1 /* 2 * Copyright (C) 2014 The Android Open Source Project 3 * Copyright (c) 1997, 2006, Oracle and/or its affiliates. All rights reserved. 4 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 5 * 6 * This code is free software; you can redistribute it and/or modify it 7 * under the terms of the GNU General Public License version 2 only, as 8 * published by the Free Software Foundation. Oracle designates this 9 * particular file as subject to the "Classpath" exception as provided 10 * by Oracle in the LICENSE file that accompanied this code. 11 * 12 * This code is distributed in the hope that it will be useful, but WITHOUT 13 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 14 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 15 * version 2 for more details (a copy is included in the LICENSE file that 16 * accompanied this code). 17 * 18 * You should have received a copy of the GNU General Public License version 19 * 2 along with this work; if not, write to the Free Software Foundation, 20 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 21 * 22 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 23 * or visit www.oracle.com if you need additional information or have any 24 * questions. 25 */ 26 27 /* 28 * (C) Copyright Taligent, Inc. 1996-1998 - All Rights Reserved 29 * (C) Copyright IBM Corp. 1996-1998 - All Rights Reserved 30 * 31 * The original version of this source code and documentation is copyrighted 32 * and owned by Taligent, Inc., a wholly-owned subsidiary of IBM. These 33 * materials are provided under terms of a License Agreement between Taligent 34 * and Sun. This technology is protected by multiple US and International 35 * patents. This notice and attribution to Taligent may not be removed. 36 * Taligent is a registered trademark of Taligent, Inc. 37 * 38 */ 39 40 package java.text; 41 42 import java.util.Locale; 43 44 import libcore.icu.ICU; 45 46 /** 47 * The <code>Collator</code> class performs locale-sensitive 48 * <code>String</code> comparison. You use this class to build 49 * searching and sorting routines for natural language text. 50 * 51 * <p> 52 * <code>Collator</code> is an abstract base class. Subclasses 53 * implement specific collation strategies. One subclass, 54 * <code>RuleBasedCollator</code>, is currently provided with 55 * the Java Platform and is applicable to a wide set of languages. Other 56 * subclasses may be created to handle more specialized needs. 57 * 58 * <p> 59 * Like other locale-sensitive classes, you can use the static 60 * factory method, <code>getInstance</code>, to obtain the appropriate 61 * <code>Collator</code> object for a given locale. You will only need 62 * to look at the subclasses of <code>Collator</code> if you need 63 * to understand the details of a particular collation strategy or 64 * if you need to modify that strategy. 65 * 66 * <p> 67 * The following example shows how to compare two strings using 68 * the <code>Collator</code> for the default locale. 69 * <blockquote> 70 * <pre> 71 * // Compare two strings in the default locale 72 * Collator myCollator = Collator.getInstance(); 73 * if( myCollator.compare("abc", "ABC") < 0 ) 74 * System.out.println("abc is less than ABC"); 75 * else 76 * System.out.println("abc is greater than or equal to ABC"); 77 * </pre> 78 * </blockquote> 79 * 80 * <p> 81 * You can set a <code>Collator</code>'s <em>strength</em> property 82 * to determine the level of difference considered significant in 83 * comparisons. Four strengths are provided: <code>PRIMARY</code>, 84 * <code>SECONDARY</code>, <code>TERTIARY</code>, and <code>IDENTICAL</code>. 85 * The exact assignment of strengths to language features is 86 * locale dependant. For example, in Czech, "e" and "f" are considered 87 * primary differences, while "e" and "ě" are secondary differences, 88 * "e" and "E" are tertiary differences and "e" and "e" are identical. 89 * The following shows how both case and accents could be ignored for 90 * US English. 