parser.java

This is a resource based on j2me embedded,if you dont understand,you can connection with me .

				return ("CDATA".equals(str) == true)? 'c': '?';
			case 'F':
				return ("FIXED".equals(str) == true)? 'F': '?';
			default:
				break;
			}
			break;

		case 6:	// IDREFS, ENTITY
			switch (mBuff[1]) {
			case 'I':
				return ("IDREFS".equals(str) == true)? 'R': '?';
			case 'E':
				return ("ENTITY".equals(str) == true)? 'n': '?';
			default:
				break;
			}
			break;

		case 7:	// NMTOKEN, IMPLIED, ATTLIST, ELEMENT
			switch (mBuff[1]) {
			case 'I':
				return ("IMPLIED".equals(str) == true)? 'I': '?';
			case 'N':
				return ("NMTOKEN".equals(str) == true)? 't': '?';
			case 'A':
				return ("ATTLIST".equals(str) == true)? 'a': '?';
			case 'E':
				return ("ELEMENT".equals(str) == true)? 'e': '?';
			default:
				break;
			}
			break;

		case 8:	// ENTITIES, NMTOKENS, NOTATION, REQUIRED
			switch (mBuff[2]) {
			case 'N':
				return ("ENTITIES".equals(str) == true)? 'N': '?';
			case 'M':
				return ("NMTOKENS".equals(str) == true)? 'T': '?';
			case 'O':
				return ("NOTATION".equals(str) == true)? 'o': '?';
			case 'E':
				return ("REQUIRED".equals(str) == true)? 'Q': '?';
			default:
				break;
			}
			break;

		default:
			break;
		}
		return '?';
	}

	/**
	 * Reads a single or double quotted string in to the buffer.
	 *
	 * This method resolves entities inside a string unless the parser 
	 * parses DTD.
	 *
	 * @param flag 'c' - CDATA, 'i' - non CDATA, ' ' or '-' - no normalization.
	 * @exception SAXException 
	 * @exception IOException 
	 */
	private void bqstr(char flag)
		throws SAXException, IOException
	{
		Input inp = mInp;	// remember the original input
		mBuffIdx  = -1;
		bappend((char)0);	// default offset to the colon char
		char ch;
		for (short st = 0; st >= 0;) {
			ch = next();
			switch (st) {
			case 0:		// read a single or double quote
				switch (ch) {
				case ' ':
				case '\n':
				case '\r':
				case '\t':
					break;

				case '\'':
					st = 2;	// read a single quoted string
					break;

				case '\"':
					st = 3;	// read a double quoted string
					break;

				default:
					panic(FAULT);
					break;
				}
				break;

			case 2:		// read a single quoted string
			case 3:		// read a double quoted string
				switch (ch) {
				case '\'':
					if ((st == 2) && (mInp == inp))
						st = -1;
					else
						bappend(ch);
					break;

				case '\"':
					if ((st == 3) && (mInp == inp))
						st = -1;
					else
						bappend(ch);
					break;

				case '&':
					ent(' ');
					break;

				case '%':
					pent(flag);
					break;

				case '\r':		// EOL processing [#2.11]
					if (flag != ' ') {
						if (next() != '\n')
							back();
						ch = '\n';
					}
				default:
					//		This implements attribute value normalization as 
					//		described in the XML specification [#3.3.3].
					switch (flag) {
					case 'i':	// non CDATA normalization
						switch (ch) {
						case ' ':
						case '\n':
						case '\t':
							if ((mBuffIdx > 0) && (mBuff[mBuffIdx] != ' '))
								bappend(' ');
							break;

						default:
							bappend(ch);
							break;
						}
						break;

					case 'c':	// CDATA normalization
						switch (ch) {
						case '\n':
						case '\t':
							bappend(' ');
							break;

						default:
							bappend(ch);
							break;
						}
						break;

					default:	// no normalization
						bappend(ch);
						break;
					}
					break;
				}
				break;

			default:
				panic(FAULT);
			}
		}
		//		There is maximum one space at the end of the string in
		//		i-mode (non CDATA normalization) and it has to be removed.
		if ((flag == 'i') && (mBuff[mBuffIdx] == ' '))
			mBuffIdx -= 1;
	}

