libjava/java/text/BreakIterator.java

   1 /* BreakIterator.java -- Breaks text into elements
   2    Copyright (C) 1998, 1999, 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., 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  * The standard iterators created by the factory methods in this
  53  * class will be valid upon creation.  That is, their methods will
  54  * not cause exceptions if called before you call setText().
  55  *
  56  * @author Tom Tromey (tromey@cygnus.com)
  57  * @author Aaron M. Renn (arenn@urbanophile.com)
  58  * @date March 19, 1999
  59  */
  60 /* Written using "Java Class Libraries", 2nd edition, plus online
  61  * API docs for JDK 1.2 beta from http://www.javasoft.com.
  62  * Status:  Believed complete and correct to 1.1.
  63  */
  64 public abstract class BreakIterator implements Cloneable
  65 {
  66   /**
  67    * This value is returned by the <code>next()</code> and
  68    * <code>previous</code> in order to indicate that the end of the
  69    * text has been reached.
  70    */
  71   // The value was discovered by writing a test program.
  72   public static final int DONE = -1;
  73
  74   /**
  75    * This method initializes a new instance of <code>BreakIterator</code>.
  76    * This protected constructor is available to subclasses as a default
  77    * no-arg superclass constructor.
  78    */
  79   protected BreakIterator ()
  80   {
  81   }
  82
  83   /**
  84    * Create a clone of this object.
  85    */
  86   public Object clone ()
  87   {
  88     try
  89       {
  90         return super.clone();
  91       }
  92     catch (CloneNotSupportedException e)
  93       {
  94         return null;
  95       }
  96   }
  97
  98   /**
  99    * This method returns the index of the current text element boundary.
 100    *
 101    * @return The current text boundary.
 102    */
 103   public abstract int current ();
 104
 105   /**
 106    * This method returns the first text element boundary in the text being
 107    * iterated over.
 108    *
 109    * @return The first text boundary.
 110    */
 111   public abstract int first ();
 112
 113   /**
 114    * This methdod returns the offset of the text element boundary following
 115    * the specified offset.
 116    *
 117    * @param offset The text index from which to find the next text boundary.
 118    *
 119    * @param The next text boundary following the specified index.
 120    */
 121   public abstract int following (int pos);
 122
 123   /**
 124    * This method returns a list of locales for which instances of
 125    * <code>BreakIterator</code> are available.
 126    *
 127    * @return A list of available locales
 128    */
 129   public static synchronized Locale[] getAvailableLocales ()
 130   {
 131     Locale[] l = new Locale[1];
 132     l[0] = Locale.US;
 133     return l;
 134   }
 135
 136   private static BreakIterator getInstance (String type, Locale loc)
 137   {
 138     String className;
 139     try
 140       {
 141         ResourceBundle res
 142           = ResourceBundle.getBundle("gnu.java.locale.LocaleInformation",
 143                                      loc, ClassLoader.getSystemClassLoader());
 144         className = res.getString(type);
 145       }
 146     catch (MissingResourceException x)
 147       {
 148         return null;
 149       }
 150     try
 151       {
 152         Class k = Class.forName(className);
 153         return (BreakIterator) k.newInstance();
 154       }
 155     catch (ClassNotFoundException x1)
 156       {
 157         return null;
 158       }
 159     catch (InstantiationException x2)
 160       {
 161         return null;
 162       }
 163     catch (IllegalAccessException x3)
 164       {
 165         return null;
 166       }
 167   }
 168
 169   /**
 170    * This method returns an instance of <code>BreakIterator</code> that will
 171    * iterate over characters as defined in the default locale.
 172    *
 173    * @return A <code>BreakIterator</code> instance for the default locale.
 174    */
 175   public static BreakIterator getCharacterInstance ()
 176   {
 177     return getCharacterInstance (Locale.getDefault());
 178   }
 179
 180   /**
 181    * This method returns an instance of <code>BreakIterator</code> that will
 182    * iterate over characters as defined in the specified locale.  If the
 183    * desired locale is not available, the default locale is used.
 184    *
 185    * @param locale The desired locale.
 186    *
 187    * @return A <code>BreakIterator</code> instance for the default locale.
 188    */
 189   public static BreakIterator getCharacterInstance (Locale loc)
 190   {
 191     BreakIterator r = getInstance ("CharacterIterator", loc);
 192     if (r == null)
 193       r = new gnu.java.text.CharacterBreakIterator ();
 194     return r;
 195   }
 196
 197   /**
 198    * This method returns an instance of <code>BreakIterator</code> that will
 199    * iterate over line breaks as defined in the default locale.
 200    *
 201    * @return A <code>BreakIterator</code> instance for the default locale.
 202    */
 203   public static BreakIterator getLineInstance ()
 204   {
 205     return getLineInstance (Locale.getDefault());
 206   }
 207
 208   /**
 209    * This method returns an instance of <code>BreakIterator</code> that will
 210    * iterate over line breaks as defined in the specified locale.  If the
 211    * desired locale is not available, the default locale is used.