91 * <blockquote> 92 * <pre> 93 * //Get the Collator for US English and set its strength to PRIMARY 94 * Collator usCollator = Collator.getInstance(Locale.US); 95 * usCollator.setStrength(Collator.PRIMARY); 96 * if( usCollator.compare("abc", "ABC") == 0 ) { 97 * System.out.println("Strings are equivalent"); 98 * } 99 * </pre> 100 * </blockquote> 101 * <p> 102 * For comparing <code>String</code>s exactly once, the <code>compare</code> 103 * method provides the best performance. When sorting a list of 104 * <code>String</code>s however, it is generally necessary to compare each 105 * <code>String</code> multiple times. In this case, <code>CollationKey</code>s 106 * provide better performance. The <code>CollationKey</code> class converts 107 * a <code>String</code> to a series of bits that can be compared bitwise 108 * against other <code>CollationKey</code>s. A <code>CollationKey</code> is 109 * created by a <code>Collator</code> object for a given <code>String</code>. 110 * <br> 111 * <strong>Note:</strong> <code>CollationKey</code>s from different 112 * <code>Collator</code>s can not be compared. See the class description 113 * for {@link CollationKey} 114 * for an example using <code>CollationKey</code>s. 115 * 116 * @see RuleBasedCollator 117 * @see CollationKey 118 * @see CollationElementIterator 119 * @see Locale 120 * @author Helena Shih, Laura Werner, Richard Gillam 121 */ 122 123 public abstract class Collator 124 implements java.util.Comparator<Object>, Cloneable { 125 /** 126 * Collator strength value. When set, only PRIMARY differences are 127 * considered significant during comparison. The assignment of strengths 128 * to language features is locale dependant. A common example is for 129 * different base letters ("a" vs "b") to be considered a PRIMARY difference. 130 * 131 * @see java.text.Collator#setStrength 132 * @see java.text.Collator#getStrength 133 */ 134 public final static int PRIMARY = 0; 135 /** 136 * Collator strength value. When set, only SECONDARY and above differences are 137 * considered significant during comparison. The assignment of strengths 138 * to language features is locale dependant. A common example is for 139 * different accented forms of the same base letter ("a" vs "\u00E4") to be 140 * considered a SECONDARY difference. 141 * 142 * @see java.text.Collator#setStrength 143 * @see java.text.Collator#getStrength 144 */ 145 public final static int SECONDARY = 1; 146 /** 147 * Collator strength value. When set, only TERTIARY and above differences are 148 * considered significant during comparison. The assignment of strengths 149 * to language features is locale dependant. A common example is for 150 * case differences ("a" vs "A") to be considered a TERTIARY difference. 151 * 152 * @see java.text.Collator#setStrength 153 * @see java.text.Collator#getStrength 154 */ 155 public final static int TERTIARY = 2; 156 157 /** 158 * Collator strength value. When set, all differences are 159 * considered significant during comparison. The assignment of strengths 160 * to language features is locale dependant. A common example is for control 161 * characters ("\u0001" vs "\u0002") to be considered equal at the 162 * PRIMARY, SECONDARY, and TERTIARY levels but different at the IDENTICAL 163 * level. Additionally, differences between pre-composed accents such as 164 * "\u00C0" (A-grave) and combining accents such as "A\u0300" 165 * (A, combining-grave) will be considered significant at the IDENTICAL 166 * level if decomposition is set to NO_DECOMPOSITION. 167 */ 168 public final static int IDENTICAL = 3; 169 170 /** 171 * Decomposition mode value. With NO_DECOMPOSITION 172 * set, accented characters will not be decomposed for collation. This 173 * is the default setting and provides the fastest collation but 174 * will only produce correct results for languages that do not use accents. 