| GNU Classpath (0.95) | |
| Frames | No Frames |
1: /* Collator.java -- Perform locale dependent String comparisons. 2: Copyright (C) 1998, 1999, 2000, 2001, 2004, 2005 Free Software Foundation, Inc. 3: 4: This file is part of GNU Classpath. 5: 6: GNU Classpath is free software; you can redistribute it and/or modify 7: it under the terms of the GNU General Public License as published by 8: the Free Software Foundation; either version 2, or (at your option) 9: any later version. 10: 11: GNU Classpath is distributed in the hope that it will be useful, but 12: WITHOUT ANY WARRANTY; without even the implied warranty of 13: MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 14: General Public License for more details. 15: 16: You should have received a copy of the GNU General Public License 17: along with GNU Classpath; see the file COPYING. If not, write to the 18: Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 19: 02110-1301 USA. 20: 21: Linking this library statically or dynamically with other modules is 22: making a combined work based on this library. Thus, the terms and 23: conditions of the GNU General Public License cover the whole 24: combination. 25: 26: As a special exception, the copyright holders of this library give you 27: permission to link this library with independent modules to produce an 28: executable, regardless of the license terms of these independent 29: modules, and to copy and distribute the resulting executable under 30: terms of your choice, provided that you also meet, for each linked 31: independent module, the terms and conditions of the license of that 32: module. An independent module is a module which is not derived from 33: or based on this library. If you modify this library, you may extend 34: this exception to your version of the library, but you are not 35: obligated to do so. If you do not wish to do so, delete this 36: exception statement from your version. */ 37: 38: 39: package java.text; 40: 41: import gnu.java.locale.LocaleHelper; 42: 43: import java.text.spi.CollatorProvider; 44: 45: import java.util.Comparator; 46: import java.util.Locale; 47: import java.util.MissingResourceException; 48: import java.util.ResourceBundle; 49: import java.util.ServiceLoader; 50: 51: /** 52: * This class is the abstract superclass of classes which perform 53: * locale dependent <code>String</code> comparisons. A caller requests 54: * an instance of <code>Collator</code> for a particular locale using 55: * the <code>getInstance()</code> static method in this class. That method 56: * will return a locale specific subclass of <code>Collator</code> which 57: * can be used to perform <code>String</code> comparisons for that locale. 58: * If a subclass of <code>Collator</code> cannot be located for a particular 59: * locale, a default instance for the current locale will be returned. 60: * 61: * In addition to setting the correct locale, there are two additional 62: * settings that can be adjusted to affect <code>String</code> comparisons: 63: * strength and decomposition. The strength value determines the level 64: * of signficance of character differences required for them to sort 65: * differently. (For example, whether or not capital letters are considered 66: * different from lower case letters). The decomposition value affects how 67: * variants of the same character are treated for sorting purposes. (For 68: * example, whether or not an accent is signficant or not). These settings 69: * are described in detail in the documentation for the methods and values 70: * that are related to them. 71: * 72: * @author Tom Tromey (tromey@cygnus.com) 73: * @author Aaron M. Renn (arenn@urbanophile.com) 74: * @date March 18, 1999 75: */ 76: public abstract class Collator implements Comparator<Object>, Cloneable 77: { 78: /** 79: * This constant is a strength value which indicates that only primary 80: * differences between characters will be considered signficant. As an 81: * example, two completely different English letters such as 'a' and 'b' 82: * are considered to have a primary difference. 83: */ 84: public static final int PRIMARY = 0; 85: 86: /** 87: * This constant is a strength value which indicates that only secondary 88: * or primary differences between characters will be considered 89: * significant. An example of a secondary difference between characters 90: * are instances of the same letter with different accented forms. 91: */ 92: public static final int SECONDARY = 1; 93: 94: /** 95: * This constant is a strength value which indicates that tertiary, 96: * secondary, and primary differences will be considered during sorting. 97: * An example of a tertiary difference is capitalization of a given letter. 98: * This is the default value for the strength setting. 99: */ 100: public static final int TERTIARY = 2; 101: 102: /** 103: * This constant is a strength value which indicates that any difference 104: * at all between character values are considered significant. 105: */ 106: public static final int IDENTICAL = 3; 107: 108: /** 109: * This constant indicates that accented characters won't be decomposed 110: * when performing comparisons. This will yield the fastest results, but 111: * will only work correctly in call cases for languages which do not 112: * use accents such as English. 