code points are stored as most significant 16 1405 * bits followed by least significant 16 bits. 1406 * 1407 * @param dest pointer to buffer of destCapacity 16-bit integers. 1408 * May be NULL only if destCapacity is zero. 1409 * @param destCapacity size of dest, or zero. Must not be negative. 1410 * @param ec error code. Will be set to U_INDEX_OUTOFBOUNDS_ERROR 1411 * if n+2*m > 0x7FFF. Will be set to U_BUFFER_OVERFLOW_ERROR if 1412 * n+2*m+(m!=0?2:1) > destCapacity. 1413 * @return the total length of the serialized format, including 1414 * the header, that is, n+2*m+(m!=0?2:1), or 0 on error other 1415 * than U_BUFFER_OVERFLOW_ERROR. 1416 * @stable ICU 2.4 1417 */ 1418 int32_t serialize(uint16_t *dest, int32_t destCapacity, UErrorCode& ec) const; 1419 1420 /** 1421 * Reallocate this objects internal structures to take up the least 1422 * possible space, without changing this object's value. 1423 * A frozen set will not be modified. 1424 * @stable ICU 2.4 1425 */ 1426 virtual UnicodeSet& compact(); 1427 1428 /** 1429 * Return the class ID for this class. This is useful only for 1430 * comparing to a return value from getDynamicClassID(). For example: 1431 * <pre> 1432 * . Base* polymorphic_pointer = createPolymorphicObject(); 1433 * . if (polymorphic_pointer->getDynamicClassID() == 1434 * . Derived::getStaticClassID()) ... 1435 * </pre> 1436 * @return The class ID for all objects of this class. 1437 * @stable ICU 2.0 1438 */ 1439 static UClassID U_EXPORT2 getStaticClassID(void); 1440 1441 /** 1442 * Implement UnicodeFunctor API. 1443 * 1444 * @return The class ID for this object. All objects of a given 1445 * class have the same class ID. Objects of other classes have 1446 * different class IDs. 1447 * @stable ICU 2.4 1448 */ 1449 virtual UClassID getDynamicClassID(void) const; 1450 1451 private: 1452 1453 // Private API for the USet API 1454 1455 friend class USetAccess; 1456 1457 int32_t getStringCount() const; 1458 1459 const UnicodeString* getString(int32_t index) const; 1460 1461 //---------------------------------------------------------------- 1462 // RuleBasedTransliterator support 1463 //---------------------------------------------------------------- 1464 1465 private: 1466 1467 /** 1468 * Returns <tt>true</tt> if this set contains any character whose low byte 1469 * is the given value. This is used by <tt>RuleBasedTransliterator</tt> for 1470 * indexing. 1471 */ 1472 virtual UBool matchesIndexValue(uint8_t v) const; 1473 1474 private: 1475 friend class RBBIRuleScanner; 1476 1477 //---------------------------------------------------------------- 1478 // Implementation: Clone as thawed (see ICU4J Freezable) 1479 //---------------------------------------------------------------- 1480 1481 UnicodeSet(const UnicodeSet& o, UBool /* asThawed */); 1482 1483 //---------------------------------------------------------------- 1484 // Implementation: Pattern parsing 1485 //---------------------------------------------------------------- 1486 1487 void applyPatternIgnoreSpace(const UnicodeString& pattern, 1488 ParsePosition& pos, 1489 const SymbolTable* symbols, 1490 UErrorCode& status); 1491 1492 void applyPattern(RuleCharacterIterator& chars, 1493 const SymbolTable* symbols, 1494 UnicodeString& rebuiltPat, 1495 uint32_t options, 1496 UnicodeSet& (UnicodeSet::*caseClosure)(int32_t attribute), 1497 UErrorCode& ec); 1498 1499 //---------------------------------------------------------------- 1500 // Implementation: Utility methods 1501 //---------------------------------------------------------------- 1502 1503 void ensureCapacity(int32_t newLen, UErrorCode& ec); 1504 1505 void ensureBufferCapacity(int32_t newLen, UErrorCode& ec); 1506 1507 void swapBuffers(void); 1508 1509 UBool allocateStrings(UErrorCode &status); 1510 1511 UnicodeString& _toPattern(UnicodeString& result, 1512 UBool escapeUnprintable) const; 1513 1514 UnicodeString& _generatePattern(UnicodeString& result, 1515 UBool escapeUnprintable) const; 1516 1517 static void _appendToPat(UnicodeString& buf, const UnicodeString& s, UBool escapeUnprintable); 1518 1519 static void _appendToPat(UnicodeString& buf, UChar32 c, UBool escapeUnprintable); 1520 1521 //---------------------------------------------------------------- 1522 // Implementation: Fundamental operators 1523 //---------------------------------------------------------------- 1524 1525 void exclusiveOr(const UChar32* other, int32_t otherLen, int8_t polarity); 1526 1527 void add(const UChar32* other, int32_t otherLen, int8_t polarity); 1528 1529 void retain(const UChar32* other, int32_t otherLen, int8_t polarity); 1530 1531 /** 1532 * Return true if the given position, in the given pattern, appears 1533 * to be the start of a property set pattern [:foo:], \\p{foo}, or 1534 * \\P{foo}, or \\N{name}. 