	/**
	 * Reports characters and empties the parser's buffer.
	 */
	private void bflash()
		throws SAXException
	{
		if (mBuffIdx >= 0) {
			//		Textual data has been read
			mHand.characters(mBuff, 0, (mBuffIdx + 1));
			mBuffIdx = -1;
		}
	}

	/**
	 * Appends a character to parser's buffer.
	 *
	 * @param ch The character to append to the buffer.
	 */
	private void bappend(char ch)
	{
		try {
			mBuffIdx++;
			mBuff[mBuffIdx] = ch;
		} catch (Exception exp) {
			//		Double the buffer size
			char buff[] = new char[mBuff.length << 1];
			System.arraycopy(mBuff, 0, buff, 0, mBuff.length);
			mBuff			= buff;
			mBuff[mBuffIdx]	= ch;
		}
	}

	/**
	 * Recognizes the built-in entities <i>lt</i>, <i>gt</i>, <i>amp</i>, 
	 * <i>apos</i>, <i>quot</i>. 
	 * The initial state is 0x100. Any state belowe 0x100 is a built-in 
	 * entity replacement character. 
	 *
	 * @param ch the next character of an entity name.
	 */
	private void eappend(char ch)
	{
		switch (mESt) {
		case 0x100:	// "l" or "g" or "a" or "q"
			switch (ch) {
			case 'l':	mESt	= 0x101; break;
			case 'g':	mESt	= 0x102; break;
			case 'a':	mESt	= 0x103; break;
			case 'q':	mESt	= 0x107; break;
			default:	mESt	= 0x200; break;
			}
			break;
		case 0x101:	// "lt"
			mESt = (ch == 't')? '<': (char)0x200;
			break;
		case 0x102:	// "gt"
			mESt = (ch == 't')? '>': (char)0x200;
			break;
		case 0x103:	// "am" or "ap"
			switch (ch) {
			case 'm':	mESt	= 0x104; break;
			case 'p':	mESt	= 0x105; break;
			default:	mESt	= 0x200; break;
			}
			break;
		case 0x104:	// "amp"
			mESt = (ch == 'p')? '&': (char)0x200;
			break;
		case 0x105:	// "apo"
			mESt = (ch == 'o')? (char)0x106: (char)0x200;
			break;
		case 0x106:	// "apos"
			mESt = (ch == 's')? '\'': (char)0x200;
			break;
		case 0x107:	// "qu"
			mESt = (ch == 'u')? (char)0x108: (char)0x200;
			break;
		case 0x108:	// "quo"
			mESt = (ch == 'o')? (char)0x109: (char)0x200;
			break;
		case 0x109:	// "quot"
			mESt = (ch == 't')? '\"': (char)0x200;
			break;
		case '<':	// "lt"
		case '>':	// "gt"
		case '&':	// "amp"
		case '\'':	// "apos"
		case '\"':	// "quot"
			mESt	= 0x200;
		default:
			break;
		}
	}