113: */ 114: public static final int NO_DECOMPOSITION = 0; 115: 116: /** 117: * This constant indicates that only characters which are canonical variants 118: * in Unicode 2.0 will be decomposed prior to performing comparisons. This 119: * will cause accented languages to be sorted correctly. This is the 120: * default decomposition value. 121: */ 122: public static final int CANONICAL_DECOMPOSITION = 1; 123: 124: /** 125: * This constant indicates that both canonical variants and compatibility 126: * variants in Unicode 2.0 will be decomposed prior to performing 127: * comparisons. This is the slowest mode, but is required to get the 128: * correct sorting for certain languages with certain special formats. 129: */ 130: public static final int FULL_DECOMPOSITION = 2; 131: 132: /** 133: * This method initializes a new instance of <code>Collator</code> to have 134: * the default strength (TERTIARY) and decomposition 135: * (CANONICAL_DECOMPOSITION) settings. This constructor is protected and 136: * is for use by subclasses only. Non-subclass callers should use the 137: * static <code>getInstance()</code> methods of this class to instantiate 138: * <code>Collation</code> objects for the desired locale. 139: */ 140: protected Collator () 141: { 142: strength = TERTIARY; 143: decmp = CANONICAL_DECOMPOSITION; 144: } 145: 146: /** 147: * This method compares the two <code>String</code>'s and returns an 148: * integer indicating whether or not the first argument is less than, 149: * equal to, or greater than the second argument. The comparison is 150: * performed according to the rules of the locale for this 151: * <code>Collator</code> and the strength and decomposition rules in 152: * effect. 153: * 154: * @param source The first object to compare 155: * @param target The second object to compare 156: * 157: * @return A negative integer if str1 < str2, 0 if str1 == str2, or 158: * a positive integer if str1 > str2. 159: */ 160: public abstract int compare (String source, String target); 161: 162: /** 163: * This method compares the two <code>Object</code>'s and returns an 164: * integer indicating whether or not the first argument is less than, 165: * equal to, or greater than the second argument. These two objects 166: * must be <code>String</code>'s or an exception will be thrown. 167: * 168: * @param o1 The first object to compare 169: * @param o2 The second object to compare 170: * 171: * @return A negative integer if obj1 < obj2, 0 if obj1 == obj2, or 172: * a positive integer if obj1 > obj2. 173: * 174: * @exception ClassCastException If the arguments are not instances 175: * of <code>String</code>. 176: */ 177: public int compare (Object o1, Object o2) 178: { 179: return compare ((String) o1, (String) o2); 180: } 181: 182: /** 183: * This method tests the specified object for equality against this 184: * object. This will be true if and only if the following conditions are 185: * met: 186: * <ul> 187: * <li>The specified object is not <code>null</code>.</li> 188: * <li>The specified object is an instance of <code>Collator</code>.</li> 189: * <li>The specified object has the same strength and decomposition 190: * settings as this object.</li> 191: * </ul> 192: * 193: * @param obj The <code>Object</code> to test for equality against 194: * this object. 195: * 196: * @return <code>true</code> if the specified object is equal to 197: * this one, <code>false</code> otherwise. 198: */ 199: public boolean equals (Object obj) 200: { 201: if (! (obj instanceof Collator)) 202: return false; 203: Collator c = (Collator) obj; 204: return decmp == c.decmp && strength == c.strength; 205: } 206: 207: /** 208: * This method tests whether the specified <code>String</code>'s are equal 209: * according to the collation rules for the locale of this object and 210: * the current strength and decomposition settings. 211: * 212: * @param source The first <code>String</code> to compare 213: * @param target The second <code>String</code> to compare 214: * 215: * @return <code>true</code> if the two strings are equal, 216: * <code>false</code> otherwise. 217: */ 218: public boolean equals (String source, String target) 219: { 220: return compare (source, target) == 0; 221: } 222: 223: /** 224: * This method returns a copy of this <code>Collator</code> object. 225: * 226: * @return A duplicate of this object. 227: */ 228: public Object clone () 229: { 230: try 231: { 232: return super.clone (); 233: } 234: catch (CloneNotSupportedException _) 235: { 236: return null; 237: } 238: } 239: 240: /** 241: * This method returns an array of <code>Locale</code> objects which is 242: * the list of locales for which <code>Collator</code> objects exist. 243: * 244: * @return The list of locales for which <code>Collator</code>'s exist. 245: */ 246: public static synchronized Locale[] getAvailableLocales () 247: { 248: return LocaleHelper.getCollatorLocales(); 249: } 250: 251: /** 252: * This method transforms the specified <code>String</code> into a 253: * <code>CollationKey</code> for faster comparisons. This is useful when 254: * comparisons against a string might be performed multiple times, such 255: * as during a sort operation. 