1 /* 2 *************************************************************************** 3 * Copyright (C) 1999-2013, 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 // Forward Declarations. 26 void U_CALLCONV UnicodeSet_initInclusion(int32_t src, UErrorCode &status); /**< @internal */ 27 28 class BMPSet; 29 class ParsePosition; 30 class RBBIRuleScanner; 31 class SymbolTable; 32 class UnicodeSetStringSpan; 33 class UVector; 34 class RuleCharacterIterator; 35 36 /** 37 * A mutable set of Unicode characters and multicharacter strings. Objects of this class 38 * represent <em>character classes</em> used in regular expressions. 39 * A character specifies a subset of Unicode code points. Legal 40 * code points are U+0000 to U+10FFFF, inclusive. 41 * 42 * <p>The UnicodeSet class is not designed to be subclassed. 43 * 44 * <p><code>UnicodeSet</code> supports two APIs. The first is the 45 * <em>operand</em> API that allows the caller to modify the value of 46 * a <code>UnicodeSet</code> object. It conforms to Java 2's 47 * <code>java.util.Set</code> interface, although 48 * <code>UnicodeSet</code> does not actually implement that 49 * interface. All methods of <code>Set</code> are supported, with the 50 * modification that they take a character range or single character 51 * instead of an <code>Object</code>, and they take a 52 * <code>UnicodeSet</code> instead of a <code>Collection</code>. The 53 * operand API may be thought of in terms of boolean logic: a boolean 54 * OR is implemented by <code>add</code>, a boolean AND is implemented 55 * by <code>retain</code>, a boolean XOR is implemented by 56 * <code>complement</code> taking an argument, and a boolean NOT is 57 * implemented by <code>complement</code> with no argument. In terms 58 * of traditional set theory function names, <code>add</code> is a 59 * union, <code>retain</code> is an intersection, <code>remove</code> 60 * is an asymmetric difference, and <code>complement</code> with no 61 * argument is a set complement with respect to the superset range 62 * <code>MIN_VALUE-MAX_VALUE</code> 63 * 64 * <p>The second API is the 65 * <code>applyPattern()</code>/<code>toPattern()</code> API from the 66 * <code>java.text.Format</code>-derived classes. Unlike the 67 * methods that add characters, add categories, and control the logic 68 * of the set, the method <code>applyPattern()</code> sets all 69 * attributes of a <code>UnicodeSet</code> at once, based on a 70 * string pattern. 71 * 72 * <p><b>Pattern syntax</b></p> 73 * 74 * Patterns are accepted by the constructors and the 75 * <code>applyPattern()</code> methods and returned by the 76 * <code>toPattern()</code> method. These patterns follow a syntax 77 * similar to that employed by version 8 regular expression character 78 * classes. Here are some simple examples: 79 * 80 * \htmlonly<blockquote>\endhtmlonly 81 * <table> 82 * <tr align="top"> 83 * <td nowrap valign="top" align="left"><code>[]</code></td> 84 * <td valign="top">No characters</td> 85 * </tr><tr align="top"> 86 * <td nowrap valign="top" align="left"><code>[a]</code></td> 87 * <td valign="top">The character 'a'</td> 88 * </tr><tr align="top"> 89 * <td nowrap valign="top" align="left"><code>[ae]</code></td> 90 * <td valign="top">The characters 'a' and 'e'</td> 91 * </tr> 92 * <tr> 93 * <td nowrap valign="top" align="left"><code>[a-e]</code></td> 94 * <td valign="top">The characters 'a' through 'e' inclusive, in Unicode code 95 * point order</td> 96 * </tr> 97 * <tr> 98 * <td nowrap valign="top" align="left"><code>[\\u4E01]</code></td> 99 * <td valign="top">The character U+4E01</td> 100 * </tr> 101 * <tr> 102 * <td nowrap valign="top" align="left"><code>[a{ab}{ac}]</code></td> 103 * <td valign="top">The character 'a' and the multicharacter strings "ab" and 104 * "ac"</td> 105 * </tr> 106 * <tr> 107 * <td nowrap valign="top" align="left"><code>[\\p{Lu}]</code></td> 108 * <td valign="top">All characters in the general category Uppercase Letter</td> 109 * </tr> 110 * </table> 111 * \htmlonly</blockquote>\endhtmlonly 112 * 113 * Any character may be preceded by a backslash in order to remove any special 114 * meaning. White space characters, as defined by UCharacter.isWhitespace(), are 115 * ignored, unless they are escaped. 116 * 117 * <p>Property patterns specify a set of characters having a certain 118 * property as defined by the Unicode standard. Both the POSIX-like 119 * "[:Lu:]" and the Perl-like syntax "\\p{Lu}" are recognized. For a 120 * complete list of supported property patterns, see the User's Guide 121 * for UnicodeSet at 122 * <a href="http://icu-project.org/userguide/unicodeSet.html"> 123 * http://icu-project.org/userguide/unicodeSet.html</a>. 124 * Actual determination of property data is defined by the underlying 125 * Unicode database as implemented by UCharacter. 126 * 127 * <p>Patterns specify individual characters, ranges of characters, and 128 * Unicode property sets. When elements are concatenated, they 129 * specify their union. To complement a set, place a '^' immediately 130 * after the opening '['. Property patterns are inverted by modifying 131 * their delimiters; "[:^foo]" and "\\P{foo}". In any other location, 132 * '^' has no special meaning. 133 * 134 * <p>Ranges are indicated by placing two a '-' between two 135 * characters, as in "a-z". This specifies the range of all 136 * characters from the left to the right, in Unicode order. If the 137 * left character is greater than or equal to the 138 * right character it is a syntax error. If a '-' occurs as the first 139 * character after the opening '[' or '[^', or if it occurs as the 140 * last character before the closing ']', then it is taken as a 141 * literal. Thus "[a\-b]", "[-ab]", and "[ab-]" all indicate the same 142 * set of three characters, 'a', 'b', and '-'. 143 * 144 * <p>Sets may be intersected using the '&' operator or the asymmetric 145 * set difference may be taken using the '-' operator, for example, 146 * "[[:L:]&[\\u0000-\\u0FFF]]" indicates the set of all Unicode letters 147 * with values less than 4096. Operators ('&' and '|') have equal 148 * precedence and bind left-to-right. Thus 149 * "[[:L:]-[a-z]-[\\u0100-\\u01FF]]" is equivalent to 150 * "[[[:L:]-[a-z]]-[\\u0100-\\u01FF]]". This only really matters for 151 * difference; intersection is commutative. 152 * 153 * <table> 154 * <tr valign=top><td nowrap><code>[a]</code><td>The set containing 'a' 155 * <tr valign=top><td nowrap><code>[a-z]</code><td>The set containing 'a' 156 * through 'z' and all letters in between, in Unicode order 157 * <tr valign=top><td nowrap><code>[^a-z]</code><td>The set containing 158 * all characters but 'a' through 'z', 159 * that is, U+0000 through 'a'-1 and 'z'+1 through U+10FFFF 160 * <tr valign=top><td nowrap><code>[[<em>pat1</em>][<em>pat2</em>]]</code> 161 * <td>The union of sets specified by <em>pat1</em> and <em>pat2</em> 162 * <tr valign=top><td nowrap><code>[[<em>pat1</em>]&[<em>pat2</em>]]</code> 163 * <td>The intersection of sets specified by <em>pat1</em> and <em>pat2</em> 164 * <tr valign=top><td nowrap><code>[[<em>pat1</em>]-[<em>pat2</em>]]</code> 165 * <td>The asymmetric difference of sets specified by <em>pat1</em> and 166 * <em>pat2</em> 167 * <tr valign=top><td nowrap><code>[:Lu:] or \\p{Lu}</code> 168 * <td>The set of characters having the specified 169 * Unicode property; in 170 * this case, Unicode uppercase letters 171 * <tr valign=top><td nowrap><code>[:^Lu:] or \\P{Lu}</code> 172 * <td>The set of characters <em>not</em> having the given 173 * Unicode property 174 * </table> 175 * 176 * <p><b>Warning</b>: you cannot add an empty string ("") to a UnicodeSet.