1 /* 2 * Copyright (c) 1997, 2012, Oracle and/or its affiliates. All rights reserved. 3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 4 * 5 * This code is free software; you can redistribute it and/or modify it 6 * under the terms of the GNU General Public License version 2 only, as 7 * published by the Free Software Foundation. Oracle designates this 8 * particular file as subject to the "Classpath" exception as provided 9 * by Oracle in the LICENSE file that accompanied this code. 10 * 11 * This code is distributed in the hope that it will be useful, but WITHOUT 12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 14 * version 2 for more details (a copy is included in the LICENSE file that 15 * accompanied this code). 16 * 17 * You should have received a copy of the GNU General Public License version 18 * 2 along with this work; if not, write to the Free Software Foundation, 19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 20 * 21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 22 * or visit www.oracle.com if you need additional information or have any 23 * questions. 24 */ 25 26 /* 27 * (C) Copyright Taligent, Inc. 1996-1998 - All Rights Reserved 28 * (C) Copyright IBM Corp. 1996-1998 - All Rights Reserved 29 * 30 * The original version of this source code and documentation is copyrighted 31 * and owned by Taligent, Inc., a wholly-owned subsidiary of IBM. These 32 * materials are provided under terms of a License Agreement between Taligent 33 * and Sun. This technology is protected by multiple US and International 34 * patents. This notice and attribution to Taligent may not be removed. 35 * Taligent is a registered trademark of Taligent, Inc. 36 * 37 */ 38 39 package java.text; 40 41 import java.lang.ref.SoftReference; 42 import java.text.spi.CollatorProvider; 43 import java.util.Locale; 44 import java.util.ResourceBundle; 45 import java.util.concurrent.ConcurrentHashMap; 46 import java.util.concurrent.ConcurrentMap; 47 import sun.util.locale.provider.LocaleProviderAdapter; 48 import sun.util.locale.provider.LocaleServiceProviderPool; 49 50 51 /** 52 * The <code>Collator</code> class performs locale-sensitive 53 * <code>String</code> comparison. You use this class to build 54 * searching and sorting routines for natural language text. 55 * 56 * <p> 57 * <code>Collator</code> is an abstract base class. Subclasses 58 * implement specific collation strategies. One subclass, 59 * <code>RuleBasedCollator</code>, is currently provided with 60 * the Java Platform and is applicable to a wide set of languages. Other 61 * subclasses may be created to handle more specialized needs. 62 * 63 * <p> 64 * Like other locale-sensitive classes, you can use the static 65 * factory method, <code>getInstance</code>, to obtain the appropriate 66 * <code>Collator</code> object for a given locale. You will only need 67 * to look at the subclasses of <code>Collator</code> if you need 68 * to understand the details of a particular collation strategy or 69 * if you need to modify that strategy. 70 * 71 * <p> 72 * The following example shows how to compare two strings using 73 * the <code>Collator</code> for the default locale. 74 * <blockquote> 75 * <pre>{@code 76 * // Compare two strings in the default locale 77 * Collator myCollator = Collator.getInstance(); 78 * if( myCollator.compare("abc", "ABC") < 0 ) 79 * System.out.println("abc is less than ABC"); 80 * else 81 * System.out.println("abc is greater than or equal to ABC"); 82 * }</pre> 83 * </blockquote> 84 * 85 * <p> 86 * You can set a <code>Collator</code>'s <em>strength</em> property 87 * to determine the level of difference considered significant in 88 * comparisons. Four strengths are provided: <code>PRIMARY</code>, 89 * <code>SECONDARY</code>, <code>TERTIARY</code>, and <code>IDENTICAL</code>. 90 * The exact assignment of strengths to language features is 91 * locale dependant. For example, in Czech, "e" and "f" are considered 92 * primary differences, while "e" and "ě" are secondary differences, 93 * "e" and "E" are tertiary differences and "e" and "e" are identical. 