256: * 257: * @param source The <code>String</code> to convert. 258: * 259: * @return A <code>CollationKey</code> for the specified <code>String</code>. 260: */ 261: public abstract CollationKey getCollationKey (String source); 262: 263: /** 264: * This method returns the current decomposition setting for this 265: * object. This * will be one of NO_DECOMPOSITION, 266: * CANONICAL_DECOMPOSITION, or * FULL_DECOMPOSITION. See the 267: * documentation for those constants for an * explanation of this 268: * setting. 269: * 270: * @return The current decomposition setting. 271: */ 272: public synchronized int getDecomposition () 273: { 274: return decmp; 275: } 276: 277: /** 278: * This method returns an instance of <code>Collator</code> for the 279: * default locale. 280: * 281: * @return A <code>Collator</code> for the default locale. 282: */ 283: public static Collator getInstance () 284: { 285: return getInstance (Locale.getDefault()); 286: } 287: 288: /** 289: * This method returns an instance of <code>Collator</code> for the 290: * specified locale. If no <code>Collator</code> exists for the desired 291: * locale, the fallback procedure described in 292: * {@link java.util.spi.LocaleServiceProvider} is invoked. 293: * 294: * @param loc The desired locale to load a <code>Collator</code> for. 295: * 296: * @return A <code>Collator</code> for the requested locale 297: */ 298: public static Collator getInstance (Locale loc) 299: { 300: String pattern; 301: try 302: { 303: ResourceBundle res = 304: ResourceBundle.getBundle("gnu.java.locale.LocaleInformation", 305: loc, ClassLoader.getSystemClassLoader()); 306: return new RuleBasedCollator(res.getString("collation_rules")); 307: } 308: catch (MissingResourceException x) 309: { 310: /* This means runtime support for the locale 311: * is not available, so we check providers. */ 312: } 313: catch (ParseException x) 314: { 315: throw (InternalError)new InternalError().initCause(x); 316: } 317: for (CollatorProvider p : ServiceLoader.load(CollatorProvider.class)) 318: { 319: for (Locale l : p.getAvailableLocales()) 320: { 321: if (l.equals(loc)) 322: { 323: Collator c = p.getInstance(loc); 324: if (c != null) 325: return c; 326: break; 327: } 328: } 329: } 330: if (loc.equals(Locale.ROOT)) 331: { 332: try 333: { 334: return new RuleBasedCollator("<0<1<2<3<4<5<6<7<8<9<A,a<b,B<c," + 335: "C<d,D<e,E<f,F<g,G<h,H<i,I<j,J<k,K" + 336: "<l,L<m,M<n,N<o,O<p,P<q,Q<r,R<s,S<t,"+ 337: "T<u,U<v,V<w,W<x,X<y,Y<z,Z"); 338: } 339: catch (ParseException x) 340: { 341: throw (InternalError)new InternalError().initCause(x); 342: } 343: } 344: return getInstance(LocaleHelper.getFallbackLocale(loc)); 345: } 346: 347: /** 348: * This method returns the current strength setting for this object. This 349: * will be one of PRIMARY, SECONDARY, TERTIARY, or IDENTICAL. See the 350: * documentation for those constants for an explanation of this setting. 351: * 352: * @return The current strength setting. 353: */ 354: public synchronized int getStrength () 355: { 356: return strength; 357: } 358: 359: /** 360: * This method returns a hash code value for this object. 361: * 362: * @return A hash value for this object. 363: */ 364: public abstract int hashCode (); 365: 366: /** 367: * This method sets the decomposition setting for this object to the 368: * specified value. This must be one of NO_DECOMPOSITION, 369: * CANONICAL_DECOMPOSITION, or FULL_DECOMPOSITION. Otherwise an 370: * exception will be thrown. See the documentation for those 371: * contants for an explanation of this setting. 372: * 373: * @param mode The new decomposition setting. 374: * 375: * @exception IllegalArgumentException If the requested 376: * decomposition setting is not valid. 377: */ 378: public synchronized void setDecomposition (int mode) 379: { 380: if (mode != NO_DECOMPOSITION 381: && mode != CANONICAL_DECOMPOSITION 382: && mode != FULL_DECOMPOSITION) 383: throw new IllegalArgumentException (); 384: decmp = mode; 385: } 386: 387: /** 388: * This method sets the strength setting for this object to the specified 389: * value. This must be one of PRIMARY, SECONDARY, TERTIARY, or IDENTICAL. 390: * Otherwise an exception is thrown. See the documentation for these 391: * constants for an explanation of this setting. 392: * 393: * @param strength The new strength setting. 394: * 395: * @exception IllegalArgumentException If the requested strength 396: * setting value is not valid. 397: */ 398: public synchronized void setStrength (int strength) 399: { 400: if (strength != PRIMARY && strength != SECONDARY 401: && strength != TERTIARY && strength != IDENTICAL) 402: throw new IllegalArgumentException (); 403: this.strength = strength; 404: } 405: 406: // Decompose a single character and append results to the buffer. 407: // FIXME: for libgcj this is a native method which handles 408: // decomposition. For Classpath, for now, it does nothing. 409: final void decomposeCharacter (char c, StringBuffer buf) 410: { 411: buf.append (c); 412: } 413: 414: /** 415: * This is the current collation decomposition setting. 416: */ 417: int decmp; 418: 419: /** 420: * This is the current collation strength setting. 421: */ 422: int strength; 423: }
| GNU Classpath (0.95) |