175 * 176 * @see java.text.Collator#getDecomposition 177 * @see java.text.Collator#setDecomposition 178 */ 179 public final static int NO_DECOMPOSITION = 0; 180 181 /** 182 * Decomposition mode value. With CANONICAL_DECOMPOSITION 183 * set, characters that are canonical variants according to Unicode 184 * standard will be decomposed for collation. This should be used to get 185 * correct collation of accented characters. 186 * <p> 187 * CANONICAL_DECOMPOSITION corresponds to Normalization Form D as 188 * described in 189 * <a href="http://www.unicode.org/unicode/reports/tr15/tr15-23.html">Unicode 190 * Technical Report #15</a>. 191 * 192 * @see java.text.Collator#getDecomposition 193 * @see java.text.Collator#setDecomposition 194 */ 195 public final static int CANONICAL_DECOMPOSITION = 1; 196 197 /** 198 * Decomposition mode value. With FULL_DECOMPOSITION 199 * set, both Unicode canonical variants and Unicode compatibility variants 200 * will be decomposed for collation. This causes not only accented 201 * characters to be collated, but also characters that have special formats 202 * to be collated with their norminal form. For example, the half-width and 203 * full-width ASCII and Katakana characters are then collated together. 204 * FULL_DECOMPOSITION is the most complete and therefore the slowest 205 * decomposition mode. 206 * <p> 207 * FULL_DECOMPOSITION corresponds to Normalization Form KD as 208 * described in 209 * <a href="http://www.unicode.org/unicode/reports/tr15/tr15-23.html">Unicode 210 * Technical Report #15</a>. 211 * 212 * @see java.text.Collator#getDecomposition 213 * @see java.text.Collator#setDecomposition 214 */ 215 public final static int FULL_DECOMPOSITION = 2; 216 217 android.icu.text.Collator icuColl; 218 Collator(android.icu.text.Collator icuColl)219 Collator(android.icu.text.Collator icuColl) { 220 this.icuColl = icuColl; 221 } 222 223 /** 224 * Default constructor. This constructor is 225 * protected so subclasses can get access to it. Users typically create 226 * a Collator sub-class by calling the factory method getInstance. 227 * 228 * @see java.text.Collator#getInstance 229 */ Collator()230 protected Collator() { 231 icuColl = android.icu.text.RuleBasedCollator.getInstance(Locale.getDefault()); 232 } 233 234 /** 235 * Gets the Collator for the current default locale. 236 * The default locale is determined by java.util.Locale.getDefault. 237 * 238 * @return the Collator for the default locale.(for example, en_US) 239 * @see java.util.Locale#getDefault 240 */ getInstance()241 public static synchronized Collator getInstance() { 242 return getInstance(Locale.getDefault()); 243 } 244 245 /** 246 * Gets the Collator for the desired locale. 247 * 248 * @param desiredLocale the desired locale. 249 * @return the Collator for the desired locale. 250 * @see java.util.Locale 251 * @see java.util.ResourceBundle 252 */ getInstance(Locale desiredLocale)253 public static synchronized Collator getInstance(Locale desiredLocale) { 254 if (desiredLocale == null) { 255 throw new NullPointerException("locale == null"); 256 } 257 return new RuleBasedCollator((android.icu.text.RuleBasedCollator) 258 android.icu.text.Collator.getInstance(desiredLocale)); 259 } 260 261 /** 262 * Returns an array of all locales for which the 263 * <code>getInstance</code> methods of this class can return 264 * localized instances. 265 * The returned array represents the union of locales supported 266 * by the Java runtime and by installed 267 * {@link java.text.spi.CollatorProvider CollatorProvider} implementations. 268 * It must contain at least a Locale instance equal to 269 * {@link java.util.Locale#US Locale.US}. 270 * 271 * @return An array of locales for which localized 272 * <code>Collator</code> instances are available. 273 */ getAvailableLocales()274 public static synchronized Locale[] getAvailableLocales() { 275 return ICU.getAvailableCollatorLocales(); 276 } 277 278 /** 279 * Returns a new collator with the same decomposition mode and 280 * strength value as this collator. 281 * 282 * @return a shallow copy of this collator. 