simplexmlparser.java

处理PDF

                    previousCharacter = character;
                    text.append('&').append(entity.toString());
                    entity.setLength(0);
                }
                else {
                    entity.append((char)character);
                }
                break;
            // We are processing the quoted right-hand side of an element's attribute.
			case QUOTE:
                if (html && quoteCharacter == ' ' && character == '>') {
                    flush();
                    processTag(true);
                    initTag();
                    state = restoreState();
                }
                else if (html && quoteCharacter == ' ' && Character.isWhitespace((char)character)) {
                	flush();
                    state = TAG_EXAMINED;
                }
                else if (html && quoteCharacter == ' ') {
                    text.append((char)character);
                }
                else if(character == quoteCharacter) {
                	flush();
                    state = TAG_EXAMINED;
                } else if(" \r\n\u0009".indexOf(character)>=0) {
                    text.append(' ');
                } else if(character == '&') {
                    saveState(state);
                    state = ENTITY;
                    entity.setLength(0);
                } else {
                    text.append((char)character);
                }
                break;
                
			case ATTRIBUTE_KEY:
                if(Character.isWhitespace((char)character)) {
                    flush();
                    state = ATTRIBUTE_EQUAL;
                } else if(character == '=') {
                	flush();
                    state = ATTRIBUTE_VALUE;
                } else if (html && character == '>') {
                    text.setLength(0);
                    processTag(true);
                    initTag();
                    state = restoreState();
                } else {
                    text.append((char)character);
                }
                break;
                
			case ATTRIBUTE_EQUAL:
                if(character == '=') {
                    state = ATTRIBUTE_VALUE;
                } else if(Character.isWhitespace((char)character)) {
                    // empty
                } else if (html && character == '>') {
                    text.setLength(0);
                    processTag(true);
                    initTag();
                    state = restoreState();
                } else if (html && character == '/') {
                    flush();
                    state = SINGLE_TAG;
                } else if (html) {
                    flush();
                    text.append((char)character);
                    state = ATTRIBUTE_KEY;
                } else {
                    throwException("Error in attribute processing.");
                }
                break;
                
			case ATTRIBUTE_VALUE:
                if(character == '"' || character == '\'') {
                    quoteCharacter = character;
                    state = QUOTE;
                } else if(Character.isWhitespace((char)character)) {
                    // empty
                } else if (html && character == '>') {
                    flush();
                    processTag(true);
                    initTag();
                    state = restoreState();
                } else if (html) {
                    text.append((char)character);
                    quoteCharacter = ' ';
                    state = QUOTE;
                } else {
                    throwException("Error in attribute processing");
                }
                break;
            }
        }
    }

    /**
     * Gets a state from the stack
     * @return the previous state
     */
    private int restoreState() {
        if(!stack.empty())
            return ((Integer)stack.pop()).intValue();
        else
            return UNKNOWN;
    }
    /**
     * Adds a state to the stack.
     * @param	s	a state to add to the stack
     */
    private void saveState(int s) {
    	stack.push(new Integer(s));
    }
    /**
     * Flushes the text that is currently in the buffer.
     * The text can be ignored, added to the document
     * as content or as comment,... depending on the current state.
     */
    private void flush() {
    	switch(state){
    	case TEXT:
    	case CDATA:
            if(text.length() > 0) {
                doc.text(text.toString());
            }
            break;
    	case COMMENT:
        	if (comment != null) {
                comment.comment(text.toString());
            }
        	break;
    	case ATTRIBUTE_KEY:
            attributekey = text.toString();
            if (html)
                attributekey = attributekey.toLowerCase();
    		break;
    	case QUOTE:
    	case ATTRIBUTE_VALUE:
        	attributevalue = text.toString();
            attributes.put(attributekey,attributevalue);
            break;
    	default:
    		// do nothing
    	}
        text.setLength(0);
    }
    /**
     * Initialized the tag name and attributes.
     */
    private void initTag() {
        tag = null;
        attributes = new HashMap();
    }
    /** Sets the name of the tag. */
    private void doTag() {
    	if(tag == null)
    		tag = text.toString();
    	if (html)
    		tag = tag.toLowerCase();
    	text.setLength(0);
    }
    /**
     * processes the tag.
     * @param start	if true we are dealing with a tag that has just been opened; if false we are closing a tag.
     */
    private void processTag(boolean start) {
    	if (start) {
    		nested++;
    		doc.startElement(tag,attributes);
    	}
    	else {
            nested--;
            doc.endElement(tag);
    	}
    }
    /** Throws an exception */
    private void throwException(String s) throws IOException {
        throw new IOException(s+" near line " + lines + ", column " + columns);
    }
    
    /**
     * Parses the XML document firing the events to the handler.
     * @param doc the document handler
     * @param r the document. The encoding is already resolved. The reader is not closed
     * @throws IOException on error
     */
    public static void parse(SimpleXMLDocHandler doc, SimpleXMLDocHandlerComment comment, Reader r, boolean html) throws IOException {
    	SimpleXMLParser parser = new SimpleXMLParser(doc, comment, html);
    	parser.go(r);
    }
    
