libjava/java/text/BreakIterator.java

   1 /* BreakIterator.java -- Breaks text into elements
   2    Copyright (C) 1998, 1999, 2001 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., 59 Temple Place, Suite 330, Boston, MA
  19 02111-1307 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 java.util.Locale;
  42 import java.util.MissingResourceException;
  43 import java.util.ResourceBundle;
  44
  45 /**
  46  * This class iterates over text elements such as words, lines, sentences,
  47  * and characters.  It can only iterate over one of these text elements at
  48  * a time.  An instance of this class configured for the desired iteration
  49  * type is created by calling one of the static factory methods, not
  50  * by directly calling a constructor.
  51  *
  52  * @author Tom Tromey <tromey@cygnus.com>
  53  * @author Aaron M. Renn (arenn@urbanophile.com)
  54  * @date March 19, 1999
  55  */
  56 /* Written using "Java Class Libraries", 2nd edition, plus online
  57  * API docs for JDK 1.2 beta from http://www.javasoft.com.
  58  * Status:  Believed complete and correct to 1.1.
  59  */
  60 public abstract class BreakIterator implements Cloneable
  61 {
  62   /**
  63    * This value is returned by the <code>next()</code> and
  64    * <code>previous</code> in order to indicate that the end of the
  65    * text has been reached.
  66    */
  67   // The value was discovered by writing a test program.
  68   public static final int DONE = -1;
  69
  70   /**
  71    * This method initializes a new instance of <code>BreakIterator</code>.
  72    * This protected constructor is available to subclasses as a default
  73    * no-arg superclass constructor.
  74    */
  75   protected BreakIterator ()
  76   {
  77   }
  78
  79   /**
  80    * Create a clone of this object.
  81    */
  82   public Object clone ()
  83   {
  84     try
  85       {
  86         return super.clone();
  87       }
  88     catch (CloneNotSupportedException e)
  89       {
  90         return null;
  91       }
  92   }
  93
  94   /**
  95    * This method returns the index of the current text element boundary.
  96    *
  97    * @return The current text boundary.
  98    */
  99   public abstract int current ();
 100
 101   /**
 102    * This method returns the first text element boundary in the text being
 103    * iterated over.
 104    *
 105    * @return The first text boundary.
 106    */
 107   public abstract int first ();
 108
 109   /**
 110    * This methdod returns the offset of the text element boundary following
 111    * the specified offset.
 112    *
 113    * @param offset The text index from which to find the next text boundary.
 114    *
 115    * @param The next text boundary following the specified index.
 116    */
 117   public abstract int following (int pos);
 118
 119   /**
 120    * This method returns a list of locales for which instances of
 121    * <code>BreakIterator</code> are available.
 122    *
 123    * @return A list of available locales
 124    */
 125   public static synchronized Locale[] getAvailableLocales ()
 126   {
 127     Locale[] l = new Locale[1];
 128     l[0] = Locale.US;
 129     return l;
 130   }
 131
 132   private static BreakIterator getInstance (String type, Locale loc)
 133   {
 134     String className;
 135     try
 136       {
 137         ResourceBundle res
 138           = ResourceBundle.getBundle("gnu.java.locale.LocaleInformation",
 139                                      loc);
 140         className = res.getString(type);
 141       }
 142     catch (MissingResourceException x)
 143       {
 144         return null;
 145       }
 146     try
 147       {
 148         Class k = Class.forName(className);
 149         return (BreakIterator) k.newInstance();
 150       }
 151     catch (ClassNotFoundException x1)
 152       {
 153         return null;
 154       }
 155     catch (InstantiationException x2)
 156       {
 157         return null;
 158       }
 159     catch (IllegalAccessException x3)
 160       {
 161         return null;
 162       }
 163   }
 164
 165   /**
 166    * This method returns an instance of <code>BreakIterator</code> that will
 167    * iterate over characters as defined in the default locale.
 168    *
 169    * @return A <code>BreakIterator</code> instance for the default locale.
 170    */
 171   public static BreakIterator getCharacterInstance ()
 172   {
 173     return getCharacterInstance (Locale.getDefault());
 174   }
 175
 176   /**
 177    * This method returns an instance of <code>BreakIterator</code> that will
 178    * iterate over characters as defined in the specified locale.  If the
 179    * desired locale is not available, the default locale is used.
 180    *
 181    * @param locale The desired locale.
 182    *
 183    * @return A <code>BreakIterator</code> instance for the default locale.
 184    */
 185   public static BreakIterator getCharacterInstance (Locale loc)
 186   {
 187     BreakIterator r = getInstance ("CharacterIterator", loc);
 188     if (r == null)
 189       r = new gnu.java.text.CharacterBreakIterator ();
 190     return r;
 191   }
 192
 193   /**
 194    * This method returns an instance of <code>BreakIterator</code> that will
 195    * iterate over line breaks as defined in the default locale.
 196    *
 197    * @return A <code>BreakIterator</code> instance for the default locale.