283 * @see java.lang.Cloneable 284 */ 285 @Override clone()286 public Object clone() { 287 try { 288 Collator clone = (Collator) super.clone(); 289 clone.icuColl = (android.icu.text.Collator) icuColl.clone(); 290 return clone; 291 } catch (CloneNotSupportedException e) { 292 throw new AssertionError(e); 293 } 294 } 295 296 /** 297 * Compares the source string to the target string according to the 298 * collation rules for this Collator. Returns an integer less than, 299 * equal to or greater than zero depending on whether the source String is 300 * less than, equal to or greater than the target string. See the Collator 301 * class description for an example of use. 302 * <p> 303 * For a one time comparison, this method has the best performance. If a 304 * given String will be involved in multiple comparisons, CollationKey.compareTo 305 * has the best performance. See the Collator class description for an example 306 * using CollationKeys. 307 * 308 * @param source the source string. 309 * @param target the target string. 310 * @return Returns an integer value. Value is less than zero if source is less than 311 * target, value is zero if source and target are equal, value is greater than zero 312 * if source is greater than target. 313 * @see java.text.CollationKey 314 * @see java.text.Collator#getCollationKey 315 */ compare(String source, String target)316 public abstract int compare(String source, String target); 317 318 /** 319 * Compares its two arguments for order. Returns a negative integer, 320 * zero, or a positive integer as the first argument is less than, equal 321 * to, or greater than the second. 322 * <p> 323 * This implementation merely returns 324 * <code> compare((String)o1, (String)o2) </code>. 325 * 326 * @return a negative integer, zero, or a positive integer as the 327 * first argument is less than, equal to, or greater than the 328 * second. 329 * @throws ClassCastException the arguments cannot be cast to Strings. 330 * @see java.util.Comparator 331 * @since 1.2 332 */ compare(Object o1, Object o2)333 public int compare(Object o1, Object o2) { 334 return compare((String) o1, (String) o2); 335 } 336 337 /** 338 * Transforms the String into a series of bits that can be compared bitwise 339 * to other CollationKeys. CollationKeys provide better performance than 340 * Collator.compare when Strings are involved in multiple comparisons. 341 * See the Collator class description for an example using CollationKeys. 342 * 343 * @param source the string to be transformed into a collation key. 344 * @return the CollationKey for the given String based on this Collator's collation 345 * rules. If the source String is null, a null CollationKey is returned. 346 * @see java.text.CollationKey 347 * @see java.text.Collator#compare 348 */ getCollationKey(String source)349 public abstract CollationKey getCollationKey(String source); 350 351 /** 352 * Convenience method for comparing the equality of two strings based on 353 * this Collator's collation rules. 354 * 355 * @param source the source string to be compared with. 356 * @param target the target string to be compared with. 357 * @return true if the strings are equal according to the collation 358 * rules. false, otherwise. 359 * @see java.text.Collator#compare 360 */ equals(String source, String target)361 public boolean equals(String source, String target) { 362 return (compare(source, target) == 0); 363 } 364 365 /** 366 * Returns this Collator's strength property. The strength property determines 367 * the minimum level of difference considered significant during comparison. 368 * See the Collator class description for an example of use. 369 * 370 * @return this Collator's current strength property. 371 * @see java.text.Collator#setStrength 372 * @see java.text.Collator#PRIMARY 373 * @see java.text.Collator#SECONDARY 374 * @see java.text.Collator#TERTIARY 375 * @see java.text.Collator#IDENTICAL 376 */ getStrength()377 public synchronized int getStrength() { 378 // The value for IDENTICAL in ICU differs from that used in this class. 