</p> 177 * 178 * <p><b>Formal syntax</b></p> 179 * 180 * \htmlonly<blockquote>\endhtmlonly 181 * <table> 182 * <tr align="top"> 183 * <td nowrap valign="top" align="right"><code>pattern := </code></td> 184 * <td valign="top"><code>('[' '^'? item* ']') | 185 * property</code></td> 186 * </tr> 187 * <tr align="top"> 188 * <td nowrap valign="top" align="right"><code>item := </code></td> 189 * <td valign="top"><code>char | (char '-' char) | pattern-expr<br> 190 * </code></td> 191 * </tr> 192 * <tr align="top"> 193 * <td nowrap valign="top" align="right"><code>pattern-expr := </code></td> 194 * <td valign="top"><code>pattern | pattern-expr pattern | 195 * pattern-expr op pattern<br> 196 * </code></td> 197 * </tr> 198 * <tr align="top"> 199 * <td nowrap valign="top" align="right"><code>op := </code></td> 200 * <td valign="top"><code>'&' | '-'<br> 201 * </code></td> 202 * </tr> 203 * <tr align="top"> 204 * <td nowrap valign="top" align="right"><code>special := </code></td> 205 * <td valign="top"><code>'[' | ']' | '-'<br> 206 * </code></td> 207 * </tr> 208 * <tr align="top"> 209 * <td nowrap valign="top" align="right"><code>char := </code></td> 210 * <td valign="top"><em>any character that is not</em><code> special<br> 211 * | ('\' </code><em>any character</em><code>)<br> 212 * | ('\\u' hex hex hex hex)<br> 213 * </code></td> 214 * </tr> 215 * <tr align="top"> 216 * <td nowrap valign="top" align="right"><code>hex := </code></td> 217 * <td valign="top"><em>any character for which 218 * </em><code>Character.digit(c, 16)</code><em> 219 * returns a non-negative result</em></td> 220 * </tr> 221 * <tr> 222 * <td nowrap valign="top" align="right"><code>property := </code></td> 223 * <td valign="top"><em>a Unicode property set pattern</em></td> 224 * </tr> 225 * </table> 226 * <br> 227 * <table border="1"> 228 * <tr> 229 * <td>Legend: <table> 230 * <tr> 231 * <td nowrap valign="top"><code>a := b</code></td> 232 * <td width="20" valign="top"> </td> 233 * <td valign="top"><code>a</code> may be replaced by <code>b</code> </td> 234 * </tr> 235 * <tr> 236 * <td nowrap valign="top"><code>a?</code></td> 237 * <td valign="top"></td> 238 * <td valign="top">zero or one instance of <code>a</code><br> 239 * </td> 240 * </tr> 241 * <tr> 242 * <td nowrap valign="top"><code>a*</code></td> 243 * <td valign="top"></td> 244 * <td valign="top">one or more instances of <code>a</code><br> 245 * </td> 246 * </tr> 247 * <tr> 248 * <td nowrap valign="top"><code>a | b</code></td> 249 * <td valign="top"></td> 250 * <td valign="top">either <code>a</code> or <code>b</code><br> 251 * </td> 252 * </tr> 253 * <tr> 254 * <td nowrap valign="top"><code>'a'</code></td> 255 * <td valign="top"></td> 256 * <td valign="top">the literal string between the quotes </td> 257 * </tr> 258 * </table> 259 * </td> 260 * </tr> 261 * </table> 262 * \htmlonly</blockquote>\endhtmlonly 263 * 264 * <p>Note: 265 * - Most UnicodeSet methods do not take a UErrorCode parameter because 266 * there are usually very few opportunities for failure other than a shortage 267 * of memory, error codes in low-level C++ string methods would be inconvenient, 268 * and the error code as the last parameter (ICU convention) would prevent 269 * the use of default parameter values. 270 * Instead, such methods set the UnicodeSet into a "bogus" state 271 * (see isBogus()) if an error occurs. 272 * 273 * @author Alan Liu 274 * @stable ICU 2.0 275 */ 276 class U_COMMON_API UnicodeSet : public UnicodeFilter { 277 278 int32_t len; // length of list used; 0 <= len <= capacity 279 int32_t capacity; // capacity of list 280 UChar32* list; // MUST be terminated with HIGH 281 BMPSet *bmpSet; // The set is frozen iff either bmpSet or stringSpan is not NULL. 282 UChar32* buffer; // internal buffer, may be NULL 283 int32_t bufferCapacity; // capacity of buffer 284 int32_t patLen; 285 286 /** 287 * The pattern representation of this set. This may not be the 288 * most economical pattern. It is the pattern supplied to 289 * applyPattern(), with variables substituted and whitespace 290 * removed. For sets constructed without applyPattern(), or 291 * modified using the non-pattern API, this string will be empty, 292 * indicating that toPattern() must generate a pattern 293 * representation from the inversion list. 294 */ 295 UChar *pat; 296 UVector* strings; // maintained in sorted order 297 UnicodeSetStringSpan *stringSpan; 298 299 private: 300 enum { // constants 301 kIsBogus = 1 // This set is bogus (i.e. not valid) 302 }; 303 uint8_t fFlags; // Bit flag (see constants above) 304 public: 305 /** 306 * Determine if this object contains a valid set. 307 * A bogus set has no value. It is different from an empty set. 308 * It can be used to indicate that no set value is available. 309 * 310 * @return TRUE if the set is valid, FALSE otherwise 311 * @see setToBogus() 312 * @stable ICU 4.0 313 */ 314 inline UBool isBogus(void) const; 315 316 /** 317 * Make this UnicodeSet object invalid. 318 * The string will test TRUE with isBogus(). 319 * 320 * A bogus set has no value. It is different from an empty set. 321 * It can be used to indicate that no set value is available. 322 * 323 * This utility function is used throughout the UnicodeSet 324 * implementation to indicate that a UnicodeSet operation failed, 325 * and may be used in other functions, 326 * especially but not exclusively when such functions do not 327 * take a UErrorCode for simplicity. 328 * 329 * @see isBogus() 330 * @stable ICU 4.0 331 */ 332 void setToBogus(); 333 334 public: 335 336 enum { 337 /** 338 * Minimum value that can be stored in a UnicodeSet. 339 * @stable ICU 2.4 340 */ 341 MIN_VALUE = 0, 342 343 /** 344 * Maximum value that can be stored in a UnicodeSet. 345 * @stable ICU 2.4 346 */ 347 MAX_VALUE = 0x10ffff 348 }; 349 350 //---------------------------------------------------------------- 351 // Constructors &c 352 //---------------------------------------------------------------- 353 354 public: 355 356 /** 357 * Constructs an empty set. 358 * @stable ICU 2.0 359 */ 360 UnicodeSet(); 361 362 /** 363 * Constructs a set containing the given range. If <code>end > 364 * start</code> then an empty set is created. 365 * 366 * @param start first character, inclusive, of range 367 * @param end last character, inclusive, of range 368 * @stable ICU 2.4 369 */ 370 UnicodeSet(UChar32 start, UChar32 end); 371 372 /** 373 * Constructs a set from the given pattern. See the class 374 * description for the syntax of the pattern language. 375 * @param pattern a string specifying what characters are in the set 376 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 377 * contains a syntax error. 378 * @stable ICU 2.0 379 */ 380 UnicodeSet(const UnicodeString& pattern, 381 UErrorCode& status); 382 383 #ifndef U_HIDE_INTERNAL_API 384 /** 385 * Constructs a set from the given pattern. See the class 386 * description for the syntax of the pattern language. 387 * @param pattern a string specifying what characters are in the set 388 * @param options bitmask for options to apply to the pattern. 389 * Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE. 390 * @param symbols a symbol table mapping variable names to values 391 * and stand-in characters to UnicodeSets; may be NULL 392 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 393 * contains a syntax error. 394 * @internal 395 */ 396 UnicodeSet(const UnicodeString& pattern, 397 uint32_t options, 398 const SymbolTable* symbols, 399 UErrorCode& status); 400 #endif /* U_HIDE_INTERNAL_API */ 401 402 /** 403 * Constructs a set from the given pattern. See the class description 404 * for the syntax of the pattern language. 405 * @param pattern a string specifying what characters are in the set 406 * @param pos on input, the position in pattern at which to start parsing. 407 * On output, the position after the last character parsed. 408 * @param options bitmask for options to apply to the pattern. 