Java Code Examples for java.text.BreakIterator#following()

/**
 * Needed to unify forward and backward searching.
 * The method assumes that s is the text assigned to words.
 */
private int findWordLimit(int index, BreakIterator words, boolean direction,
                                 String s) {
    // Fix for 4256660 and 4256661.
    // Words iterator is different from character and sentence iterators
    // in that end of one word is not necessarily start of another word.
    // Please see java.text.BreakIterator JavaDoc. The code below is
    // based on nextWordStartAfter example from BreakIterator.java.
    int last = (direction == NEXT) ? words.following(index)
                                   : words.preceding(index);
    int current = (direction == NEXT) ? words.next()
                                      : words.previous();
    while (current != BreakIterator.DONE) {
        for (int p = Math.min(last, current); p < Math.max(last, current); p++) {
            if (Character.isLetter(s.charAt(p))) {
                return last;
            }
        }
        last = current;
        current = (direction == NEXT) ? words.next()
                                      : words.previous();
    }
    return BreakIterator.DONE;
}
 

/**
 * Needed to unify forward and backward searching.
 * The method assumes that s is the text assigned to words.
 */
private int findWordLimit(int index, BreakIterator words, boolean direction,
                                 String s) {
    // Fix for 4256660 and 4256661.
    // Words iterator is different from character and sentence iterators
    // in that end of one word is not necessarily start of another word.
    // Please see java.text.BreakIterator JavaDoc. The code below is
    // based on nextWordStartAfter example from BreakIterator.java.
    int last = (direction == NEXT) ? words.following(index)
                                   : words.preceding(index);
    int current = (direction == NEXT) ? words.next()
                                      : words.previous();
    while (current != BreakIterator.DONE) {
        for (int p = Math.min(last, current); p < Math.max(last, current); p++) {
            if (Character.isLetter(s.charAt(p))) {
                return last;
            }
        }
        last = current;
        current = (direction == NEXT) ? words.next()
                                      : words.previous();
    }
    return BreakIterator.DONE;
}
 

private void testFollowing(BreakIterator bi, String text, int[] boundaries) {
    logln("testFollowing():");
    int p = 2;
    int i = 0;
    try {
        for (i = 0; i <= text.length(); i++) {  // change to <= when new BI code goes in
            if (i == boundaries[p])
                ++p;

            int b = bi.following(i);
            logln("bi.following(" + i + ") -> " + b);
            if (b != boundaries[p])
                errln("Wrong result from following() for " + i + ": expected " + boundaries[p]
                      + ", got " + b);
        }
    } catch (IllegalArgumentException illargExp) {
        errln("IllegalArgumentException caught from following() for offset: " + i);
    }
}
 

/**
 * Needed to unify forward and backward searching.
 * The method assumes that s is the text assigned to words.
 */
private int findWordLimit(int index, BreakIterator words, boolean direction,
                                 String s) {
    // Fix for 4256660 and 4256661.
    // Words iterator is different from character and sentence iterators
    // in that end of one word is not necessarily start of another word.
    // Please see java.text.BreakIterator JavaDoc. The code below is
    // based on nextWordStartAfter example from BreakIterator.java.
    int last = (direction == NEXT) ? words.following(index)
                                   : words.preceding(index);
    int current = (direction == NEXT) ? words.next()
                                      : words.previous();
    while (current != BreakIterator.DONE) {
        for (int p = Math.min(last, current); p < Math.max(last, current); p++) {
            if (Character.isLetter(s.charAt(p))) {
                return last;
            }
        }
        last = current;
        current = (direction == NEXT) ? words.next()
                                      : words.previous();
    }
    return BreakIterator.DONE;
}
 

private void testFollowing(BreakIterator bi, String text, int[] boundaries) {
    logln("testFollowing():");
    int p = 2;
    int i = 0;
    try {
        for (i = 0; i <= text.length(); i++) {  // change to <= when new BI code goes in
            if (i == boundaries[p])
                ++p;

            int b = bi.following(i);
            logln("bi.following(" + i + ") -> " + b);
            if (b != boundaries[p])
                errln("Wrong result from following() for " + i + ": expected " + boundaries[p]
                      + ", got " + b);
        }
    } catch (IllegalArgumentException illargExp) {
        errln("IllegalArgumentException caught from following() for offset: " + i);
    }
}
 

