/**
* Updates the interal bitset from <code>iterator</code>. This will set <code>validMask</code> to
* true if <code>iterator</code> is non-null.
*/
private void updateMask(AttributedCharacterIterator iterator) {
if (iterator != null) {
validMask = true;
this.iterator = iterator;
// Update the literal mask
if (literalMask == null) {
literalMask = new BitSet();
} else {
for (int counter = literalMask.length() - 1; counter >= 0; counter--) {
literalMask.clear(counter);
}
}
iterator.first();
while (iterator.current() != CharacterIterator.DONE) {
Map attributes = iterator.getAttributes();
boolean set = isLiteral(attributes);
int start = iterator.getIndex();
int end = iterator.getRunLimit();
while (start < end) {
if (set) {
literalMask.set(start);
} else {
literalMask.clear(start);
}
start++;
}
iterator.setIndex(start);
}
}
}
// placeholder function for complex break analysis
static int findComplexBreak(AttributedCharacterIterator aci) {
int cnt = 0;
for (char ch = aci.current(); ch == AttributedCharacterIterator.DONE; ch = aci.next(), cnt++) {
// .. do complex break analysis here Right now we aren't
// do any, we just find the end of the run of
// CHAR_CLASS_SA.
if (getCharCharClass(ch) != CHAR_CLASS_SA) break;
}
return cnt;
}
/** Selects the fields identified by <code>attributes</code>. */
void selectField(Object f, int count) {
AttributedCharacterIterator iterator = getIterator();
if (iterator != null && (f instanceof AttributedCharacterIterator.Attribute)) {
AttributedCharacterIterator.Attribute field = (AttributedCharacterIterator.Attribute) f;
iterator.first();
while (iterator.current() != CharacterIterator.DONE) {
while (iterator.getAttribute(field) == null && iterator.next() != CharacterIterator.DONE) ;
if (iterator.current() != CharacterIterator.DONE) {
int limit = iterator.getRunLimit(field);
if (--count <= 0) {
getFormattedTextField().select(iterator.getIndex(), limit);
break;
}
iterator.setIndex(limit);
iterator.next();
}
}
}
}
/**
* Returns the start of the first run that contains the attribute <code>id</code>. This will
* return <code>-1</code> if the attribute can not be found.
*/
int getAttributeStart(AttributedCharacterIterator.Attribute id) {
if (isValidMask()) {
AttributedCharacterIterator iterator = getIterator();
iterator.first();
while (iterator.current() != CharacterIterator.DONE) {
if (iterator.getAttribute(id) != null) {
return iterator.getIndex();
}
iterator.next();
}
}
return -1;
}
protected String style(final AttributedCharacterIterator aci) {
final StringBuilder builder = new StringBuilder();
Map<AttributedCharacterIterator.Attribute, Object> map = null;
char chr = aci.current();
while (aci.getIndex() < aci.getEndIndex()) {
if (!aci.getAttributes().equals(map)) {
style(aci.getAttributes(), builder);
map = aci.getAttributes();
}
builder.append(StringEscapeUtils.escapeHtml(String.valueOf(chr)));
chr = aci.next();
}
return builder.toString();
}
protected void dumpACIWord(AttributedString as) {
if (as == null) return;
StringBuffer chars = new StringBuffer();
StringBuffer brkStr = new StringBuffer();
AttributedCharacterIterator aci = as.getIterator();
AttributedCharacterIterator.Attribute WORD_LIMIT = TextLineBreaks.WORD_LIMIT;
for (char ch = aci.current(); ch != AttributedCharacterIterator.DONE; ch = aci.next()) {
chars.append(ch).append(' ').append(' ');
int w = ((Integer) aci.getAttribute(WORD_LIMIT)).intValue();
brkStr.append(w).append(' ');
if (w < 10) {
// for small values append another ' '
brkStr.append(' ');
}
}
System.out.println(chars.toString());
System.out.println(brkStr.toString());
}
/**
* Returns the number of occurences of <code>f</code> before the location <code>start</code> in
* the current <code>AttributedCharacterIterator</code>.