409 * Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE. 410 * @param symbols a symbol table mapping variable names to values 411 * and stand-in characters to UnicodeSets; may be NULL 412 * @param status input-output error code 413 * @stable ICU 2.8 414 */ 415 UnicodeSet(const UnicodeString& pattern, ParsePosition& pos, 416 uint32_t options, 417 const SymbolTable* symbols, 418 UErrorCode& status); 419 420 /** 421 * Constructs a set that is identical to the given UnicodeSet. 422 * @stable ICU 2.0 423 */ 424 UnicodeSet(const UnicodeSet& o); 425 426 /** 427 * Destructs the set. 428 * @stable ICU 2.0 429 */ 430 virtual ~UnicodeSet(); 431 432 /** 433 * Assigns this object to be a copy of another. 434 * A frozen set will not be modified. 435 * @stable ICU 2.0 436 */ 437 UnicodeSet& operator=(const UnicodeSet& o); 438 439 /** 440 * Compares the specified object with this set for equality. Returns 441 * <tt>true</tt> if the two sets 442 * have the same size, and every member of the specified set is 443 * contained in this set (or equivalently, every member of this set is 444 * contained in the specified set). 445 * 446 * @param o set to be compared for equality with this set. 447 * @return <tt>true</tt> if the specified set is equal to this set. 448 * @stable ICU 2.0 449 */ 450 virtual UBool operator==(const UnicodeSet& o) const; 451 452 /** 453 * Compares the specified object with this set for equality. Returns 454 * <tt>true</tt> if the specified set is not equal to this set. 455 * @stable ICU 2.0 456 */ 457 UBool operator!=(const UnicodeSet& o) const; 458 459 /** 460 * Returns a copy of this object. All UnicodeFunctor objects have 461 * to support cloning in order to allow classes using 462 * UnicodeFunctors, such as Transliterator, to implement cloning. 463 * If this set is frozen, then the clone will be frozen as well. 464 * Use cloneAsThawed() for a mutable clone of a frozen set. 465 * @see cloneAsThawed 466 * @stable ICU 2.0 467 */ 468 virtual UnicodeFunctor* clone() const; 469 470 /** 471 * Returns the hash code value for this set. 472 * 473 * @return the hash code value for this set. 474 * @see Object#hashCode() 475 * @stable ICU 2.0 476 */ 477 virtual int32_t hashCode(void) const; 478 479 /** 480 * Get a UnicodeSet pointer from a USet 481 * 482 * @param uset a USet (the ICU plain C type for UnicodeSet) 483 * @return the corresponding UnicodeSet pointer. 484 * 485 * @stable ICU 4.2 486 */ 487 inline static UnicodeSet *fromUSet(USet *uset); 488 489 /** 490 * Get a UnicodeSet pointer from a const USet 491 * 492 * @param uset a const USet (the ICU plain C type for UnicodeSet) 493 * @return the corresponding UnicodeSet pointer. 494 * 495 * @stable ICU 4.2 496 */ 497 inline static const UnicodeSet *fromUSet(const USet *uset); 498 499 /** 500 * Produce a USet * pointer for this UnicodeSet. 501 * USet is the plain C type for UnicodeSet 502 * 503 * @return a USet pointer for this UnicodeSet 504 * @stable ICU 4.2 505 */ 506 inline USet *toUSet(); 507 508 509 /** 510 * Produce a const USet * pointer for this UnicodeSet. 511 * USet is the plain C type for UnicodeSet 512 * 513 * @return a const USet pointer for this UnicodeSet 514 * @stable ICU 4.2 515 */ 516 inline const USet * toUSet() const; 517 518 519 //---------------------------------------------------------------- 520 // Freezable API 521 //---------------------------------------------------------------- 522 523 /** 524 * Determines whether the set has been frozen (made immutable) or not. 525 * See the ICU4J Freezable interface for details. 526 * @return TRUE/FALSE for whether the set has been frozen 527 * @see freeze 528 * @see cloneAsThawed 529 * @stable ICU 3.8 530 */ 531 inline UBool isFrozen() const; 532 533 /** 534 * Freeze the set (make it immutable). 535 * Once frozen, it cannot be unfrozen and is therefore thread-safe 536 * until it is deleted. 537 * See the ICU4J Freezable interface for details. 538 * Freezing the set may also make some operations faster, for example 539 * contains() and span(). 540 * A frozen set will not be modified. (It remains frozen.) 541 * @return this set. 542 * @see isFrozen 543 * @see cloneAsThawed 544 * @stable ICU 3.8 545 */ 546 UnicodeFunctor *freeze(); 547 548 /** 549 * Clone the set and make the clone mutable. 550 * See the ICU4J Freezable interface for details. 551 * @return the mutable clone 552 * @see freeze 553 * @see isFrozen 554 * @stable ICU 3.8 555 */ 556 UnicodeFunctor *cloneAsThawed() const; 557 558 //---------------------------------------------------------------- 559 // Public API 560 //---------------------------------------------------------------- 561 562 /** 563 * Make this object represent the range <code>start - end</code>. 564 * If <code>end > start</code> then this object is set to an 565 * an empty range. 566 * A frozen set will not be modified. 567 * 568 * @param start first character in the set, inclusive 569 * @param end last character in the set, inclusive 570 * @stable ICU 2.4 571 */ 572 UnicodeSet& set(UChar32 start, UChar32 end); 573 574 /** 575 * Return true if the given position, in the given pattern, appears 576 * to be the start of a UnicodeSet pattern. 577 * @stable ICU 2.4 578 */ 579 static UBool resemblesPattern(const UnicodeString& pattern, 580 int32_t pos); 581 582 /** 583 * Modifies this set to represent the set specified by the given 584 * pattern, ignoring Unicode Pattern_White_Space characters. 585 * See the class description for the syntax of the pattern language. 586 * A frozen set will not be modified. 587 * @param pattern a string specifying what characters are in the set 588 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 589 * contains a syntax error. 590 * <em> Empties the set passed before applying the pattern.</em> 591 * @return a reference to this 592 * @stable ICU 2.0 593 */ 594 UnicodeSet& applyPattern(const UnicodeString& pattern, 595 UErrorCode& status); 596 597 #ifndef U_HIDE_INTERNAL_API 598 /** 599 * Modifies this set to represent the set specified by the given 600 * pattern, optionally ignoring Unicode Pattern_White_Space characters. 601 * See the class description for the syntax of the pattern language. 602 * A frozen set will not be modified. 603 * @param pattern a string specifying what characters are in the set 604 * @param options bitmask for options to apply to the pattern. 605 * Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE. 606 * @param symbols a symbol table mapping variable names to 607 * values and stand-ins to UnicodeSets; may be NULL 608 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 609 * contains a syntax error. 610 *<em> Empties the set passed before applying the pattern.</em> 611 * @return a reference to this 612 * @internal 613 */ 614 UnicodeSet& applyPattern(const UnicodeString& pattern, 615 uint32_t options, 616 const SymbolTable* symbols, 617 UErrorCode& status); 618 #endif /* U_HIDE_INTERNAL_API */ 619 620 /** 621 * Parses the given pattern, starting at the given position. The 622 * character at pattern.charAt(pos.getIndex()) must be '[', or the 623 * parse fails. Parsing continues until the corresponding closing 624 * ']'. If a syntax error is encountered between the opening and 625 * closing brace, the parse fails. Upon return from a successful 626 * parse, the ParsePosition is updated to point to the character 627 * following the closing ']', and a StringBuffer containing a 628 * pairs list for the parsed pattern is returned. This method calls 629 * itself recursively to parse embedded subpatterns. 630 *<em> Empties the set passed before applying the pattern.</em> 631 * A frozen set will not be modified. 632 * 633 * @param pattern the string containing the pattern to be parsed. 634 * The portion of the string from pos.getIndex(), which must be a 635 * '[', to the corresponding closing ']', is parsed. 636 * @param pos upon entry, the position at which to being parsing. 637 * The character at pattern.charAt(pos.getIndex()) must be a '['. 