	/**
	 * Sets up a new input source on the top of the input stack.
	 * Noet, the first byte returned by the entity's byte stream has to be the 
	 * first byte in the entity. However, the parser does not expect the byte 
	 * order mask in both cases when encoding is provided by the input source.
	 *
	 * @param is A new input source to set up.
	 * @exception IOException If any IO errors occur.
	 * @exception SAXException If the input source cannot be read.
	 */
	private void setinp(InputSource is)
		throws SAXException, IOException
	{
		Reader reader = null;
		mChIdx   = 0;
		mChLen   = 0;
		mChars   = mInp.chars;
		mInp.src = null;
		if (is.getCharacterStream() != null) {
			//		Ignore encoding in the xml text decl. 
			reader = is.getCharacterStream();
			xml(reader);
		} else if (is.getByteStream() != null) {
			if (is.getEncoding() != null) {
				//		Ignore encoding in the xml text decl.
				String encoding = is.getEncoding().toUpperCase();
				if (encoding.equals("UTF-16"))
					reader = bom(is.getByteStream(), 'U');	// UTF-16 [#4.3.3]
				else
					reader = enc(encoding, is.getByteStream());
				xml(reader);
			} else {
				//		Get encoding from BOM or the xml text decl.
				reader = bom(is.getByteStream(), ' ');
				if (reader == null) {
					//		Encoding is defined by the xml text decl.
					reader = enc("UTF-8", is.getByteStream());
					reader = enc(xml(reader), is.getByteStream());
				} else {
					//		Encoding is defined by the BOM.
					xml(reader);
				}
			}
		} else {
			//		There is no support for public/system identifiers.
			panic(FAULT);
		}
		mInp.src   = reader;
		mInp.pubid = is.getPublicId();
		mInp.sysid = is.getSystemId();
	}

	/**
	 * Determines the entity encoding.
	 *
	 * This method gets encoding from Byte Order Mask [#4.3.3] if any. 
	 * Note, the first byte returned by the entity's byte stream has 
	 * to be the first byte in the entity. Also, there is no support 
	 * for UCS-4.
	 *
	 * @param is A byte stream of the entity.
	 * @param hint An encoding hint, character U means UTF-16.
	 * @return a reader constructed from the BOM or UTF-8 by default.
	 * @exception SAXException 
	 * @exception IOException 
	 */
	private Reader bom(InputStream is, char hint)
		throws SAXException, IOException
	{
		int val = is.read();
		switch (val) {
		case 0xef:		// UTF-8
			if (hint == 'U')	// must be UTF-16
				panic(FAULT);
			if (is.read() != 0xbb) 
				panic(FAULT);
			if (is.read() != 0xbf) 
				panic(FAULT);
			return new ReaderUTF8(is);

		case 0xfe:		// UTF-16, big-endian
			if (is.read() != 0xff) 
				panic(FAULT);
			return new ReaderUTF16(is, 'b');

		case 0xff:		// UTF-16, little-endian
			if (is.read() != 0xfe) 
				panic(FAULT);
			return new ReaderUTF16(is, 'l');

		case -1:
			mChars[mChIdx++] = EOS;
			return new ReaderUTF8(is);

		default:
			if (hint == 'U')	// must be UTF-16
				panic(FAULT);
			//		Read the rest of UTF-8 character
			switch (val & 0xf0) {
			case 0xc0:
			case 0xd0:
				mChars[mChIdx++] = (char)(((val & 0x1f) << 6) | (is.read() & 0x3f));
				break;

			case 0xe0:
				mChars[mChIdx++] = (char)(((val & 0x0f) << 12) | 
					((is.read() & 0x3f) << 6) | (is.read() & 0x3f));
				break;

			case 0xf0:	// UCS-4 character
				throw new UnsupportedEncodingException();

			default:
				mChars[mChIdx++] = (char)val;
				break;
			}
			return null;
		}
	}

	/**
	 * Parses the xml text declaration.
	 *
	 * This method gets encoding from the xml text declaration [#4.3.1] if any. 
	 * The method assumes the buffer (mChars) is big enough to accomodate whole 
	 * xml text declaration.
	 *
	 * @param reader is entity reader.
	 * @return The xml text declaration encoding or default UTF-8 encoding.
	 * @exception SAXException 
	 * @exception IOException 
	 */
	private String xml(Reader reader)
		throws SAXException, IOException
	{
		String	str	= null;
		String	enc	= "UTF-8";
		char	ch;
		int		val;
		short	st;
		//		Read the xml text declaration into the buffer
		if (mChIdx != 0) {
			//		The bom method have read ONE char into the buffer. 
			st = (short)((mChars[0] == '<')? 1: -1);
		} else {
			st = 0;
		}
		while (st >= 0) {
			ch = ((val = reader.read()) >= 0)? (char)val: EOS;
			mChars[mChIdx++] = ch;
			switch (st) {
			case 0:		// read '<' of xml declaration
				switch (ch) {
				case '<':
					st = 1;
					break;

				case 0xfeff:	// the byte order mask
					ch = ((val = reader.read()) >= 0)? (char)val: EOS;
					mChars[mChIdx - 1] = ch;
					st = (short)((ch == '<')? 1: -1);
					break;

				default:
					st = -1;
					break;