    /**
     * Parses the XML document firing the events to the handler.
     * @param doc the document handler
     * @param in the document. The encoding is deduced from the stream. The stream is not closed
     * @throws IOException on error
     */    
    public static void parse(SimpleXMLDocHandler doc, InputStream in) throws IOException {
        byte b4[] = new byte[4];
        int count = in.read(b4);
        if (count != 4)
            throw new IOException("Insufficient length.");
        String encoding = getEncodingName(b4);
        String decl = null;
        if (encoding.equals("UTF-8")) {
            StringBuffer sb = new StringBuffer();
            int c;
            while ((c = in.read()) != -1) {
                if (c == '>')
                    break;
                sb.append((char)c);
            }
            decl = sb.toString();
        }
        else if (encoding.equals("CP037")) {
            ByteArrayOutputStream bi = new ByteArrayOutputStream();
            int c;
            while ((c = in.read()) != -1) {
                if (c == 0x6e) // that's '>' in ebcdic
                    break;
                bi.write(c);
            }
            decl = new String(bi.toByteArray(), "CP037");
        }
        if (decl != null) {
            decl = getDeclaredEncoding(decl);
            if (decl != null)
                encoding = decl;
        }
        parse(doc, new InputStreamReader(in, IanaEncodings.getJavaEncoding(encoding)));
    }
    
    private static String getDeclaredEncoding(String decl) {
        if (decl == null)
            return null;
        int idx = decl.indexOf("encoding");
        if (idx < 0)
            return null;
        int idx1 = decl.indexOf('"', idx);
        int idx2 = decl.indexOf('\'', idx);
        if (idx1 == idx2)
            return null;
        if ((idx1 < 0 && idx2 > 0) || (idx2 > 0 && idx2 < idx1)) {
            int idx3 = decl.indexOf('\'', idx2 + 1);
            if (idx3 < 0)
                return null;
            return decl.substring(idx2 + 1, idx3);
        }
        if ((idx2 < 0 && idx1 > 0) || (idx1 > 0 && idx1 < idx2)) {
            int idx3 = decl.indexOf('"', idx1 + 1);
            if (idx3 < 0)
                return null;
            return decl.substring(idx1 + 1, idx3);
        }
        return null;
    }
    
    public static void parse(SimpleXMLDocHandler doc,Reader r) throws IOException {
        parse(doc, null, r, false);
    }
    
    /**
     * Escapes a string with the appropriated XML codes.
     * @param s the string to be escaped
     * @param onlyASCII codes above 127 will always be escaped with &amp;#nn; if <CODE>true</CODE>
     * @return the escaped string
     */    
    public static String escapeXML(String s, boolean onlyASCII) {
        char cc[] = s.toCharArray();
        int len = cc.length;
        StringBuffer sb = new StringBuffer();
        for (int k = 0; k < len; ++k) {
            int c = cc[k];
            switch (c) {
                case '<':
                    sb.append("&lt;");
                    break;
                case '>':
                    sb.append("&gt;");
                    break;
                case '&':
                    sb.append("&amp;");
                    break;
                case '"':
                    sb.append("&quot;");
                    break;
                case '\'':
                    sb.append("&apos;");
                    break;
                default:
                    if (onlyASCII && c > 127)
                        sb.append("&#").append(c).append(';');
                    else
                        sb.append((char)c);
            }
        }
        return sb.toString();
    }
    /**
     * Returns the IANA encoding name that is auto-detected from
     * the bytes specified, with the endian-ness of that encoding where appropriate.
     * (method found in org.apache.xerces.impl.XMLEntityManager, originally published
     * by the Apache Software Foundation under the Apache Software License; now being
     * used in iText under the MPL)
     * @param b4    The first four bytes of the input.
     * @return an IANA-encoding string
     */
    private static String getEncodingName(byte[] b4) {
        
        // UTF-16, with BOM
        int b0 = b4[0] & 0xFF;
        int b1 = b4[1] & 0xFF;
        if (b0 == 0xFE && b1 == 0xFF) {
            // UTF-16, big-endian
            return "UTF-16BE";
        }
        if (b0 == 0xFF && b1 == 0xFE) {
            // UTF-16, little-endian
            return "UTF-16LE";
        }
        
        // UTF-8 with a BOM
        int b2 = b4[2] & 0xFF;
        if (b0 == 0xEF && b1 == 0xBB && b2 == 0xBF) {
            return "UTF-8";
        }
        
        // other encodings
        int b3 = b4[3] & 0xFF;
        if (b0 == 0x00 && b1 == 0x00 && b2 == 0x00 && b3 == 0x3C) {
            // UCS-4, big endian (1234)
            return "ISO-10646-UCS-4";
        }
        if (b0 == 0x3C && b1 == 0x00 && b2 == 0x00 && b3 == 0x00) {
            // UCS-4, little endian (4321)
            return "ISO-10646-UCS-4";
        }
        if (b0 == 0x00 && b1 == 0x00 && b2 == 0x3C && b3 == 0x00) {
            // UCS-4, unusual octet order (2143)
            // REVISIT: What should this be?
            return "ISO-10646-UCS-4";
        }
        if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x00) {
            // UCS-4, unusual octet order (3412)
            // REVISIT: What should this be?
            return "ISO-10646-UCS-4";
        }
        if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x3F) {
            // UTF-16, big-endian, no BOM
            // (or could turn out to be UCS-2...
            // REVISIT: What should this be?
            return "UTF-16BE";
        }
        if (b0 == 0x3C && b1 == 0x00 && b2 == 0x3F && b3 == 0x00) {
            // UTF-16, little-endian, no BOM
            // (or could turn out to be UCS-2...
            return "UTF-16LE";
        }
        if (b0 == 0x4C && b1 == 0x6F && b2 == 0xA7 && b3 == 0x94) {
            // EBCDIC
            // a la xerces1, return CP037 instead of EBCDIC here
            return "CP037";
        }
        
        // default encoding
        return "UTF-8";
    }
}