638 * Upon return from a successful parse, pos.getIndex() is either 639 * the character after the closing ']' of the parsed pattern, or 640 * pattern.length() if the closing ']' is the last character of 641 * the pattern string. 642 * @param options bitmask for options to apply to the pattern. 643 * Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE. 644 * @param symbols a symbol table mapping variable names to 645 * values and stand-ins to UnicodeSets; may be NULL 646 * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern 647 * contains a syntax error. 648 * @return a reference to this 649 * @stable ICU 2.8 650 */ 651 UnicodeSet& applyPattern(const UnicodeString& pattern, 652 ParsePosition& pos, 653 uint32_t options, 654 const SymbolTable* symbols, 655 UErrorCode& status); 656 657 /** 658 * Returns a string representation of this set. If the result of 659 * calling this function is passed to a UnicodeSet constructor, it 660 * will produce another set that is equal to this one. 661 * A frozen set will not be modified. 662 * @param result the string to receive the rules. Previous 663 * contents will be deleted. 664 * @param escapeUnprintable if TRUE then convert unprintable 665 * character to their hex escape representations, \\uxxxx or 666 * \\Uxxxxxxxx. Unprintable characters are those other than 667 * U+000A, U+0020..U+007E. 668 * @stable ICU 2.0 669 */ 670 virtual UnicodeString& toPattern(UnicodeString& result, 671 UBool escapeUnprintable = FALSE) const; 672 673 /** 674 * Modifies this set to contain those code points which have the given value 675 * for the given binary or enumerated property, as returned by 676 * u_getIntPropertyValue. Prior contents of this set are lost. 677 * A frozen set will not be modified. 678 * 679 * @param prop a property in the range UCHAR_BIN_START..UCHAR_BIN_LIMIT-1 680 * or UCHAR_INT_START..UCHAR_INT_LIMIT-1 681 * or UCHAR_MASK_START..UCHAR_MASK_LIMIT-1. 682 * 683 * @param value a value in the range u_getIntPropertyMinValue(prop).. 684 * u_getIntPropertyMaxValue(prop), with one exception. If prop is 685 * UCHAR_GENERAL_CATEGORY_MASK, then value should not be a UCharCategory, but 686 * rather a mask value produced by U_GET_GC_MASK(). This allows grouped 687 * categories such as [:L:] to be represented. 688 * 689 * @param ec error code input/output parameter 690 * 691 * @return a reference to this set 692 * 693 * @stable ICU 2.4 694 */ 695 UnicodeSet& applyIntPropertyValue(UProperty prop, 696 int32_t value, 697 UErrorCode& ec); 698 699 /** 700 * Modifies this set to contain those code points which have the 701 * given value for the given property. Prior contents of this 702 * set are lost. 703 * A frozen set will not be modified. 704 * 705 * @param prop a property alias, either short or long. The name is matched 706 * loosely. See PropertyAliases.txt for names and a description of loose 707 * matching. If the value string is empty, then this string is interpreted 708 * as either a General_Category value alias, a Script value alias, a binary 709 * property alias, or a special ID. Special IDs are matched loosely and 710 * correspond to the following sets: 711 * 712 * "ANY" = [\\u0000-\\U0010FFFF], 713 * "ASCII" = [\\u0000-\\u007F], 714 * "Assigned" = [:^Cn:]. 715 * 716 * @param value a value alias, either short or long. The name is matched 717 * loosely. See PropertyValueAliases.txt for names and a description of 718 * loose matching. In addition to aliases listed, numeric values and 719 * canonical combining classes may be expressed numerically, e.g., ("nv", 720 * "0.5") or ("ccc", "220"). The value string may also be empty. 721 * 722 * @param ec error code input/output parameter 723 * 724 * @return a reference to this set 725 * 726 * @stable ICU 2.4 727 */ 728 UnicodeSet& applyPropertyAlias(const UnicodeString& prop, 729 const UnicodeString& value, 730 UErrorCode& ec); 731 732 /** 733 * Returns the number of elements in this set (its cardinality). 734 * Note than the elements of a set may include both individual 735 * codepoints and strings. 736 * 737 * @return the number of elements in this set (its cardinality). 738 * @stable ICU 2.0 739 */ 740 virtual int32_t size(void) const; 741 742 /** 743 * Returns <tt>true</tt> if this set contains no elements. 744 * 745 * @return <tt>true</tt> if this set contains no elements. 746 * @stable ICU 2.0 747 */ 748 virtual UBool isEmpty(void) const; 749 750 /** 751 * Returns true if this set contains the given character. 752 * This function works faster with a frozen set. 753 * @param c character to be checked for containment 754 * @return true if the test condition is met 755 * @stable ICU 2.0 756 */ 757 virtual UBool contains(UChar32 c) const; 758 759 /** 760 * Returns true if this set contains every character 761 * of the given range. 762 * @param start first character, inclusive, of the range 763 * @param end last character, inclusive, of the range 764 * @return true if the test condition is met 765 * @stable ICU 2.0 766 */ 767 virtual UBool contains(UChar32 start, UChar32 end) const; 768 769 /** 770 * Returns <tt>true</tt> if this set contains the given 771 * multicharacter string. 772 * @param s string to be checked for containment 773 * @return <tt>true</tt> if this set contains the specified string 774 * @stable ICU 2.4 775 */ 776 UBool contains(const UnicodeString& s) const; 777 778 /** 779 * Returns true if this set contains all the characters and strings 780 * of the given set. 781 * @param c set to be checked for containment 782 * @return true if the test condition is met 783 * @stable ICU 2.4 784 */ 785 virtual UBool containsAll(const UnicodeSet& c) const; 786 787 /** 788 * Returns true if this set contains all the characters 789 * of the given string. 790 * @param s string containing characters to be checked for containment 791 * @return true if the test condition is met 792 * @stable ICU 2.4 793 */ 794 UBool containsAll(const UnicodeString& s) const; 795 796 /** 797 * Returns true if this set contains none of the characters 798 * of the given range. 799 * @param start first character, inclusive, of the range 800 * @param end last character, inclusive, of the range 801 * @return true if the test condition is met 802 * @stable ICU 2.4 803 */ 804 UBool containsNone(UChar32 start, UChar32 end) const; 805 806 /** 807 * Returns true if this set contains none of the characters and strings 808 * of the given set. 809 * @param c set to be checked for containment 810 * @return true if the test condition is met 811 * @stable ICU 2.4 812 */ 813 UBool containsNone(const UnicodeSet& c) const; 814 815 /** 816 * Returns true if this set contains none of the characters 817 * of the given string. 818 * @param s string containing characters to be checked for containment 819 * @return true if the test condition is met 820 * @stable ICU 2.4 821 */ 822 UBool containsNone(const UnicodeString& s) const; 823 824 /** 825 * Returns true if this set contains one or more of the characters 826 * in the given range. 827 * @param start first character, inclusive, of the range 828 * @param end last character, inclusive, of the range 829 * @return true if the condition is met 830 * @stable ICU 2.4 831 */ 832 inline UBool containsSome(UChar32 start, UChar32 end) const; 833 834 /** 835 * Returns true if this set contains one or more of the characters 836 * and strings of the given set. 837 * @param s The set to be checked for containment 838 * @return true if the condition is met 839 * @stable ICU 2.4 840 */ 841 inline UBool containsSome(const UnicodeSet& s) const; 842 843 /** 844 * Returns true if this set contains one or more of the characters 845 * of the given string. 846 * @param s string containing characters to be checked for containment 847 * @return true if the condition is met 848 * @stable ICU 2.4 849 */ 850 inline UBool containsSome(const UnicodeString& s) const; 851 852 /** 853 * Returns the length of the initial substring of the input string which 854 * consists only of characters and strings that are contained in this set 855 * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE), 856 * or only of characters and strings that are not contained 857 * in this set (USET_SPAN_NOT_CONTAINED). 