/**
* Reads in translated plurals forms that are separated by NULL.
*/
private ArrayList <StringValue> readPluralForms(int length)
{
ArrayList <StringValue> list = new ArrayList <StringValue>();
StringValue sb = new UnicodeBuilderValue();
for (; length > 0; length--)
{
int ch = _in.readChar();
if (ch > 0)
{
sb.append((char)ch);
}
else if (ch == 0)
{
list.add(sb);
sb = new UnicodeBuilderValue();
}
else
{
break;
}
}
list.add(sb);
return(list);
}
public StringValue decodeUnicode(StringValue str)
{
UnicodeBuilderValue sb = new UnicodeBuilderValue();
decodeUnicode(str, sb);
return(sb);
}
public CharSequence decode(Env env, StringValue str)
{
if (str.isUnicode())
{
return(str);
}
UnicodeBuilderValue sb = new UnicodeBuilderValue();
decodeUnicode(str, sb);
return(sb);
}
public override void decodeUnicode(StringValue str, UnicodeBuilderValue sb)
{
ByteBuffer in = ByteBuffer.wrap(str.toBytes());
TempCharBuffer tempBuf = TempCharBuffer.allocate();
try {
CharBuffer @out = CharBuffer.wrap(tempBuf.getBuffer());
while (in.hasRemaining())
{
CoderResult coder = _decoder.decode(in, @out, false);
if (!fill(sb, in, @out, coder))
{
return;
}
@out.clear();
}
CoderResult coder = _decoder.decode(in, @out, true);
if (!fill(sb, in, @out, coder))
{
return;
}
@out.clear();
coder = _decoder.flush(out);
fill(sb, in, @out, coder);
return;
}
finally {
TempCharBuffer.free(tempBuf);
}
}
public abstract void decodeUnicode(StringValue str, UnicodeBuilderValue sb);
static UnicodeValue fromUtf8(StringValue source)
{
UnicodeValue target = new UnicodeBuilderValue();
int len = source.length();
for (int i = 0; i < len; i++)
{
char ch = source[i];
if (ch < 0x80)
{
target.append(ch);
}
else if ((ch & 0xe0) == 0xc0)
{
if (len <= i + 1)
{
log.fine(L.l("Regexp: bad UTF-8 sequence, saw EOF"));
return(null);
}
char ch2 = source[++i];
target.append((char)(((ch & 0x1f) << 6)
+ (ch2 & 0x3f)));
}
else if ((ch & 0xf0) == 0xe0)
{
if (len <= i + 2)
{
log.fine(L.l("Regexp: bad UTF-8 sequence, saw EOF"));
return(null);
}
char ch2 = source[++i];
char ch3 = source[++i];
target.append((char)(((ch & 0x0f) << 12)
+ ((ch2 & 0x3f) << 6)
+ (ch3 & 0x3f)));
}
else
{
if (i + 3 >= len)
{
log.fine(L.l("Regexp: bad UTF-8 sequence, saw EOF"));
return(null);
}
char ch2 = source[++i];
char ch3 = source[++i];
char ch4 = source[++i];
int codePoint = ((ch & 0x07) << 18)
+ ((ch2 & 0x3F) << 12)
+ ((ch3 & 0x3F) << 6)
+ (ch4 & 0x3F);
int high = ((codePoint - 0x10000) >> 10) + 0xD800;
int low = (codePoint & 0x3FF) + 0xDC00;
target.append((char)high);
target.append((char)low);
}
}
return(target);
}
protected bool fill(UnicodeBuilderValue sb, ByteBuffer in,
public override void decodeUnicode(StringValue str, UnicodeBuilderValue sb)
{
int len = str.length();
for (int i = 0; i < len; i++)
{
int ch = str[i];
if (ch <= 0x7F)
{
sb.append((char)ch);
}
else if (0xC2 <= ch && ch <= 0xDF)
{
int ch2;
if (i + 1 < len &&
0x80 <= (ch2 = str[i + 1]) && ch2 <= 0xBF)
{
i++;
int code = ((ch - 0xC0) << 6) + (ch2 - 0x80);
sb.append((char)code);
}
else if (_isIgnoreErrors)
{
}
else if (_replacement != null)
{
sb.append(_replacement);
}
else if (_isAllowMalformedOut)
{
sb.append((char)ch);
}
else
{
return;
}
}
else if (0xE0 <= ch && ch <= 0xEF)
{
int ch2;
int ch3;
if (i + 2 < len &&
0x80 <= (ch2 = str[i + 1]) && ch2 <= 0xBF &&
0x80 <= (ch3 = str[i + 2]) && ch3 <= 0xBF)
{
i += 2;
int code = ((ch - 0xE0) << 12)
+ ((ch2 - 0x80) << 6)
+ (ch3 - 0x80);
if (0xD800 <= code && code <= 0xDBFF)
{
code &= 0xFFFFF;
int high = 0xD800 + (code >> 10);
int low = 0xDC00 + (code & 0x3FF);
sb.append((char)high);
sb.append((char)low);
}
else
{
sb.append((char)code);
}
}
else if (_isIgnoreErrors)
{
}
else if (_replacement != null)
{
sb.append(_replacement);
}
else if (_isAllowMalformedOut)
{
sb.append((char)ch);
}
else
{
return;
}
}
else if (0xF0 <= ch && ch <= 0xF4)
{
int ch2;
int ch3;
int ch4;
if (i + 3 < len &&
0x80 <= (ch2 = str[i + 1]) && ch2 <= 0xBF &&
0x80 <= (ch3 = str[i + 2]) && ch3 <= 0xBF &&
0x80 <= (ch4 = str[i + 3]) && ch4 <= 0xBF)
{
i += 3;
int code = ((ch - 0xF0) << 18)
+ ((ch2 - 0x80) << 12)
+ ((ch3 - 0x80) << 6)
+ (ch4 - 0x80);
if (code > 0xFFFF || 0xD800 <= code && code <= 0xDBFF)
{
code &= 0xFFFFF;
int high = 0xD800 + code >> 10;
int low = 0xDC00 + code & 0x3FF;
sb.append((char)high);
sb.append((char)low);
}
else
{
sb.append((char)code);
}
}
else if (_isIgnoreErrors)
{
}
else if (_replacement != null)
{
sb.append(_replacement);
}
else if (_isAllowMalformedOut)
{
sb.append((char)ch);
}
else
{
return;
}
}
else if (_isIgnoreErrors)
{
}
else if (_replacement != null)
{
sb.append(_replacement);
}
else if (_isAllowMalformedOut)
{
sb.append((char)ch);
}
else
{
return;
}
}
/*
* Utf8Reader reader = new Utf8Reader(str);
*
* int ch;
*
* while ((ch = reader.read()) >= 0) {
* if (ch == ERROR_CHARACTER) {
* _hasError = true;
*
* if (_isIgnoreErrors) {
* }
* else if (_replacement != null)
* sb.append(_replacement);
* else
* return sb;
* }
* else
* sb.append((char) ch);
* }
*/
return;
}