/**
 * Needed to unify forward and backward searching.
 * The method assumes that s is the text assigned to words.
 */
private int findWordLimit(int index, BreakIterator words, boolean direction,
                                 String s) {
    // Fix for 4256660 and 4256661.
    // Words iterator is different from character and sentence iterators
    // in that end of one word is not necessarily start of another word.
    // Please see java.text.BreakIterator JavaDoc. The code below is
    // based on nextWordStartAfter example from BreakIterator.java.
    int last = (direction == NEXT) ? words.following(index)
                                   : words.preceding(index);
    int current = (direction == NEXT) ? words.next()
                                      : words.previous();
    while (current != BreakIterator.DONE) {
        for (int p = Math.min(last, current); p < Math.max(last, current); p++) {
            if (Character.isLetter(s.charAt(p))) {
                return last;
            }
        }
        last = current;
        current = (direction == NEXT) ? words.next()
                                      : words.previous();
    }
    return BreakIterator.DONE;
}
 

/**
 * Needed to unify forward and backward searching.
 * The method assumes that s is the text assigned to words.
 */
private int findWordLimit(int index, BreakIterator words, boolean direction,
                                 String s) {
    // Fix for 4256660 and 4256661.
    // Words iterator is different from character and sentence iterators
    // in that end of one word is not necessarily start of another word.
    // Please see java.text.BreakIterator JavaDoc. The code below is
    // based on nextWordStartAfter example from BreakIterator.java.
    int last = (direction == NEXT) ? words.following(index)
                                   : words.preceding(index);
    int current = (direction == NEXT) ? words.next()
                                      : words.previous();
    while (current != BreakIterator.DONE) {
        for (int p = Math.min(last, current); p < Math.max(last, current); p++) {
            if (Character.isLetter(s.charAt(p))) {
                return last;
            }
        }
        last = current;
        current = (direction == NEXT) ? words.next()
                                      : words.previous();
    }
    return BreakIterator.DONE;
}
 

public static int getWordStart(JTextComponent c, int offs) throws BadLocationException {
  Element line = Optional.ofNullable(Utilities.getParagraphElement(c, offs))
      .orElseThrow(() -> new BadLocationException("No word at " + offs, offs));
  Document doc = c.getDocument();
  int lineStart = line.getStartOffset();
  int lineEnd = Math.min(line.getEndOffset(), doc.getLength());
  int offs2 = offs;
  Segment seg = SegmentCache.getSharedSegment();
  doc.getText(lineStart, lineEnd - lineStart, seg);
  if (seg.count > 0) {
    BreakIterator words = BreakIterator.getWordInstance(c.getLocale());
    words.setText(seg);
    int wordPosition = seg.offset + offs - lineStart;
    if (wordPosition >= words.last()) {
      wordPosition = words.last() - 1;
      words.following(wordPosition);
      offs2 = lineStart + words.previous() - seg.offset;
    } else {
      words.following(wordPosition);
      offs2 = lineStart + words.previous() - seg.offset;
      for (int i = offs; i > offs2; i--) {
        char ch = seg.charAt(i - seg.offset);
        if (ch == '_' || ch == '-') {
          offs2 = i + 1;
          break;
        }
      }
    }
  }
  SegmentCache.releaseSharedSegment(seg);
  return offs2;
}
 

public static int getWordEnd(JTextComponent c, int offs) throws BadLocationException {
  Element line = Optional.ofNullable(Utilities.getParagraphElement(c, offs))
      .orElseThrow(() -> new BadLocationException("No word at " + offs, offs));
  Document doc = c.getDocument();
  int lineStart = line.getStartOffset();
  int lineEnd = Math.min(line.getEndOffset(), doc.getLength());
  int offs2 = offs;