858 * See USetSpanCondition for details. 859 * Similar to the strspn() C library function. 860 * Unpaired surrogates are treated according to contains() of their surrogate code points. 861 * This function works faster with a frozen set and with a non-negative string length argument. 862 * @param s start of the string 863 * @param length of the string; can be -1 for NUL-terminated 864 * @param spanCondition specifies the containment condition 865 * @return the length of the initial substring according to the spanCondition; 866 * 0 if the start of the string does not fit the spanCondition 867 * @stable ICU 3.8 868 * @see USetSpanCondition 869 */ 870 int32_t span(const UChar *s, int32_t length, USetSpanCondition spanCondition) const; 871 872 /** 873 * Returns the end of the substring of the input string according to the USetSpanCondition. 874 * Same as <code>start+span(s.getBuffer()+start, s.length()-start, spanCondition)</code> 875 * after pinning start to 0<=start<=s.length(). 876 * @param s the string 877 * @param start the start index in the string for the span operation 878 * @param spanCondition specifies the containment condition 879 * @return the exclusive end of the substring according to the spanCondition; 880 * the substring s.tempSubStringBetween(start, end) fulfills the spanCondition 881 * @stable ICU 4.4 882 * @see USetSpanCondition 883 */ 884 inline int32_t span(const UnicodeString &s, int32_t start, USetSpanCondition spanCondition) const; 885 886 /** 887 * Returns the start of the trailing substring of the input string which 888 * consists only of characters and strings that are contained in this set 889 * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE), 890 * or only of characters and strings that are not contained 891 * in this set (USET_SPAN_NOT_CONTAINED). 892 * See USetSpanCondition for details. 893 * Unpaired surrogates are treated according to contains() of their surrogate code points. 894 * This function works faster with a frozen set and with a non-negative string length argument. 895 * @param s start of the string 896 * @param length of the string; can be -1 for NUL-terminated 897 * @param spanCondition specifies the containment condition 898 * @return the start of the trailing substring according to the spanCondition; 899 * the string length if the end of the string does not fit the spanCondition 900 * @stable ICU 3.8 901 * @see USetSpanCondition 902 */ 903 int32_t spanBack(const UChar *s, int32_t length, USetSpanCondition spanCondition) const; 904 905 /** 906 * Returns the start of the substring of the input string according to the USetSpanCondition. 907 * Same as <code>spanBack(s.getBuffer(), limit, spanCondition)</code> 908 * after pinning limit to 0<=end<=s.length(). 909 * @param s the string 910 * @param limit the exclusive-end index in the string for the span operation 911 * (use s.length() or INT32_MAX for spanning back from the end of the string) 912 * @param spanCondition specifies the containment condition 913 * @return the start of the substring according to the spanCondition; 914 * the substring s.tempSubStringBetween(start, limit) fulfills the spanCondition 915 * @stable ICU 4.4 916 * @see USetSpanCondition 917 */ 918 inline int32_t spanBack(const UnicodeString &s, int32_t limit, USetSpanCondition spanCondition) const; 919 920 /** 921 * Returns the length of the initial substring of the input string which 922 * consists only of characters and strings that are contained in this set 923 * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE), 924 * or only of characters and strings that are not contained 925 * in this set (USET_SPAN_NOT_CONTAINED). 926 * See USetSpanCondition for details. 927 * Similar to the strspn() C library function. 928 * Malformed byte sequences are treated according to contains(0xfffd). 929 * This function works faster with a frozen set and with a non-negative string length argument. 930 * @param s start of the string (UTF-8) 931 * @param length of the string; can be -1 for NUL-terminated 932 * @param spanCondition specifies the containment condition 933 * @return the length of the initial substring according to the spanCondition; 934 * 0 if the start of the string does not fit the spanCondition 935 * @stable ICU 3.8 936 * @see USetSpanCondition 937 */ 938 int32_t spanUTF8(const char *s, int32_t length, USetSpanCondition spanCondition) const; 939 940 /** 941 * Returns the start of the trailing substring of the input string which 942 * consists only of characters and strings that are contained in this set 943 * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE), 944 * or only of characters and strings that are not contained 945 * in this set (USET_SPAN_NOT_CONTAINED). 946 * See USetSpanCondition for details. 947 * Malformed byte sequences are treated according to contains(0xfffd). 948 * This function works faster with a frozen set and with a non-negative string length argument. 949 * @param s start of the string (UTF-8) 950 * @param length of the string; can be -1 for NUL-terminated 951 * @param spanCondition specifies the containment condition 952 * @return the start of the trailing substring according to the spanCondition; 953 * the string length if the end of the string does not fit the spanCondition 954 * @stable ICU 3.8 955 * @see USetSpanCondition 956 */ 957 int32_t spanBackUTF8(const char *s, int32_t length, USetSpanCondition spanCondition) const; 958 959 /** 960 * Implement UnicodeMatcher::matches() 961 * @stable ICU 2.4 962 */ 963 virtual UMatchDegree matches(const Replaceable& text, 964 int32_t& offset, 965 int32_t limit, 966 UBool incremental); 967 968 private: 969 /** 970 * Returns the longest match for s in text at the given position. 971 * If limit > start then match forward from start+1 to limit 972 * matching all characters except s.charAt(0). If limit < start, 973 * go backward starting from start-1 matching all characters 974 * except s.charAt(s.length()-1). This method assumes that the 975 * first character, text.charAt(start), matches s, so it does not 976 * check it. 977 * @param text the text to match 978 * @param start the first character to match. In the forward 979 * direction, text.charAt(start) is matched against s.charAt(0). 980 * In the reverse direction, it is matched against 981 * s.charAt(s.length()-1). 982 * @param limit the limit offset for matching, either last+1 in 983 * the forward direction, or last-1 in the reverse direction, 984 * where last is the index of the last character to match. 985 * @param s 986 * @return If part of s matches up to the limit, return |limit - 987 * start|. If all of s matches before reaching the limit, return 988 * s.length(). If there is a mismatch between s and text, return 989 * 0 990 */ 991 static int32_t matchRest(const Replaceable& text, 992 int32_t start, int32_t limit, 993 const UnicodeString& s); 994 995 /** 996 * Returns the smallest value i such that c < list[i]. Caller 997 * must ensure that c is a legal value or this method will enter 998 * an infinite loop. This method performs a binary search. 999 * @param c a character in the range MIN_VALUE..MAX_VALUE 1000 * inclusive 1001 * @return the smallest integer i in the range 0..len-1, 1002 * inclusive, such that c < list[i] 1003 */ 1004 int32_t findCodePoint(UChar32 c) const; 1005 1006 public: 1007 1008 /** 1009 * Implementation of UnicodeMatcher API. Union the set of all 1010 * characters that may be matched by this object into the given 1011 * set. 1012 * @param toUnionTo the set into which to union the source characters 1013 * @stable ICU 2.4 1014 */ 1015 virtual void addMatchSetTo(UnicodeSet& toUnionTo) const; 1016 1017 /** 1018 * Returns the index of the given character within this set, where 1019 * the set is ordered by ascending code point. If the character 1020 * is not in this set, return -1. The inverse of this method is 1021 * <code>charAt()</code>. 