 198    */
 199   public static BreakIterator getLineInstance ()
 200   {
 201     return getLineInstance (Locale.getDefault());
 202   }
 203
 204   /**
 205    * This method returns an instance of <code>BreakIterator</code> that will
 206    * iterate over line breaks as defined in the specified locale.  If the
 207    * desired locale is not available, the default locale is used.
 208    *
 209    * @param locale The desired locale.
 210    *
 211    * @return A <code>BreakIterator</code> instance for the default locale.
 212    */
 213   public static BreakIterator getLineInstance (Locale loc)
 214   {
 215     BreakIterator r = getInstance ("LineIterator", loc);
 216     if (r == null)
 217       r = new gnu.java.text.LineBreakIterator ();
 218     return r;
 219   }
 220
 221   /**
 222    * This method returns an instance of <code>BreakIterator</code> that will
 223    * iterate over sentences as defined in the default locale.
 224    *
 225    * @return A <code>BreakIterator</code> instance for the default locale.
 226    */
 227   public static BreakIterator getSentenceInstance ()
 228   {
 229     return getSentenceInstance (Locale.getDefault());
 230   }
 231
 232   /**
 233    * This method returns an instance of <code>BreakIterator</code> that will
 234    * iterate over sentences as defined in the specified locale.  If the
 235    * desired locale is not available, the default locale is used.
 236    *
 237    * @param locale The desired locale.
 238    *
 239    * @return A <code>BreakIterator</code> instance for the default locale.
 240    */
 241   public static BreakIterator getSentenceInstance (Locale loc)
 242   {
 243     BreakIterator r = getInstance ("SentenceIterator", loc);
 244     if (r == null)
 245       r = new gnu.java.text.SentenceBreakIterator ();
 246     return r;
 247   }
 248
 249   /**
 250    * This method returns the text this object is iterating over as a
 251    * <code>CharacterIterator</code>.
 252    *
 253    * @param The text being iterated over.
 254    */
 255   public abstract CharacterIterator getText ();
 256
 257   /**
 258    * This method returns an instance of <code>BreakIterator</code> that will
 259    * iterate over words as defined in the default locale.
 260    *
 261    * @return A <code>BreakIterator</code> instance for the default locale.
 262    */
 263   public static BreakIterator getWordInstance ()
 264   {
 265     return getWordInstance (Locale.getDefault());
 266   }
 267
 268   /**
 269    * This method returns an instance of <code>BreakIterator</code> that will
 270    * iterate over words as defined in the specified locale.  If the
 271    * desired locale is not available, the default locale is used.
 272    *
 273    * @param locale The desired locale.
 274    *
 275    * @return A <code>BreakIterator</code> instance for the default locale.
 276    */
 277   public static BreakIterator getWordInstance (Locale loc)
 278   {
 279     BreakIterator r = getInstance ("WordIterator", loc);
 280     if (r == null)
 281       r = new gnu.java.text.WordBreakIterator ();
 282     return r;
 283   }
 284
 285   /**
 286    * This method tests whether or not the specified position is a text
 287    * element boundary.
 288    *
 289    * @param offset The text position to test.
 290    *
 291    * @return <code>true</code> if the position is a boundary,
 292    * <code>false</code> otherwise.
 293    */
 294   public boolean isBoundary (int pos)
 295   {
 296     if (pos == 0)
 297       return true;
 298     return following (pos - 1) == pos;
 299   }
 300
 301   /**
 302    * This method returns the last text element boundary in the text being
 303    * iterated over.
 304    *
 305    * @return The last text boundary.
 306    */
 307   public abstract int last ();
 308
 309   /**
 310    * This method returns the text element boundary following the current
 311    * text position.
 312    *
 313    * @return The next text boundary.
 314    */
 315   public abstract int next ();
 316
 317   /**
 318    * This method returns the n'th text element boundary following the current
 319    * text position.
 320    *
 321    * @param n The number of text element boundaries to skip.
 322    *
 323    * @return The next text boundary.
 324    */
 325   public abstract int next (int n);
 326
 327   /**
 328    * This methdod returns the offset of the text element boundary preceding
 329    * the specified offset.
 330    *
 331    * @param offset The text index from which to find the preceding
 332    *               text boundary.
 333    *
 334    * @returns The next text boundary preceding the specified index.
 335    */
 336   public int preceding (int pos)
 337   {
 338     if (following (pos) == DONE)
 339       last ();
 340     while (previous () >= pos)
 341       ;
 342     return current ();
 343   }
 344
 345   /**
 346    * This method returns the text element boundary preceding the current
 347    * text position.
 348    *
 349    * @return The previous text boundary.
 350    */
 351   public abstract int previous ();
 352
 353   /**
 354    * This method sets the text string to iterate over.
 355    *
 356    * @param str The <code>String</code> to iterate over.
 357    */
 358   public void setText (String newText)
 359   {
 360     setText (new StringCharacterIterator (newText));
 361   }
 362
 363   /**
 364    * This method sets the text to iterate over from the specified
 365    * <code>CharacterIterator</code>.
 366    *
 367    * @param ci The desired <code>CharacterIterator</code>.
 368    */
 369   public abstract void setText (CharacterIterator newText);
 370 }