  Segment seg = SegmentCache.getSharedSegment();
  doc.getText(lineStart, lineEnd - lineStart, seg);
  if (seg.count > 0) {
    BreakIterator words = BreakIterator.getWordInstance(c.getLocale());
    words.setText(seg);
    int wordPosition = offs - lineStart + seg.offset;
    if (wordPosition >= words.last()) {
      wordPosition = words.last() - 1;
    }
    offs2 = lineStart + words.following(wordPosition) - seg.offset;

    for (int i = offs; i < offs2; i++) {
      char ch = seg.charAt(i - seg.offset);
      if (ch == '_' || ch == '-') {
        offs2 = i;
        break;
      }
    }
  }
  SegmentCache.releaseSharedSegment(seg);
  return offs2;
}
 

/**
 * Tries to find the word at the given offset.
 * 
 * @param line
 *            the line
 * @param offset
 *            the offset
 * @return the word or <code>null</code> if none
 */
protected static IRegion findWordRegion(String line, int offset)
{
	BreakIterator breakIter = BreakIterator.getWordInstance();
	breakIter.setText(line);

	int start = breakIter.preceding(offset);
	if (start == BreakIterator.DONE)
		start = 0;

	int end = breakIter.following(offset);
	if (end == BreakIterator.DONE)
		end = line.length();

	if (breakIter.isBoundary(offset))
	{
		if (end - offset > offset - start)
		{
			start = offset;
		}
		else
		{
			end = offset;
		}
	}

	if (end == start)
	{
		return new Region(start, 0);
	}
	return new Region(start, end - start);
}
 

/**
 * Returns the Segment at <code>index</code> representing either
 * the paragraph or sentence as identified by <code>part</code>, or
 * null if a valid paragraph/sentence can't be found. The offset
 * will point to the start of the word/sentence in the array, and
 * the modelOffset will point to the location of the word/sentence
 * in the model.
 */
private IndexedSegment getSegmentAt(int part, int index)
    throws BadLocationException {

    IndexedSegment seg = getParagraphElementText(index);
    if (seg == null) {
        return null;
    }
    BreakIterator iterator;
    switch (part) {
    case AccessibleText.WORD:
        iterator = BreakIterator.getWordInstance(getLocale());
        break;
    case AccessibleText.SENTENCE:
        iterator = BreakIterator.getSentenceInstance(getLocale());
        break;
    default:
        return null;
    }
    seg.first();
    iterator.setText(seg);
    int end = iterator.following(index - seg.modelOffset + seg.offset);
    if (end == BreakIterator.DONE) {
        return null;
    }
    if (end > seg.offset + seg.count) {
        return null;
    }
    int begin = iterator.previous();
    if (begin == BreakIterator.DONE ||
        begin >= seg.offset + seg.count) {
        return null;
    }
    seg.modelOffset = seg.modelOffset + begin - seg.offset;
    seg.offset = begin;
    seg.count = end - begin;
    return seg;
}
 

/**
 * Returns the Segment at <code>index</code> representing either
 * the paragraph or sentence as identified by <code>part</code>, or
 * null if a valid paragraph/sentence can't be found. The offset
 * will point to the start of the word/sentence in the array, and
 * the modelOffset will point to the location of the word/sentence
 * in the model.
 */
private IndexedSegment getSegmentAt(int part, int index)
    throws BadLocationException {