1022 * @return an index from 0..size()-1, or -1 1023 * @stable ICU 2.4 1024 */ 1025 int32_t indexOf(UChar32 c) const; 1026 1027 /** 1028 * Returns the character at the given index within this set, where 1029 * the set is ordered by ascending code point. If the index is 1030 * out of range, return (UChar32)-1. The inverse of this method is 1031 * <code>indexOf()</code>. 1032 * @param index an index from 0..size()-1 1033 * @return the character at the given index, or (UChar32)-1. 1034 * @stable ICU 2.4 1035 */ 1036 UChar32 charAt(int32_t index) const; 1037 1038 /** 1039 * Adds the specified range to this set if it is not already 1040 * present. If this set already contains the specified range, 1041 * the call leaves this set unchanged. If <code>end > start</code> 1042 * then an empty range is added, leaving the set unchanged. 1043 * This is equivalent to a boolean logic OR, or a set UNION. 1044 * A frozen set will not be modified. 1045 * 1046 * @param start first character, inclusive, of range to be added 1047 * to this set. 1048 * @param end last character, inclusive, of range to be added 1049 * to this set. 1050 * @stable ICU 2.0 1051 */ 1052 virtual UnicodeSet& add(UChar32 start, UChar32 end); 1053 1054 /** 1055 * Adds the specified character to this set if it is not already 1056 * present. If this set already contains the specified character, 1057 * the call leaves this set unchanged. 1058 * A frozen set will not be modified. 1059 * @stable ICU 2.0 1060 */ 1061 UnicodeSet& add(UChar32 c); 1062 1063 /** 1064 * Adds the specified multicharacter to this set if it is not already 1065 * present. If this set already contains the multicharacter, 1066 * the call leaves this set unchanged. 1067 * Thus "ch" => {"ch"} 1068 * <br><b>Warning: you cannot add an empty string ("") to a UnicodeSet.</b> 1069 * A frozen set will not be modified. 1070 * @param s the source string 1071 * @return this object, for chaining 1072 * @stable ICU 2.4 1073 */ 1074 UnicodeSet& add(const UnicodeString& s); 1075 1076 private: 1077 /** 1078 * @return a code point IF the string consists of a single one. 1079 * otherwise returns -1. 1080 * @param s string to test 1081 */ 1082 static int32_t getSingleCP(const UnicodeString& s); 1083 1084 void _add(const UnicodeString& s); 1085 1086 public: 1087 /** 1088 * Adds each of the characters in this string to the set. Thus "ch" => {"c", "h"} 1089 * If this set already any particular character, it has no effect on that character. 1090 * A frozen set will not be modified. 1091 * @param s the source string 1092 * @return this object, for chaining 1093 * @stable ICU 2.4 1094 */ 1095 UnicodeSet& addAll(const UnicodeString& s); 1096 1097 /** 1098 * Retains EACH of the characters in this string. Note: "ch" == {"c", "h"} 1099 * If this set already any particular character, it has no effect on that character. 1100 * A frozen set will not be modified. 1101 * @param s the source string 1102 * @return this object, for chaining 1103 * @stable ICU 2.4 1104 */ 1105 UnicodeSet& retainAll(const UnicodeString& s); 1106 1107 /** 1108 * Complement EACH of the characters in this string. Note: "ch" == {"c", "h"} 1109 * If this set already any particular character, it has no effect on that character. 1110 * A frozen set will not be modified. 1111 * @param s the source string 1112 * @return this object, for chaining 1113 * @stable ICU 2.4 1114 */ 1115 UnicodeSet& complementAll(const UnicodeString& s); 1116 1117 /** 1118 * Remove EACH of the characters in this string. Note: "ch" == {"c", "h"} 1119 * If this set already any particular character, it has no effect on that character. 1120 * A frozen set will not be modified. 1121 * @param s the source string 1122 * @return this object, for chaining 1123 * @stable ICU 2.4 1124 */ 1125 UnicodeSet& removeAll(const UnicodeString& s); 1126 1127 /** 1128 * Makes a set from a multicharacter string. Thus "ch" => {"ch"} 1129 * <br><b>Warning: you cannot add an empty string ("") to a UnicodeSet.</b> 1130 * @param s the source string 1131 * @return a newly created set containing the given string. 1132 * The caller owns the return object and is responsible for deleting it. 1133 * @stable ICU 2.4 1134 */ 1135 static UnicodeSet* U_EXPORT2 createFrom(const UnicodeString& s); 1136 1137 1138 /** 1139 * Makes a set from each of the characters in the string. Thus "ch" => {"c", "h"} 1140 * @param s the source string 1141 * @return a newly created set containing the given characters 1142 * The caller owns the return object and is responsible for deleting it. 1143 * @stable ICU 2.4 1144 */ 1145 static UnicodeSet* U_EXPORT2 createFromAll(const UnicodeString& s); 1146 1147 /** 1148 * Retain only the elements in this set that are contained in the 1149 * specified range. If <code>end > start</code> then an empty range is 1150 * retained, leaving the set empty. This is equivalent to 1151 * a boolean logic AND, or a set INTERSECTION. 1152 * A frozen set will not be modified. 1153 * 1154 * @param start first character, inclusive, of range to be retained 1155 * to this set. 1156 * @param end last character, inclusive, of range to be retained 1157 * to this set. 1158 * @stable ICU 2.0 1159 */ 1160 virtual UnicodeSet& retain(UChar32 start, UChar32 end); 1161 1162 1163 /** 1164 * Retain the specified character from this set if it is present. 1165 * A frozen set will not be modified. 1166 * @stable ICU 2.0 1167 */ 1168 UnicodeSet& retain(UChar32 c); 1169 1170 /** 1171 * Removes the specified range from this set if it is present. 1172 * The set will not contain the specified range once the call 1173 * returns. If <code>end > start</code> then an empty range is 1174 * removed, leaving the set unchanged. 1175 * A frozen set will not be modified. 1176 * 1177 * @param start first character, inclusive, of range to be removed 1178 * from this set. 1179 * @param end last character, inclusive, of range to be removed 1180 * from this set. 1181 * @stable ICU 2.0 1182 */ 1183 virtual UnicodeSet& remove(UChar32 start, UChar32 end); 1184 1185 /** 1186 * Removes the specified character from this set if it is present. 1187 * The set will not contain the specified range once the call 1188 * returns. 1189 * A frozen set will not be modified. 1190 * @stable ICU 2.0 1191 */ 1192 UnicodeSet& remove(UChar32 c); 1193 1194 /** 1195 * Removes the specified string from this set if it is present. 1196 * The set will not contain the specified character once the call 1197 * returns. 1198 * A frozen set will not be modified. 1199 * @param s the source string 1200 * @return this object, for chaining 1201 * @stable ICU 2.4 1202 */ 1203 UnicodeSet& remove(const UnicodeString& s); 1204 1205 /** 1206 * Inverts this set. This operation modifies this set so that 1207 * its value is its complement. This is equivalent to 1208 * <code>complement(MIN_VALUE, MAX_VALUE)</code>. 1209 * A frozen set will not be modified. 1210 * @stable ICU 2.0 1211 */ 1212 virtual UnicodeSet& complement(void); 1213 1214 /** 1215 * Complements the specified range in this set. Any character in 1216 * the range will be removed if it is in this set, or will be 1217 * added if it is not in this set. If <code>end > start</code> 1218 * then an empty range is complemented, leaving the set unchanged. 1219 * This is equivalent to a boolean logic XOR. 1220 * A frozen set will not be modified. 1221 * 1222 * @param start first character, inclusive, of range to be removed 1223 * from this set. 1224 * @param end last character, inclusive, of range to be removed 1225 * from this set. 1226 * @stable ICU 2.0 1227 */ 1228 virtual UnicodeSet& complement(UChar32 start, UChar32 end); 1229 1230 /** 1231 * Complements the specified character in this set. The character 1232 * will be removed if it is in this set, or will be added if it is 1233 * not in this set. 1234 * A frozen set will not be modified. 