1535 */ 1536 static UBool resemblesPropertyPattern(const UnicodeString& pattern, 1537 int32_t pos); 1538 1539 static UBool resemblesPropertyPattern(RuleCharacterIterator& chars, 1540 int32_t iterOpts); 1541 1542 /** 1543 * Parse the given property pattern at the given parse position 1544 * and set this UnicodeSet to the result. 1545 * 1546 * The original design document is out of date, but still useful. 1547 * Ignore the property and value names: 1548 * http://source.icu-project.org/repos/icu/icuhtml/trunk/design/unicodeset_properties.html 1549 * 1550 * Recognized syntax: 1551 * 1552 * [:foo:] [:^foo:] - white space not allowed within "[:" or ":]" 1553 * \\p{foo} \\P{foo} - white space not allowed within "\\p" or "\\P" 1554 * \\N{name} - white space not allowed within "\\N" 1555 * 1556 * Other than the above restrictions, Unicode Pattern_White_Space characters are ignored. 1557 * Case is ignored except in "\\p" and "\\P" and "\\N". In 'name' leading 1558 * and trailing space is deleted, and internal runs of whitespace 1559 * are collapsed to a single space. 1560 * 1561 * We support binary properties, enumerated properties, and the 1562 * following non-enumerated properties: 1563 * 1564 * Numeric_Value 1565 * Name 1566 * Unicode_1_Name 1567 * 1568 * @param pattern the pattern string 1569 * @param ppos on entry, the position at which to begin parsing. 1570 * This should be one of the locations marked '^': 1571 * 1572 * [:blah:] \\p{blah} \\P{blah} \\N{name} 1573 * ^ % ^ % ^ % ^ % 1574 * 1575 * On return, the position after the last character parsed, that is, 1576 * the locations marked '%'. If the parse fails, ppos is returned 1577 * unchanged. 1578 * @param ec status 1579 * @return a reference to this. 1580 */ 1581 UnicodeSet& applyPropertyPattern(const UnicodeString& pattern, 1582 ParsePosition& ppos, 1583 UErrorCode &ec); 1584 1585 void applyPropertyPattern(RuleCharacterIterator& chars, 1586 UnicodeString& rebuiltPat, 1587 UErrorCode& ec); 1588 1589 static const UnicodeSet* getInclusions(int32_t src, UErrorCode &status); 1590 1591 /** 1592 * A filter that returns TRUE if the given code point should be 1593 * included in the UnicodeSet being constructed. 1594 */ 1595 typedef UBool (*Filter)(UChar32 codePoint, void* context); 1596 1597 /** 1598 * Given a filter, set this UnicodeSet to the code points 1599 * contained by that filter. The filter MUST be 1600 * property-conformant. That is, if it returns value v for one 1601 * code point, then it must return v for all affiliated code 1602 * points, as defined by the inclusions list. See 1603 * getInclusions(). 1604 * src is a UPropertySource value. 1605 */ 1606 void applyFilter(Filter filter, 1607 void* context, 1608 int32_t src, 1609 UErrorCode &status); 1610 1611 /** 1612 * Set the new pattern to cache. 1613 */ 1614 void setPattern(const UnicodeString& newPat); 1615 /** 1616 * Release existing cached pattern. 1617 */ 1618 void releasePattern(); 1619 1620 friend class UnicodeSetIterator; 1621 }; 1622 1623 1624 1625 inline UBool UnicodeSet::operator!=(const UnicodeSet& o) const { 1626 return !operator==(o); 1627 } 1628 1629 inline UBool UnicodeSet::isFrozen() const { 1630 return (UBool)(bmpSet!=NULL || stringSpan!=NULL); 1631 } 1632 1633 inline UBool UnicodeSet::containsSome(UChar32 start, UChar32 end) const { 1634 return !containsNone(start, end); 1635 } 1636 1637 inline UBool UnicodeSet::containsSome(const UnicodeSet& s) const { 1638 return !containsNone(s); 1639 } 1640 1641 inline UBool UnicodeSet::containsSome(const UnicodeString& s) const { 1642 return !containsNone(s); 1643 } 1644 1645 inline UBool UnicodeSet::isBogus() const { 1646 return (UBool)(fFlags & kIsBogus); 1647 } 1648 1649 inline UnicodeSet *UnicodeSet::fromUSet(USet *uset) { 1650 return reinterpret_cast<UnicodeSet *>(uset); 1651 } 1652 1653 inline const UnicodeSet *UnicodeSet::fromUSet(const USet *uset) { 1654 return reinterpret_cast<const UnicodeSet *>(uset); 1655 } 1656 1657 inline USet *UnicodeSet::toUSet() { 1658 return reinterpret_cast<USet *>(this); 1659 } 1660 1661 inline const USet *UnicodeSet::toUSet() const { 1662 return reinterpret_cast<const USet *>(this); 1663 } 1664 1665 inline int32_t UnicodeSet::span(const UnicodeString &s, int32_t start, USetSpanCondition spanCondition) const { 1666 int32_t sLength=s.length(); 1667 if(start<0) { 1668 start=0; 1669 } else if(start>sLength) { 1670 start=sLength; 1671 } 1672 return start+span(s.getBuffer()+start, sLength-start, spanCondition); 1673 } 1674 1675 inline int32_t UnicodeSet::spanBack(const UnicodeString &s, int32_t limit, USetSpanCondition spanCondition) const { 1676 int32_t sLength=s.length(); 1677 if(limit<0) { 1678 limit=0; 1679 } else if(limit>sLength) { 1680 limit=sLength; 1681 } 1682 return spanBack(s.getBuffer(), limit, spanCondition); 1683 } 1684 1685 U_NAMESPACE_END 1686 1687 #endif 1688