				}
				break;

			case 1:		// read '?' of xml declaration
				st = (short)((ch == '?')? 2: -1);
				break;

			case 2:		// read 'x' of xml declaration
				st = (short)(((ch == 'x') || (ch == 'X'))? 3: -1);
				break;

			case 3:		// read 'm' of xml declaration
				st = (short)(((ch == 'm') || (ch == 'M'))? 4: -1);
				break;

			case 4:		// read 'l' of xml declaration
				st = (short)(((ch == 'l') || (ch == 'L'))? 5: -1);
				break;

			case 5:		// read white space after 'xml'
				switch (ch) {
				case ' ':
				case '\t':
				case '\r':
				case '\n':
					st = 6;
					break;

				default:
					st = -1;
					break;
				}
				break;

			case 6:		// read content of xml declaration
				if (ch == '?')
					st = 7;
				break;

			case 7:		// read '>' after '?' of xml declaration
				st = (short)((ch != '>')? 6: -2);
				break;

			default:
				panic(FAULT);
				break;
			}
		}
		mChLen = mChIdx;
		mChIdx = 0;
		//		If there is no xml text declaration, the encoding is default.
		if (st == -1) {
			return enc;
		}
		mChIdx = 5;		// the first white space after "<?xml"
		//		Parse the xml text declaration
		for (st = 0; st >= 0;) {
			ch = next();
			switch (st) {
			case 0:		// skip spaces after the xml declaration name
				if (chtyp(ch) != ' ') {
					back();
					st = 1;
				}
				break;

			case 1:		// read xml declaration version
			case 2:		// read xml declaration encoding or standalone
			case 3:		// read xml declaration standalone
				switch (chtyp(ch)) {
				case 'a':
				case 'A':
				case '_':
					back();
					str	= name(false).toLowerCase();
					if ("version".equals(str) == true) {
						if (st != 1)
							panic(FAULT);
						if ("1.0".equals(eqstr('=')) != true)
							panic(FAULT);
						st = 2;
					} else if ("encoding".equals(str) == true) {
						if (st != 2)
							panic(FAULT);
						enc = eqstr('=').toUpperCase();
						st  = 3;
					} else if ("standalone".equals(str) == true) {
						if (st == 1)
							panic(FAULT);
						str = eqstr('=');
						// BUG: the value has to be analized by the parser [#5.1]
						st  = 4;
					} else {
						panic(FAULT);
					}
					break;

				case ' ':
					break;

				case '?':
					if (st == 1)
						panic(FAULT);
					back();
					st = 4;
					break;

				default:
					panic(FAULT);
				}
				break;

			case 4:		// end of xml declaration
				switch (chtyp(ch)) {
				case '?':
					if (next() != '>')
						panic(FAULT);
					if (mSt == 0)		// the begining of the document
						mSt	= 1;		// misc before DTD
					st = -1;
					break;

				case ' ':
					break;

				default:
					panic(FAULT);
				}
				break;

			default:
				panic(FAULT);
			}
		}
		return enc;
	}

	/**
	 * Sets up the document reader.
	 *
	 * @param name an encoding name.
	 * @param is the document byte input stream.
	 * @return a reader constructed from encoding name and input stream.
	 * @exception UnsupportedEncodingException 
	 */
	private Reader enc(String name, InputStream is)
		throws java.io.UnsupportedEncodingException
	{
		//		DO NOT CLOSE curren