1235 * @stable ICU 2.0 1236 */ 1237 UnicodeSet& complement(UChar32 c); 1238 1239 /** 1240 * Complement the specified string in this set. 1241 * The set will not contain the specified string once the call 1242 * returns. 1243 * <br><b>Warning: you cannot add an empty string ("") to a UnicodeSet.</b> 1244 * A frozen set will not be modified. 1245 * @param s the string to complement 1246 * @return this object, for chaining 1247 * @stable ICU 2.4 1248 */ 1249 UnicodeSet& complement(const UnicodeString& s); 1250 1251 /** 1252 * Adds all of the elements in the specified set to this set if 1253 * they're not already present. This operation effectively 1254 * modifies this set so that its value is the <i>union</i> of the two 1255 * sets. The behavior of this operation is unspecified if the specified 1256 * collection is modified while the operation is in progress. 1257 * A frozen set will not be modified. 1258 * 1259 * @param c set whose elements are to be added to this set. 1260 * @see #add(UChar32, UChar32) 1261 * @stable ICU 2.0 1262 */ 1263 virtual UnicodeSet& addAll(const UnicodeSet& c); 1264 1265 /** 1266 * Retains only the elements in this set that are contained in the 1267 * specified set. In other words, removes from this set all of 1268 * its elements that are not contained in the specified set. This 1269 * operation effectively modifies this set so that its value is 1270 * the <i>intersection</i> of the two sets. 1271 * A frozen set will not be modified. 1272 * 1273 * @param c set that defines which elements this set will retain. 1274 * @stable ICU 2.0 1275 */ 1276 virtual UnicodeSet& retainAll(const UnicodeSet& c); 1277 1278 /** 1279 * Removes from this set all of its elements that are contained in the 1280 * specified set. This operation effectively modifies this 1281 * set so that its value is the <i>asymmetric set difference</i> of 1282 * the two sets. 1283 * A frozen set will not be modified. 1284 * 1285 * @param c set that defines which elements will be removed from 1286 * this set. 1287 * @stable ICU 2.0 1288 */ 1289 virtual UnicodeSet& removeAll(const UnicodeSet& c); 1290 1291 /** 1292 * Complements in this set all elements contained in the specified 1293 * set. Any character in the other set will be removed if it is 1294 * in this set, or will be added if it is not in this set. 1295 * A frozen set will not be modified. 1296 * 1297 * @param c set that defines which elements will be xor'ed from 1298 * this set. 1299 * @stable ICU 2.4 1300 */ 1301 virtual UnicodeSet& complementAll(const UnicodeSet& c); 1302 1303 /** 1304 * Removes all of the elements from this set. This set will be 1305 * empty after this call returns. 1306 * A frozen set will not be modified. 1307 * @stable ICU 2.0 1308 */ 1309 virtual UnicodeSet& clear(void); 1310 1311 /** 1312 * Close this set over the given attribute. For the attribute 1313 * USET_CASE, the result is to modify this set so that: 1314 * 1315 * 1. For each character or string 'a' in this set, all strings or 1316 * characters 'b' such that foldCase(a) == foldCase(b) are added 1317 * to this set. 1318 * 1319 * 2. For each string 'e' in the resulting set, if e != 1320 * foldCase(e), 'e' will be removed. 1321 * 1322 * Example: [aq\\u00DF{Bc}{bC}{Fi}] => [aAqQ\\u00DF\\uFB01{ss}{bc}{fi}] 1323 * 1324 * (Here foldCase(x) refers to the operation u_strFoldCase, and a 1325 * == b denotes that the contents are the same, not pointer 1326 * comparison.) 1327 * 1328 * A frozen set will not be modified. 1329 * 1330 * @param attribute bitmask for attributes to close over. 1331 * Currently only the USET_CASE bit is supported. Any undefined bits 1332 * are ignored. 1333 * @return a reference to this set. 1334 * @stable ICU 4.2 1335 */ 1336 UnicodeSet& closeOver(int32_t attribute); 1337 1338 /** 1339 * Remove all strings from this set. 1340 * 1341 * @return a reference to this set. 1342 * @stable ICU 4.2 1343 */ 1344 virtual UnicodeSet &removeAllStrings(); 1345 1346 /** 1347 * Iteration method that returns the number of ranges contained in 1348 * this set. 1349 * @see #getRangeStart 1350 * @see #getRangeEnd 1351 * @stable ICU 2.4 1352 */ 1353 virtual int32_t getRangeCount(void) const; 1354 1355 /** 1356 * Iteration method that returns the first character in the 1357 * specified range of this set. 1358 * @see #getRangeCount 1359 * @see #getRangeEnd 1360 * @stable ICU 2.4 1361 */ 1362 virtual UChar32 getRangeStart(int32_t index) const; 1363 1364 /** 1365 * Iteration method that returns the last character in the 1366 * specified range of this set. 1367 * @see #getRangeStart 1368 * @see #getRangeEnd 1369 * @stable ICU 2.4 1370 */ 1371 virtual UChar32 getRangeEnd(int32_t index) const; 1372 1373 /** 1374 * Serializes this set into an array of 16-bit integers. Serialization 1375 * (currently) only records the characters in the set; multicharacter 1376 * strings are ignored. 1377 * 1378 * The array has following format (each line is one 16-bit 1379 * integer): 1380 * 1381 * length = (n+2*m) | (m!=0?0x8000:0) 1382 * bmpLength = n; present if m!=0 1383 * bmp[0] 1384 * bmp[1] 1385 * ... 1386 * bmp[n-1] 1387 * supp-high[0] 1388 * supp-low[0] 1389 * supp-high[1] 1390 * supp-low[1] 1391 * ... 1392 * supp-high[m-1] 1393 * supp-low[m-1] 1394 * 1395 * The array starts with a header. After the header are n bmp 1396 * code points, then m supplementary code points. Either n or m 1397 * or both may be zero. n+2*m is always <= 0x7FFF. 1398 * 1399 * If there are no supplementary characters (if m==0) then the 1400 * header is one 16-bit integer, 'length', with value n. 1401 * 1402 * If there are supplementary characters (if m!=0) then the header 1403 * is two 16-bit integers. The first, 'length', has value 1404 * (n+2*m)|0x8000. The second, 'bmpLength', has value n. 1405 * 1406 * After the header the code points are stored in ascending order. 1407 * Supplementary code points are stored as most significant 16 1408 * bits followed by least significant 16 bits. 1409 * 1410 * @param dest pointer to buffer of destCapacity 16-bit integers. 1411 * May be NULL only if destCapacity is zero. 1412 * @param destCapacity size of dest, or zero. Must not be negative. 1413 * @param ec error code. Will be set to U_INDEX_OUTOFBOUNDS_ERROR 1414 * if n+2*m > 0x7FFF. Will be set to U_BUFFER_OVERFLOW_ERROR if 1415 * n+2*m+(m!=0?2:1) > destCapacity. 1416 * @return the total length of the serialized format, including 1417 * the header, that is, n+2*m+(m!=0?2:1), or 0 on error other 1418 * than U_BUFFER_OVERFLOW_ERROR. 1419 * @stable ICU 2.4 1420 */ 1421 int32_t serialize(uint16_t *dest, int32_t destCapacity, UErrorCode& ec) const; 1422 1423 /** 1424 * Reallocate this objects internal structures to take up the least 1425 * possible space, without changing this object's value. 1426 * A frozen set will not be modified. 1427 * @stable ICU 2.4 1428 */ 1429 virtual UnicodeSet& compact(); 1430 1431 /** 1432 * Return the class ID for this class. This is useful only for 1433 * comparing to a return value from getDynamicClassID(). For example: 1434 * <pre> 1435 * . Base* polymorphic_pointer = createPolymorphicObject(); 1436 * . if (polymorphic_pointer->getDynamicClassID() == 1437 * . Derived::getStaticClassID()) ... 1438 * </pre> 1439 * @return The class ID for all objects of this class. 