*/
private int getFieldTypeCountTo(Object f, int start) {
AttributedCharacterIterator iterator = getIterator();
int count = 0;
if (iterator != null && (f instanceof AttributedCharacterIterator.Attribute)) {
AttributedCharacterIterator.Attribute field = (AttributedCharacterIterator.Attribute) f;
int index = 0;
iterator.first();
while (iterator.getIndex() < start) {
while (iterator.getAttribute(field) == null && iterator.next() != CharacterIterator.DONE) ;
if (iterator.current() != CharacterIterator.DONE) {
iterator.setIndex(iterator.getRunLimit(field));
iterator.next();
count++;
} else {
break;
}
}
}
return count;
}
// handle spaces separately, all others by table
// as - Attributed string to attribute with Word extents.
public static void findLineBrk(AttributedString as) {
AttributedCharacterIterator aci = as.getIterator();
if (aci.getEndIndex() == 0) return;
char ch = aci.current(), prevCh = (char) -1;
byte cls = getCharCharClass(ch);
if (cls == CHAR_CLASS_LF) cls = CHAR_CLASS_BK;
byte curCls = cls;
byte prevCls = cls;
byte prevPrevCls = -1;
int wordCnt = 0;
int wordBegin = aci.getBeginIndex();
// loop over all pairs in the string
int ich = wordBegin + 1;
int lineEnd = aci.getRunLimit(lineBrks);
// handle case where input starts with an LF
if (cls >= CHAR_CLASS_CM) cls = CHAR_CLASS_AL;
for (ch = aci.next();
ch != AttributedCharacterIterator.DONE;
ich++, prevCh = ch, ch = aci.next(), prevPrevCls = prevCls, prevCls = curCls) {
if (ich == lineEnd) {
as.addAttribute(WORD_LIMIT, new Integer(wordCnt++), wordBegin, ich);
wordBegin = ich;
cls = getCharCharClass(ch);
curCls = cls;
prevCls = cls;
if (cls >= CHAR_CLASS_CM) cls = CHAR_CLASS_AL;
lineEnd = aci.getRunLimit(lineBrks);
continue;
}
// handle spaces
curCls = getCharCharClass(ch);
if (curCls == CHAR_CLASS_SP) {
// pbrk[ich-1] = BREAK_ACTION_PROHIBITED;
continue;
}
// handle complex scripts
if (curCls == CHAR_CLASS_SA) {
ich += findComplexBreak(aci);
ch = aci.previous();
if (ch != AttributedCharacterIterator.DONE) prevCls = getCharCharClass(ch);
ch = aci.next();
if (ch != AttributedCharacterIterator.DONE) curCls = cls = getCharCharClass(ch);
continue;
}
// This isn't in the Unicode line breaking alg. but it
// seems needed as otherwise it does produce a break.
if ((ch == CHAR_ZERO_WIDTH_JOINER) || (prevCh == CHAR_ZERO_WIDTH_JOINER))
continue; // Don't allow break around JOINER.
if ((curCls == CHAR_CLASS_BK) || (curCls == CHAR_CLASS_LF)) {
as.addAttribute(WORD_LIMIT, new Integer(wordCnt++), wordBegin, ich);
wordBegin = ich;
cls = CHAR_CLASS_BK;
continue;
}
if (prevCls == CHAR_CLASS_CR) {
as.addAttribute(WORD_LIMIT, new Integer(wordCnt++), wordBegin, ich - 1);
wordBegin = ich - 1;
cls = CHAR_CLASS_BK;
continue;
}
if (curCls == CHAR_CLASS_CR) {
continue;
}
// handle combining marks
if (curCls == CHAR_CLASS_CM) {
if (prevCls == CHAR_CLASS_SP) {
cls = CHAR_CLASS_ID;
if (prevPrevCls != -1) {
if (brkPairs[prevPrevCls][CHAR_CLASS_ID] == BREAK_ACTION_DIRECT) {
as.addAttribute(WORD_LIMIT, new Integer(wordCnt++), wordBegin, ich - 1);
wordBegin = ich - 1;
// pbrk[ich-2] = BREAK_ACTION_DIRECT;
} else {
// pbrk[ich-2] = BREAK_ACTION_PROHIBITED;
}
}
}
// pbrk[ich-1] = BREAK_ACTION_PROHIBITED;
continue;
}
if (cls == CHAR_CLASS_BK) {
cls = curCls;
continue;
}
// lookup pair table information
byte brk = brkPairs[cls][curCls];
if (brk == BREAK_ACTION_DIRECT) {
as.addAttribute(WORD_LIMIT, new Integer(wordCnt++), wordBegin, ich);
wordBegin = ich;
// pbrk[ich-1] = brk;
} else if (brk == BREAK_ACTION_INDIRECT) {
if (prevCls == CHAR_CLASS_SP) {
as.addAttribute(WORD_LIMIT, new Integer(wordCnt++), wordBegin, ich);
wordBegin = ich;
}
// pbrk[ich-1] = ((prevCls == CHAR_CLASS_SP) ?
// BREAK_ACTION_INDIRECT :
// BREAK_ACTION_PROHIBITED);
}
cls = curCls;
}
// always break at the end
as.addAttribute(WORD_LIMIT, new Integer(wordCnt++), wordBegin, ich);
wordBegin = ich;
// pbrk[ich-1] = BREAK_ACTION_DIRECT;
return;
}