94 * The following shows how both case and accents could be ignored for 95 * US English. 96 * <blockquote> 97 * <pre> 98 * //Get the Collator for US English and set its strength to PRIMARY 99 * Collator usCollator = Collator.getInstance(Locale.US); 100 * usCollator.setStrength(Collator.PRIMARY); 101 * if( usCollator.compare("abc", "ABC") == 0 ) { 102 * System.out.println("Strings are equivalent"); 103 * } 104 * </pre> 105 * </blockquote> 106 * <p> 107 * For comparing <code>String</code>s exactly once, the <code>compare</code> 108 * method provides the best performance. When sorting a list of 109 * <code>String</code>s however, it is generally necessary to compare each 110 * <code>String</code> multiple times. In this case, <code>CollationKey</code>s 111 * provide better performance. The <code>CollationKey</code> class converts 112 * a <code>String</code> to a series of bits that can be compared bitwise 113 * against other <code>CollationKey</code>s. A <code>CollationKey</code> is 114 * created by a <code>Collator</code> object for a given <code>String</code>. 115 * <br> 116 * <strong>Note:</strong> <code>CollationKey</code>s from different 117 * <code>Collator</code>s can not be compared. See the class description 118 * for {@link CollationKey} 119 * for an example using <code>CollationKey</code>s. 120 * 121 * @see RuleBasedCollator 122 * @see CollationKey 123 * @see CollationElementIterator 124 * @see Locale 125 * @author Helena Shih, Laura Werner, Richard Gillam 126 */ 127 128 public abstract class Collator 129 implements java.util.Comparator<Object>, Cloneable 130 { 131 /** 132 * Collator strength value. When set, only PRIMARY differences are 133 * considered significant during comparison. The assignment of strengths 134 * to language features is locale dependant. A common example is for 135 * different base letters ("a" vs "b") to be considered a PRIMARY difference. 136 * @see java.text.Collator#setStrength 137 * @see java.text.Collator#getStrength 138 */ 139 public final static int PRIMARY = 0; 140 /** 141 * Collator strength value. When set, only SECONDARY and above differences are 142 * considered significant during comparison. The assignment of strengths 143 * to language features is locale dependant. A common example is for 144 * different accented forms of the same base letter ("a" vs "\u00E4") to be 145 * considered a SECONDARY difference. 146 * @see java.text.Collator#setStrength 147 * @see java.text.Collator#getStrength 148 */ 149 public final static int SECONDARY = 1; 150 /** 151 * Collator strength value. When set, only TERTIARY and above differences are 152 * considered significant during comparison. The assignment of strengths 153 * to language features is locale dependant. A common example is for 154 * case differences ("a" vs "A") to be considered a TERTIARY difference. 155 * @see java.text.Collator#setStrength 156 * @see java.text.Collator#getStrength 157 */ 158 public final static int TERTIARY = 2; 159 160 /** 161 * Collator strength value. When set, all differences are 162 * considered significant during comparison. The assignment of strengths 163 * to language features is locale dependant. A common example is for control 164 * characters ("\u0001" vs "\u0002") to be considered equal at the 165 * PRIMARY, SECONDARY, and TERTIARY levels but different at the IDENTICAL 166 * level. Additionally, differences between pre-composed accents such as 167 * "\u00C0" (A-grave) and combining accents such as "A\u0300" 168 * (A, combining-grave) will be considered significant at the IDENTICAL 169 * level if decomposition is set to NO_DECOMPOSITION. 170 */ 171 public final static int IDENTICAL = 3; 172 173 /** 174 * Decomposition mode value. With NO_DECOMPOSITION 175 * set, accented characters will not be decomposed for collation. This 176 * is the default setting and provides the fastest collation but 177 * will only produce correct results for languages that do not use accents. 