1440 * @stable ICU 2.0 1441 */ 1442 static UClassID U_EXPORT2 getStaticClassID(void); 1443 1444 /** 1445 * Implement UnicodeFunctor API. 1446 * 1447 * @return The class ID for this object. All objects of a given 1448 * class have the same class ID. Objects of other classes have 1449 * different class IDs. 1450 * @stable ICU 2.4 1451 */ 1452 virtual UClassID getDynamicClassID(void) const; 1453 1454 private: 1455 1456 // Private API for the USet API 1457 1458 friend class USetAccess; 1459 1460 int32_t getStringCount() const; 1461 1462 const UnicodeString* getString(int32_t index) const; 1463 1464 //---------------------------------------------------------------- 1465 // RuleBasedTransliterator support 1466 //---------------------------------------------------------------- 1467 1468 private: 1469 1470 /** 1471 * Returns <tt>true</tt> if this set contains any character whose low byte 1472 * is the given value. This is used by <tt>RuleBasedTransliterator</tt> for 1473 * indexing. 1474 */ 1475 virtual UBool matchesIndexValue(uint8_t v) const; 1476 1477 private: 1478 friend class RBBIRuleScanner; 1479 1480 //---------------------------------------------------------------- 1481 // Implementation: Clone as thawed (see ICU4J Freezable) 1482 //---------------------------------------------------------------- 1483 1484 UnicodeSet(const UnicodeSet& o, UBool /* asThawed */); 1485 1486 //---------------------------------------------------------------- 1487 // Implementation: Pattern parsing 1488 //---------------------------------------------------------------- 1489 1490 void applyPatternIgnoreSpace(const UnicodeString& pattern, 1491 ParsePosition& pos, 1492 const SymbolTable* symbols, 1493 UErrorCode& status); 1494 1495 void applyPattern(RuleCharacterIterator& chars, 1496 const SymbolTable* symbols, 1497 UnicodeString& rebuiltPat, 1498 uint32_t options, 1499 UnicodeSet& (UnicodeSet::*caseClosure)(int32_t attribute), 1500 UErrorCode& ec); 1501 1502 //---------------------------------------------------------------- 1503 // Implementation: Utility methods 1504 //---------------------------------------------------------------- 1505 1506 void ensureCapacity(int32_t newLen, UErrorCode& ec); 1507 1508 void ensureBufferCapacity(int32_t newLen, UErrorCode& ec); 1509 1510 void swapBuffers(void); 1511 1512 UBool allocateStrings(UErrorCode &status); 1513 1514 UnicodeString& _toPattern(UnicodeString& result, 1515 UBool escapeUnprintable) const; 1516 1517 UnicodeString& _generatePattern(UnicodeString& result, 1518 UBool escapeUnprintable) const; 1519 1520 static void _appendToPat(UnicodeString& buf, const UnicodeString& s, UBool escapeUnprintable); 1521 1522 static void _appendToPat(UnicodeString& buf, UChar32 c, UBool escapeUnprintable); 1523 1524 //---------------------------------------------------------------- 1525 // Implementation: Fundamental operators 1526 //---------------------------------------------------------------- 1527 1528 void exclusiveOr(const UChar32* other, int32_t otherLen, int8_t polarity); 1529 1530 void add(const UChar32* other, int32_t otherLen, int8_t polarity); 1531 1532 void retain(const UChar32* other, int32_t otherLen, int8_t polarity); 1533 1534 /** 1535 * Return true if the given position, in the given pattern, appears 1536 * to be the start of a property set pattern [:foo:], \\p{foo}, or 1537 * \\P{foo}, or \\N{name}. 1538 */ 1539 static UBool resemblesPropertyPattern(const UnicodeString& pattern, 1540 int32_t pos); 1541 1542 static UBool resemblesPropertyPattern(RuleCharacterIterator& chars, 1543 int32_t iterOpts); 1544 1545 /** 1546 * Parse the given property pattern at the given parse position 1547 * and set this UnicodeSet to the result. 1548 * 1549 * The original design document is out of date, but still useful. 1550 * Ignore the property and value names: 1551 * http://source.icu-project.org/repos/icu/icuhtml/trunk/design/unicodeset_properties.html 1552 * 1553 * Recognized syntax: 1554 * 1555 * [:foo:] [:^foo:] - white space not allowed within "[:" or ":]" 1556 * \\p{foo} \\P{foo} - white space not allowed within "\\p" or "\\P" 1557 * \\N{name} - white space not allowed within "\\N" 1558 * 1559 * Other than the above restrictions, Unicode Pattern_White_Space characters are ignored. 1560 * Case is ignored except in "\\p" and "\\P" and "\\N". In 'name' leading 1561 * and trailing space is deleted, and internal runs of whitespace 1562 * are collapsed to a single space. 1563 * 1564 * We support binary properties, enumerated properties, and the 1565 * following non-enumerated properties: 1566 * 1567 * Numeric_Value 1568 * Name 1569 * Unicode_1_Name 1570 * 1571 * @param pattern the pattern string 1572 * @param ppos on entry, the position at which to begin parsing. 1573 * This should be one of the locations marked '^': 1574 * 1575 * [:blah:] \\p{blah} \\P{blah} \\N{name} 1576 * ^ % ^ % ^ % ^ % 1577 * 1578 * On return, the position after the last character parsed, that is, 1579 * the locations marked '%'. If the parse fails, ppos is returned 1580 * unchanged. 1581 * @param ec status 1582 * @return a reference to this. 1583 */ 1584 UnicodeSet& applyPropertyPattern(const UnicodeString& pattern, 1585 ParsePosition& ppos, 1586 UErrorCode &ec); 1587 1588 void applyPropertyPattern(RuleCharacterIterator& chars, 1589 UnicodeString& rebuiltPat, 1590 UErrorCode& ec); 1591 1592 friend void U_CALLCONV UnicodeSet_initInclusion(int32_t src, UErrorCode &status); 1593 static const UnicodeSet* getInclusions(int32_t src, UErrorCode &status); 1594 1595 /** 1596 * A filter that returns TRUE if the given code point should be 1597 * included in the UnicodeSet being constructed. 1598 */ 1599 typedef UBool (*Filter)(UChar32 codePoint, void* context); 1600 1601 /** 1602 * Given a filter, set this UnicodeSet to the code points 1603 * contained by that filter. The filter MUST be 1604 * property-conformant. That is, if it returns value v for one 1605 * code point, then it must return v for all affiliated code 1606 * points, as defined by the inclusions list. See 1607 * getInclusions(). 1608 * src is a UPropertySource value. 1609 */ 1610 void applyFilter(Filter filter, 1611 void* context, 1612 int32_t src, 1613 UErrorCode &status); 1614 1615 /** 1616 * Set the new pattern to cache. 1617 */ 1618 void setPattern(const UnicodeString& newPat); 1619 /** 1620 * Release existing cached pattern. 1621 */ 1622 void releasePattern(); 1623 1624 friend class UnicodeSetIterator; 1625 }; 1626 1627 1628 1629 inline UBool UnicodeSet::operator!=(const UnicodeSet& o) const { 1630 return !operator==(o); 1631 } 1632 1633 inline UBool UnicodeSet::isFrozen() const { 1634 return (UBool)(bmpSet!=NULL || stringSpan!=NULL); 1635 } 1636 1637 inline UBool UnicodeSet::containsSome(UChar32 start, UChar32 end) const { 1638 return !containsNone(start, end); 1639 } 1640 1641 inline UBool UnicodeSet::containsSome(const UnicodeSet& s) const { 1642 return !containsNone(s); 1643 } 1644 1645 inline UBool UnicodeSet::containsSome(const UnicodeString& s) const { 1646 return !containsNone(s); 1647 } 1648 1649 inline UBool UnicodeSet::isBogus() const { 1650 return (UBool)(fFlags & kIsBogus); 1651 } 1652 1653 inline UnicodeSet *UnicodeSet::fromUSet(USet *uset) { 1654 return reinterpret_cast<UnicodeSet *>(uset); 1655 } 1656 1657 inline const UnicodeSet *UnicodeSet::fromUSet(const USet *uset) { 1658 return reinterpret_cast<const UnicodeSet *>(uset); 1659 } 1660 1661 inline USet *UnicodeSet::toUSet() { 1662 return reinterpret_cast<USet *>(this); 1663 } 1664 1665 inline const USet *UnicodeSet::toUSet() const { 1666 return reinterpret_cast<const USet *>(this); 1667 } 1668 1669 inline int32_t UnicodeSet::span(const UnicodeString &s, int32_t start, USetSpanCondition spanCondition) const { 1670 int32_t sLength=s.length(); 1671 if(start<0) { 1672 start=0; 1673 } else if(start>sLength) { 1674 start=sLength; 1675 } 1676 return start+span(s.getBuffer()+start, sLength-start, spanCondition); 1677 } 1678 1679 inline int32_t UnicodeSet::spanBack(const UnicodeString &s, int32_t limit, USetSpanCondition spanCondition) const { 1680 int32_t sLength=s.length(); 1681 if(limit<0) { 1682 limit=0; 1683 } else if(limit>sLength) { 1684 limit=sLength; 1685 } 1686 return spanBack(s.getBuffer(), limit, spanCondition); 1687 } 1688 1689 U_NAMESPACE_END 1690 1691 #endif 1692