 212    *
 213    * @param locale The desired locale.
 214    *
 215    * @return A <code>BreakIterator</code> instance for the default locale.
 216    */
 217   public static BreakIterator getLineInstance (Locale loc)
 218   {
 219     BreakIterator r = getInstance ("LineIterator", loc);
 220     if (r == null)
 221       r = new gnu.java.text.LineBreakIterator ();
 222     return r;
 223   }
 224
 225   /**
 226    * This method returns an instance of <code>BreakIterator</code> that will
 227    * iterate over sentences as defined in the default locale.
 228    *
 229    * @return A <code>BreakIterator</code> instance for the default locale.
 230    */
 231   public static BreakIterator getSentenceInstance ()
 232   {
 233     return getSentenceInstance (Locale.getDefault());
 234   }
 235
 236   /**
 237    * This method returns an instance of <code>BreakIterator</code> that will
 238    * iterate over sentences as defined in the specified locale.  If the
 239    * desired locale is not available, the default locale is used.
 240    *
 241    * @param locale The desired locale.
 242    *
 243    * @return A <code>BreakIterator</code> instance for the default locale.
 244    */
 245   public static BreakIterator getSentenceInstance (Locale loc)
 246   {
 247     BreakIterator r = getInstance ("SentenceIterator", loc);
 248     if (r == null)
 249       r = new gnu.java.text.SentenceBreakIterator ();
 250     return r;
 251   }
 252
 253   /**
 254    * This method returns the text this object is iterating over as a
 255    * <code>CharacterIterator</code>.
 256    *
 257    * @param The text being iterated over.
 258    */
 259   public abstract CharacterIterator getText ();
 260
 261   /**
 262    * This method returns an instance of <code>BreakIterator</code> that will
 263    * iterate over words as defined in the default locale.
 264    *
 265    * @return A <code>BreakIterator</code> instance for the default locale.
 266    */
 267   public static BreakIterator getWordInstance ()
 268   {
 269     return getWordInstance (Locale.getDefault());
 270   }
 271
 272   /**
 273    * This method returns an instance of <code>BreakIterator</code> that will
 274    * iterate over words as defined in the specified locale.  If the
 275    * desired locale is not available, the default locale is used.
 276    *
 277    * @param locale The desired locale.
 278    *
 279    * @return A <code>BreakIterator</code> instance for the default locale.
 280    */
 281   public static BreakIterator getWordInstance (Locale loc)
 282   {
 283     BreakIterator r = getInstance ("WordIterator", loc);
 284     if (r == null)
 285       r = new gnu.java.text.WordBreakIterator ();
 286     return r;
 287   }
 288
 289   /**
 290    * This method tests whether or not the specified position is a text
 291    * element boundary.
 292    *
 293    * @param offset The text position to test.
 294    *
 295    * @return <code>true</code> if the position is a boundary,
 296    * <code>false</code> otherwise.
 297    */
 298   public boolean isBoundary (int pos)
 299   {
 300     if (pos == 0)
 301       return true;
 302     return following (pos - 1) == pos;
 303   }
 304
 305   /**
 306    * This method returns the last text element boundary in the text being
 307    * iterated over.
 308    *
 309    * @return The last text boundary.
 310    */
 311   public abstract int last ();
 312
 313   /**
 314    * This method returns the text element boundary following the current
 315    * text position.
 316    *
 317    * @return The next text boundary.
 318    */
 319   public abstract int next ();
 320
 321   /**
 322    * This method returns the n'th text element boundary following the current
 323    * text position.
 324    *
 325    * @param n The number of text element boundaries to skip.
 326    *
 327    * @return The next text boundary.
 328    */
 329   public abstract int next (int n);
 330
 331   /**
 332    * This methdod returns the offset of the text element boundary preceding
 333    * the specified offset.
 334    *
 335    * @param offset The text index from which to find the preceding
 336    *               text boundary.
 337    *
 338    * @returns The next text boundary preceding the specified index.
 339    */
 340   public int preceding (int pos)
 341   {
 342     if (following (pos) == DONE)
 343       last ();
 344     while (previous () >= pos)
 345       ;
 346     return current ();
 347   }
 348
 349   /**
 350    * This method returns the text element boundary preceding the current
 351    * text position.
 352    *
 353    * @return The previous text boundary.
 354    */
 355   public abstract int previous ();
 356
 357   /**
 358    * This method sets the text string to iterate over.
 359    *
 360    * @param str The <code>String</code> to iterate over.
 361    */
 362   public void setText (String newText)
 363   {
 364     setText (new StringCharacterIterator (newText));
 365   }
 366
 367   /**
 368    * This method sets the text to iterate over from the specified
 369    * <code>CharacterIterator</code>.
 370    *
 371    * @param ci The desired <code>CharacterIterator</code>.
 372    */
 373   public abstract void setText (CharacterIterator newText);
 374 }