178 * @see java.text.Collator#getDecomposition 179 * @see java.text.Collator#setDecomposition 180 */ 181 public final static int NO_DECOMPOSITION = 0; 182 183 /** 184 * Decomposition mode value. With CANONICAL_DECOMPOSITION 185 * set, characters that are canonical variants according to Unicode 186 * standard will be decomposed for collation. This should be used to get 187 * correct collation of accented characters. 188 * <p> 189 * CANONICAL_DECOMPOSITION corresponds to Normalization Form D as 190 * described in 191 * <a href="http://www.unicode.org/unicode/reports/tr15/tr15-23.html">Unicode 192 * Technical Report #15</a>. 193 * @see java.text.Collator#getDecomposition 194 * @see java.text.Collator#setDecomposition 195 */ 196 public final static int CANONICAL_DECOMPOSITION = 1; 197 198 /** 199 * Decomposition mode value. With FULL_DECOMPOSITION 200 * set, both Unicode canonical variants and Unicode compatibility variants 201 * will be decomposed for collation. This causes not only accented 202 * characters to be collated, but also characters that have special formats 203 * to be collated with their norminal form. For example, the half-width and 204 * full-width ASCII and Katakana characters are then collated together. 205 * FULL_DECOMPOSITION is the most complete and therefore the slowest 206 * decomposition mode. 207 * <p> 208 * FULL_DECOMPOSITION corresponds to Normalization Form KD as 209 * described in 210 * <a href="http://www.unicode.org/unicode/reports/tr15/tr15-23.html">Unicode 211 * Technical Report #15</a>. 212 * @see java.text.Collator#getDecomposition 213 * @see java.text.Collator#setDecomposition 214 */ 215 public final static int FULL_DECOMPOSITION = 2; 216 217 /** 218 * Gets the Collator for the current default locale. 219 * The default locale is determined by java.util.Locale.getDefault. 220 * @return the Collator for the default locale.(for example, en_US) 221 * @see java.util.Locale#getDefault 222 */ 223 public static synchronized Collator getInstance() { 224 return getInstance(Locale.getDefault()); 225 } 226 227 /** 228 * Gets the Collator for the desired locale. 229 * @param desiredLocale the desired locale. 230 * @return the Collator for the desired locale. 231 * @see java.util.Locale 232 * @see java.util.ResourceBundle 233 */ 234 public static Collator getInstance(Locale desiredLocale) { 235 SoftReference<Collator> ref = cache.get(desiredLocale); 236 Collator result = (ref != null) ? ref.get() : null; 237 if (result == null) { 238 LocaleProviderAdapter adapter; 239 adapter = LocaleProviderAdapter.getAdapter(CollatorProvider.class, 240 desiredLocale); 241 CollatorProvider provider = adapter.getCollatorProvider(); 242 result = provider.getInstance(desiredLocale); 243 if (result == null) { 244 result = LocaleProviderAdapter.forJRE() 245 .getCollatorProvider().getInstance(desiredLocale); 246 } 247 while (true) { 248 if (ref != null) { 249 // Remove the empty SoftReference if any 250 cache.remove(desiredLocale, ref); 251 } 252 ref = cache.putIfAbsent(desiredLocale, new SoftReference<>(result)); 253 if (ref == null) { 254 break; 255 } 256 Collator cachedColl = ref.get(); 257 if (cachedColl != null) { 258 result = cachedColl; 259 break; 260 } 261 } 262 } 263 return (Collator) result.clone(); // make the world safe 264 } 265 266 /** 267 * Compares the source string to the target string according to the 268 * collation rules for this Collator. Returns an integer less than, 269 * equal to or greater than zero depending on whether the source String is 270 * less than, equal to or greater than the target string. See the Collator 271 * class description for an example of use. 272 * <p> 273 * For a one time comparison, this method has the best performance. If a 274 * given String will be involved in multiple comparisons, CollationKey.compareTo 275 * has the best performance. See the Collator class description for an example 276 * using CollationKeys. 277 * @param source the source string. 278 * @param target the target string. 279 * @return Returns an integer value. Value is less than zero if source is less than 280 * target, value is zero if source and target are equal, value is greater than zero 281 * if source is greater than target. 