    IndexedSegment seg = getParagraphElementText(index);
    if (seg == null) {
        return null;
    }
    BreakIterator iterator;
    switch (part) {
    case AccessibleText.WORD:
        iterator = BreakIterator.getWordInstance(getLocale());
        break;
    case AccessibleText.SENTENCE:
        iterator = BreakIterator.getSentenceInstance(getLocale());
        break;
    default:
        return null;
    }
    seg.first();
    iterator.setText(seg);
    int end = iterator.following(index - seg.modelOffset + seg.offset);
    if (end == BreakIterator.DONE) {
        return null;
    }
    if (end > seg.offset + seg.count) {
        return null;
    }
    int begin = iterator.previous();
    if (begin == BreakIterator.DONE ||
        begin >= seg.offset + seg.count) {
        return null;
    }
    seg.modelOffset = seg.modelOffset + begin - seg.offset;
    seg.offset = begin;
    seg.count = end - begin;
    return seg;
}
 

private void testFollowing(BreakIterator bi, String text, String boundaries, List<Integer> indexes) {
  String message = "Text: " + text;
  for (Integer index : indexes) {
    int got = bi.following(index);
    if (index == boundaries.length()) {
      assertEquals(message, BreakIterator.DONE, got);
      assertEquals(boundaries.lastIndexOf('^'), bi.current());
      continue;
    }
    assertEquals(message + " index:" + index, boundaries.indexOf('^', index + 1), got);
  }
}
 

/**
 * Returns the Segment at <code>index</code> representing either
 * the paragraph or sentence as identified by <code>part</code>, or
 * null if a valid paragraph/sentence can't be found. The offset
 * will point to the start of the word/sentence in the array, and
 * the modelOffset will point to the location of the word/sentence
 * in the model.
 */
private IndexedSegment getSegmentAt(int part, int index)
    throws BadLocationException {

    IndexedSegment seg = getParagraphElementText(index);
    if (seg == null) {
        return null;
    }
    BreakIterator iterator;
    switch (part) {
    case AccessibleText.WORD:
        iterator = BreakIterator.getWordInstance(getLocale());
        break;
    case AccessibleText.SENTENCE:
        iterator = BreakIterator.getSentenceInstance(getLocale());
        break;
    default:
        return null;
    }
    seg.first();
    iterator.setText(seg);
    int end = iterator.following(index - seg.modelOffset + seg.offset);
    if (end == BreakIterator.DONE) {
        return null;
    }
    if (end > seg.offset + seg.count) {
        return null;
    }
    int begin = iterator.previous();
    if (begin == BreakIterator.DONE ||
        begin >= seg.offset + seg.count) {
        return null;
    }
    seg.modelOffset = seg.modelOffset + begin - seg.offset;
    seg.offset = begin;
    seg.count = end - begin;
    return seg;
}
 

/**
 * Returns the Segment at <code>index</code> representing either
 * the paragraph or sentence as identified by <code>part</code>, or
 * null if a valid paragraph/sentence can't be found. The offset
 * will point to the start of the word/sentence in the array, and
 * the modelOffset will point to the location of the word/sentence
 * in the model.
 */
private IndexedSegment getSegmentAt(int part, int index)
    throws BadLocationException {

    IndexedSegment seg = getParagraphElementText(index);
    if (seg == null) {
        return null;
    }
    BreakIterator iterator;
    switch (part) {
    case AccessibleText.WORD:
        iterator = BreakIterator.getWordInstance(getLocale());
        break;
    case AccessibleText.SENTENCE:
        iterator = BreakIterator.getSentenceInstance(getLocale());
        break;
    default:
        return null;
    }
    seg.first();
    iterator.setText(seg);
    int end = iterator.following(index - seg.modelOffset + seg.offset);
    if (end == BreakIterator.DONE) {
        return null;
    }
    if (end > seg.offset + seg.count) {
        return null;
    }
    int begin = iterator.previous();
    if (begin == BreakIterator.DONE ||
        begin >= seg.offset + seg.count) {
        return null;
    }
    seg.modelOffset = seg.modelOffset + begin - seg.offset;
    seg.offset = begin;
    seg.count = end - begin;
    return seg;
}
 

@Override
protected IRegion findWord(IDocument document, int offset) {
	try {
		IRegion line = document.getLineInformationOfOffset(offset);

		if (offset == line.getOffset() + line.getLength())
			return null;