379 int value = icuColl.getStrength(); 380 return (value == android.icu.text.Collator.IDENTICAL) ? IDENTICAL : value; 381 } 382 383 /** 384 * Sets this Collator's strength property. The strength property determines 385 * the minimum level of difference considered significant during comparison. 386 * See the Collator class description for an example of use. 387 * 388 * @param newStrength the new strength value. 389 * @throws IllegalArgumentException If the new strength value is not one of 390 * PRIMARY, SECONDARY, TERTIARY or IDENTICAL. 391 * @see Collator#getStrength 392 * @see Collator#PRIMARY 393 * @see Collator#SECONDARY 394 * @see Collator#TERTIARY 395 * @see Collator#IDENTICAL 396 */ setStrength(int newStrength)397 public synchronized void setStrength(int newStrength) { 398 // The ICU value for IDENTICAL differs from that defined in this class. 399 if (newStrength == IDENTICAL) { 400 newStrength = android.icu.text.Collator.IDENTICAL; 401 } 402 icuColl.setStrength(newStrength); 403 } 404 405 /** 406 * Get the decomposition mode of this Collator. Decomposition mode 407 * determines how Unicode composed characters are handled. Adjusting 408 * decomposition mode allows the user to select between faster and more 409 * complete collation behavior. 410 * <p>The three values for decomposition mode are: 411 * <UL> 412 * <LI>NO_DECOMPOSITION, 413 * <LI>CANONICAL_DECOMPOSITION 414 * <LI>FULL_DECOMPOSITION. 415 * </UL> 416 * See the documentation for these three constants for a description 417 * of their meaning. 418 * 419 * @return the decomposition mode 420 * @see java.text.Collator#setDecomposition 421 * @see java.text.Collator#NO_DECOMPOSITION 422 * @see java.text.Collator#CANONICAL_DECOMPOSITION 423 * @see java.text.Collator#FULL_DECOMPOSITION 424 */ getDecomposition()425 public synchronized int getDecomposition() { 426 return decompositionMode_ICU_Java(icuColl.getDecomposition()); 427 } 428 429 /** 430 * Set the decomposition mode of this Collator. See getDecomposition 431 * for a description of decomposition mode. 432 * 433 * @param decompositionMode the new decomposition mode. 434 * @throws IllegalArgumentException If the given value is not a valid decomposition 435 * mode. 436 * @see java.text.Collator#getDecomposition 437 * @see java.text.Collator#NO_DECOMPOSITION 438 * @see java.text.Collator#CANONICAL_DECOMPOSITION 439 * @see java.text.Collator#FULL_DECOMPOSITION 440 */ setDecomposition(int decompositionMode)441 public synchronized void setDecomposition(int decompositionMode) { 442 icuColl.setDecomposition(decompositionMode_Java_ICU(decompositionMode)); 443 } 444 decompositionMode_Java_ICU(int mode)445 private int decompositionMode_Java_ICU(int mode) { 446 switch (mode) { 447 case Collator.CANONICAL_DECOMPOSITION: 448 return android.icu.text.Collator.CANONICAL_DECOMPOSITION; 449 case Collator.NO_DECOMPOSITION: 450 return android.icu.text.Collator.NO_DECOMPOSITION; 451 } 452 throw new IllegalArgumentException("Bad mode: " + mode); 453 } 454 decompositionMode_ICU_Java(int mode)455 private int decompositionMode_ICU_Java(int mode) { 456 int javaMode = mode; 457 switch (mode) { 458 case android.icu.text.Collator.NO_DECOMPOSITION: 459 javaMode = Collator.NO_DECOMPOSITION; 460 break; 461 case android.icu.text.Collator.CANONICAL_DECOMPOSITION: 462 javaMode = Collator.CANONICAL_DECOMPOSITION; 463 break; 464 } 465 return javaMode; 466 } 467 468 /** 469 * Compares the equality of two Collators. 470 * 471 * @param that the Collator to be compared with this. 472 * @return true if this Collator is the same as that Collator; 473 * false otherwise. 474 */ equals(Object that)475 public boolean equals(Object that) { 476 if (this == that) return true; 477 if (that == null) return false; 478 if (getClass() != that.getClass()) return false; 479 Collator other = (Collator) that; 480 return icuColl == null ? other.icuColl == null : icuColl.equals(other.icuColl); 481 } 482 483 /** 484 * Generates the hash code for this Collator. 485 */ hashCode()486 abstract public int hashCode(); 487 } 488