282 * @see java.text.CollationKey 283 * @see java.text.Collator#getCollationKey 284 */ 285 public abstract int compare(String source, String target); 286 287 /** 288 * Compares its two arguments for order. Returns a negative integer, 289 * zero, or a positive integer as the first argument is less than, equal 290 * to, or greater than the second. 291 * <p> 292 * This implementation merely returns 293 * <code> compare((String)o1, (String)o2) </code>. 294 * 295 * @return a negative integer, zero, or a positive integer as the 296 * first argument is less than, equal to, or greater than the 297 * second. 298 * @exception ClassCastException the arguments cannot be cast to Strings. 299 * @see java.util.Comparator 300 * @since 1.2 301 */ 302 @Override 303 public int compare(Object o1, Object o2) { 304 return compare((String)o1, (String)o2); 305 } 306 307 /** 308 * Transforms the String into a series of bits that can be compared bitwise 309 * to other CollationKeys. CollationKeys provide better performance than 310 * Collator.compare when Strings are involved in multiple comparisons. 311 * See the Collator class description for an example using CollationKeys. 312 * @param source the string to be transformed into a collation key. 313 * @return the CollationKey for the given String based on this Collator's collation 314 * rules. If the source String is null, a null CollationKey is returned. 315 * @see java.text.CollationKey 316 * @see java.text.Collator#compare 317 */ 318 public abstract CollationKey getCollationKey(String source); 319 320 /** 321 * Convenience method for comparing the equality of two strings based on 322 * this Collator's collation rules. 323 * @param source the source string to be compared with. 324 * @param target the target string to be compared with. 325 * @return true if the strings are equal according to the collation 326 * rules. false, otherwise. 327 * @see java.text.Collator#compare 328 */ 329 public boolean equals(String source, String target) 330 { 331 return (compare(source, target) == Collator.EQUAL); 332 } 333 334 /** 335 * Returns this Collator's strength property. The strength property determines 336 * the minimum level of difference considered significant during comparison. 337 * See the Collator class description for an example of use. 338 * @return this Collator's current strength property. 339 * @see java.text.Collator#setStrength 340 * @see java.text.Collator#PRIMARY 341 * @see java.text.Collator#SECONDARY 342 * @see java.text.Collator#TERTIARY 343 * @see java.text.Collator#IDENTICAL 344 */ 345 public synchronized int getStrength() 346 { 347 return strength; 348 } 349 350 /** 351 * Sets this Collator's strength property. The strength property determines 352 * the minimum level of difference considered significant during comparison. 353 * See the Collator class description for an example of use. 354 * @param newStrength the new strength value. 355 * @see java.text.Collator#getStrength 356 * @see java.text.Collator#PRIMARY 357 * @see java.text.Collator#SECONDARY 358 * @see java.text.Collator#TERTIARY 359 * @see java.text.Collator#IDENTICAL 360 * @exception IllegalArgumentException If the new strength value is not one of 361 * PRIMARY, SECONDARY, TERTIARY or IDENTICAL. 362 */ 363 public synchronized void setStrength(int newStrength) { 364 if ((newStrength != PRIMARY) && 365 (newStrength != SECONDARY) && 366 (newStrength != TERTIARY) && 367 (newStrength != IDENTICAL)) { 368 throw new IllegalArgumentException("Incorrect comparison level."); 369 } 370 strength = newStrength; 371 } 372 373 /** 374 * Get the decomposition mode of this Collator. Decomposition mode 375 * determines how Unicode composed characters are handled. Adjusting 376 * decomposition mode allows the user to select between faster and more 377 * complete collation behavior. 378 * <p>The three values for decomposition mode are: 379 * <UL> 380 * <LI>NO_DECOMPOSITION, 381 * <LI>CANONICAL_DECOMPOSITION 382 * <LI>FULL_DECOMPOSITION. 