		BreakIterator breakIter = createBreakIterator();
		CharacterIterator characterIterator = new DocumentCharacterIterator(document);
		breakIter.setText(characterIterator);
		int start = breakIter.preceding(offset);
		if (start == BreakIterator.DONE)
			start = line.getOffset();

		int end = breakIter.following(offset);
		if (end == BreakIterator.DONE)
			end = line.getOffset() + line.getLength();

		if (breakIter.isBoundary(offset)) {
			if (end - offset > offset - start)
				start = offset;
			else
				end = offset;
		}

		if (end == start)
			return null;
		return new Region(start, end - start);
	} catch (BadLocationException e) {
		return null;
	}
}
 

/** Assumes that {@code getArea().getLength != 0} is true and {@link BreakIterator#setText(String)} has been called */
private int calculatePositionViaBreakingForwards(int numOfBreaks, BreakIterator breakIterator, int position) {
    breakIterator.following(position);
    for (int i = 1; i < numOfBreaks; i++) {
        breakIterator.next(numOfBreaks);
    }
    return breakIterator.current();
}
 

@Override // to select words containing underscores
public void selectWord()
{
    if ( getLength() == 0 ) return;

    CaretSelectionBind<?,?,?> csb = getCaretSelectionBind();
    int paragraph = csb.getParagraphIndex();
    int position = csb.getColumnPosition(); 
    
    String paragraphText = getText( paragraph );
    BreakIterator breakIterator = BreakIterator.getWordInstance( getLocale() );
    breakIterator.setText( paragraphText );

    breakIterator.preceding( position );
    int start = breakIterator.current();
    
    while ( start > 0 && paragraphText.charAt( start-1 ) == '_' )
    {
        if ( --start > 0 && ! breakIterator.isBoundary( start-1 ) )
        {
            breakIterator.preceding( start );
            start = breakIterator.current();
        }
    }
    
    breakIterator.following( position );
    int end = breakIterator.current();
    int len = paragraphText.length();
    
    while ( end < len && paragraphText.charAt( end ) == '_' )
    {
        if ( ++end < len && ! breakIterator.isBoundary( end+1 ) )
        {
            breakIterator.following( end );
            end = breakIterator.current();
        }
        // For some reason single digits aren't picked up so ....
        else if ( Character.isDigit( paragraphText.charAt( end ) ) )
        {
            end++;
        }
    }
    
    csb.selectRange( paragraph, start, paragraph, end );
}
 

/**
 * Returns the character index of the next line break.  If the next
 * character is wider than <code>width</code> this method will return
 * <code>start</code> - the caller should check for this case.
 *
 * @param text  the text (<code>null</code> not permitted).
 * @param start  the start index.
 * @param width  the target display width.
 * @param iterator  the word break iterator.
 * @param measurer  the text measurer.
 *
 * @return The index of the next line break.
 */
private static int nextLineBreak(String text, int start, float width, 
        BreakIterator iterator, TextMeasurer measurer) {

    // this method is (loosely) based on code in JFreeReport's
    // TextParagraph class
    int current = start;
    int end;
    float x = 0.0f;
    boolean firstWord = true;
    int newline = text.indexOf('\n', start);
    if (newline < 0) {
        newline = Integer.MAX_VALUE;
    }
    while (((end = iterator.following(current)) != BreakIterator.DONE)) {
        x += measurer.getStringWidth(text, current, end);
        if (x > width) {
            if (firstWord) {
                while (measurer.getStringWidth(text, start, end) > width) {
                    end--;
                    if (end <= start) {
                        return end;
                    }
                }
                return end;
            }
            else {
                end = iterator.previous();
                return end;
            }
        }
        else {
            if (end > newline) {
                return newline;
            }
        }
        // we found at least one word that fits ...
        firstWord = false;
        current = end;
    }
    return BreakIterator.DONE;
}
 

private int getNextWordBoundary(CharSequence text, int selectionEnd) {
    BreakIterator boundary = BreakIterator.getWordInstance();
    boundary.setText(text.toString());
    int next = boundary.following(selectionEnd);
    return (next == BreakIterator.DONE) ? selectionEnd : next;
}