383 * </UL> 384 * See the documentation for these three constants for a description 385 * of their meaning. 386 * @return the decomposition mode 387 * @see java.text.Collator#setDecomposition 388 * @see java.text.Collator#NO_DECOMPOSITION 389 * @see java.text.Collator#CANONICAL_DECOMPOSITION 390 * @see java.text.Collator#FULL_DECOMPOSITION 391 */ 392 public synchronized int getDecomposition() 393 { 394 return decmp; 395 } 396 /** 397 * Set the decomposition mode of this Collator. See getDecomposition 398 * for a description of decomposition mode. 399 * @param decompositionMode the new decomposition mode. 400 * @see java.text.Collator#getDecomposition 401 * @see java.text.Collator#NO_DECOMPOSITION 402 * @see java.text.Collator#CANONICAL_DECOMPOSITION 403 * @see java.text.Collator#FULL_DECOMPOSITION 404 * @exception IllegalArgumentException If the given value is not a valid decomposition 405 * mode. 406 */ 407 public synchronized void setDecomposition(int decompositionMode) { 408 if ((decompositionMode != NO_DECOMPOSITION) && 409 (decompositionMode != CANONICAL_DECOMPOSITION) && 410 (decompositionMode != FULL_DECOMPOSITION)) { 411 throw new IllegalArgumentException("Wrong decomposition mode."); 412 } 413 decmp = decompositionMode; 414 } 415 416 /** 417 * Returns an array of all locales for which the 418 * <code>getInstance</code> methods of this class can return 419 * localized instances. 420 * The returned array represents the union of locales supported 421 * by the Java runtime and by installed 422 * {@link java.text.spi.CollatorProvider CollatorProvider} implementations. 423 * It must contain at least a Locale instance equal to 424 * {@link java.util.Locale#US Locale.US}. 425 * 426 * @return An array of locales for which localized 427 * <code>Collator</code> instances are available. 428 */ 429 public static synchronized Locale[] getAvailableLocales() { 430 LocaleServiceProviderPool pool = 431 LocaleServiceProviderPool.getPool(CollatorProvider.class); 432 return pool.getAvailableLocales(); 433 } 434 435 /** 436 * Overrides Cloneable 437 */ 438 @Override 439 public Object clone() 440 { 441 try { 442 return (Collator)super.clone(); 443 } catch (CloneNotSupportedException e) { 444 throw new InternalError(e); 445 } 446 } 447 448 /** 449 * Compares the equality of two Collators. 450 * @param that the Collator to be compared with this. 451 * @return true if this Collator is the same as that Collator; 452 * false otherwise. 453 */ 454 @Override 455 public boolean equals(Object that) 456 { 457 if (this == that) { 458 return true; 459 } 460 if (that == null) { 461 return false; 462 } 463 if (getClass() != that.getClass()) { 464 return false; 465 } 466 Collator other = (Collator) that; 467 return ((strength == other.strength) && 468 (decmp == other.decmp)); 469 } 470 471 /** 472 * Generates the hash code for this Collator. 473 */ 474 @Override 475 abstract public int hashCode(); 476 477 /** 478 * Default constructor. This constructor is 479 * protected so subclasses can get access to it. Users typically create 480 * a Collator sub-class by calling the factory method getInstance. 481 * @see java.text.Collator#getInstance 482 */ 483 protected Collator() 484 { 485 strength = TERTIARY; 486 decmp = CANONICAL_DECOMPOSITION; 487 } 488 489 private int strength = 0; 490 private int decmp = 0; 491 private static final ConcurrentMap<Locale, SoftReference<Collator>> cache 492 = new ConcurrentHashMap<>(); 493 494 // 495 // FIXME: These three constants should be removed. 496 // 497 /** 498 * LESS is returned if source string is compared to be less than target 499 * string in the compare() method. 500 * @see java.text.Collator#compare 501 */ 502 final static int LESS = -1; 503 /** 504 * EQUAL is returned if source string is compared to be equal to target 505 * string in the compare() method. 506 * @see java.text.Collator#compare 507 */ 508 final static int EQUAL = 0; 509 /** 510 * GREATER is returned if source string is compared to be greater than 511 * target string in the compare() method. 512 * @see java.text.Collator#compare 513 */ 514 final static int GREATER = 1; 515 }